Refine
Year of publication
Document Type
- Doctoral Thesis (939) (remove)
Language
- English (939) (remove)
Has Fulltext
- yes (939)
Keywords
- Visualisierung (16)
- Visualization (9)
- finite element method (9)
- Infrarotspektroskopie (8)
- Deep Learning (7)
- Finite-Elemente-Methode (7)
- Optimization (7)
- Algebraische Geometrie (6)
- Numerische Strömungssimulation (6)
- Simulation (6)
Faculty / Organisational entity
- Kaiserslautern - Fachbereich Mathematik (278)
- Kaiserslautern - Fachbereich Informatik (218)
- Kaiserslautern - Fachbereich Maschinenbau und Verfahrenstechnik (143)
- Kaiserslautern - Fachbereich Chemie (79)
- Kaiserslautern - Fachbereich Elektrotechnik und Informationstechnik (64)
- Kaiserslautern - Fachbereich Biologie (53)
- Kaiserslautern - Fachbereich Sozialwissenschaften (26)
- Landau - Fachbereich Natur- und Umweltwissenschaften (22)
- Kaiserslautern - Fachbereich Wirtschaftswissenschaften (19)
- Kaiserslautern - Fachbereich Physik (9)
Destructive diseases of the lung like lung cancer or fibrosis are still often lethal. Also in case of fibrosis in the liver, the only possible cure is transplantation.
In this thesis, we investigate 3D micro computed synchrotron radiation (SR\( \mu \)CT) images of capillary blood vessels in mouse lungs and livers. The specimen show so-called compensatory lung growth as well as different states of pulmonary and hepatic fibrosis.
During compensatory lung growth, after resecting part of the lung, the remaining part compensates for this loss by extending into the empty space. This process is accompanied by an active vessel growing.
In general, the human lung can not compensate for such a loss. Thus, understanding this process in mice is important to improve treatment options in case of diseases like lung cancer.
In case of fibrosis, the formation of scars within the organ's tissue forces the capillary vessels to grow to ensure blood supply.
Thus, the process of fibrosis as well as compensatory lung growth can be accessed by considering the capillary architecture.
As preparation of 2D microscopic images is faster, easier, and cheaper compared to SR\( \mu \)CT images, they currently form the basis of medical investigation. Yet, characteristics like direction and shape of objects can only properly be analyzed using 3D imaging techniques. Hence, analyzing SR\( \mu \)CT data provides valuable additional information.
For the fibrotic specimen, we apply image analysis methods well-known from material science. We measure the vessel diameter using the granulometry distribution function and describe the inter-vessel distance by the spherical contact distribution. Moreover, we estimate the directional distribution of the capillary structure. All features turn out to be useful to characterize fibrosis based on the deformation of capillary vessels.
It is already known that the most efficient mechanism of vessel growing forms small torus-shaped holes within the capillary structure, so-called intussusceptive pillars. Analyzing their location and number strongly contributes to the characterization of vessel growing. Hence, for all three applications, this is of great interest. This thesis provides the first algorithm to detect intussusceptive pillars in SR\( \mu \)CT images. After segmentation of raw image data, our algorithm works automatically and allows for a quantitative evaluation of a large amount of data.
The analysis of SR\( \mu \)CT data using our pillar algorithm as well as the granulometry, spherical contact distribution, and directional analysis extends the current state-of-the-art in medical studies. Although it is not possible to replace certain 3D features by 2D features without losing information, our results could be used to examine 2D features approximating the 3D findings reasonably well.
The various uses of fiber-reinforced composites, for example in the enclosures of planes, boats and cars, generates the demand for a detailed analysis of these materials. The final goal is to optimize fibrous materials by the means of “virtual material design”. New fibrous materials are virtually created as realizations of a stochastic model and evaluated with physical simulations. In that way, materials can be optimized for specific use cases, without constructing expensive prototypes or performing mechanical experiments. In order to design a practically fabricable material, the stochastic model is first adapted to an existing material and then slightly modified. The virtual reconstruction of the existing material requires a precise knowledge of the geometry of its microstructure. The first part of this thesis describes a fiber quantification method by the means of local measurements of the fiber radius and orientation. The combination of a sparse chord length transform and inertia moments leads to an efficient and precise new algorithm. It outperforms existing approaches with the possibility to treat different fiber radii within one sample, with high precision in continuous space and comparably fast computing time. This local quantification method can be directly applied on gray value images by adapting the directional distance transforms on gray values. In this work, several approaches of this kind are developed and evaluated. Further characterization of the fiber system requires a segmentation of each single fiber. Using basic morphological operators with specific structuring elements, it is possible to derive a probability for each pixel describing if the pixel belongs to a fiber core in a region without overlapping fibers. Tracking high probabilities leads to a partly reconstruction of the fiber cores in non crossing regions. These core parts are then reconnected over critical regions, if they fulfill certain conditions ensuring the affiliation to the same fiber. In the second part of this work, we develop a new stochastic model for dense systems of non overlapping fibers with a controllable level of bending. Existing approaches in the literature have at least one weakness in either achieving high volume fractions, producing non overlapping fibers, or controlling the bending or the orientation distribution. This gap can be bridged by our stochastic model, which operates in two steps. Firstly, a random walk with the multivariate von Mises-Fisher orientation distribution defines bent fibers. Secondly, a force-biased packing approach arranges them in a non overlapping configuration. Furthermore, we provide the estimation of all parameters needed for the fitting of this model to a real microstructure. Finally, we simulate the macroscopic behavior of different microstructures to derive their mechanical and thermal properties. This part is mostly supported by existing software and serves as a summary of physical simulation applied to random fiber systems. The application on a glass fiber reinforced polymer proves the quality of the reconstruction by our stochastic model, as the effective properties match for both the real microstructure and the realizations of the fitted model. This thesis includes all steps to successfully perform virtual material design on various data sets. With novel and efficient algorithms it contributes to the science of analysis and modeling of fiber reinforced materials.
The fifth-generation (5G) of wireless networks promises to bring new advances, such as a huge increase in mobile data rates, a plunge in communications latency, and an increase in the quality of experience perceived by users that can cope with the ever-increasing demand in Internet traffic. However, the high cost of capital and operational expenditure (CAPEX/OPEX) of the new 5G network and the lack of a killer application hinder its rapid adoption. In this context, Mobile Network Operators (MNOs) have turned their attention to the following idea: opening up their infrastructure so that vertical businesses can leverage the new 5G network to improve their primary businesses and develop new ones. However, deploying multiple isolated vertical applications on top of the same infrastructure poses unique challenges that must be addressed. In this thesis, we provide critical contributions to developing 5G networks to accommodate different vertical applications in an isolated, flexible, and automated manner. This thesis contributions spawn on three main areas: (i) the development of an integrated fronthaul and backhaul network, (ii) the development of a network slicing overbooking algorithm, and (iii) the development of a method to mitigate the noisy neighbors' problem in a vRAN deployment.
A building-block model reveals new insights into the biogenesis of yeast mitochondrial ribosomes
(2020)
Most of the mitochondrial proteins in yeast are encoded in the nuclear genome, get synthesized by cytosolic ribosomes and are imported via TOM and TIM23 into the matrix or other subcompartments of mitochondria. The mitochondrial DNA in yeast however also encodes a small set of 8 proteins from which most are hydrophobic membrane proteins and build core components of the OXPHOS complexes. They get synthesized by mitochondrial ribosomes which are descendants of bacterial ribosomes and still have some similarities to them. On the other hand, mitochondrial ribosomes experienced various structural and functional changes during evolution that specialized them for the synthesis of the mitochondrial encoded membrane proteins. The mitoribosome contains mitochondria-specific ribosomal proteins and replaced the bacterial 5S rRNA by mitochondria-specific proteins and rRNA extensions. Furthermore, the mitoribosome is tethered to the inner mitochondrial membrane to facilitate a co-translational insertion of newly synthesized proteins. Thus, also the assembly process of mitoribosomes differs from that of bacteria and is to date not well understood.
Therefore, the biogenesis of mitochondrial ribosomes in yeast should be investigated. To this end, a strain was generated in which the gene of the mitochondrial RNA-polymerase RPO41 is under control of an inducible GAL10-promoter. Since the scaffold of ribosomes is built by ribosomal RNAs, the depletion of the RNA-polymerase subsequently leads to a loss of mitochondrial ribosomes. Reinduction of Rpo41 initiates the assembly of new mitoribosomes, which makes this strain an attractive model to study mitoribosome biogenesis.
Initially, the effects of Rpo41 depletion on cellular and mitochondrial physiology was investigated. Upon Rpo41 depletion, growth on respiratory glycerol medium was inhibited. Furthermore, mitochondrial ribosomal 21S and 15S rRNA was diminished and mitochondrial translation was almost completely absent. Also, mitochondrial DNA was strongly reduced due to the fact that mtDNA replication requires RNA primers that get synthesized by Rpo41.
Next, the effect of reinduction of Rpo41 on mitochondria was tested. Time course experiments showed that mitochondrial translation can partially recover from 48h Rpo41 depletion within a timeframe of 4.5h. Sucrose gradient sedimentation experiments further showed that the mitoribosomal constitution was comparable to wildtype control samples during the time course of 4.5h of reinduction, suggesting that the ribosome assembly is not fundamentally altered in Gal-Rpo41 mitochondria. In addition, the depletion time was found to be critical for recovery of mitochondrial translation and mitochondrial RNA levels. It was observed that after 36h of Rpo41 depletion, the rRNA levels and mitochondrial translation recovered to almost 100%, but only within a time course of 10h.
Finally, mitochondria from Gal-Rpo41 cells isolated after different timepoints of reinduction were used to perform complexome profiling and the assembly of mitochondrial protein complexes was investigated. First, the steady state conditions and the assembly process of mitochondrial respiratory chain complexes were monitored. The individual respiratory chain complexes and the super-complexes of complex III, complex IV and complex V were observed. Furthermore, it was seen that they recovered from Rpo41 depletion within 4.5h of reinduction. Complexome profiles of the mitoribosomal small and large subunit discovered subcomplexes of mitoribosomal proteins that were assumed to form prior to their incorporation into assembly intermediates. The complexome profiles after reinduction indeed showed the formation of these subcomplexes before formation of the fully assembled subunit. In the mitochondrial LSU one subcomplex builds the membrane facing protuberance and a second subcomplex forms the central protuberance. In contrast to the preassembled subcomplexes, proteins that were involved in early assembly steps were exclusively found in the fully assembled subunit. Proteins that assemble at the periphery of the mitoribosome during intermediate and late assembly steps where found in soluble form suggesting a pool of unassembled proteins that supply assembly intermediates with proteins.
Taken together, the findings of this thesis suggest a so far unknow building-block model for mitoribosome assembly in which characteristic structures of the yeast mitochondrial ribosome form preassembled subcomplexes prior to their incorporation into the mitoribosome.
A Consistent Large Eddy Approach for Lattice Boltzmann Methods and its Application to Complex Flows
(2015)
Lattice Boltzmann Methods have shown to be promising tools for solving fluid flow problems. This is related to the advantages of these methods, which are among others, the simplicity in handling complex geometries and the high efficiency in calculating transient flows. Lattice Boltzmann Methods are mesoscopic methods, based on discrete particle dynamics. This is in contrast to conventional Computational Fluid Dynamics methods, which are based on the solution of the continuum equations. Calculations of turbulent flows in engineering depend in general on modeling, since resolving of all turbulent scales is and will be in near future far beyond the computational possibilities. One of the most auspicious modeling approaches is the large eddy simulation, in which the large, inhomogeneous turbulence structures are directly computed and the smaller, more homogeneous structures are modeled.
In this thesis, a consistent large eddy approach for the Lattice Boltzmann Method is introduced. This large eddy model includes, besides a subgrid scale model, appropriate boundary conditions for wall resolved and wall modeled calculations. It also provides conditions for turbulent domain inlets. For the case of wall modeled simulations, a two layer wall model is derived in the Lattice Boltzmann context. Turbulent inlet conditions are achieved by means of a synthetic turbulence technique within the Lattice Boltzmann Method.
The proposed approach is implemented in the Lattice Boltzmann based CFD package SAM-Lattice, which has been created in the course of this work. SAM-Lattice is feasible of the calculation of incompressible or weakly compressible, isothermal flows of engineering interest in complex three dimensional domains. Special design targets of SAM-Lattice are high automatization and high performance.
Validation of the suggested large eddy Lattice Boltzmann scheme is performed for pump intake flows, which have not yet been treated by LBM. Even though, this numerical method is very suitable for this kind of vortical flows in complicated domains. In general, applications of LBM to hydrodynamic engineering problems are rare. The results of the pump intake validation cases reveal that the proposed numerical approach is able to represent qualitatively and quantitatively the very complex flows in the intakes. The findings provided in this thesis can serve as the basis for a broader application of LBM in hydrodynamic engineering problems.
Beamforming performs spatial filtering to preserve the signal from given directions of interest while suppressing interfering signals and noise arriving from other directions.
For example, a microphone array equipped with beamforming algorithm could preserve the sound coming from a target speaker and suppress sounds coming from other speakers.
Beamformer has been widely used in many applications such as radar, sonar, communication, and acoustic systems.
A data-independent beamformer is the beamformer whose coefficients are independent on sensor signals, it normally uses less computation since the coefficients are computed once. Moreover, its coefficients are derived from the well-defined statistical models, then it produces less artifacts. The major drawback of this beamforming class is its limitation to the interference suppression.
On the other hand, an adaptive beamformer is a beamformer whose coefficients depend on or adapt to sensor signals. It is capable of suppressing the interference better than a data-independent beamforming but it suffers from either too much distortion of the signal of interest or less noise reduction when the updating rate of coefficients does not synchronize with the changing rate of the noise model. Besides, it is computationally intensive since the coefficients need to be updated frequently.
In acoustic applications, the bandwidth of signals of interest extends over several octaves, but we always expect that the characteristic of the beamformer is invariant with regard to the bandwidth of interest. This can be achieved by the so-called broadband beamforming.
Since the beam pattern of conventional beamformers depends on the frequency of the signal, it is common to use a dense and uniform array for the broadband beamforming to guarantee some essential performances together, such as frequency-independence, less sensitive to white noise, high directivity factor or high front-to-back ratio. In this dissertation, we mainly focus on the sparse array of which the aim is to use fewer sensors in the array,
while simultaneously assuring several important performances of the beamformer.
In the past few decades, many design methodologies for sparse arrays have been proposed and were applied in a variety of practical applications.
Although good results were presented, there are still some restrictions, such as the number of sensors is large, the designed beam pattern must be fixed, the steering ability is limited and the computational complexity is high.
In this work, two novel approaches for the sparse array design taking a hypothesized uniform array as a basis are proposed, that is, one for data-independent beamformers and the another for adaptive beamformers.
As an underlying component of the proposed methods, the dissertation introduces some new insights into the uniform array with broadband beamforming. In this context, a function formulating the relations between the sensor coefficients and its beam pattern over frequency is proposed. The function mainly contains the coordinate transform and inverse Fourier transform.
Furthermore, from the bijection of the function and broadband beamforming perspective, we propose the lower and upper bounds for the inter-distance of sensors. Within these bounds, the function is a bijective function that can be utilized to design the uniform array with broadband beamforming.
For data-independent beamforming, many studies have focused on optimization procedures to seek the sparse array deployment. This dissertation presents an alternative approach to determine the location of sensors.
Starting with a weight spectrum of a virtual dense and uniform array, some techniques are used, such as analyzing a weight spectrum to determine the critical sensors, applying the clustering technique to group the sensors into different groups and selecting representative sensors for each group.
After the sparse array deployment is specified, the optimization technique is applied to find the beamformer coefficients. The proposed method helps to save the computation time in the design phase and its beamformer performance outperforms other state-of-the-art methods in several aspects such as the higher white noise gain, higher directivity factor or more frequency-independence.
For adaptive beamforming, the dissertation attempts to design a versatile sparse microphone array that can be used for different beam patterns.
Furthermore, we aim to reduce the number of microphones in the sparse array while ensuring that its performance can continue to compete with a highly dense and uniform array in terms of broadband beamforming.
An irregular microphone array in a planar surface with the maximum number of distinct distances between the microphones is proposed.
It is demonstrated that the irregular microphone array is well-suited to sparse recovery algorithms that are used to solve underdetermined systems with subject to sparse solutions. Here, a sparse solution is the sound source's spatial spectrum that need to be reconstructed from microphone signals.
From the reconstructed sound sources, a method for array interpolation is presented to obtain an interpolated dense and uniform microphone array that performs well with broadband beamforming.
In addition, two alternative approaches for generalized sidelobe canceler (GSC) beamformer are proposed. One is the data-independent beamforming variant, the other is the adaptive beamforming variant. The GSC decomposes beamforming into two paths: The upper path is to preserve the desired signal, the lower path is to suppress the desired signal. From a beam pattern viewpoint, we propose an improvement for GSC, that is, instead of using the blocking matrix in the lower path to suppress the desired signal, we design a beamformer that contains the nulls at the look direction and at some other directions. Both approaches are simple beamforming design methods and they can be applied to either sparse array or uniform array.
Lastly, a new technique for direction-of-arrival (DOA) estimation based on the annihilating filter is also presented in this dissertation.
It is based on the idea of finite rate of innovation to reconstruct the stream of Diracs, that is, identifying an annihilating filter/locator filter for a few uniform samples and the position of the Diracs are then related to the roots of the filter. Here, an annihilating filter is the filter that suppresses the signal, since its coefficient vector is always orthogonal to every frame of signal.
In the DOA context, we regard an active source as a Dirac associated with the arrival direction, then the directions of active sources can be derived from the roots of the annihilating filter. However,
the DOA obtained by this method is sensitive to noise and the number of DOAs is limited.
To address these issues, the dissertation proposes a robust method to design the annihilating filter and to increase the degree-of-freedom of the measurement system (more active sources can be detected) via observing multiple data frames.
Furthermore, we also analyze the performance of DOA with diffuse noise and propose an extended multiple signal classification algorithm that takes diffuse noise into account. In the simulation,
it shows, that in the case of diffuse noise, only the extended multiple signal classification algorithm can estimate the DOAs properly.
The growing computational power enables the establishment of the Population Balance Equation (PBE)
to model the steady state and dynamic behavior of multiphase flow unit operations. Accordingly, the twophase
flow
behavior inside liquid-liquid extraction equipment is characterized by different factors. These
factors include: interactions among droplets (breakage and coalescence), different time scales due to the
size distribution of the dispersed phase, and micro time scales of the interphase diffusional mass transfer
process. As a result of this, the general PBE has no well known analytical solution and therefore robust
numerical solution methods with low computational cost are highly admired.
In this work, the Sectional Quadrature Method of Moments (SQMOM) (Attarakih, M. M., Drumm, C.,
Bart, H.-J. (2009). Solution of the population balance equation using the Sectional Quadrature Method of
Moments (SQMOM). Chem. Eng. Sci. 64, 742-752) is extended to take into account the continuous flow
systems in spatial domain. In this regard, the SQMOM is extended to solve the spatially distributed
nonhomogeneous bivariate PBE to model the hydrodynamics and physical/reactive mass transfer
behavior of liquid-liquid extraction equipment. Based on the extended SQMOM, two different steady
state and dynamic simulation algorithms for hydrodynamics and mass transfer behavior of liquid-liquid
extraction equipment are developed and efficiently implemented. At the steady state modeling level, a
Spatially-Mixed SQMOM (SM-SQMOM) algorithm is developed and successfully implemented in a onedimensional
physical spatial domain. The integral spatial numerical flux is closed using the mean mass
droplet diameter based on the One Primary and One Secondary Particle Method (OPOSPM which is the
simplest case of the SQMOM). On the other hand the hydrodynamics integral source terms are closed
using the analytical Two-Equal Weight Quadrature (TEqWQ). To avoid the numerical solution of the
droplet rise velocity, an analytical solution based on the algebraic velocity model is derived for the
particular case of unit velocity exponent appearing in the droplet swarm model. In addition to this, the
source term due to mass transport is closed using OPOSPM. The resulting system of ordinary differential
equations with respect to space is solved using the MATLAB adaptive Runge–Kutta method (ODE45). At
the dynamic modeling level, the SQMOM is extended to a one-dimensional physical spatial domain and
resolved using the finite volume method. To close the mathematical model, the required quadrature nodes
and weights are calculated using the analytical solution based on the Two Unequal Weights Quadrature
(TUEWQ) formula. By applying the finite volume method to the spatial domain, a semi-discreet ordinary
differential equation system is obtained and solved. Both steady state and dynamic algorithms are
extensively validated at analytical, numerical, and experimental levels. At the numerical level, the
predictions of both algorithms are validated using the extended fixed pivot technique as implemented in
PPBLab software (Attarakih, M., Alzyod, S., Abu-Khader, M., Bart, H.-J. (2012). PPBLAB: A new
multivariate population balance environment for particulate system modeling and simulation. Procedia
Eng. 42, pp. 144-562). At the experimental validation level, the extended SQMOM is successfully used
to model the steady state hydrodynamics and physical and reactive mass transfer behavior of agitated
liquid-liquid extraction columns under different operating conditions. In this regard, both models are
found efficient and able to follow liquid extraction column behavior during column scale-up, where three
column diameters were investigated (DN32, DN80, and DN150). To shed more light on the local
interactions among the contacted phases, a reduced coupled PBE and CFD framework is used to model
the hydrodynamic behavior of pulsed sieve plate columns. In this regard, OPOSPM is utilized and
implemented in FLUENT 18.2 commercial software as a special case of the SQMOM. The dropletdroplet
interactions
(breakage
and
coalescence)
are
taken
into
account
using
OPOSPM,
while
the
required
information
about
the
velocity
field
and
energy
dissipation
is
calculated
by
the
CFD
model.
In
addition
to
this,
the proposed coupled OPOSPM-CFD framework is extended to include the mass transfer. The
proposed framework is numerically tested and the results are compared with the published experimental
data. The required breakage and coalescence parameters to perform the 2D-CFD simulation are estimated
using PPBLab software, where a 1D-CFD simulation using a multi-sectional gird is performed. A very
good agreement is obtained at the experimental and the numerical validation levels.
The present thesis describes the development and the evaluation of a design procedure of inducer with arbitrary meridional and blade shape. This special type of pump impeller, which is usually mounted upstream of a main pump impeller, is employed in many applications demanding the realization of low NPSH values. An inducer basically increases suction performance by producing mostly a small pressure rise while allowing for a greater degree of cavitation, that is the formation of vapor bubbles, at its inlet than a conventional pump impeller. This is achieved by specially designed blade channels promoting the collapse of the produced vapor bubbles.
The main focus of the present thesis is the description of the design method, which enables the generation of the three-dimensional blade geometry. The method is based on a parametric representation of the geometry considering the particular requirements for inducers and the publicly available design practice. Within this approach the sequence of design steps is adapted from the classical design process of mixed flow and radial impellers. As a consequence leading and trailing edge blade angles are determined based on simplifications and certain empirical assumptions for multiple blade sections and are used to design the blade camber curves. Along the camber curves the blade profile is generated following a thickness distribution that has to be prescribed. A special feature of the newly developed method is that arbitrary shaped, asymmetric thickness distributions can be realized.
Due to the detailed description of the design and calculation steps a fully comprehensible procedure is outlined, which covers the development of inducer bladings from an initial set of duty parameters to the final three-dimensional blade geometry.
The components involved in the design procedure are tested by designing two exemplary inducers and they are assessed by comparison with numerical simulations. Functioning of these inducers in the real application is finally demonstrated with water tests.
The main result of this dissertation is a design software for inducers allowing for the design of three-dimensional, asymmetrically profiled bladings. The developed software is free of commercial third-party libraries. As a consequence a program is available that can be modified and extended as desired. As potential future development goals inducers with splitter and tandem blades as well as an integrated design of inducer and impeller are proposed.
Medical cyber-physical systems (MCPS) emerged as an evolution of the relations between connected health systems, healthcare providers, and modern medical devices. Such systems combine independent medical devices at runtime in order to render new patient monitoring/control functionalities, such as physiological closed loops for controlling drug infusion or optimization of alarms. Despite the advances regarding alarm precision, healthcare providers still struggle with alarm flooding caused by the limited risk assessment models. Furthermore, these limitations also impose severe barriers on the adoption of automated supervision through autonomous actions, such as safety interlocks for avoiding overdosage. The literature has focused on the verification of safety parameters to assure the safety of treatment at runtime and thus optimize alarms and automated actions. Such solutions have relied on the definition of actuation ranges based on thresholds for a few monitored parameters. Given the very dynamic nature of the relevant context conditions (e.g., the patient’s condition, treatment details, system configurations, etc.), fixed thresholds are a weak means for assessing the current risk. This thesis presents an approach for enabling dynamic risk assessment for cooperative MCPS based on an adaptive Bayesian Networks (BN) model. The main aim of the approach is to support continuous runtime risk assessment of the current situation based on relevant context and system information. The presented approach comprises (i) a dynamic risk analysis constituent, which corresponds to the elicitation of relevant risk parameters, risk metric building, and risk metric management; and (ii) a runtime risk classification constituent, which aims to analyze the current situation risk, establish risk classes, and identify and deploy mitigation measures. The proposed approach was evaluated and its feasibility proved by means of simulated experiments guided by an international team of medical experts with a focus on the requirements of efficacy, efficiency, and availability of patient treatment.
This work deals with the simulation of the micro-cutting process of titanium. For this
purpose, a suitable crystal-plastic material model is developed and efficient implemen-
tations are investigated to simulate the micro-cutting process. Several challenges arise
for the material model. On the one hand, the low symmetry hexagonal close-packed
crystal structure of titanium has to be considered. On the other hand, large defor-
mations and strains occur during the machining process. Another important part is
the algorithm for the determination of the active slip systems, which has a significant
influence on the stability of the simulation. In order to obtain a robust implemen-
tation, different aspects, such as the algorithm for the determination of the active
slip systems, the method for mesh separation between chip and workpiece as well as
the hardening process are investigated, and different approaches are compared. The
developed crystal-plastic material model and the selected implementations are first
validated and investigated using illustrative examples. The presented simulations of
the micro-cutting process show the influence of different machining parameters on the
process. Finally, the influence of a real microstructure on the plastic deformation and
the cutting force during the process is shown.
A prime motivation for using XML to directly represent pieces of information is the ability of supporting ad-hoc or 'schema-later' settings. In such scenarios, modeling data under loose data constraints is essential. Of course, the flexibility of XML comes at a price: the absence of a rigid, regular, and homogeneous structure makes many aspects of data management more challenging. Such malleable data formats can also lead to severe information quality problems, because the risk of storing inconsistent and incorrect data is greatly increased. A prominent example of such problems is the appearance of the so-called fuzzy duplicates, i.e., multiple and non-identical representations of a real-world entity. Similarity joins correlating XML document fragments that are similar can be used as core operators to support the identification of fuzzy duplicates. However, similarity assessment is especially difficult on XML datasets because structure, besides textual information, may exhibit variations in document fragments representing the same real-world entity. Moreover, similarity computation is substantially more expensive for tree-structured objects and, thus, is a serious performance concern. This thesis describes the design and implementation of an effective, flexible, and high-performance XML-based similarity join framework. As main contributions, we present novel structure-conscious similarity functions for XML trees - either considering XML structure in isolation or combined with textual information -, mechanisms to support the selection of relevant information from XML trees and organization of this information into a suitable format for similarity calculation, and efficient algorithms for large-scale identification of similar, set-represented objects. Finally, we validate the applicability of our techniques by integrating our framework into a native XML database management system; in this context we address several issues around the integration of similarity operations into traditional database architectures.
This thesis presents a novel, generic framework for information segmentation in document images.
A document image contains different types of information, for instance, text (machine printed/handwritten), graphics, signatures, and stamps.
It is necessary to segment information in documents so that to process such segmented information only when required in automatic document processing workflows.
The main contribution of this thesis is the conceptualization and implementation of an information segmentation framework that is based on part-based features.
The generic nature of the presented framework makes it applicable to a variety of documents (technical drawings, magazines, administrative, scientific, and academic documents) digitized using different methods (scanners, RGB cameras, and hyper-spectral imaging (HSI) devices).
A highlight of the presented framework is that it does not require large training sets, rather a few training samples (for instance, four pages) lead to high performance, i.e., better than previously existing methods.
In addition, the presented framework is simple and can be adapted quickly to new problem domains.
This thesis is divided into three major parts on the basis of document digitization method (scanned, hyper-spectral imaging, and camera captured) used.
In the area of scanned document images, three specific contributions have been realized.
The first of them is in the domain of signature segmentation in administrative documents.
In some workflows, it is very important to check the document authenticity before processing the actual content.
This can be done based on the available seal of authenticity, e.g., signatures.
However, signature verification systems expect pre-segmented signature image, while signatures are usually a part of document.
To use signature verification systems on document images, it is necessary to first segment signatures in documents.
This thesis shows that the presented framework can be used to segment signatures in administrative documents.
The system based on the presented framework is tested on a publicly available dataset where it outperforms the state-of-the-art methods and successfully segmented all signatures, while less than half of the found signatures are false positives.
This shows that it can be applied for practical use.
The second contribution in the area of scanned document images is segmentation of stamps in administrative documents.
A stamp also serves as a seal for documents authenticity.
However, the location of stamp on the document can be more arbitrary than a signature depending on the person sealing the document.
This thesis shows that a system based on our generic framework is able to extract stamps of any arbitrary shape and color.
The evaluation of the presented system on a publicly available dataset shows that it is also able to segment black stamps (that were not addressed in the past) with a recall and precision of 83% and 73%, respectively.
%Furthermore, to segment colored stamps, this thesis presents a novel feature set which is based on intensity gradient, is able to extract unseen, colored, arbitrary shaped, textual as well as graphical stamps, and outperforms the state-of-the-art methods.
The third contribution in the scanned document images is in the domain of information segmentation in technical drawings (architectural floorplans, maps, circuit diagrams, etc.) containing usually a large amount of graphics and comparatively less textual components. Further, as in technical drawings, text is overlapping with graphics.
Thus, automatic analysis of technical drawings uses text/graphics segmentation as a pre-processing step.
This thesis presents a method based on our generic information segmentation framework that is able to detect the text, which is touching graphical components in architectural floorplans and maps.
Evaluation of the method on a publicly available dataset of architectural floorplans shows that it is able to extract almost all touching text components with precision and recall of 71% and 95%, respectively.
This means that almost all of the touching text components are successfully extracted.
In the area of hyper-spectral document images, two contributions have been realized.
Unlike normal three channels RGB images, hyper-spectral images usually have multiple channels that range from ultraviolet to infrared regions including the visible region.
First, this thesis presents a novel automatic method for signature segmentation from hyper-spectral document images (240 spectral bands between 400 - 900 nm).
The presented method is based on a part-based key point detection technique, which does not use any structural information, but relies only on the spectral response of the document regardless of ink color and intensity.
The presented method is capable of segmenting (overlapping and non-overlapping) signatures from varying backgrounds like, printed text, tables, stamps, logos, etc.
Importantly, the presented method can extract signature pixels and not just the bounding boxes.
This is substantial when signatures are overlapping with text and/or other objects in image. Second, this thesis presents a new dataset comprising of 300 documents scanned using a high-resolution hyper-spectral scanner. Evaluation of the presented signature segmentation method on this hyper-spectral dataset shows that it is able to extract signature pixels with the precision and recall of 100% and 79%, respectively.
Further contributions have been made in the area of camera captured document images. A major problem in the development of Optical Character Recognition (OCR) systems for camera captured document images is the lack of labeled camera captured document images datasets. In the first place, this thesis presents a novel, generic, method for automatic ground truth generation/labeling of document images. The presented method builds large-scale (i.e., millions of images) datasets of labeled camera captured / scanned documents without any human intervention. The method is generic and can be used for automatic ground truth generation of (scanned and/or camera captured) documents in any language, e.g., English, Russian, Arabic, Urdu. The evaluation of the presented method, on two different datasets in English and Russian, shows that 99.98% of the images are correctly labeled in every case.
Another important contribution in the area of camera captured document images is the compilation of a large dataset comprising 1 million word images (10 million character images), captured in a real camera-based acquisition environment, along with the word and character level ground truth. The dataset can be used for training as well as testing of character recognition systems for camera-captured documents. Various benchmark tests are performed to analyze the behavior of different open source OCR systems on camera captured document images. Evaluation results show that the existing OCRs, which already get very high accuracies on scanned documents, fail on camera captured document images.
Using the presented camera-captured dataset, a novel character recognition system is developed which is based on a variant of recurrent neural networks, i.e., Long Short Term Memory (LSTM) that outperforms all of the existing OCR engines on camera captured document images with an accuracy of more than 95%.
Finally, this thesis provides details on various tasks that have been performed in the area closely related to information segmentation. This includes automatic analysis and sketch based retrieval of architectural floor plan images, a novel scheme for online signature verification, and a part-based approach for signature verification. With these contributions, it has been shown that part-based methods can be successfully applied to document image analysis.
For many years real-time task models have focused the timing constraints on execution windows defined by earliest start times and deadlines for feasibility.
However, the utility of some application may vary among scenarios which yield correct behavior, and maximizing this utility improves the resource utilization.
For example, target sensitive applications have a target point where execution results in maximized utility, and an execution window for feasibility.
Execution around this point and within the execution window is allowed, albeit at lower utility.
The intensity of the utility decay accounts for the importance of the application.
Examples of such applications include multimedia and control; multimedia application are very popular nowadays and control applications are present in every automated system.
In this thesis, we present a novel real-time task model which provides for easy abstractions to express the timing constraints of target sensitive RT applications: the gravitational task model.
This model uses a simple gravity pendulum (or bob pendulum) system as a visualization model for trade-offs among target sensitive RT applications.
We consider jobs as objects in a pendulum system, and the target points as the central point.
Then, the equilibrium state of the physical problem is equivalent to the best compromise among jobs with conflicting targets.
Analogies with well-known systems are helpful to fill in the gap between application requirements and theoretical abstractions used in task models.
For instance, the so-called nature algorithms use key elements of physical processes to form the basis of an optimization algorithm.
Examples include the knapsack problem, traveling salesman problem, ant colony optimization, and simulated annealing.
We also present a few scheduling algorithms designed for the gravitational task model which fulfill the requirements for on-line adaptivity.
The scheduling of target sensitive RT applications must account for timing constraints, and the trade-off among tasks with conflicting targets.
Our proposed scheduling algorithms use the equilibrium state concept to order the execution sequence of jobs, and compute the deviation of jobs from their target points for increased system utility.
The execution sequence of jobs in the schedule has a significant impact on the equilibrium of jobs, and dominates the complexity of the problem --- the optimum solution is NP-hard.
We show the efficacy of our approach through simulations results and 3 target sensitive RT applications enhanced with the gravitational task model.
This thesis is concerned with the modeling of the solid-solid phase transformation, such as the martensitic transformation. The allotropes austenite and martensite are important for industry applications. As a result of its ductility, austenite is desired in the bulk, as opposed to martensite, which desired in the near surface region. The phase field method is used to model the phase transformation by minimizing the free energy. It consists of a mechanical part, due to elastic strain and a chemical part, due to the martensitic transformation. The latter is temperature dependent. Therefore, a temperature dependent separation potential is presented here. To accommodate multiple orientation variants, a multivariant phase field model is employed. Using the Khachaturyan approach, the effective material parameters can be used to describe a constitutive model. This however, renders the nodal residual vector and elemental tangent matrix directly dependent on the phase, making a generalization complicated. An easier approach is the use of the Voigt/Taylor homogenization, in which the energy and their derivatives are interpolated creating an interface for material law of the individual phases.
Numerical Godeaux surfaces are minimal surfaces of general type with the smallest possible numerical invariants. It is known that the torsion group of a numerical Godeaux surface is cyclic of order \(m\leq 5\). A full classification has been given for the cases \(m=3,4,5\) by the work of Reid and Miyaoka. In each case, the corresponding moduli space is 8-dimensional and irreducible.
There exist explicit examples of numerical Godeaux surfaces for the orders \(m=1,2\), but a complete classification for these surfaces is still missing.
In this thesis we present a construction method for numerical Godeaux surfaces which is based on homological algebra and computer algebra and which arises from an experimental approach by Schreyer. The main idea is to consider the canonical ring \(R(X)\) of a numerical Godeaux surface \(X\) as a module over some graded polynomial ring \(S\). The ring \(S\) is chosen so that \(R(X)\) is finitely generated as an \(S\)-module and a Gorenstein \(S\)-algebra of codimension 3. We prove that the canonical ring of any numerical Godeaux surface, considered as an \(S\)-module, admits a minimal free resolution whose middle map is alternating. Moreover, we show that a partial converse of this statement is true under some additional conditions.
Afterwards we use these results to construct (canonical rings of) numerical Godeaux surfaces. Hereby, we restrict our study to surfaces whose bicanonical system has no fixed component but 4 distinct base points, in the following referred to as marked numerical Godeaux surfaces.
The particular interest of this thesis lies on marked numerical Godeaux surfaces whose torsion group is trivial. For these surfaces we study the fibration of genus 4 over \(\mathbb{P}^1\) induced by the bicanonical system. Catanese and Pignatelli showed that the general fibre is non-hyperelliptic and that the number \(\tilde{h}\) of hyperelliptic fibres is bounded by 3. The two explicit constructions of numerical Godeaux surfaces with a trivial torsion group due to Barlow and Craighero-Gattazzo, respectively, satisfy \(\tilde{h} = 2\).
With the method from this thesis, we construct an 8-dimensional family of numerical Godeaux surfaces with a trivial torsion group and whose general element satisfy \(\tilde{h}=0\).
Furthermore, we establish a criterion for the existence of hyperelliptic fibres in terms of a minimal free resolution of \(R(X)\). Using this criterion, we verify experimentally the
existence of a numerical Godeaux surface with \(\tilde{h}=1\).
Data is the new gold and serves as a key to answer the five W’s (Who, What, Where, When, Why) and How’s of any business. Companies are now mining data more than ever and one of the most important aspects while analyzing this data is to detect anomalous patterns to identify critical patterns and points. To tackle the vital aspects of timeseries analysis, this thesis presents a novel hybrid framework that stands on three pillars: Anomaly Detection, Uncertainty Estimation,
and Interpretability and Explainability.
The first pillar is comprised of contributions in the area of time-series anomaly detection. Deep Anomaly Detection for Time-series (DeepAnT), a novel deep learning-based anomaly detection method, lies at the foundation of the proposed hybrid framework and addresses the inadequacy of traditional anomaly detection methods. To the best of the author’s knowledge, Convolutional Neural Network (CNN) was used for the first time in Deep Anomaly Detection for Time-series (DeepAnT) to robustly detect multiple types of anomalies in the tricky
and continuously changing time-series data. To further improve the anomaly detection performance, a fusion-based method, Fusion of
Statistical and Deep Learning for Anomaly Detection (FuseAD) is proposed. This method aims to combine the strengths of existing wellfounded
statistical methods and powerful data-driven methods.
In the second pillar of this framework, a hybrid approach that combines the high accuracy of the deterministic models with the posterior distribution approximation of Bayesian neural networks is proposed.
In the third pillar of the proposed framework, mechanisms to enable both HOW and WHY parts are presented.
Embedded systems have become ubiquitous in everyday life, and especially in the automotive industry. New applications challenge their design by introducing a new class of problems that are based on a detailed analysis of the environmental situation. Situation analysis systems rely on models and algorithms of the domain of computational geometry. The basic model is usually an Euclidean plane, which contains polygons to represent the objects of the environment. Usual implementations of computational geometry algorithms cannot be directly used for safety-critical systems. First, a strict analysis of their correctness is indispensable and second, nonfunctional requirements with respect to the limited resources must be considered. This thesis proposes a layered approach to a polygon-processing system. On top of rational numbers, a geometry kernel is formalised at first. Subsequently, geometric primitives form a second layer of abstraction that is used for plane sweep and polygon algorithms. These layers do not only divide the whole system into manageable parts but make it possible to model problems and reason about them at the appropriate level of abstraction. This structure is used for the verification as well as the implementation of the developed polygon-processing library.
In the filling process of a car tank, the formation of foam plays an unwanted role, as it may prevent the tank from being completely filled or at least delay the filling. Therefore it is of interest to optimize the geometry of the tank using numerical simulation in such a way that the influence of the foam is minimized. In this dissertation, we analyze the behaviour of the foam mathematically on the mezoscopic scale, that is for single lamellae. The most important goals are on the one hand to gain a deeper understanding of the interaction of the relevant physical effects, on the other hand to obtain a model for the simulation of the decay of a lamella which can be integrated in a global foam model. In the first part of this work, we give a short introduction into the physical properties of foam and find that the Marangoni effect is the main cause for its stability. We then develop a mathematical model for the simulation of the dynamical behaviour of a lamella based on an asymptotic analysis using the special geometry of the lamella. The result is a system of nonlinear partial differential equations (PDE) of third order in two spatial and one time dimension. In the second part, we analyze this system mathematically and prove an existence and uniqueness result for a simplified case. For some special parameter domains the system can be further simplified, and in some cases explicit solutions can be derived. In the last part of the dissertation, we solve the system using a finite element approach and discuss the results in detail.
The detection and characterisation of undesired lead structures on shaft surfaces is a concern in production and quality control of rotary shaft lip-type sealing systems. The potential lead structures are generally divided into macro and micro lead based on their characteristics and formation. Macro lead measurement methods exist and are widely applied. This work describes a method to characterise micro lead on ground shaft surfaces. Micro lead is known as the deviation of main orientation of the ground micro texture from circumferential direction. Assessing the orientation of microscopic structures with arc minute accuracy with regard to circumferential direction requires exact knowledge of both the shaft’s orientation and the direction of surface texture. The shaft’s circumferential direction is found by calibration. Measuring systems and calibration procedures capable of calibrating shaft axis orientation with high accuracy and low uncertainty are described. The measuring systems employ areal-topographic measuring instruments suited for evaluating texture orientation. A dedicated evaluation scheme for texture orientation is based on the Radon transform of these topographies and parametrised for the application. Combining the calibration of circumferential direction with the evaluation of texture orientation the method enables the measurement of micro lead on ground shaft surfaces.
1,3-Diynes are frequently found as an important structural motif in natural products, pharmaceuticals and bioactive compounds, electronic and optical materials and supramolecular molecules. Copper and palladium complexes are widely used to prepare 1,3-diynes by homocoupling of terminal alkynes; albeit the potential of nickel complexes towards the same is essentially unexplored. Although a detailed study on the reported nickel-acetylene chemistry has not been carried out, a generalized mechanism featuring a nickel(II)/nickel(0) catalytic cycle has been proposed. In the present work, a detailed mechanistic aspect of the nickel-mediated homocoupling reaction of terminal alkynes is investigated through the isolation and/or characterization of key intermediates from both the stoichiometric and the catalytic reactions. A nickel(II) complex [Ni(L-N4Me2)(MeCN)2](ClO4)2 (1) containing a tetradentate N,N′-dimethyl-2,11-diaza[3.3](2,6)pyridinophane (L-N4Me2) as ligand was used as catalyst for homocoupling of terminal alkynes by employing oxygen as oxidant at room temperature. A series of dinuclear nickel(I) complexes bridged by a 1,3-diyne ligand have been isolated from stoichiometric reaction between [Ni(L-N4Me2)(MeCN)2](ClO4)2 (1) and lithium acetylides. The dinuclear nickel(I)-diyne complexes [{Ni(L-N4Me2)}2(RC4R)](ClO4)2 (2) were well characterized by X-ray crystal structures, various spectroscopic methods, SQUID and DFT calculation. The complexes not only represent as a key intermediate in aforesaid catalytic reaction, but also describe the first structurally characterized dinuclear nickel(I)-diyne complexes. In addition, radical trapping and low temperature UV-Vis-NIR experiments in the formation of the dinuclear nickel(I)-diyne confirm that the reactions occurring during the reduction of nickel(II) to nickel(I) and C-C bond formation of 1,3-diyne follow non-radical concerted mechanism. Furthermore, spectroscopic investigation on the reactivity of the dinuclear nickel(I)-diyne complex towards molecular oxygen confirmed the formation of a mononuclear nickel(I)-diyne species [Ni(L-N4Me2)(RC4R)]+ (4) and a mononuclear nickel(III)-peroxo species [Ni(L-N4Me2)(O2)]+ (5) which were converted to free 1,3-diyne and an unstable dinuclear nickel(II) species [{Ni(L-N4Me2)}2(O2)]2+ (6). A mononuclear nickel(I)-alkyne complex [Ni(L-N4Me2)(PhC2Ph)](ClO4).MeOH (3) and the mononuclear nickel(III)-peroxo species [Ni(L-N4Me2)(O2)]+ (5) were isolated/generated and characterized to confirm the formulation of aforementioned mononuclear nickel(I)-diyne and mononuclear nickel(III)-peroxo species. Spectroscopic experiments on the catalytic reaction mixture also confirm the presence of aforesaid intermediates. Results of both stoichiometric and catalytic reactions suggested an intriguing mechanism involving nickel(II)/nickel(I)/nickel(III) oxidation states in contrast to the reported nickel(II)/nickel(0) catalytic cycle. These findings are expected to open a new paradigm towards nickel-catalyzed organic transformations.
Nowadays, accounting, charging and billing users' network resource consumption are commonly used for the purpose of facilitating reasonable network usage, controlling congestion, allocating cost, gaining revenue, etc. In traditional IP traffic accounting systems, IP addresses are used to identify the corresponding consumers of the network resources. However, there are some situations in which IP addresses cannot be used to identify users uniquely, for example, in multi-user systems. In these cases, network resource consumption can only be ascribed to the owners of these hosts instead of corresponding real users who have consumed the network resources. Therefore, accurate accountability in these systems is practically impossible. This is a flaw of the traditional IP address based IP traffic accounting technique. This dissertation proposes a user based IP traffic accounting model which can facilitate collecting network resource usage information on the basis of users. With user based IP traffic accounting, IP traffic can be distinguished not only by IP addresses but also by users. In this dissertation, three different schemes, which can achieve the user based IP traffic accounting mechanism, are discussed in detail. The inband scheme utilizes the IP header to convey the user information of the corresponding IP packet. The Accounting Agent residing in the measured host intercepts IP packets passing through it. Then it identifies the users of these IP packets and inserts user information into the IP packets. With this mechanism, a meter located in a key position of the network can intercept the IP packets tagged with user information, extract not only statistic information, but also IP addresses and user information from the IP packets to generate accounting records with user information. The out-of-band scheme is a contrast scheme to the in-band scheme. It also uses an Accounting Agent to intercept IP packets and identify the users of IP traffic. However, the user information is transferred through a separated channel, which is different from the corresponding IP packets' transmission. The Multi-IP scheme provides a different solution for identifying users of IP traffic. It assigns each user in a measured host a unique IP address. Through that, an IP address can be used to identify a user uniquely without ambiguity. This way, traditional IP address based accounting techniques can be applied to achieve the goal of user based IP traffic accounting. In this dissertation, a user based IP traffic accounting prototype system developed according to the out-of-band scheme is also introduced. The application of user based IP traffic accounting model in the distributed computing environment is also discussed.
The present situation of control engineering in the context of automated production can be described as a tension field between its desired outcome and its actual consideration. On the one hand, the share of control engineering compared to the other engineering domains has significantly increased within the last decades due to rising automation degrees of production processes and equipment. On the other hand, the control engineering domain is still underrepresented within the production engineering process. Another limiting factor constitutes a lack of methods and tools to decrease the amount of software engineering efforts and to permit the development of innovative automation applications that ideally support the business requirements.
This thesis addresses this challenging situation by means of the development of a new control engineering methodology. The foundation is built by concepts from computer science to promote structuring and abstraction mechanisms for the software development. In this context, the key sources for this thesis are the paradigm of Service-oriented Architecture and concepts from Model-driven Engineering. To mold these concepts into an integrated engineering procedure, ideas from Systems Engineering are applied. The overall objective is to develop an engineering methodology to improve the efficiency of control engineering by a higher adaptability of control software and decreased programming efforts by reuse.
Cloud Computing, or the Cloud, became one of the most used technologies in today's world, right after its possibilities had been figured out. It is a renowned technology that enables ubiquitous access to tasks that need collaboration or remote monitoring. It is widely used in daily lives as well as the industry. The paradigm uses Internet Technologies which rely on best-effort communication. Best-effort communication limits the applicability of the technology in the domains where the timing is critical. Edge Computing is a paradigm that is seen as a complementary technology to the Cloud. It is expected to solve the Quality of Service (QoS) and latency problems that are raised due to the increased count of connected devices, and the physical distance between the infrastructure and devices. The Edge Computing adds a new tier between Information Technology (IT) and Operational Technology (OT) and brings the computing power close to the source of the data. Computing power near devices reduces the dependency to the Internet; hence, in case of a network failure, the computation can still continue. Close proximity deployments also enable the application of Edge Computing in the areas where real-timeliness is necessary. Computation and communication in Edge Computing are performed via Edge Servers. This thesis suggests a standardized and hardware-independent software reference architecture for Edge Servers that can be realized as a framework on servers, to be used on domains where the timing is critical. The suggested architecture is scalable, extensible, modular, multi-user supported, and decentralized. In decentralized systems, several precautions must be taken into consideration, such as latencies, delays, and available resources of the neighbouring servers. The resulting architecture evaluates these factors and enables real-time execution. It also hides the complexity of low-level communication and automates the collaboration between Edge Servers to enable seamless offloading in case of a need due to lack of resources. The thesis also validates an exemplary instance of the architecture with at framework, called Real-Time Execution Framework (RTEF), with multiple scenarios. The tasks used are resource-demanding and requested to be executed on an Edge Server in an Edge Network comprising multiple Edge Servers. The servers can make decisions by evaluating their availabilities, and determine the optimal location to execute the task, without causing deadline misses. Even under a heavy load, the decisions made by the servers to execute the tasks on time were correct, and the concept is proven.
A Multi-Phase Flow Model Incorporated with Population Balance Equation in a Meshfree Framework
(2011)
This study deals with the numerical solution of a meshfree coupled model of Computational Fluid Dynamics (CFD) and Population Balance Equation (PBE) for liquid-liquid extraction columns. In modeling the coupled hydrodynamics and mass transfer in liquid extraction columns one encounters multidimensional population balance equation that could not be fully resolved numerically within a reasonable time necessary for steady state or dynamic simulations. For this reason, there is an obvious need for a new liquid extraction model that captures all the essential physical phenomena and still tractable from computational point of view. This thesis discusses a new model which focuses on discretization of the external (spatial) and internal coordinates such that the computational time is drastically reduced. For the internal coordinates, the concept of the multi-primary particle method; as a special case of the Sectional Quadrature Method of Moments (SQMOM) is used to represent the droplet internal properties. This model is capable of conserving the most important integral properties of the distribution; namely: the total number, solute and volume concentrations and reduces the computational time when compared to the classical finite difference methods, which require many grid points to conserve the desired physical quantities. On the other hand, due to the discrete nature of the dispersed phase, a meshfree Lagrangian particle method is used to discretize the spatial domain (extraction column height) using the Finite Pointset Method (FPM). This method avoids the extremely difficult convective term discretization using the classical finite volume methods, which require a lot of grid points to capture the moving fronts propagating along column height.
A Multi-Sensor Intelligent Assistance System for Driver Status Monitoring and Intention Prediction
(2017)
Advanced sensing systems, sophisticated algorithms, and increasing computational resources continuously enhance the advanced driver assistance systems (ADAS). To date, despite that some vehicle based approaches to driver fatigue/drowsiness detection have been realized and deployed, objectively and reliably detecting the fatigue/drowsiness state of driver without compromising driving experience still remains challenging. In general, the choice of input sensorial information is limited in the state-of-the-art work. On the other hand, smart and safe driving, as representative future trends in the automotive industry worldwide, increasingly demands the new dimensional human-vehicle interactions, as well as the associated behavioral and bioinformatical data perception of driver. Thus, the goal of this research work is to investigate the employment of general and custom 3D-CMOS sensing concepts for the driver status monitoring, and to explore the improvement by merging/fusing this information with other salient customized information sources for gaining robustness/reliability. This thesis presents an effective multi-sensor approach with novel features to driver status monitoring and intention prediction aimed at drowsiness detection based on a multi-sensor intelligent assistance system -- DeCaDrive, which is implemented on an integrated soft-computing system with multi-sensing interfaces in a simulated driving environment. Utilizing active illumination, the IR depth camera of the realized system can provide rich facial and body features in 3D in a non-intrusive manner. In addition, steering angle sensor, pulse rate sensor, and embedded impedance spectroscopy sensor are incorporated to aid in the detection/prediction of driver's state and intention. A holistic design methodology for ADAS encompassing both driver- and vehicle-based approaches to driver assistance is discussed in the thesis as well. Multi-sensor data fusion and hierarchical SVM techniques are used in DeCaDrive to facilitate the classification of driver drowsiness levels based on which a warning can be issued in order to prevent possible traffic accidents. The realized DeCaDrive system achieves up to 99.66% classification accuracy on the defined drowsiness levels, and exhibits promising features such as head/eye tracking, blink detection, gaze estimation that can be utilized in human-vehicle interactions. However, the driver's state of "microsleep" can hardly be reflected in the sensor features of the implemented system. General improvements on the sensitivity of sensory components and on the system computation power are required to address this issue. Possible new features and development considerations for DeCaDrive are discussed as well in the thesis aiming to gain market acceptance in the future.
The simulation of cutting process challenges established methods due to large deformations and topological changes. In this work a particle finite element method (PFEM) is presented, which combines the benefits of discrete modeling techniques and methods based on continuum mechanics. A crucial part of the PFEM is the detection of the boundary of a set of particles. The impact of this boundary detection method on the structural integrity is examined and a relation of the key parameter of the method to the eigenvalues of strain tensors is elaborated. The influence of important process parameters on the cutting force is studied and a comparison to an empirical relation is presented.
The dissertation is concerned with the numerical solution of Fokker-Planck equations in high dimensions arising in the study of dynamics of polymeric liquids. Traditional methods based on tensor product structure are not applicable in high dimensions for the number of nodes required to yield a fixed accuracy increases exponentially with the dimension; a phenomenon often referred to as the curse of dimension. Particle methods or finite point set methods are known to break the curse of dimension. The Monte Carlo method (MCM) applied to such problems are 1/sqrt(N) accurate, where N is the cardinality of the point set considered, independent of the dimension. Deterministic version of the Monte Carlo method called the quasi Monte Carlo method (QMC) are quite effective in integration problems and accuracy of the order of 1/N can be achieved, up to a logarithmic factor. However, such a replacement cannot be carried over to particle simulations due to the correlation among the quasi-random points. The method proposed by Lecot (C.Lecot and F.E.Khettabi, Quasi-Monte Carlo simulation of diffusion, Journal of Complexity, 15 (1999), pp.342-359) is the only known QMC approach, but it not only leads to large particle numbers but also the proven order of convergence is 1/N^(2s) in dimension s. We modify the method presented there, in such a way that the new method works with reasonable particle numbers even in high dimensions and has better order of convergence. Though the provable order of convergence is 1/sqrt(N), the results show less variance and thus the proposed method still slightly outperforms standard MCM.
This thesis is concerned with a phase field model for martensitic transformations in metastable austenitic steels. Within the phase field approach an order parameter is introduced to indicate whether the present phase is austenite or martensite. The evolving microstructure is described by the evolution of the order parameter, which is assumed to follow the time-dependent Ginzburg-Landau equation. The elastic phase field model is enhanced in two different ways to take further phenomena into account. First, dislocation movement is considered by a crystal plasticity setting. Second, the elastic model for martensitic transformations is combined with a phase field model for fracture. Finite element simulations are used to study the single effects separately which contribute to the microstructure formation.
Human interferences within the Earth System are accelerating, leading to major impacts and feedback that we are just beginning to understand. Summarized under the term 'global change' these impacts put human and natural systems under ever-increasing stress and impose a threat to human well-being, particularly in the Global South. Global governance bodies have acknowledged that decisive measures have to be taken to mitigate the causes and to adapt to these new conditions. Nevertheless, neither current international nor national pledges and measures reach the effectiveness needed to sustain global human well-being under accelerating global change. On the contrary, competing interests are not only paralyzing the international debate but also playing an increasingly important role in debates over social fragmentation and societal polarization on national and local scales. This interconnectedness of the natural and the social system and its impact on social phenomena such as cooperation and conflicts need to be understood better, to strengthen social resilience to future disturbances, drive societal transformation towards socially desirable futures while at the same time avoiding path dependencies along continuing colonial continuities. As a case example, this thesis provides insights into southwestern Amazonia, where the intertwined challenges of human contribution to global change in all its dimensions, as well as human adaptation and mitigation attempts to the imposed changes become exaggeratedly visible. As such, southwestern Amazonia with its high social, economic, and biological diversity is a good example to study the deep interrelations of humans with nature and the consequences these relations have on social cohesion amid an ecological crisis.
Therefore, this thesis takes a social-ecological perspective on conflicts and social cohesion. Social cohesion is in a wider sense understood as the way "how members of a society, group, or organization relate to each other and work together" (Dany and Dijkzeul 2022, p. 12). In particular in contexts of violence, conflicts, and fragility, little has been investigated on the role of social cohesion to govern public goods and build resilience for (future) environmental crises. At the same time, governments and international decision-makers more and more acknowledge the role of social cohesion _ comprising both relations between social groups and between groups and the state _ to build upon resilience against crises. Facing uncertainty in how natural and social systems react to certain disturbances and shocks, the governance of potential tipping points, is an additional challenge for the governance of social-ecological systems (SES). Therefore, this thesis asks: "How does governance shape pathways towards cooperative or conflictive social-ecological tipping points?" The results of this thesis can be distinguished into theoretical/conceptual results and empirical results. Initial systematic literature research on the nexus of climate change, land use, and conflict revealed, an extensive body of literature on direct effects, for example, drought-related land use conflicts, with diverging opinions on whether global warming increases the risk for conflicts or not. Adding the perspective of indirect implications, we further identified research gaps, and also a lack of policy recognition, concerning the negative externalities on land use and conflict through climate mitigation and adaptation measures. On a conceptual note, taking a social cohesion perspective into the analysis is beneficial to shift the focus from the problem-oriented perspective of vulnerabilities and conflicts to global change and potential resulting conflicts to a solution-oriented perspective of enhancing agency and resilience to strengthen collaboration. The developed Social Cohesion Conceptual Model and the related analytical framework facilitate the incorporation of societal dynamics into the analysis of SES dynamics. In addition, the elaborated Tipping Multiverse Framework took up this idea and enhanced it with a more detailed perspective on the soil ecosystem and the household livelihood system to identify entry points to potential social-ecological tipping cascades. As such, the Tipping Multiverse Framework offered two matrices that can advance the understanding of regional SES by identifying core processes, functioning, and links in each TE and thus provide entry points to identify potential tipping cascades across SES sub-systems. The exemplified application of these two frameworks on southwestern Amazonia shows the analytical potential of both proposed frameworks in advancing the understanding of social-ecological tipping points and potential tipping cascades in a regional SES.
On an empirical note, zooming in on questions of governance by applying a political ecology lens to human security, we find that 'glocal' resource governance often reproduces, amplifies, or creates power imbalances and divisions on and between different scales. Our results show that the winners of resource extraction are mostly found at the national and international scale while local communities receive little benefit and are left vulnerable to externalities. Hence, our study contributes to the existing research by stressing the importance of one underlying question: "governance by whom and for whom?" This question raised the demand to understand the underlying dynamics of resource governance and resulting conflicts. Therefore, we aimed at analyzing how (environmental) institutions influence the major drivers of social-ecological conflicts over land in and around three protected areas, Tambopata (Peru), the Extractive Reserve Chico Mendes (Brazil), and Manuripi (Bolivia). We found that state institutions, in particular, have the following effects on key conflict drivers: Overlapping responsibilities of governance institutions and limited enforcement of regulations protecting and empowering rural and disadvantaged populations, enabling external actors to (illegally) access and control resources in the protected areas. Consequently, the already fragile social contract between the residents of the protected area and its surrounding areas and the central state is further weakened by the expanding influence of criminal organizations that oppose the state's authority. For state institutions to avoid aggravating these conflict drivers but instead better manage them or even contribute to conflict prevention and mitigation, a transformation from reactive to reflexive institutions and the development of new reflexive governance competencies is needed.
This need for reflexive governance becomes particularly visible when sudden disturbances or shocks impact the SES. Our analysis of the impacts of the COVID-19 pandemic on the interconnections of land use change, ecosystem services, human agency, conflict, and cooperation that the pandemic has had a severe influence on the human security of marginalized social groups in southwestern Amazonia. Civil society actions have been an essential strategy in the fight against COVID-19, not just in the health sector but also in the economic, political, social, and cultural realms. However, our research also showed that the pandemic has consolidated and partly renewed criminal structures, while the already weak state has fallen further behind due to additional tasks managing the pandemic and other disasters such as floods.
In conclusion, it can be said that the reflexivity of governance is crucial to foster cooperation and preventing conflicts in the realm of social-ecological systems. By not only reacting to already occurring changes but also reflecting upon potential future changes, governance can shape transformation pathways away from the detrimental and towards life-sustaining pathways. It can do so, by exercising agency across scales to avoid the crossing of detrimental social-ecological tipping points but rather to trigger life-sustaining tipping points that contribute to global social-ecological well-being.
Ecotoxicology is the science that researches effects of toxicants on biological entities. Following the famous toxicological principle formulated 1538 by von Hohenheim, known as Paracelsus, thereby generally all chemicals are able to act as toxicants. Unlike human toxicology that focuses on toxic effects on individuals and populations of one species, Homo sapiens, ecotoxicology is not constrained in its scope of biological entities. It is interested in toxic effects on individuals and populations of any species (excluding humans), and on communities and entire ecosystems (Walker et al., 2012; Köhler & Triebskorn, 2013; Newman 2014). One example of where the ecological foundation of ecotoxicology manifests itself are indirect effects, which are effects on biological entities that are not directly caused by chemicals but instead are mediated by ecological interactions and environmental conditions (Walker et al., 2012). With this large scope, ecotoxicology is an inter- and multidisciplinary science that links chemical, biological and environmental knowledge.
With millions of species and at least 100,000 chemicals that potentially interact with them in the environment (Wang et al., 2021), ecotoxicology has a large ground to cover. Among these sheer numbers, there are some groups that are of special importance regarding their potential environmental impact. Pesticides are one group of chemicals that have a large, if not the largest, ecotoxicological relevance: they are toxic for biological entities, sometimes in very low concentrations , and they are used in large amounts and globally (Bernhardt et al., 2017). The high toxicity of pesticides, much higher than that of most other groups of chemicals, is a result of their intended use: they are designed to reduce detrimental effects of, e.g., insects, plants or fungi on agriculture by controlling respective populations, often, and in the sense of their Latin name, through induced lethality (Walker et al., 2012). However, they act not specific enough to be toxic only for the intended species that are considered pests, but also show toxicity towards species living in habitats next to pesticide-treated areas. The widespread agricultural use of pesticides, on the other hand, is a result of their work-and-cost-efficiency for securing yields, but also results in exposure of ecosystems at a global scale (Sharma et al., 2019). In summary, pesticides can be abstractly seen as toxicity intentionally applied to agricultural areas, unintentionally also exposing organisms in non-agricultural areas to toxicity.
The risks of pesticide use for ecosystems have led major jurisdictions, like the United States of America (US) and the European Union (EU), to enact elaborated regulatory processes that require a registration of pesticides prior use (EFSA, 2013; EPA, 2011; Stehle & Schulz, 2015b). A by-product of these registration processes are regulatory threshold levels (RTL) which can be used for scientific risk analysis outside the regulatory process (Stehle & Schulz, 2015a). The RTL for an organism group is basically derived from the most sensitive effect concentrations found in standardized toxicity tests for species representative for the group, multiplied by a safety factor, although specifics differ among regulatory processes. Conceptually, they mark the threshold that separates environmental concentrations associated with acceptable risk (concentrations below the RTL) from concentrations associated with unacceptable risk (concentrations above the RTL).
Due to the high degree of procedural standardization in the derivation of RTLs, they have been found as a good measure to make the toxicities of different pesticides comparable, and they were employed in a series of studies to characterize environmental pesticide concentrations (e.g., Stehle & Schulz, 2015a; Stehle et al., 2018; Wolfram et al., 2018; Wolfram et al., 2021; Schulz et al., 2021, also, in Appendix B; Bub et al., 2023, also, in Appendix C). RTL reflect, for instance, that insecticides show regulatory unacceptable concentrations towards fish between 3 ng/L (deltamethrin, a pyrethroid) and 110 mg/L (imidacloprid, a neonicotinoid), a range of nine orders of magnitude. At the same, imidacloprid is very toxic to pollinators (RTL of 1.52 ng/organism), while more than 95% of all of the insecticides, with regulatory unacceptable concentrations among insecticides ranging as high as 1,6 mg/organism, indicating a toxicity six orders of magnitude lower than that of imidacloprid.
At large-scales, ecotoxicology deals with pesticide impacts on a national (e.g., Bub et al., 2023; Douglas & Tooker, 2015; Hallmann et al., 2014; Schulz et al., 2021; Stehle et al., 2019; Wolfram et al., 2018), continental (Wolfram et al., 2021) or the global scale (Stehle & Schulz, 2015a; Stehle et al., 2018). This maximization of considered scale is in line with the general tendency of ecotoxicology towards larger scales, but generally requires new methodological and conceptual approaches. Historically, individual chemicals and groups of chemicals have been identified that mark, caused by their immense release into the environment, main disruptors of processes in the Earth system, like greenhouses gases for the climate change, chlorofluorocarbons for the depletion of the atmosphere’s ozone layer, dichlorodiphenyl-trichloroethane and other organochlorides for bioaccumulation in food webs and declines in bird populations, etc., but for other phenomena, like declines in biodiversity or numbers of insect species (Outhwaite et al., 2020; Seibold et al., 2019; Vörösmarty et al., 2010), the active part of chemical pollution is only understood to a much lesser extent. There are indicators that pesticides may play a major role
This dissertation contributes to the research of large-scale risks of pesticide use, and of large-scale ecotoxicology in general, in several ways (Figure 1). In Chapter 2, it presents a labeled property graph, the MAGIC graph (Meta-Analysis of the Global Impact of Chemicals graph), as a solution to the methodological issues that arise when increasing amounts of data from more and more sources are combined for analysis (Bub et al., 2019; also, in Appendix A). The MAGIC graph is able to link chemical information from different sources, even if these sources use different nomenclatures. This enables analyses that incorporate toxicological data, like thousands of RTLs (for different organism groups and jurisdictions) for hundreds of pesticides, and information on pesticide use and chemical classes. The MAGIC graph is implemented in a way that allows it to be organically extended by additional chemical, biological and environmental data, and eventually scaled to all chemicals of environmental interest.
Chapter 3 shows, how the combination of the linked pesticide data with a systemic consideration of pesticide use supports the interpretation of pesticide risks in the US (Schulz et al., 2021; also, in Appendix B). This systemic approach includes a new measure, the total applied toxicity (TAT), which integrates used pesticide amounts and pesticide toxicities, and the consideration of pesticide use as a complex system whose state and evolution can be visualized in phase-space plots. The combination of the described methods and concepts led to a novel view on pesticide risks in the US and can provide a framework for future ecotoxicological research at large scales.
Chapter 4 displays the results of the methods and concepts of the US pesticide risk analysis applied to Germany (Bub et al., 2023; also, in Appendix C). A pesticide risk analysis of Germany is of special importance in the context of the EU’s goal to drastically reduce pesticide risks (European Commission, 2020) and Germany being one of the important agricultural producers in the EU. A comparison of the results for Germany to those for the US did also allow to evaluate the impact of scale and differing RTLs, information that can help other ecotoxicological large-scale assessments. Chapter 5 adds a conclusion and an outlook.
In this thesis, we investigate a statistical model for precipitation time series recorded at a single site. The sequence of observations consists of rainfall amounts aggregated over time periods of fixed duration. As the properties of this sequence depend strongly on the length of the observation intervals, we follow the approach of Rodriguez-Iturbe et. al. [1] and use an underlying model for rainfall intensity in continuous time. In this idealized representation, rainfall occurs in clusters of rectangular cells, and each observations is treated as the sum of cell contributions during a given time period. Unlike the previous work, we use a multivariate lognormal distribution for the temporal structure of the cells and clusters. After formulating the model, we develop a Markov-Chain Monte-Carlo algorithm for fitting it to a given data set. A particular problem we have to deal with is the need to estimate the unobserved intensity process alongside the parameter of interest. The performance of the algorithm is tested on artificial data sets generated from the model. [1] I. Rodriguez-Iturbe, D. R. Cox, and Valerie Isham. Some models for rainfall based on stochastic point processes. Proc. R. Soc. Lond. A, 410:269-288, 1987.
The main goal of this work is to model size effects, as they occur in materials with an intrinsic microstructure at the consideration of specimens that are not by orders larger than this microstructure. The micromorphic continuum theory as a generalized continuum theory is well suited to account for the occuring size effects. Thereby additional degrees of freedoms capture the independent deformations of these microstructures, while they provide additional balance equation. In this thesis, the deformational and configurational mechanics of the micromorphic continuum is exploited in a finite-deformation setting. A constitutive and numerical framework is developed, in which also the material-force method is advanced. Furthermore the multiscale modelling of thin material layers with a heterogeneous substructure is of interest. To this end, a computational homogenization framework is developed, which allows to obtain the constitutive relation between traction and separation based on the properties of the underlying micromorphic mesostructure numerically in a nested solution scheme. Within the context of micromorphic continuum mechanics, concepts of both gradient and micromorphic plasticity are developed by systematically varying key ingredients of the respective formulations.
The interest of the exploration of new hydrocarbon fields as well as deep geothermal reservoirs is permanently growing. The analysis of seismic data specific for such exploration projects is very complex and requires the deep knowledge in geology, geophysics, petrology, etc from interpreters, as well as the ability of advanced tools that are able to recover some particular properties. There again the existing wavelet techniques have a huge success in signal processing, data compression, noise reduction, etc. They enable to break complicate functions into many simple pieces at different scales and positions that makes detection and interpretation of local events significantly easier.
In this thesis mathematical methods and tools are presented which are applicable to the seismic data postprocessing in regions with non-smooth boundaries. We provide wavelet techniques that relate to the solutions of the Helmholtz equation. As application we are interested in seismic data analysis. A similar idea to construct wavelet functions from the limit and jump relations of the layer potentials was first suggested by Freeden and his Geomathematics Group.
The particular difficulty in such approaches is the formulation of limit and
jump relations for surfaces used in seismic data processing, i.e., non-smooth
surfaces in various topologies (for example, uniform and
quadratic). The essential idea is to replace the concept of parallel surfaces known for a smooth regular surface by certain appropriate substitutes for non-smooth surfaces.
By using the jump and limit relations formulated for regular surfaces, Helmholtz wavelets can be introduced that recursively approximate functions on surfaces with edges and corners. The exceptional point is that the construction of wavelets allows the efficient implementation in form of
a tree algorithm for the fast numerical computation of functions on the boundary.
In order to demonstrate the
applicability of the Helmholtz FWT, we study a seismic image obtained by the reverse time migration which is based on a finite-difference implementation. In fact, regarding the requirements of such migration algorithms in filtering and denoising the wavelet decomposition is successfully applied to this image for the attenuation of low-frequency
artifacts and noise. Essential feature is the space localization property of
Helmholtz wavelets which numerically enables to discuss the velocity field in
pointwise dependence. Moreover, the multiscale analysis leads us to reveal additional geological information from optical features.
The present thesis describes the development and validation of a viscosity adaption method for the numerical simulation of non-Newtonian fluids on the basis of the Lattice Boltzmann Method (LBM), as well as the development and verification of the related software bundle SAM-Lattice.
By now, Lattice Boltzmann Methods are established as an alternative approach to classical computational fluid dynamics
methods. The LBM has been shown to be an accurate and efficient tool for the numerical simulation of weakly compressible or incompressible fluids. Fields of application reach from turbulent simulations through thermal problems to acoustic calculations among others. The transient nature of the method and the need for a regular grid based, non body conformal discretization makes the LBM ideally suitable for simulations involving complex solids. Such geometries are common, for instance, in the food processing industry, where fluids are mixed by static mixers or agitators. Those fluid flows are often laminar and non-Newtonian.
This work is motivated by the immense practical use of the Lattice Boltzmann Method, which is limited due to stability issues. The stability of the method is mainly influenced by the discretization and the viscosity of the fluid. Thus, simulations of non-Newtonian fluids, whose kinematic viscosity depend on the shear rate, are problematic. Several authors have shown that the LBM is capable of simulating those fluids. However, the vast majority of the simulations in the literature are carried out for simple geometries and/or moderate shear rates, where the LBM is still stable. Special care has to be taken for practical non-Newtonian Lattice Boltzmann simulations in order to keep them stable. A straightforward way is to truncate the modeled viscosity range by numerical stability criteria. This is an effective approach, but from the physical point of view the viscosity bounds are chosen arbitrarily. Moreover, these bounds depend on and vary with the grid and time step size and, therefore, with the simulation Mach number, which is freely chosen at the start of the simulation. Consequently, the modeled viscosity range may not fit to the actual range of the physical problem, because the correct simulation Mach number is unknown a priori. A way around is, to perform precursor simulations on a fixed grid to determine a possible time step size and simulation Mach number, respectively. These precursor simulations can be time consuming and expensive, especially for complex cases and a number of operating points. This makes the LBM unattractive for use in practical simulations of non-Newtonian fluids.
The essential novelty of the method, developed in the course of this thesis, is that the numerically modeled viscosity range is consistently adapted to the actual physically exhibited viscosity range through change of the simulation time step and the simulation Mach number, respectively, while the simulation is running. The algorithm is robust, independent of the Mach number the simulation was started with, and applicable for stationary flows as well as transient flows. The method for the viscosity adaption will be referred to as the "viscosity adaption method (VAM)" and the combination with LBM leads to the "viscosity adaptive LBM (VALBM)".
Besides the introduction of the VALBM, a goal of this thesis is to offer assistance in the spirit of a theory guide to students and assistant researchers concerning the theory of the Lattice Boltzmann Method and its implementation in SAM-Lattice. In Chapter 2, the mathematical foundation of the LBM is given and the route from the BGK approximation of the Boltzmann equation to the Lattice Boltzmann (BGK) equation is delineated in detail.
The derivation is restricted to isothermal flows only. Restrictions of the method, such as low Mach number flows are highlighted and the accuracy of the method is discussed.
SAM-Lattice is a C++ software bundle developed by the author and his colleague Dipl.-Ing. Andreas Schneider. It is a highly automated package for the simulation of isothermal flows of incompressible or weakly compressible fluids in 3D on the basis of the Lattice Boltzmann Method. By the time of writing of this thesis, SAM-Lattice comprises 5 components. The main components are the highly automated lattice generator SamGenerator and the Lattice Boltzmann solver SamSolver. Postprocessing is done with ParaSam, which is our extension of the
open source visualization software ParaView. Additionally, domain decomposition for MPI
parallelism is done by SamDecomposer, which makes use of the graph partitioning library MeTiS. Finally, all mentioned components can be controlled through a user friendly GUI (SamLattice) implemented by the author using QT, including features to visually track output data.
In Chapter 3, some fundamental aspects on the implementation of the main components, including the corresponding flow charts will be discussed. Actual details on the implementation are given in the comprehensive programmers guides to SamGenerator and SamSolver.
In order to ensure the functionality of the implementation of SamSolver, the solver is verified in Chapter 4 for Stokes's First Problem, the suddenly accelerated plate, and for Stokes's Second Problem, the oscillating plate, both for Newtonian fluids. Non-Newtonian fluids are modeled in SamSolver with the power-law model according to Ostwald de Waele. The implementation for non-Newtonian fluids is verified for the Hagen-Poiseuille channel flow in conjunction with a convergence analysis of the method. At the same time, the local grid refinement as it is implemented in SamSolver, is verified. Finally, the verification of higher order boundary conditions is done for the 3D Hagen-Poiseuille pipe flow for both Newtonian and non-Newtonian fluids.
In Chapter 5, the theory of the viscosity adaption method is introduced. For the adaption process, a target collision frequency or target simulation Mach number must be chosen and the distributions must be rescaled according to the modified time step size. A convenient choice is one of the stability bounds. The time step size for the adaption step is deduced from the target collision frequency \(\Omega_t\) and the currently minimal or maximal shear rate in the system, while obeying auxiliary conditions for the simulation Mach number. The adaption is done in the collision step of the Lattice Boltzmann algorithm. We use the transformation matrices of the MRT model to map from distribution space to moment space and vice versa. The actual scaling of the distributions is conducted on the back mapping, because we use the transformation matrix on the basis of the new adaption time step size. It follows an additional rescaling of the non-equilibrium part of the distributions, because of the form of the definition for the discrete stress tensor in the LBM context. For that reason it is clear, that the VAM is applicable for the SRT model as well as the MRT model, where there is virtually no extra cost in the latter case. Also, in Chapter 5, the multi level treatment will be discussed.
Depending on the target collision frequency and the target Mach number, the VAM can be used to optimally use the viscosity range that can be modeled within the stability bounds or it can be used to drastically accelerate the simulation. This is shown in Chapter 6. The viscosity adaptive LBM is verified in the stationary case for the Hagen-Poiseuille channel flow and in the transient case for the Wormersley flow, i.e., the pulsatile 3D Hagen-Poiseuille pipe flow. Although, the VAM is used here for fluids that can be modeled with the power-law approach, the implementation of the VALBM is straightforward for other non-Newtonian models, e.g., the Carreau-Yasuda or Cross model. In the same chapter, the VALBM is validated for the case of a propeller viscosimeter developed at the chair SAM. To this end, the experimental data of the torque on the impeller of three shear thinning non-Newtonian liquids serve for the validation. The VALBM shows excellent agreement with experimental data for all of the investigated fluids and in every operating point. For reasons of comparison, a series of standard LBM simulations is carried out with different simulation Mach numbers, which partly show errors of several hundred percent. Moreover, in Chapter 7, a sensitivity analysis on the parameters used within the VAM is conducted for the simulation of the propeller viscosimeter.
Finally, the accuracy of non-Newtonian Lattice Boltzmann simulations with the SRT and the MRT model is analyzed in detail. Previous work for Newtonian fluids indicate that depending on the numerical value of the collision frequency \(\Omega\), additional artificial viscosity is introduced due to the finite difference scheme, which negatively influences the accuracy. For the non-Newtonian case, an error estimate in the form of a functional is derived on the basis of a series expansion of the Lattice Boltzmann equation. This functional can be solved analytically for the case of the Hagen-Poiseuille channel flow of non-Newtonian fluids. The estimation of the error minimum is excellent in regions where the \(\Omega\) error is the dominant source of error as opposed to the compressibility error.
Result of this dissertation is a verified and validated software bundle on the basis of the viscosity adaptive Lattice Boltzmann Method. The work restricts itself on the simulation of isothermal, laminar flows with small Mach numbers. As further research goals, the testing of the VALBM with minimal error estimate and the investigation of the VALBM in the case of turbulent flows is suggested.
The subject of this thesis is the probabilistic reliability assessment of notched metallic components under periodic constant-amplitude loads with respect to the failure mode of high-cycle fatigue. The latter refers to the crack initiation within the considered component caused by a high number, typically millions, of load cycles characterized by their small magnitude in terms of the material's static strength. In order to estimate the probability of failure due to high-cycle fatigue for a specified component under given loads, a new empirical model based on weakest-link theory is developed which describes a probabilistic and component specific constant-life diagram with respect to the anticipated design life. A conventional, non-probabilistic constant-life diagram reflects a discrete design boundary in terms of mean stress and stress amplitude, typically based on test results with respect to unnotched coupons made from the material of interest. Its application to the design of a notched component is established by identifying the stress conditions at the component's hot spot with those acting in the smooth coupons during the tests, and comparing those hot-spot conditions with the design boundary described in the constant-life diagram. Disregarded influences, such as notch and statistical size effect have to be incorporated by respective correction factors. The proposed probabilistic model on the other hand describes a continuous field of failure probabilities in the design stress plane, taking into account not only the hot-spot stresses, but the entire cyclic stress field acting throughout the component. In this way, the methodology directly accounts for notch and statistical size effects. Responsible for providing this greater scope is the weakest-link concept, which represents a non-local stochastic approach for quantifying the failure probability of loaded solids. The four model parameters can be calibrated with fatigue test data sets containing entirely unrelated test results on arbitrary specimen geometries, obliterating the constraining need for test data following staircase or probit schemes. This work contains the formulation, analysis, validation and application of the proposed model. After its introduction and a comparison with existing methods, it is analyzed in terms of its numerical properties when applied to finite element models, its efficient calibration and the corresponding model uncertainty. The validation is split into two parts. In a first analysis, the model is fitted to test data, containing results on several types of notched specimens, reflecting predominantly elastic material behavior. In a second step, this restriction is lifted and the model is used in order to predict the failure behavior of notched test specimens experiencing notch root plasticity due to high mean stresses. In both validation studies, the derived model predictions are, for the most part, well in line with the experimentally observed failure behavior of the test specimens. Finally, the applicability of the proposed probabilistic methodology in a design context is demonstrated on the example of a gas turbine compressor blade and the corresponding compressor stage.
This dissertation is intended to transport the theory of Serre functors into the context of A-infinity-categories. We begin with an introduction to multicategories and closed multicategories, which form a framework in which the theory of A-infinity-categories is developed. We prove that (unital) A-infinity-categories constitute a closed symmetric multicategory. We define the notion of A-infinity-bimodule similarly to Tradler and show that it is equivalent to an A-infinity-functor of two arguments which takes values in the differential graded category of complexes of k-modules, where k is a commutative ground ring. Serre A-infinity-functors are defined via A-infinity-bimodules following ideas of Kontsevich and Soibelman. We prove that a unital closed under shifts A-infinity-category over a field admits a Serre A-infinity-functor if and only if its homotopy category admits an ordinary Serre functor. The proof uses categories and Serre functors enriched in the homotopy category of complexes of k-modules. Another important ingredient is an A-infinity-version of the Yoneda Lemma.
Computer-based simulation and visualization of acoustics of a virtual scene can aid during the design process of concert halls, lecture rooms, theaters, or living rooms. Because, not only the visual aspect of the room is important, but also its acoustics. In factory floors noise reduction is important since noise is hazardous to health. Despite the obvious dissimilarity between our aural and visual senses, many techniques required for the visualization of photo-realistic images and for the auralization of acoustic environments are quite similar. Both applications can be served by geometric methods such as particle- and ray tracing if we neglect a number of less important effects. By means of the simulation of room acoustics we want to predict the acoustic properties of a virtual model. For auralization, a pulse response filter needs to be assembled for each pair of source and listener positions. The convolution of this filter with an anechoic source signal provides the signal received at the listener position. Hence, the pulse response filter must contain all reverberations (echos) of a unit pulse, including their frequency decompositions due to absorption at different surface materials. For the room acoustic simulation a method named phonon tracing, since it is based on particles, is developed. The approach computes the energy or pressure decomposition for each particle (phonon) sent out from a sound source and uses this in a second pass (phonon collection) to construct the response filters for different listeners. This step can be performed in different precision levels. During the tracing step particle paths and additional information are stored in a so called phonon map. Using this map several sound visualization approaches were developed. From the visualization, the effect of different materials on the spectral energy / pressure distribution can be observed. The first few reflections already show whether certain frequency bands are rapidly absorbed. The absorbing materials can be identified and replaced in the virtual model, improving the overall acoustic quality of the simulated room. Furthermore an insight into the pressure / energy received at the listener position is possible. The phonon tracing algorithm as well as several sound visualization approaches are integrated into a common system utilizing Virtual Reality technologies in order to facilitate the immersion into the virtual scene. The system is a prototype developed within a project at the University of Kaiserslautern and is still a subject of further improvements. It consists of a stereoscopic back-projection system for visual rendering as well as professional audio equipment for auralization purposes.
Three dimensional (3d) point data is used in industry for measurement and reverse engineering. Precise point data is usually acquired with triangulating laser scanners or high precision structured light scanners. Lower precision point data is acquired by real-time structured light devices or by stereo matching with multiple cameras. The basic principle of all these methods is the so-called triangulation of 3d coordinates from two dimensional (2d) camera images.
This dissertation contributes a method for multi-camera stereo matching that uses a system of four synchronized cameras. A GPU based stereo matching method is presented to achieve a high quality reconstruction at interactive frame rates. Good depth resolution is achieved by allowing large disparities between the images. A multi level approach on the GPU allows a fast processing of these large disparities. In reverse engineering, hand-held laser scanners are used for the scanning of complex shaped objects. The operator of the scanner can scan complex regions slower, multiple times, or from multiple angles to achieve a higher point density. Traditionally, computer aided design (CAD) geometry is reconstructed in a separate step after the scanning. Errors or missing parts in the scan prevent a successful reconstruction. The contribution of this dissertation is an on-line algorithm that allows the reconstruction during the scanning of an object. Scanned points are added to the reconstruction and improve it on-line. The operator can detect the areas in the scan where the reconstruction needs additional data.
First, the point data is thinned out using an octree based data structure. Local normals and principal curvatures are estimated for the reduced set of points. These local geometric values are used for segmentation using a region growing approach. Implicit quadrics are fitted to these segments. The canonical form of the quadrics provides the parameters of basic geometric primitives.
An improved approach uses so called accumulated means of local geometric properties to perform segmentation and primitive reconstruction in a single step. Local geometric values can be added and removed on-line to these means to get a stable estimate over a complete segment. By estimating the shape of the segment it is decided which local areas are added to a segment. An accumulated score estimates the probability for a segment to belong to a certain type of geometric primitive. A boundary around the segment is reconstructed using a growing algorithm that ensures that the boundary is closed and avoids self intersections.
At present the standardization of third generation (3G) mobile radio systems is the subject of worldwide research activities. These systems will cope with the market demand for high data rate services and the system requirement for exibility concerning the offered services and the transmission qualities. However, there will be de ciencies with respect to high capacity, if 3G mobile radio systems exclusively use single antennas. Very promising technique developed for increasing the capacity of 3G mobile radio systems the application is adaptive antennas. In this thesis, the benefits of using adaptive antennas are investigated for 3G mobile radio systems based on Time Division CDMA (TD-CDMA), which forms part of the European 3G mobile radio air interface standard adopted by the ETSI, and is intensively studied within the standardization activities towards a worldwide 3G air interface standard directed by the 3GPP (3rd Generation Partnership Project). One of the most important issues related to adaptive antennas is the analysis of the benefits of using adaptive antennas compared to single antennas. In this thesis, these bene ts are explained theoretically and illustrated by computer simulation results for both data detection, which is performed according to the joint detection principle, and channel estimation, which is applied according to the Steiner estimator, in the TD-CDMA uplink. The theoretical explanations are based on well-known solved mathematical problems. The simulation results illustrating the benefits of adaptive antennas are produced by employing a novel simulation concept, which offers a considerable reduction of the simulation time and complexity, as well as increased exibility concerning the use of different system parameters, compared to the existing simulation concepts for TD-CDMA. Furthermore, three novel techniques are presented which can be used in systems with adaptive antennas for additionally improving the system performance compared to single antennas. These techniques concern the problems of code-channel mismatch, of user separation in the spatial domain, and of intercell interference, which, as it is shown in the thesis, play a critical role on the performance of TD-CDMA with adaptive antennas. Finally, a novel approach for illustrating the performance differences between the uplink and downlink of TD-CDMA based mobile radio systems in a straightforward manner is presented. Since a cellular mobile radio system with adaptive antennas is considered, the ultimate goal is the investigation of the overall system efficiency rather than the efficiency of a single link. In this thesis, the efficiency of TD-CDMA is evaluated through its spectrum efficiency and capacity, which are two closely related performance measures for cellular mobile radio systems. Compared to the use of single antennas, the use of adaptive antennas allows impressive improvements of both spectrum efficiency and capacity. Depending on the mobile radio channel model and the user velocity, improvement factors range from six to 10.7 for the spectrum efficiency, and from 6.7 to 12.6 for the spectrum capacity of TD-CDMA. Thus, adaptive antennas constitute a promising technique for capacity increase of future mobile communications systems.
Adaptive Extraction and Representation of Geometric Structures from Unorganized 3D Point Sets
(2009)
The primary emphasis of this thesis concerns the extraction and representation of intrinsic properties of three-dimensional (3D) unorganized point clouds. The points establishing a point cloud as it mainly emerges from LiDaR (Light Detection and Ranging) scan devices or by reconstruction from two-dimensional (2D) image series represent discrete samples of real world objects. Depending on the type of scenery the data is generated from the resulting point cloud may exhibit a variety of different structures. Especially, in the case of environmental LiDaR scans the complexity of the corresponding point clouds is relatively high. Hence, finding new techniques allowing the efficient extraction and representation of the underlying structural entities becomes an important research issue of recent interest. This thesis introduces new methods regarding the extraction and visualization of structural features like surfaces and curves (e.g. ridge-lines, creases) from 3D (environmental) point clouds. One main part concerns the extraction of curve-like features from environmental point data sets. It provides a new method supporting a stable feature extraction by incorporating a probability-based point classification scheme that characterizes individual points regarding their affiliation to surface-, curve- and volume-like structures. Another part is concerned with the surface reconstruction from (environmental) point clouds exhibiting objects that are more or less complex. A new method providing multi-resolutional surface representations from regular point clouds is discussed. Following the applied principles of this approach a volumetric surface reconstruction method based on the proposed classification scheme is introduced. It allows the reconstruction of surfaces from highly unstructured and noisy point data sets. Furthermore, contributions in the field of reconstructing 3D point clouds from 2D image series are provided. In addition, a discussion concerning the most important properties of (environmental) point clouds with respect to feature extraction is presented.
Real-time systems are systems that have to react correctly to stimuli from the environment within given timing constraints.
Today, real-time systems are employed everywhere in industry, not only in safety-critical systems but also in, e.g., communication, entertainment, and multimedia systems.
With the advent of multicore platforms, new challenges on the efficient exploitation of real-time systems have arisen:
First, there is the need for effective scheduling algorithms that feature low overheads to improve the use of the computational resources of real-time systems.
The goal of these algorithms is to ensure timely execution of tasks, i.e., to provide runtime guarantees.
Additionally, many systems require their scheduling algorithm to flexibly react to unforeseen events.
Second, the inherent parallelism of multicore systems leads to contention for shared hardware resources and complicates system analysis.
At any time, multiple applications run with varying resource requirements and compete for the scarce resources of the system.
As a result, there is a need for an adaptive resource management.
Achieving and implementing an effective and efficient resource management is a challenging task.
The main goal of resource management is to guarantee a minimum resource availability to real-time applications.
A further goal is to fulfill global optimization objectives, e.g., maximization of the global system performance, or the user perceived quality of service.
In this thesis, we derive methods based on the slot shifting algorithm.
Slot shifting provides flexible scheduling of time-constrained applications and can react to unforeseen events in time-triggered systems.
For this reason, we aim at designing slot shifting based algorithms targeted for multicore systems to tackle the aforementioned challenges.
The main contribution of this thesis is to present two global slot shifting algorithms targeted for multicore systems.
Additionally, we extend slot shifting algorithms to improve their runtime behavior, or to handle non-preemptive firm aperiodic tasks.
In a variety of experiments, the effectiveness and efficiency of the algorithms are evaluated and confirmed.
Finally, the thesis presents an implementation of a slot-shifting-based logic into a resource management framework for multicore systems.
Thus, the thesis closes the circle and successfully bridges the gap between real-time scheduling theory and real-world implementations.
We prove applicability of the slot shifting algorithm to effectively and efficiently perform adaptive resource management on multicore systems.
Adjoint-Based Shape Optimization and Optimal Control with Applications to Microchannel Systems
(2021)
Optimization problems constrained by partial differential equations (PDEs) play an important role in many areas of science and engineering. They often arise in the optimization of technological applications, where the underlying physical effects are modeled by PDEs. This thesis investigates such problems in the context of shape optimization and optimal control with microchannel systems as novel applications. Such systems are used, e.g., as cooling systems, heat exchangers, or chemical reactors as their high surface-to-volume ratio, which results in beneficial heat and mass transfer characteristics, allows them to excel in these settings. Additionally, this thesis considers general PDE constrained optimization problems with particular regard to their efficient solution.
As our first application, we study a shape optimization problem for a microchannel cooling system: We rigorously analyze this problem, prove its shape differentiability, and calculate the corresponding shape derivative. Afterwards, we consider the numerical optimization of the cooling system for which we employ a hierarchy of reduced models derived via porous medium modeling and a dimension reduction technique. A comparison of the models in this context shows that the reduced models approximate the original one very accurately while requiring substantially less computational resources.
Our second application is the optimization of a chemical microchannel reactor for the Sabatier process using techniques from PDE constrained optimal control. To treat this problem, we introduce two models for the reactor and solve a parameter identification problem to determine the necessary kinetic reaction parameters for our models. Thereafter, we consider the optimization of the reactor's operating conditions with the objective of improving its product yield, which shows considerable potential for enhancing the design of the reactor.
To provide efficient solution techniques for general shape optimization problems, we introduce novel nonlinear conjugate gradient methods for PDE constrained shape optimization and analyze their performance on several well-established benchmark problems. Our results show that the proposed methods perform very well, making them efficient and appealing gradient-based shape optimization algorithms.
Finally, we continue recent software-based developments for PDE constrained optimization and present our novel open-source software package cashocs. Our software implements and automates the adjoint approach and, thus, facilitates the solution of general PDE constrained shape optimization and optimal control problems. Particularly, we highlight our software's user-friendly interface, straightforward applicability, and mesh independent behavior.
In recent years the field of polymer tribology experienced a tremendous development
leading to an increased demand for highly sophisticated in-situ measurement methods.
Therefore, advanced measurement techniques were developed and established
in this study. Innovative approaches based on dynamic thermocouple, resistive electrical
conductivity, and confocal distance measurement methods were developed in
order to in-situ characterize both the temperature at sliding interfaces and real contact
area, and furthermore the thickness of transfer films. Although dynamic thermocouple
and real contact area measurement techniques were already used in similar
applications for metallic sliding pairs, comprehensive modifications were necessary to
meet the specific demands and characteristics of polymers and composites since
they have significantly different thermal conductivities and contact kinematics. By using
tribologically optimized PEEK compounds as reference a new measurement and
calculation model for the dynamic thermocouple method was set up. This method
allows the determination of hot spot temperatures for PEEK compounds, and it was
found that they can reach up to 1000 °C in case of short carbon fibers present in the
polymer. With regard to the non-isotropic characteristics of the polymer compound,
the contact situation between short carbon fibers and steel counterbody could be
successfully monitored by applying a resistive measurement method for the real contact
area determination. Temperature compensation approaches were investigated
for the transfer film layer thickness determination, resulting in in-situ measurements
with a resolution of ~0.1 μm. In addition to a successful implementation of the measurement
systems, failure mechanism processes were clarified for the PEEK compound
used. For the first time in polymer tribology the behavior of the most interesting
system parameters could be monitored simultaneously under increasing load
conditions. It showed an increasing friction coefficient, wear rate, transfer film layer
thickness, and specimen overall temperature when frictional energy exceeded the
thermal transport capabilities of the specimen. In contrast, the real contact area between
short carbon fibers and steel decreased due to the separation effect caused by
the transfer film layer. Since the sliding contact was more and more matrix dominated,
the hot spot temperatures on the fibers dropped, too. The results of this failure
mechanism investigation already demonstrate the opportunities which the new
measurement techniques provide for a deeper understanding of tribological processes,
enabling improvements in material composition and application design.
If gradient based derivative algorithms are used to improve industrial products by reducing their target functions, the derivatives need to be exact.
The last percent of possible improvement, like the efficiency of a turbine, can only be gained if the derivatives are consistent with the solution process that is used in the simulation software.
It is problematic that the development of the simulation software is an ongoing process which leads to the use of approximated derivatives.
If a derivative computation is implemented manually, it will be inconsistent after some time if it is not updated.
This thesis presents a generalized approach which differentiates the whole simulation software with Algorithmic Differentiation (AD), and guarantees a correct and consistent derivative computation after each change to the software.
For this purpose, the variable tagging technique is developed.
The technique checks at run-time if all dependencies, which are used by the derivative algorithms, are correct.
Since it is also necessary to check the correctness of the implementation, a theorem is developed which describes how AD derivatives can be compared.
This theorem is used to develop further methods that can detect and correct errors.
All methods are designed such that they can be applied in real world applications and are used within industrial configurations.
The process described above yields consistent and correct derivatives but the efficiency can still be improved.
This is done by deriving new derivative algorithms.
A fixed-point iterator approach, with a consistent derivation, yields all state of the art algorithms and produces two new algorithms.
These two new algorithms include all implementation details and therefore they produce consistent derivative results.
For detecting hot spots in the application, the state of the art techniques are presented and extended.
The data management is changed such that the performance of the software is affected only marginally when quantities, like the number of input and output variables or the memory consumption, are computed for the detection.
The hot spots can be treated with techniques like checkpointing or preaccumulation.
How these techniques change the time and memory consumption is analyzed and it is shown how they need to be used in selected AD tools.
As a last step, the used AD tools are analyzed in more detail.
The major implementation strategies for operator overloading AD tools are presented and implementation improvements for existing AD tools are discussed.
The discussion focuses on a minimal memory consumption and makes it possible to compare AD tools on a theoretical level.
The new AD tool CoDiPack is based on these findings and its design and concepts are presented.
The improvements and findings in this thesis make it possible, that an automatic, consistent and correct derivative is generated in an efficient way for industrial applications.
Automated theorem proving is a search problem and, by its undecidability, a very difficult one. The challenge in the development of a practically successful prover is the mapping of the extensively developed theory into a program that runs efficiently on a computer. Starting from a level-based system model for automated theorem provers, in this work we present different techniques that are important for the development of powerful equational theorem provers. The contributions can be divided into three areas: Architecture. We present a novel prover architecture that is based on a set-based compression scheme. With moderate additional computational costs we achieve a substantial reduction of the memory requirements. Further wins are architectural clarity, the easy provision of proof objects, and a new way to parallelize a prover which shows respectable speed-ups in practice. The compact representation paves the way to new applications of automated equational provers in the area of verification systems. Algorithms. To improve the speed of a prover we need efficient solutions for the most time-consuming sub-tasks. We demonstrate improvements of several orders of magnitude for two of the most widely used term orderings, LPO and KBO. Other important contributions are a novel generic unsatisfiability test for ordering constraints and, based on that, a sufficient ground reducibility criterion with an excellent cost-benefit ratio. Redundancy avoidance. The notion of redundancy is of central importance to justify simplifying inferences which are used to prune the search space. In our experience with unfailing completion, the usual notion of redundancy is not strong enough. In the presence of associativity and commutativity, the provers often get stuck enumerating equations that are permutations of each other. By extending and refining the proof ordering, many more equations can be shown redundant. Furthermore, our refinement of the unfailing completion approach allows us to use redundant equations for simplification without the need to consider them for generating inferences. We describe the efficient implementation of several redundancy criteria and experimentally investigate their influence on the proof search. The combination of these techniques results in a considerable improvement of the practical performance of a prover, which we demonstrate with extensive experiments for the automated theorem prover Waldmeister. The progress achieved allows the prover to solve problems that were previously out of reach. This considerably enhances the potential of the prover and opens up the way for new applications.
Stochastic Network Calculus (SNC) emerged from two branches in the late 90s:
the theory of effective bandwidths and its predecessor the Deterministic Network
Calculus (DNC). As such SNC’s goal is to analyze queueing networks and support
their design and control.
In contrast to queueing theory, which strives for similar goals, SNC uses in-
equalities to circumvent complex situations, such as stochastic dependencies or
non-Poisson arrivals. Leaving the objective to compute exact distributions behind,
SNC derives stochastic performance bounds. Such a bound would, for example,
guarantee a system’s maximal queue length that is violated by a known small prob-
ability only.
This work includes several contributions towards the theory of SNC. They are
sorted into four main contributions:
(1) The first chapters give a self-contained introduction to deterministic net-
work calculus and its two branches of stochastic extensions. The focus lies on the
notion of network operations. They allow to derive the performance bounds and
simplifying complex scenarios.
(2) The author created the first open-source tool to automate the steps of cal-
culating and optimizing MGF-based performance bounds. The tool automatically
calculates end-to-end performance bounds, via a symbolic approach. In a second
step, this solution is numerically optimized. A modular design allows the user to
implement their own functions, like traffic models or analysis methods.
(3) The problem of the initial modeling step is addressed with the development
of a statistical network calculus. In many applications the properties of included
elements are mostly unknown. To that end, assumptions about the underlying
processes are made and backed by measurement-based statistical methods. This
thesis presents a way to integrate possible modeling errors into the bounds of SNC.
As a byproduct a dynamic view on the system is obtained that allows SNC to adapt
to non-stationarities.
(4) Probabilistic bounds are fundamentally different from deterministic bounds:
While deterministic bounds hold for all times of the analyzed system, this is not
true for probabilistic bounds. Stochastic bounds, although still valid for every time
t, only hold for one time instance at once. Sample path bounds are only achieved by
using Boole’s inequality. This thesis presents an alternative method, by adapting
the theory of extreme values.
(5) A long standing problem of SNC is the construction of stochastic bounds
for a window flow controller. The corresponding problem for DNC had been solved
over a decade ago, but remained an open problem for SNC. This thesis presents
two methods for a successful application of SNC to the window flow controller.
Toxicology, the study of the adverse effects of chemicals and physical agents on living organisms, is a critical process in chemical and drug development. The low throughput, high costs, limited predictivity and ethical concerns related to traditional animal-based toxicity studies render them impractical to assess the growing number and complexity of both existing and new compounds and their formulations. These factors together with the increasing implementation of more demanding regulations, evidence the current need to develop innovative, reliable, cost effective and high throughput toxicological methods.
The use of metabolomics in vitro presents the powerful combination of a human relevant system with a multiparametric approach that allows assessing multiple endpoints in a single biological sample. Applying metabolomics in a cell-based system offers an alternative to both, the ethical concerns and relevance of animal testing and the restraining nature of single endpoint evaluations characteristic of conventional toxicological in vitro assays. However, there are still challenges that hamper the expansion of metabolomics beyond a research tool to a feasible and implementable technology for toxicology assessment.
The aim of this dissertation is to advance the applications of in vitro metabolomics in toxicology by addressing three major challenges that have limited its widespread implementation in the field. In chapter 2 the restrictive high cost and low throughput of in vitro metabolomics was addressed through the development, standardization and proof of concept of a high throughput targeted LC-MS/MS in vitro metabolomics platform for the characterization of hepatotoxicity. In chapter 3, the use of the developed in vitro metabolomics system was expanded beyond hazard identification, to its implementation for deriving dose- and time response metrics that were shown useful for Point of departure (PoD) estimations for human risk assessment. Finally, in chapter 4 in order to increase the reliance and confidence of using in vitro metabolomics data for risk assessment, the human relevance of the metabolomics in vitro assays was attempted to be improved by the implementation and evaluation of in vitro metabolomics in a hiPSCs-derived 3D liver organoid system.
The work developed here demonstrates the suitable of in vitro metabolomics for mechanistic-based hazard identification and risk assessment. By advancing the applications of metabolomics in toxicology, this work has significantly contributed to the aim of toxicology of the 21st century for a human-relevant non-animal toxicological testing, supporting the toxicology task of protecting human health and the environment.
The recently established technologies in the areas of distributed measurement and intelligent
information processing systems, e.g., Cyber Physical Systems (CPS), Ambient
Intelligence/Ambient Assisted Living systems (AmI/AAL), the Internet of Things
(IoT), and Industry 4.0 have increased the demand for the development of intelligent
integrated multi-sensory systems as to serve rapid growing markets [1, 2]. These increase
the significance of complex measurement systems, that incorporate numerous advanced
methodological implementations including electronics circuit, signal processing,
and multi-sensory information fusion. In particular, in multi-sensory cognition applications,
to design such systems, the skill-required tasks, e.g., method selection, parameterization,
model analysis, and processing chain construction are elaborated with immense
effort, which conventionally are done manually by the expert designer. Moreover, the
strong technological competition imposes even more complicated design problems with
multiple constraints, e.g., cost, speed, power consumption,
exibility, and reliability.
Thus, the conventional human expert based design approach may not be able to cope
with the increasing demand in numbers, complexity, and diversity. To alleviate the issue,
the design automation approach has been the topic for numerous research works [3-14]
and has been commercialized to several products [15-18]. Additionally, the dynamic
adaptation of intelligent multi-sensor systems is the potential solution for developing
dependable and robust systems. Intrinsic evolution approach and self-x properties [19],
which include self-monitoring, -calibrating/trimming, and -healing/repairing, are among
the best candidates for the issue. Motivated from the ongoing research trends and based
on the background of our research work [12, 13] among the pioneers in this topic, the
research work of the thesis contributes to the design automation of intelligent integrated
multi-sensor systems.
In this research work, the Design Automation for Intelligent COgnitive system with self-
X properties, the DAICOX, architecture is presented with the aim of tackling the design
effort and to providing high quality and robust solutions for multi-sensor intelligent
systems. Therefore, the DAICOX architecture is conceived with the defined goals as
listed below.
Perform front to back complete processing chain design with automated method
selection and parameterization,
Provide a rich choice of pattern recognition methods to the design method pool,
Associate design information via interactive user interface and visualization along
with intuitive visual programming,
Deliver high quality solutions outperforming conventional approaches by using
multi-objective optimization,
Gain the adaptability, reliability and robustness of designed solutions with self-x
properties,
Derived from the goals, several scientific methodological developments and implementations,
particularly in the areas of pattern recognition and computational intelligence,
will be pursued as part of the DAICOX architecture in the research work of this thesis.
The method pool is aimed to contain a rich choice of methods and algorithms covering
data acquisition and sensor configuration, signal processing and feature computation,
dimensionality reduction, and classification. These methods will be selected and parameterized
automatically by the DAICOX design optimization to construct a multi-sensory
cognition processing chain. A collection of non-parametric feature quality assessment
functions for the purpose of Dimensionality Reduction (DR) process will be presented.
In addition, to standard DR methods, the variations of feature selection method, in
particular, feature weighting will be proposed. Three different classification categories
shall be incorporated in the method pool. Hierarchical classification approach will be
proposed and developed to serve as a multi-sensor fusion architecture at the decision
level. Beside multi-class classification, one-class classification methods, e.g., One-Class
SVM and NOVCLASS will be presented to extend functionality of the solutions, in particular,
anomaly and novelty detection. DAICOX is conceived to effectively handle the
problem of method selection and parameter setting for a particular application yielding
high performance solutions. The processing chain construction tasks will be carried
out by meta-heuristic optimization methods, e.g., Genetic Algorithms (GA) and Particle
Swarm Optimization (PSO), with multi-objective optimization approach and model
analysis for robust solutions. In addition, to the automated system design mechanisms,
DAICOX will facilitate the design tasks with intuitive visual programming and various
options of visualization. Design database concept of DAICOX is aimed to allow the
reusability and extensibility of the designed solutions gained from previous knowledge.
Thus, the cooperative design of machine and knowledge from the design expert can also
be utilized for obtaining fully enhanced solutions. In particular, the integration of self-x
properties as well as intrinsic optimization into the system is proposed to gain enduring
reliability and robustness. Hence, DAICOX will allow the inclusion of dynamically
reconfigurable hardware instances to the designed solutions in order to realize intrinsic
optimization and self-x properties.
As a result from the research work in this thesis, a comprehensive intelligent multisensor
system design architecture with automated method selection, parameterization,
and model analysis is developed with compliance to open-source multi-platform software.It is integrated with an intuitive design environment, which includes visual programming
concept and design information visualizations. Thus, the design effort is minimized as
investigated in three case studies of different application background, e.g., food analysis
(LoX), driving assistance (DeCaDrive), and magnetic localization. Moreover, DAICOX
achieved better quality of the solutions compared to the manual approach in all cases,
where the classification rate was increased by 5.4%, 0.06%, and 11.4% in the LoX,
DeCaDrive, and magnetic localization case, respectively. The design time was reduced
by 81.87% compared to the conventional approach by using DAICOX in the LoX case
study. At the current state of development, a number of novel contributions of the thesis
are outlined below.
Automated processing chain construction and parameterization for the design of
signal processing and feature computation.
Novel dimensionality reduction methods, e.g., GA and PSO based feature selection
and feature weighting with multi-objective feature quality assessment.
A modification of non-parametric compactness measure for feature space quality
assessment.
Decision level sensor fusion architecture based on proposed hierarchical classification
approach using, i.e., H-SVM.
A collection of one-class classification methods and a novel variation, i.e.,
NOVCLASS-R.
Automated design toolboxes supporting front to back design with automated
model selection and information visualization.
In this research work, due to the complexity of the task, neither all of the identified goals
have been comprehensively reached yet nor has the complete architecture definition been
fully implemented. Based on the currently implemented tools and frameworks, ongoing
development of DAICOX is pursuing towards the complete architecture. The potential
future improvements are the extension of method pool with a richer choice of methods
and algorithms, processing chain breeding via graph based evolution approach, incorporation
of intrinsic optimization, and the integration of self-x properties. According to
these features, DAICOX will improve its aptness in designing advanced systems to serve
the increasingly growing technologies of distributed intelligent measurement systems, in
particular, CPS and Industrie 4.0.
Advantage of Filtering for Portfolio Optimization in Financial Markets with Partial Information
(2016)
In a financial market we consider three types of investors trading with a finite
time horizon with access to a bank account as well as multliple stocks: the
fully informed investor, the partially informed investor whose only source of
information are the stock prices and an investor who does not use this infor-
mation. The drift is modeled either as following linear Gaussian dynamics
or as being a continuous time Markov chain with finite state space. The
optimization problem is to maximize expected utility of terminal wealth.
The case of partial information is based on the use of filtering techniques.
Conditions to ensure boundedness of the expected value of the filters are
developed, in the Markov case also for positivity. For the Markov modulated
drift, boundedness of the expected value of the filter relates strongly to port-
folio optimization: effects are studied and quantified. The derivation of an
equivalent, less dimensional market is presented next. It is a type of Mutual
Fund Theorem that is shown here.
Gains and losses eminating from the use of filtering are then discussed in
detail for different market parameters: For infrequent trading we find that
both filters need to comply with the boundedness conditions to be an advan-
tage for the investor. Losses are minimal in case the filters are advantageous.
At an increasing number of stocks, again boundedness conditions need to be
met. Losses in this case depend strongly on the added stocks. The relation
of boundedness and portfolio optimization in the Markov model leads here to
increasing losses for the investor if the boundedness condition is to hold for
all numbers of stocks. In the Markov case, the losses for different numbers
of states are negligible in case more states are assumed then were originally
present. Assuming less states leads to high losses. Again for the Markov
model, a simplification of the complex optimal trading strategy for power
utility in the partial information setting is shown to cause only minor losses.
If the market parameters are such that shortselling and borrowing constraints
are in effect, these constraints may lead to big losses depending on how much
effect the constraints have. They can though also be an advantage for the
investor in case the expected value of the filters does not meet the conditions
for boundedness.
All results are implemented and illustrated with the corresponding numerical
findings.
The present thesis describes the experimental performance determination and numerical
modeling of an aerostatic porous bearing made of an orthotropically layered ceramic
composite material (CMC). The high temperature resistance, low thermal expansion and
high reusability of this material makes it eminently suitable for use in highly stressed
fluid-film bearing applications.
The work involves the development of an aerostatic journal bearing made of porous,
orthotropically layered carbon fiber-reinforced carbon composite (C/C) and the design
of a journal bearing test rig, which contained additional aerostatic support bearings and
six optical laser triangulation sensors. The sensor system enabled the measurement of
lubricant film thickness and shaft misalignment. As a result of the slight air lubrication
clearance of 30 μm, the focus was on low concentricity and the determination of shaft
misalignments.
The preliminary tests included the determination of the permeability of the porous material
and the applicability of Darcy’s law. A scan of the inner surface of the porous bushing
revealed a characteristic grooved structure, which can be attributed to the layered structure
of the material. Bearing tests were conducted up to a rotational speed of 8000 rpm and a
pressure ratio of 5 to 7. No significant effect of rotational speed on load-carrying capacity
and gas consumption was observed in this operating range. The examined operating points
did not indicate any sign of the occurrence of the pneumatic hammer. A temporary load of
below 90N on the bearing and an eccentricity ratio below 0.8 did not cause any significant
wear on the shaft.
Four numerical models, based on Reynolds’ lubricant film equation and Darcy’s law were
developed. The models were gradually extended with consideration of shaft misalignment,
the compressibility of the gas, the geometry of the pressure supply chamber and the
embedding of the groove structure. The models were validated with external publications
and the performed tests.
Numerous studies have investigated aerostatic porous bearings made of sintered metal
and graphite. Current computational approaches to determine a fast preliminary design
reached max. deviations of approximately 20 - 24% compared to experimental tests. One
of the central claims of this research was to extend this area of investigation by porous,
othotropically layered bearings made of C/C. The developed extended Full-Darcy model
achieved a maximum deviation in the load-carrying capacity of 21.6% and in the gas
consumption of 23.5%.
This study demonstrates the applicability of a resistant material from the aerospace field
(reusable thrust chambers made of CMC) for highly stressed and durable fluid-film bearings.
Furthermore, a numerical model for the computation and design of these bearings was
developed and validated.
In recent years, the Internet has become a major source of visual information exchange. Popular social platforms have reported an average of 80 million photo uploads a day. These images, are often accompanied with a user provided text one-liner, called an image caption. Deep Learning techniques have made significant advances towards automatic generation of factual image captions. However, captions generated by humans are much more than mere factual image descriptions. This work takes a step towards enhancing a machine's ability to generate image captions with human-like properties. We name this field as Affective Image Captioning, to differentiate it from the other areas of research focused on generating factual descriptions.
To deepen our understanding of human generated captions, we first perform a large-scale Crowd-Sourcing study on a subset of Yahoo Flickr Creative Commons 100 Million Dataset (YFCC100M). Three thousand random image-caption pairs were evaluated by native English speakers w.r.t different dimensions like focus, intent, emotion, meaning, and visibility. Our findings indicate three important underlying properties of human captions: subjectivity, sentiment, and variability. Based on these results, we develop Deep Learning models to address each of these dimensions.
To address the subjectivity dimension, we propose the Focus-Aspect-Value (FAV) model (along with a new task of aspect-detection) to structure the process of capturing subjectivity. We also introduce a novel dataset, aspects-DB, following this way of modeling. To implement the model, we propose a novel architecture called Tensor Fusion. Our experiments show that Tensor Fusion outperforms the state-of-the-art cross residual networks (XResNet) in aspect-detection.
Towards the sentiment dimension, we propose two models:Concept & Syntax Transition Network (CAST) and Show & Tell with Emotions (STEM). The CAST model uses a graphical structure to generate sentiment. The STEM model uses a neural network to inject adjectives into a neutral caption. Achieving a high score of 93% with human evaluation, these models were selected as the top-3 at the ACMMM Grand Challenge 2016.
To address the last dimension, variability, we take a generative approach called Generative Adversarial Networks (GAN) along with multimodal fusion. Our modified GAN, with two discriminators, is trained using Reinforcement Learning. We also show that it is possible to control the properties of the generated caption-variations with an external signal. Using sentiment as the external signal, we show that we can easily outperform state-of-the-art sentiment caption models.
Aflatoxins, a group of mycotoxins produced by various mold species within the genus Aspergillus, have been extensively investigated for their potential to contaminate food and feed, rendering them unfit for consumption. Nevertheless, the role of aflatoxins as environmental contaminants in soil, which represents their natural habitat, remains a relatively unexplored area in aflatoxin research. This knowledge gap can be attributed, in part, to the methodological challenges associated with detecting aflatoxins in soil. The main objective of this PhD project was to develop and validate an analytical method that allows monitoring of aflatoxins in soil, and scrutinize the mechanisms and extent of occurrence of aflatoxins in soil, the processes governing their dissipation, and their impact on the soil microbiome and associated soil functions. By utilizing an efficient extraction solvent mixture comprising acetonitrile and water, coupled with an ultrasonication step, recoveries of 78% to 92% were achieved, enabling reliable determination of trace levels in soil ranging from 0.5 to 20 µg kg-1. However, in a field trial conducted in a high-risk model region for aflatoxin contamination in Sub-Saharan Africa, no aflatoxins were detected using this procedure, underscoring the complexities of field monitoring. These challenges encompassed rapid degradation, spatial heterogeneity, and seasonal fluctuations in aflatoxin occurrence. Degradation experiments revealed the importance of microbial and photochemical processes in the dissipation of aflatoxins in soil with half-lives of 20 - 65 days. The rate of dissipation was found to be influenced by soil properties, most notably soil texture and the initial concentration of aflatoxins in the soil. An exposure study provided evidence that aflatoxins do not pose a substantial threat to the soil microbiome, encompassing microbial biomass, activity, and catabolic functionality. This was particularly evident in clayey soils, where the toxicity of aflatoxins diminished significantly due to their strong binding to clay minerals. However, several critical questions remain unanswered, emphasizing the necessity for further research to attain a more comprehensive understanding of the ecological importance of aflatoxins. Future research should prioritize the challenges associated with field monitoring of aflatoxins, elucidate the mechanisms responsible for the dissipation of aflatoxins in soil during microbial and photochemical degradation, and investigate the ecological consequences of aflatoxins in regions heavily affected by aflatoxins, taking into account the interactions between aflatoxins and environmental and anthropogenic stressors. Addressing these questions contributes to a comprehensive understanding of the environmental impact of aflatoxins in soil, ultimately contributing to more effective strategies for aflatoxin management in agriculture.
Organizational routines constitute how work is accomplished in organizations. This dissertation thesis draws on recent routine research and is anchored in the field of organization theory. The thesis consists of four separate manuscripts that contribute to related research fields such as agility or coordination research from a routine perspective while also extending routine dynamics research. Recent routine dynamics research offers a wide perspective on how situated actions within and across routines unfold as emergent accomplishments. This allows us to analyze other organization research phenomena, such as agility and coordination. Accordingly, the first and second manuscripts argue for the adoption of a very dynamic perspective on routines and the incorporation of these insights into agility and coordination research. This is followed by two empirical manuscripts that expand the routine literature based on qualitative research within agile software development. The third manuscript of this dissertation analyzes how situated actions address different temporal orientations (i.e., past, present, and future). Last, the fourth manuscript addresses the performing of roles within and through routines. In general, this dissertation contributes to overall organization research in two ways: (1) by outlining and examining how agility is enacted; (2) by highlighting that actions are performed flexibly to consider the situation at hand.
This thesis contains the mathematical treatment of a special class of analog microelectronic circuits called translinear circuits. The goal is to provide foundations of a new coherent synthesis approach for this class of circuits. The mathematical methods of the suggested synthesis approach come from graph theory, combinatorics, and from algebraic geometry, in particular symbolic methods from computer algebra. Translinear circuits form a very special class of analog circuits, because they rely on nonlinear device models, but still allow a very structured approach to network analysis and synthesis. Thus, translinear circuits play the role of a bridge between the "unknown space" of nonlinear circuit theory and the very well exploited domain of linear circuit theory. The nonlinear equations describing the behavior of translinear circuits possess a strong algebraic structure that is nonetheless flexible enough for a wide range of nonlinear functionality. Furthermore, translinear circuits offer several technical advantages like high functional density, low supply voltage and insensitivity to temperature. This unique profile is the reason that several authors consider translinear networks as the key to systematic synthesis methods for nonlinear circuits. The thesis proposes the usage of a computer-generated catalog of translinear network topologies as a synthesis tool. The idea to compile such a catalog has grown from the observation that on the one hand, the topology of a translinear network must satisfy strong constraints which severely limit the number of "admissible" topologies, in particular for networks with few transistors, and on the other hand, the topology of a translinear network already fixes its essential behavior, at least for static networks, because the so-called translinear principle requires the continuous parameters of all transistors to be the same. Even though the admissible topologies are heavily restricted, it is a highly nontrivial task to compile such a catalog. Combinatorial techniques have been adapted to undertake this task. In a catalog of translinear network topologies, prototype network equations can be stored along with each topology. When a circuit with a specified behavior is to be designed, one can search the catalog for a network whose equations can be matched with the desired behavior. In this context, two algebraic problems arise: To set up a meaningful equation for a network in the catalog, an elimination of variables must be performed, and to test whether a prototype equation from the catalog and a specified equation of desired behavior can be "matched", a complex system of polynomial equations must be solved, where the solutions are restricted to a finite set of integers. Sophisticated algorithms from computer algebra are applied in both cases to perform the symbolic computations. All mentioned algorithms have been implemented using C++, Singular, and Mathematica, and are successfully applied to actual design problems of humidity sensor circuitry at Analog Microelectronics GmbH, Mainz. As result of the research conducted, an exhaustive catalog of all static formal translinear networks with at most eight transistors is available. The application for the humidity sensor system proves the applicability of the developed synthesis approach. The details and implementations of the algorithms are worked out only for static networks, but can easily be adopted for dynamic networks as well. While the implementation of the combinatorial algorithms is stand-alone software written "from scratch" in C++, the implementation of the algebraic algorithms, namely the symbolic treatment of the network equations and the match finding, heavily rely on the sophisticated Gröbner basis engine of Singular and thus on more than a decade of experience contained in a special-purpose computer algebra system. It should be pointed out that the thesis contains the new observation that the translinear loop equations of a translinear network are precisely represented by the toric ideal of the network's translinear digraph. Altogether, this thesis confirms and strengthenes the key role of translinear circuits as systematically designable nonlinear circuits.
In the first part of this thesis we study algorithmic aspects of tropical intersection theory. We analyse how divisors and intersection products on tropical cycles can actually be computed using polyhedral geometry. The main focus is the study of moduli spaces, where the underlying combinatorics of the varieties involved allow a much more efficient way of computing certain tropical cycles. The algorithms discussed here have been implemented in an extension for polymake, a software for polyhedral computations.
In the second part we apply the algorithmic toolkit developed in the first part to the study of tropical double Hurwitz cycles. Hurwitz cycles are a higher-dimensional generalization of Hurwitz numbers, which count covers of \(\mathbb{P}^1\) by smooth curves of a given genus with a certain fixed ramification behaviour. Double Hurwitz numbers provide a strong connection between various mathematical disciplines, including algebraic geometry, representation theory and combinatorics. The tropical cycles have a rather complex combinatorial nature, so it is very difficult to study them purely "by hand". Being able to compute examples has been very helpful
in coming up with theoretical results. Our main result states that all marked and unmarked Hurwitz cycles are connected in codimension one and that for a generic choice of simple ramification points the marked cycle is a multiple of an irreducible cycle. In addition we provide computational examples to show that this is the strongest possible statement.
This thesis builds a bridge between singularity theory and computer algebra. To an isolated hypersurface singularity one can associate a regular meromorphic connection, the Gauß-Manin connection, containing a lattice, the Brieskorn lattice. The leading terms of the Brieskorn lattice with respect to the weight and V-filtration of the Gauß-Manin connection define the spectral pairs. They correspond to the Hodge numbers of the mixed Hodge structure on the cohomology of the Milnor fibre and belong to the finest known invariants of isolated hypersurface singularities. The differential structure of the Brieskorn lattice can be described by two complex endomorphisms A0 and A1 containing even more information than the spectral pairs. In this thesis, an algorithmic approach to the Brieskorn lattice in the Gauß-Manin connection is presented. It leads to algorithms to compute the complex monodromy, the spectral pairs, and the differential structure of the Brieskorn lattice. These algorithms are implemented in the computer algebra system Singular.
In modern algebraic geometry solutions of polynomial equations are studied from a qualitative point of view using highly sophisticated tools such as cohomology, \(D\)-modules and Hodge structures. The latter have been unified in Saito’s far-reaching theory of mixed Hodge modules, that has shown striking applications including vanishing theorems for cohomology. A mixed Hodge module can be seen as a special type of filtered \(D\)-module, which is an algebraic counterpart of a system of linear differential equations. We present the first algorithmic approach to Saito’s theory. To this end, we develop a Gröbner basis theory for a new class of algebras generalizing PBW-algebras.
The category of mixed Hodge modules satisfies Grothendieck’s six-functor formalism. In part these functors rely on an additional natural filtration, the so-called \(V\)-filtration. A key result of this thesis is an algorithm to compute the \(V\)-filtration in the filtered setting. We derive from this algorithm methods for the computation of (extraordinary) direct image functors under open embeddings of complements of pure codimension one subvarieties. As side results we show
how to compute vanishing and nearby cycle functors and a quasi-inverse of Kashiwara’s equivalence for mixed Hodge modules.
Describing these functors in terms of local coordinates and taking local sections, we reduce the corresponding computations to algorithms over certain bifiltered algebras. It leads us to introduce the class of so-called PBW-reduction-algebras, a generalization of the class of PBW-algebras. We establish a comprehensive Gröbner basis framework for this generalization representing the involved filtrations by weight vectors.
In modern algebraic geometry solutions of polynomial equations are studied from a qualitative point of view using highly sophisticated tools such as cohomology, \(D\)-modules and Hodge structures. The latter have been unified in Saito’s far-reaching theory of mixed Hodge modules, that has shown striking applications including vanishing theorems for cohomology. A mixed Hodge module can be seen as a special type of filtered \(D\)-module, which is an algebraic counterpart of a system of linear differential equations. We present the first algorithmic approach to Saito’s theory. To this end, we develop a Gröbner basis theory for a new class of algebras generalizing PBW-algebras.
The category of mixed Hodge modules satisfies Grothendieck’s six-functor formalism. In part these functors rely on an additional natural filtration, the so-called \(V\)-filtration. A key result of this thesis is an algorithm to compute the \(V\)-filtration in the filtered setting. We derive from this algorithm methods for the computation of (extraordinary) direct image functors under open embeddings of complements of pure codimension one subvarieties. As side results we show how to compute vanishing and nearby cycle functors and a quasi-inverse of Kashiwara’s equivalence for mixed Hodge modules.
Describing these functors in terms of local coordinates and taking local sections, we reduce the corresponding computations to algorithms over certain bifiltered algebras. It leads us to introduce the class of so-called PBW-reduction-algebras, a generalization of the class of PBW-algebras. We establish a comprehensive Gröbner basis framework for this generalization representing the involved filtrations by weight vectors.
Software is becoming increasingly concurrent: parallelization, decentralization, and reactivity necessitate asynchronous programming in which processes communicate by posting messages/tasks to others’ message/task buffers. Asynchronous programming has been widely used to build fast servers and routers, embedded systems and sensor networks, and is the basis of Web programming using Javascript. Languages such as Erlang and Scala have adopted asynchronous programming as a fundamental concept with which highly scalable and highly reliable distributed systems are built.
Asynchronous programs are challenging to implement correctly: the loose coupling between asynchronously executed tasks makes the control and data dependencies difficult to follow. Even subtle design and programming mistakes on the programs have the capability to introduce erroneous or divergent behaviors. As asynchronous programs are typically written to provide a reliable, high-performance infrastructure, there is a critical need for analysis techniques to guarantee their correctness.
In this dissertation, I provide scalable verification and testing tools to make asyn- chronous programs more reliable. I show that the combination of counter abstraction and partial order reduction is an effective approach for the verification of asynchronous systems by presenting PROVKEEPER and KUAI, two scalable verifiers for two types of asynchronous systems. I also provide a theoretical result that proves a counter-abstraction based algorithm called expand-enlarge-check, is an asymptotically optimal algorithm for the coverability problem of branching vector addition systems as which many asynchronous programs can be modeled. In addition, I present BBS and LLSPLAT, two testing tools for asynchronous programs that efficiently uncover many subtle memory violation bugs.
In the first part of the thesis we develop the theory of standard bases in free modules over (localized) polynomial rings. Given that linear equations are solvable in the coefficients of the polynomials, we introduce an algorithm to compute standard bases with respect to arbitrary (module) monomial orderings. Moreover, we take special care to principal ideal rings, allowing zero divisors. For these rings we design modified algorithms which are new and much faster than the general ones. These algorithms were motivated by current limitations in formal verification of microelectronic System-on-Chip designs. We show that our novel approach using computational algebra is able to overcome these limitations in important classes of applications coming from industrial challenges.
The second part is based on research in collaboration with Jason Morton, Bernd Sturmfels and Anne Shiu. We devise a general method to describe and compute a certain class of rank tests motivated by statistics. The class of rank tests may loosely be described as being based on computing the number of linear extensions to given partial orders. In order to apply these tests to actual data we developed two algorithms and used our implementations to apply the methodology to gene expression data created at the Stowers Institute for Medical Research. The dataset is concerned with the development of the vertebra. Our rankings proved valuable to the biologists.
This thesis, whose subject is located in the field of algorithmic commutative algebra and algebraic geometry, consists of three parts.
The first part is devoted to parallelization, a technique which allows us to take advantage of the computational power of modern multicore processors. First, we present parallel algorithms for the normalization of a reduced affine algebra A over a perfect field. Starting from the algorithm of Greuel, Laplagne, and Seelisch, we propose two approaches. For the local-to-global approach, we stratify the singular locus Sing(A) of A, compute the normalization locally at each stratum and finally reconstruct the normalization of A from the local results. For the second approach, we apply modular methods to both the global and the local-to-global normalization algorithm.
Second, we propose a parallel version of the algorithm of Gianni, Trager, and Zacharias for primary decomposition. For the parallelization of this algorithm, we use modular methods for the computationally hardest steps, such as for the computation of the associated prime ideals in the zero-dimensional case and for the standard bases computations. We then apply an innovative fast method to verify that the result is indeed a primary decomposition of the input ideal. This allows us to skip the verification step at each of the intermediate modular computations.
The proposed parallel algorithms are implemented in the open-source computer algebra system SINGULAR. The implementation is based on SINGULAR's new parallel framework which has been developed as part of this thesis and which is specifically designed for applications in mathematical research.
In the second part, we propose new algorithms for the computation of syzygies, based on an in-depth analysis of Schreyer's algorithm. Here, the main ideas are that we may leave out so-called "lower order terms" which do not contribute to the result of the algorithm, that we do not need to order the terms of certain module elements which occur at intermediate steps, and that some partial results can be cached and reused.
Finally, the third part deals with the algorithmic classification of singularities over the real numbers. First, we present a real version of the Splitting Lemma and, based on the classification theorems of Arnold, algorithms for the classification of the simple real singularities. In addition to the algorithms, we also provide insights into how real and complex singularities are related geometrically. Second, we explicitly describe the structure of the equivalence classes of the unimodal real singularities of corank 2. We prove that the equivalences are given by automorphisms of a certain shape. Based on this theorem, we explain in detail how the structure of the equivalence classes can be computed using SINGULAR and present the results in concise form. The probably most surprising outcome is that the real singularity type \(J_{10}^-\) is actually redundant.
Due to its performance, the field of deep learning has gained a lot of attention, with neural networks succeeding in areas like \( \textit{Computer Vision} \) (CV), \( \textit{Neural Language Processing} \) (NLP), and \( \textit{Reinforcement Learning} \) (RL). However, high accuracy comes at a computational cost as larger networks require longer training time and no longer fit onto a single GPU. To reduce training costs, researchers are looking into the dynamics of different optimizers, in order to find ways to make training more efficient. Resource requirements can be limited by reducing model size during training or designing more efficient models that improve accuracy without increasing network size.
This thesis combines eigenvalue computation and high-dimensional loss surface visualization to study different optimizers and deep neural network models. Eigenvectors of different eigenvalues are computed, and the loss landscape and optimizer trajectory are projected onto the plane spanned by those eigenvectors. A new parallelization method for the stochastic Lanczos method is introduced, resulting in faster computation and thus enabling high-resolution videos of the trajectory and second-order information during neural network training. Additionally, the thesis presents the loss landscape between two minima along with the eigenvalue density spectrum at intermediate points for the first time.
Secondly, this thesis presents a regularization method for \( \textit{Generative Adversarial Networks} \) (GANs) that uses second-order information. The gradient during training is modified by subtracting the eigenvector direction of the biggest eigenvalue, preventing the network from falling into the steepest minima and avoiding mode collapse. The thesis also shows the full eigenvalue density spectra of GANs during training.
Thirdly, this thesis introduces ProxSGD, a proximal algorithm for neural network training that guarantees convergence to a stationary point and unifies multiple popular optimizers. Proximal gradients are used to find a closed-form solution to the problem of training neural networks with smooth and non-smooth regularizations, resulting in better sparsity and more efficient optimization. Experiments show that ProxSGD can find sparser networks while reaching the same accuracy as popular optimizers.
Lastly, this thesis unifies sparsity and \( \textit{neural architecture search} \) (NAS) through the framework of group sparsity. Group sparsity is achieved through \( \ell_{2,1} \)-regularization during training, allowing for filter and operation pruning to reduce model size with minimal sacrifice in accuracy. By grouping multiple operations together, group sparsity can be used for NAS as well. This approach is shown to be more robust while still achieving competitive accuracies compared to state-of-the-art methods.
In this dissertation, we discuss how to price American-style options. Our aim is to study and improve the regression-based Monte Carlo methods. In order to have good benchmarks to compare with them, we also study the tree methods.
In the second chapter, we investigate the tree methods specifically. We do research firstly within the Black-Scholes model and then within the Heston model. In the Black-Scholes model, based on Müller's work, we illustrate how to price one dimensional and multidimensional American options, American Asian options, American lookback options, American barrier options and so on. In the Heston model, based on Sayer's research, we implement his algorithm to price one dimensional American options. In this way, we have good benchmarks of various American-style options and put them all in the appendix.
In the third chapter, we focus on the regression-based Monte Carlo methods theoretically and numerically. Firstly, we introduce two variations, the so called "Tsitsiklis-Roy method" and the "Longstaff-Schwartz method". Secondly, we illustrate the approximation of American option by its Bermudan counterpart. Thirdly we explain the source of low bias and high bias. Fourthly we compare these two methods using in-the-money paths and all paths. Fifthly, we examine the effect using different number and form of basis functions. Finally, we study the Andersen-Broadie method and present the lower and upper bounds.
In the fourth chapter, we study two machine learning techniques to improve the regression part of the Monte Carlo methods: Gaussian kernel method and kernel-based support vector machine. In order to choose a proper smooth parameter, we compare fixed bandwidth, global optimum and suboptimum from a finite set. We also point out that scaling the training data to [0,1] can avoid numerical difficulty. When out-of-sample paths of stock prices are simulated, the kernel method is robust and even performs better in several cases than the Tsitsiklis-Roy method and the Longstaff-Schwartz method. The support vector machine can keep on improving the kernel method and needs less representations of old stock prices during prediction of option continuation value for a new stock price.
In the fifth chapter, we switch to the hardware (FGPA) implementation of the Longstaff-Schwartz method and propose novel reversion formulas for the stock price and volatility within the Black-Scholes and Heston models. The test for this formula within the Black-Scholes model shows that the storage of data is reduced and also the corresponding energy consumption.
Nowadays one of the major objectives in geosciences is the determination of the gravitational field of our planet, the Earth. A precise knowledge of this quantity is not just interesting on its own but it is indeed a key point for a vast number of applications. The important question is how to obtain a good model for the gravitational field on a global scale. The only applicable solution - both in costs and data coverage - is the usage of satellite data. We concentrate on highly precise measurements which will be obtained by GOCE (Gravity Field and Steady State Ocean Circulation Explorer, launch expected 2006). This satellite has a gradiometer onboard which returns the second derivatives of the gravitational potential. Mathematically seen we have to deal with several obstacles. The first one is that the noise in the different components of these second derivatives differs over several orders of magnitude, i.e. a straightforward solution of this outer boundary value problem will not work properly. Furthermore we are not interested in the data at satellite height but we want to know the field at the Earth's surface, thus we need a regularization (downward-continuation) of the data. These two problems are tackled in the thesis and are now described briefly. Split Operators: We have to solve an outer boundary value problem at the height of the satellite track. Classically one can handle first order side conditions which are not tangential to the surface and second derivatives pointing in the radial direction employing integral and pseudo differential equation methods. We present a different approach: We classify all first and purely second order operators which fulfill that a harmonic function stays harmonic under their application. This task is done by using modern algebraic methods for solving systems of partial differential equations symbolically. Now we can look at the problem with oblique side conditions as if we had ordinary i.e. non-derived side conditions. The only additional work which has to be done is an inversion of the differential operator, i.e. integration. In particular we are capable to deal with derivatives which are tangential to the boundary. Auto-Regularization: The second obstacle is finding a proper regularization procedure. This is complicated by the fact that we are facing stochastic rather than deterministic noise. The main question is how to find an optimal regularization parameter which is impossible without any additional knowledge. However we could show that with a very limited number of additional information, which are obtainable also in practice, we can regularize in an asymptotically optimal way. In particular we showed that the knowledge of two input data sets allows an order optimal regularization procedure even under the hard conditions of Gaussian white noise and an exponentially ill-posed problem. A last but rather simple task is combining data from different derivatives which can be done by a weighted least squares approach using the information we obtained out of the regularization procedure. A practical application to the downward-continuation problem for simulated gravitational data is shown.
Wireless Sensor Networks (WSN) are dynamically-arranged networks typically composed of a large number of arbitrarily-distributed sensor nodes with computing capabilities contributing to –at least– one common application. The main characteristic of these networks is that of being functionally constrained due to a scarce availability of resources and strong dependence on uncontrollable environmental factors. These conditions introduce severe restrictions on the applicability of classic real-time methods aiming at guaranteeing time-bounded communications. Existing real-time solutions tend to apply concepts that were originally not conceived for sensor networks, idealizing realistic application scenarios and overlooking at important design limitations. This results in a number of misleading practices contributing to approaches of restricted validity in real-world scenarios. Amending the confrontation between WSNs and real-time objectives starts with a review of the basic fundamentals of existing approaches. In doing so, this thesis presents an alternative approach based on a generalized timeliness notion suitable to the particularities of WSNs. The new conceptual notion allows the definition of feasible real-time objectives opening a new scope of possibilities not constrained to idealized systems. The core of this thesis is based on the definition and application of Quality of Service (QoS) trade-offs between timeliness and other significant QoS metrics. The analysis of local and global trade-offs provides a step-by-step methodology identifying the correlations between these quality metrics. This association enables the definition of alternative trade-off configurations (set points) influencing the quality performance of the network at selected instants of time. With the basic grounds established, the above concepts are embedded in a simple routing protocol constituting a proof of concept for the validity of the presented analysis. Extensive evaluations under realistic scenarios are driven on simulation environments as well as real testbeds, validating the consistency of this approach.
Open distributed systems are a class of distributed systems where (i) only partial information about the environment, in which they are running, is present, (ii) new resources may become available at runtime, and (iii) a subsystem may become aware of other subsystems after some interaction. Modeling and implementing such systems correctly is a complex task due to the openness and the dynamicity aspects. One way to ensure that the resulting systems behave correctly is to utilize formal verification.
Formal verification requires an adequate semantic model of the implementation, a specification of the desired behavior, and a reasoning technique. The actor model is a semantic model that captures the challenging aspects of open distributed systems by utilizing actors as universal primitives to represent system entities and allowing them to create new actors and to communicate by sending directed messages as reply to received messages. To enable compositional reasoning, where the reasoning task is reduced to independent verification of the system parts, semantic entities at a higher level of abstraction than actors are needed.
This thesis proposes an automaton model and combines sound reasoning techniques to compositionally verify implementations of open actor systems. Based on I/O automata, the model allows automata to be created dynamically and captures dynamic changes in communication patterns. Each automaton represents either an actor or a group of actors. The specification of the desired behavior is given constructively as an automaton. As the basis for compositionality, we formalize a component notion based on the static structure of the implementation instead of the dynamic entities (the actors) occurring in the system execution. The reasoning proceeds in two stages. The first stage establishes the connection between the automata representing single actors and their implementation description by means of weakest liberal preconditions. The second stage employs this result as the basis for verifying whether a component specification is satisfied. The verification is done by building a simulation relation from the automaton representing the implementation to the component's automaton. Finally, we validate the compositional verification approach through a number of examples by proving correctness of their actor implementations with respect to system specifications.
An Efficient Automated Machine Learning Framework for Genomics and Proteomics Sequence Analysis
(2023)
Genomics and Proteomics sequence analyses are the scientific studies of understanding the language of Deoxyribonucleic Acid (DNA), Ribonucleic Acid (RNA) and protein biomolecules with an objective of controlling the production of proteins and understanding their core functionalities. It helps to detect chronic diseases in early stages, root causes of clinical changes, key genetic targets for pharmaceutical development and optimization of therapeutics for various age groups. Most Genomics and Proteomics sequence analysis work is performed using typical wet lab experimental approaches that make use of different genetic diagnostic technologies. However, these approaches are costly, time consuming, skill and labor intensive. Hence, these approaches slow down the process of developing an efficient and economical sequence analysis landscape essential to demystify a variety of cellular processes and functioning of biomolecules in living organisms. To empower manual wet lab experiment driven research, many machine learning based approaches have been developed in recent years. However, these approaches cannot be used in practical environment due to their limited performance. Considering the sensitive and inherently demanding nature of Genomics and Proteomics sequence
analysis which can have very far-reaching as well as serious repercussions on account of misdiagnosis, the main
objective of this research is to develop an efficient automated computational framework for Genomics and Proteomics sequence analysis using the predictive and prescriptive analytical powers of Artificial Intelligence (AI) to significantly improve healthcare operations.
The proposed framework is comprised of 3 main components namely sequence encoding, feature engineering and
discrete or continuous value predictor. The sequence encoding module is equipped with a variety of existing and newly developed sequence encoding algorithms that are capable of generating a rich statistical representation of DNA, RNA and protein raw sequences. The feature engineering module has diverse types of feature selection and dimensionality reduction approaches which can be used to generate the most effective feature space. Furthermore, the discrete and/or continuous value predictor module of the proposed framework contains a wide range of existing machine learning and newly developed deep learning regressors and classifiers. To evaluate the integrity and generalizability of the proposed framework, we have performed a large-scale experimentation over diverse types of Genomics and Proteomics sequence analysis tasks (i.e., DNA, RNA and proteins).
In Genomics analysis, Epigenetic modification detection is one of the key component. It helps clinical researchers and practitioners to distinguish normal cellular activities from malfunctioned ones, which can lead to diverse genetic disorders such as metabolic disorders, cancers, etc. To support this analysis, the proposed framework is used to solve the problem of DNA and Histone modification prediction where it has achieved state-of-the-art performance on 27 publicly available benchmark datasets of 17 different species with best accuracy of 97%. RNA sequence analysis is another vital component of Genomics sequence analysis where the identification of different coding and non-coding RNAs as well as their subcellular localization patterns help to demystify the functions of diverse RNAs, root causes of clinical changes, develop precision medicine and optimize therapeutics. To support this analysis, the proposed framework is utilized for non-coding RNA classification and multi-compartment RNA subcellular localization prediction. Where it achieved state-of-the-art performance on 10 publicly available benchmark datasets of Homo sapiens and Mus Musculus species with best accuracy of 98%.
Proteomics sequence analysis is essential to demystify the virus pathogenesis, host immunity responses, the way
proteins affect or are affected by the cell processes, their structure and core functionalities. To support this analysis, the proposed framework is used for host protein-protein and virus-host protein-protein interaction prediction. It has achieved state-of-the-art performance on 2 publicly available protein protein interaction datasets of Homo Sapiens and Mus Musculus species with best accuracy of 96% and 7 viral host protein protein interaction datasets of multiple hosts and viruses with best accuracy of 94%. Considering the performance and practical significance of proposed framework, we believe proposed framework will help researchers in developing cutting-edge practical applications for diverse Genomic and Proteomic sequence analyses tasks (i.e., DNA, RNA and proteins).
Multidisciplinary optimizations (MDOs) encompass optimization problems that combine different disciplines into a single optimization with the aim of converging towards a design that simultaneously fulfills multiple criteria. For example, considering both fluid and structural disciplines to obtain a shape that is not only aerodynamically efficient, but also respects structural constraints. Combined with CAD-based parametrizations, the optimization produces an improved, manufacturable shape. For turbomachinery applications, this method has been successfully applied using gradient-free optimization methods such as genetic algorithms, surrogate modeling, and others. While such algorithms can be easily applied without access to the source code, the number of iterations to converge is dependent on the number of design parameters. This results in high computational costs and limited design spaces. A competitive alternative is offered by gradient-based optimization algorithms combined with adjoint methods, where the computational complexity of the gradient calculation is no longer dependent on the number of design parameters, but rather on the number of outputs. Such methods have been extensively used in single-disciplinary aerodynamic optimizations using adjoint fluid solvers and CAD parametrizations. However, CAD-based MDOs leveraging adjoint methods are just beginning to emerge.
This thesis contributes to this field of research by setting up a CAD-based adjoint MDO framework for turbomachinery design using both fluid and structural disciplines. To achieve this, the von Kármán Institute’s existing CAD-based optimization framework cado is augmented by the development of a FEM-based structural solver which has been differentiated using the algorithmic differentiation tool CoDiPack of TU Kaiserslautern. While most of the code could be differentiated in a black-box fashion, special treatment is required for the iterative linear and eigenvalue solvers to ensure accuracy and reduce memory consumption. As a result, the solver has the capability of computing both stress and vibration gradients at a cost independent on the number of design parameters. For the presented application case of a radial turbine optimization, the full gradient calculation has a computational cost of approximately 3.14 times the cost of a primal run and the peak memory usage of approximately 2.76 times that of a primal run.
The FEM code leverages object-oriented design such that the same base structure can be reused for different purposes with minimal re-differentiation. This is demonstrated by considering a composite material test case where the gradients could be easily calculated with respect to an extended design space that includes material properties. Additionally, the structural solver is reused within a CAD-based mesh deformation, which propagates the structural FEM mesh gradients through to the CAD parameters. This closes the link between the CAD shape and FEM mesh. Finally, the MDO framework is applied by optimizing the aerodynamic efficiency of a radial turbine while respecting structural constraints.
An efficient multiscale approach is established in order to compute the macroscopic response of nonlinear composites. The micro problem is rewritten in an integral form of the Lippmann-Schwinger type and solved efficiently by Fast Fourier Transforms. Using realistic microstructure models complex nonlinear effects are reproduced and validated with measured data of fiber reinforced plastics. The micro problem is integrated in a Finite Element framework which is used to solve the macroscale. The scale coupling technique and a consistent numerical algorithm is established. The method provides an efficient way to determine the macroscopic response considering arbitrary microstructures, constitutive behaviors and loading conditions.
Optical Character Recognition (OCR) system plays an important role in digitization of data acquired as images from a variety of sources. Although the area is very well explored for Latin languages, some of the languages based on Arabic cursive script are not yet explored. It is due to many factors: Most importantly are the unavailability of proper data sets and complexities posed by cursive scripts. The Pashto language is one of such languages which needs considerable exploration towards OCR. In order to develop such an OCR system, this thesis provides a pioneering study that explores deep learning for the Pashto language in the field of OCR.
The Pashto language is spoken by more than $50$ million people across the world, and it is an active medium both for oral as well as written communication. It is associated with rich literary heritage and contains huge written collection. These written materials present contents of simple to complex nature, and layouts from hand-scribed to printed text. The Pashto language presents mainly two types of complexities (i) generic w.r.t. cursive script, (ii) specific w.r.t. Pashto language. Generic complexities are cursiveness, context dependency, and breaker character anomalies, as well as space anomalies. Pashto specific complexities are variations in shape for a single character and shape similarity for some of the additional Pashto characters. Existing research in the area of Arabic OCR did not lead to an end-to-end solution for the mentioned complexities and therefore could not be generalized to build a sophisticated OCR system for Pashto.
The contribution of this thesis spans in three levels, conceptual level, data level, and practical level. In the conceptual level, we have deeply explored the Pashto language and identified those characters, which are responsible for the challenges mentioned above. In the data level, a comprehensive dataset is introduced containing real images of hand-scribed contents. The dataset is manually transcribed and has the most frequent layout patterns associated with the Pashto language. The practical level contribution provides a bridge, in the form of a complete Pashto OCR system, and connects the outcomes of the conceptual and data levels contributions. The practical contribution comprises of skew detection, text-line segmentation, feature extraction, classification, and post-processing. The OCR module is more strengthened by using deep learning paradigm to recognize Pashto cursive script by the framework of Recursive Neural Networks (RNN). Proposed Pashto text recognition is based on Long Short-Term Memory Network (LSTM) and realizes a character recognition rate of $90.78\%$ on Pashto real hand-scribed images. All these contributions are integrated into an application to provide a flexible and generic End-to-End Pashto OCR system.
The impact of this thesis is not only specific to the Pashto language, but it is also beneficial to other cursive languages like Arabic, Urdu, and Persian e.t.c. The main reason is the Pashto character set, which is a superset of Arabic, Persian, and Urdu languages. Therefore, the conceptual contribution of this thesis provides insight and proposes solutions to almost all generic complexities associated with Arabic, Persian, and Urdu languages. For example, an anomaly caused by breaker characters is deeply analyzed, which is shared among 70 languages, mainly use Arabic script. This thesis presents a solution to this issue and is equally beneficial to almost all Arabic like languages.
The scope of this thesis has two important aspects. First, a social impact, i.e., how a society may benefit from it. The main advantages are to bring the historical and almost vanished document to life and to ensure the opportunities to explore, analyze, translate, share, and understand the contents of Pashto language globally. Second, the advancement and exploration of the technical aspects. Because, this thesis empirically explores the recognition and challenges which are solely related to the Pashto language, both regarding character-set and the materials which present such complexities. Furthermore, the conceptual and practical background of this thesis regarding complexities of Pashto language is very beneficial regarding OCR for other cursive languages.
The main theme of this thesis is the interplay between algebraic and tropical intersection
theory, especially in the context of enumerative geometry. We begin by exploiting
well-known results about tropicalizations of subvarieties of algebraic tori to give a
simple proof of Nishinou and Siebert’s correspondence theorem for rational curves
through given points in toric varieties. Afterwards, we extend this correspondence
by additionally allowing intersections with psi-classes. We do this by constructing
a tropicalization map for cycle classes on toroidal embeddings. It maps algebraic
cycle classes to elements of the Chow group of the cone complex of the toroidal
embedding, that is to weighted polyhedral complexes, which are balanced with respect
to an appropriate map to a vector space, modulo a naturally defined equivalence relation.
We then show that tropicalization respects basic intersection-theoretic operations like
intersections with boundary divisors and apply this to the appropriate moduli spaces
to obtain our correspondence theorem.
Trying to apply similar methods in higher genera inevitably confronts us with moduli
spaces which are not toroidal. This motivates the last part of this thesis, where we
construct tropicalizations of cycles on fine logarithmic schemes. The logarithmic point of
view also motivates our interpretation of tropical intersection theory as the dualization
of the intersection theory of Kato fans. This duality gives a new perspective on the
tropicalization map; namely, as the dualization of a pull-back via the characteristic
morphism of a logarithmic scheme.
A popular model for the locations of fibres or grains in composite materials
is the inhomogeneous Poisson process in dimension 3. Its local intensity function
may be estimated non-parametrically by local smoothing, e.g. by kernel
estimates. They crucially depend on the choice of bandwidths as tuning parameters
controlling the smoothness of the resulting function estimate. In this
thesis, we propose a fast algorithm for learning suitable global and local bandwidths
from the data. It is well-known, that intensity estimation is closely
related to probability density estimation. As a by-product of our study, we
show that the difference is asymptotically negligible regarding the choice of
good bandwidths, and, hence, we focus on density estimation.
There are quite a number of data-driven bandwidth selection methods for
kernel density estimates. cross-validation is a popular one and frequently proposed
to estimate the optimal bandwidth. However, if the sample size is very
large, it becomes computational expensive. In material science, in particular,
it is very common to have several thousand up to several million points.
Another type of bandwidth selection is a solve-the-equation plug-in approach
which involves replacing the unknown quantities in the asymptotically optimal
bandwidth formula by their estimates.
In this thesis, we develop such an iterative fast plug-in algorithm for estimating
the optimal global and local bandwidth for density and intensity estimation with a focus on 2- and 3-dimensional data. It is based on a detailed
asymptotics of the estimators of the intensity function and of its second
derivatives and integrals of second derivatives which appear in the formulae
for asymptotically optimal bandwidths. These asymptotics are utilised to determine
the exact number of iteration steps and some tuning parameters. For
both global and local case, fewer than 10 iterations suffice. Simulation studies
show that the estimated intensity by local bandwidth can better indicate
the variation of local intensity than that by global bandwidth. Finally, the
algorithm is applied to two real data sets from test bodies of fibre-reinforced
high-performance concrete, clearly showing some inhomogeneity of the fibre
intensity.
The cytosolic Fe65 adaptor protein family, consisting of Fe65, Fe65L1 and Fe65L2 is involved in many intracellular signaling pathways linking via its three interaction domains a continuously growing list of proteins by facilitating functional interactions. One of the most important binding partners of Fe65 family proteins is the amyloid precursor protein (APP), which plays an important role in Alzheimer Disease.
To gain deeper insights in the function of the ubiquitously expressed Fe65 and the brain enriched Fe65L1, the goal of my study was I) to analyze their putative synaptic function in vivo, II) to examine structural analysis focusing on a putative dimeric complex of Fe65, III) to consider the involvement of Fe65 in mediating LRP1 and APP intracellular trafficking in murine hippocampal neurons. By utilizing several behavioral analyses of Fe65 KO, Fe65L1 KO and Fe65/Fe65L1 DKO mice I could demonstrate that the Fe65 protein family is essential for learning and memory as well as grip strength and locomotor activity. Furthermore, immunohistological as well as protein biochemical analysis revealed that the Fe65 protein family is important for neuromuscular junction formation in the peripheral nervous system, which involves binding of APP and acting downstream of the APP signaling pathway. Via Co-immunoprecipitation analysis I could verify that Fe65 is capable to form dimers ex vivo, which exclusively occur in the cytosol and upon APP expression are shifted to membrane compartments forming trimeric complexes. The influence of the loss of Fe65 and/or Fe65L1 on APP and/or LRP1 transport characteristics in axons could not be verified, possibly conditioned by the compensatory effect of Fe65L2. However, I could demonstrate that LRP1 affects the APP transport independently of Fe65 by shifting APP into slower types of vesicles leading to changed processing and endocytosis of APP.
The outcome of my thesis advanced our understanding of the Fe65 protein family, especially its interplay with APP physiological function in synapse formation and synaptic plasticity.
The present thesis is concerned with the simulation of the loading behaviour of both hybrid lightweight structures and piezoelectric mesostructures, with a special focus on solid interfaces on the meso scale. Furthermore, an analytical review on bifurcation modes of continuum-interface problems is included. The inelastic interface behaviour is characterised by elastoplastic, viscous, damaging and fatigue-motivated models. For related numerical computations, the Finite Element Method is applied. In this context, so-called interface elements play an important role. The simulation results are reflected by numerous examples which are partially correlated to experimental data.
Various physical phenomenons with sudden transients that results into structrual changes can be modeled via
switched nonlinear differential algebraic equations (DAEs) of the type
\[
E_{\sigma}\dot{x}=A_{\sigma}x+f_{\sigma}+g_{\sigma}(x). \tag{DAE}
\]
where \(E_p,A_p \in \mathbb{R}^{n\times n}, x\mapsto g_p(x),\) is a mapping, \(p \in \{1,\cdots,P\}, P\in \mathbb{N}
f \in \mathbb{R} \rightarrow \mathbb{R}^n , \sigma: \mathbb{R} \rightarrow \{1,\cdots, P\}\).
Two related common tasks are:
Task 1: Investigate if above (DAE) has a solution and if it is unique.
Task 2: Find a connection among a solution of above (DAE) and solutions of related
partial differential equations.
In the linear case \(g(x) \equiv 0\) the task 1 has been tackeled already in a
distributional solution framework.
A main goal of the dissertation is to give contribution to task 1 for the
nonlinear case \(g(x) \not \equiv 0\) ; also contributions to the task 2 are given for
switched nonlinear DAEs arising while modeling sudden transients in water
distribution networks. In addition, this thesis contains the following further
contributions:
The notion of structured switched nonlinear DAEs has been introduced,
allowing also non regular distributions as solutions. This extend a previous
framework that allowed only piecewise smooth functions as solutions. Further six mild conditions were given to ensure existence and uniqueness of the solution within the space of piecewise smooth distribution. The main
condition, namely the regularity of the matrix pair \((E,A)\), is interpreted geometrically for those switched nonlinear DAEs arising from water network graphs.
Another contribution is the introduction of these switched nonlinear DAEs
as a simplication of the PDE model used classically for modeling water networks. Finally, with the support of numerical simulations of the PDE model it has been illustrated that this switched nonlinear DAE model is a good approximation for the PDE model in case of a small compressibility coefficient.
The development of autonomous mobile robots is a major topic of current research. As those robots must be able to react to changing environments and avoid collisions also with moving obstacles, the fulfilment of safety requirements is an important aspect. Behaviour-based systems (BBS) have proven to meet several of the properties required for these kindsof robots, such as reactivity, extensibility and re-usability of individual components. BBS consist of a number of behavioural components that individually realise simple tasks. Their interconnection allows to achieve complex robot behaviour, which implies that correct
connections are crucial. The resulting networks can get very large making them difficult to verify. This dissertation presents a novel concept for the analysis and verification of complex autonomous robot systems controlled by behaviour-based software architectures with special focus on the integration of environmental aspects into the processes.
Several analysis techniques have been investigated and adapted to the special requirements of BBS. These include a structural analysis, which is used to find constraint violations and faults in the network layout. Fault tree analysis is applied to identify root causes of hazards and the relationship of system events. For this, a technique to map the behaviour-based control network to the structure of a fault tree has been developed. Testing and data analysis are used for the detection of failures and their root causes. Here, a new concept that identifies patterns in data recorded during test runs has been introduced.
All of these methods cannot guarantee failure-free and safe robot behaviour and can never prove the absence of failures. Therefore, model checking as formal verification technique that proves a property to be correct for the given system, has been chosen to complement the set of analysis techniques. A novel concept for the integration of environmental influences into the model checking process is proposed. Environmental situations and the sensor processing chain are represented as synchronised automata similar to the modelling of the behavioural network. Tools supporting the whole verification process including the creation of formal queries in its environment have been developed.
During the verification of large behavioural networks, the scalability of the model checking approach appears as a big problem. Several approaches that deal with this problem have been investigated and the selection of slicing and abstraction methods has been justified. A concept for the application of these methods is provided, that reduces the behavioural network to the relevant parts before the actual verification process.
All techniques have been applied to the behaviour-based control system of the autonomous outdoor robot RAVON. Its complex network with more than 400 components allows for demonstrating the soundness of the presented concepts. The set of different techniques provides a fundamental basis for a comprehensive analysis and verification of BBS acting in changing environments.
This thesis is concerned with different null-models that are used in network analysis. Whenever it is of interest whether a real-world graph is exceptional regarding a particular measure, graphs from a null-model can be used to compare the real-world graph to. By analyzing an appropriate null-model, a researcher may find whether the results of the measure on the real-world graph is exceptional or not.
Deciding which null-model to use is hard and sometimes the difference between the null-models is not even considered. In this thesis, there are several results presented: First, based on simple global measures, undirected graphs are analyzed. The results for these measures indicates that it is not important which null-model is used, thus, the fastest algorithm of a null-model may be used. Next, local measures are investigated. The fastest algorithm proves to be the most complicated to analyze. The model includes multigraphs which do not meet the conditions of all the measures, thus, the measures themselves have to be altered to take care of multigraphs as well. After careful consideration, the conditions are met and the analysis shows, that the fastest is not always the best.
The same applies for directed graphs, as is shown in the last part. There, another more complex measure on graphs is introduced. I continue testing the applicability of several null-models; in the end, a set of equations proves to be fast and good enough as long as conditions regarding the degree sequence are met.
In recent decades, there has been increasing interest in analyzing the behavior of complex systems. A popular approach for analyzing such systems is a network analytic approach where the system is represented by a graph structure (Wassermann&Faust 1994, Boccaletti et al. 2006, Brandes&Erlebach 2005, Vespignani 2018): Nodes represent the system’s entities, edges their interactions. A large toolbox of network analytic methods, such as measures for structural properties (Newman 2010), centrality measures (Koschützki et al. 2005), or methods for identifying communities (Fortunato 2010), is readily available to be applied on any network structure. However, it is often overlooked that a network representation of a system and the (technically applicable) methods contain assumptions that need to be met; otherwise, the results are not interpretable or even misleading. The most important assumption of a network representation is the presence of indirect effects: If A has an impact on B, and B has an impact on C, then A has an impact on C (Zweig 2016, Brandes et al. 2013). The presence of indirect effects can be explained by ”something” flowing through the network by moving from node to node. Such network flows (or network processes) may be the propagation of information in social networks, the spread of infections, or entities using the network as infrastructure, such as in transportation networks. Also several network measures, particularly most centrality measures, assume the presence of such a network process, but additionally assume specific properties of the network processes (Borgatti 2005). Then, a centrality value indicates a node’s importance with respect to a process with these properties.
While this has been known for several years, only recently have datasets containing real-world network flows become accessible. In this context, the goal of this dissertation is to provide a better understanding of the actual behavior of real-world network processes, with a particular focus on centrality measures: If real-world network processes turn out to show different properties than those assumed by classic centrality measures, these measures might considerably under- or overestimate the importance of nodes for the actual network flow. To the best of our knowledge, there are only very few works addressing this topic.
The contributions of this thesis are therefore as follows: (i) We investigate in which aspects real-world network flows meet the assumptions contained about them in centrality measures. (ii) Since we find that the real-world flows show considerably different properties than assumed, we test to which extent the found properties can be explained by models, i.e., models based on shortest paths or random walks. (iii) We study whether the deviations from the assumed behavior have an impact on the results of centrality measures.
To this end, we introduce flow-based variants of centrality measures which are either based on the assumed behavior or on the actual behavior of the real-world network flow. This enables systematic evaluation of the impact of each assumption on the resulting rankings of centrality measures.
While–on a large scale–we observe a surprisingly large robustness of the measures against deviations in their assumptions, there are nodes whose importance is rated very differently when the real-world network flow is taken into account. (iv) As a technical contribution, we provide a method for an efficient handling of large sets of flow trajectories by summarizing them into groups of similar trajectories. (v) We furthermore present the results of an interdisciplinary research project in which the trajectories of humans in a network were analyzed in detail. In general, we are convinced that a process-driven perspective on network analysis in which the network process is considered in addition to the network representation, can help to better understand the behavior of complex systems.
Tire-soil interaction is important for the performance of off-road vehicles and the soil compaction in the agricultural field. With an analytical model, which is integrated in multibody-simulation software, and a Finite Element model, the forces and moments generated on the tire-soil contact patch were studied to analyze the tire performance. Simulations with these two models for different tire operating conditions were performed to evaluate the mechanical behaviors of an excavator tire. For the FE model validation a single wheel tester connected to an excavator arm was designed. Field tests were carried out to examine the tire vertical stiffness, the contact pressure on the tire – hard ground interface, the longitudinal/vertical force and the compaction of the sandy clay from the test field under specified operating conditions. The simulation and experimental results were compared to evaluate the model quality. The Magic Formula was used to fit the curves of longitudinal and lateral forces. A simplified tire-soil interaction model based on the fitted Magic Formula could be established and further applied to the simulation of vehicle-soil interaction.
Simplified ODE models describing blood flow rate are governed by the pressure gradient.
However, assuming the orientation of the blood flow in a human body correlates to a positive
direction, a negative pressure gradient forces the valve to shut, which stops the flow through
the valve, hence, the flow rate is zero, whereas the pressure rate is formulated by an ODE.
Presence of ODEs together with algebraic constraints and sudden changes of system characterizations
yield systems of switched differential-algebraic equations (swDAEs). Alternating
dynamics of the heart can be well modelled by means of swDAEs. Moreover, to study pulse
wave propagation in arteries and veins, PDE models have been developed. Connection between
the heart and vessels leads to coupling PDEs and swDAEs. This model motivates
to study PDEs coupled with swDAEs, for which the information exchange happens at PDE
boundaries, where swDAE provides boundary conditions to the PDE and PDE outputs serve
as inputs to swDAE. Such coupled systems occur, e.g. while modelling power grids using
telegrapher’s equations with switches, water flow networks with valves and district
heating networks with rapid consumption changes. Solutions of swDAEs might
include jumps, Dirac impulses and their derivatives of arbitrary high orders. As outputs of
swDAE read as boundary conditions of PDE, a rigorous solution framework for PDE must
be developed so that jumps, Dirac impulses and their derivatives are allowed at PDE boundaries
and in PDE solutions. This is a wider solution class than solutions of small bounded
variation (BV), for instance, used in where nonlinear hyperbolic PDEs are coupled with
ODEs. Similarly, in, the solutions to switched linear PDEs with source terms are
restricted to the class of BV. However, in the presence of Dirac impulses and their derivatives,
BV functions cannot handle the coupled systems including DAEs with index greater than one.
Therefore, hyperbolic PDEs coupled with swDAEs with index one will be studied in the BV
setting and with swDAEs whose index is greater than one will be investigated in the distributional
sense. To this end, the 1D space of piecewise-smooth distributions is extended to a 2D
piecewise-smooth distributional solution framework. 2D space of piecewise-smooth distributions
allows trace evaluations at boundaries of the PDE. Moreover, a relationship between
solutions to coupled system and switched delay DAEs is established. The coupling structure
in this thesis forms a rather general framework. In fact, any arbitrary network, where PDEs
are represented by edges and (switched) DAEs by nodes, is covered via this structure. Given
a network, by rescaling spatial domains which modifies the coefficient matrices by a constant,
each PDE can be defined on the same interval which leads to a formulation of a single
PDE whose unknown is made up of the unknowns of each PDE that are stacked over each
other with a block diagonal coefficient matrix. Likewise, every swDAE is reformulated such
that the unknowns are collected above each other and coefficient matrices compose a block
diagonal coefficient matrix so that each node in the network is expressed as a single swDAE.
The results are illustrated by numerical simulations of the power grid and simplified circulatory
system examples. Numerical results for the power grid display the evolution of jumps
and Dirac impulses caused by initial and boundary conditions as a result of instant switches.
On the other hand, the analysis and numerical results for the simplified circulatory system do
not entail a Dirac impulse, for otherwise such an entity would destroy the entire system. Yet
jumps in the flow rate in the numerical results can come about due to opening and closure of
valves, which suits clinical and physiological findings. Regarding physiological parameters,
numerical results obtained in this thesis for the simplified circulatory system agree well with
medical data and findings from literature when compared for the validation
An increasing number of nowadays tasks, such as speech recognition, image generation,
translation, classification or prediction are performed with the help of machine learning.
Especially artificial neural networks (ANNs) provide convincing results for these tasks.
The reasons for this success story are the drastic increase of available data sources in
our more and more digitalized world as well as the development of remarkable ANN
architectures. This development has led to an increasing number of model parameters
together with more and more complex models. Unfortunately, this yields a loss in the
interpretability of deployed models. However, there exists a natural desire to explain the
deployed models, not just by empirical observations but also by analytical calculations.
In this thesis, we focus on variational autoencoders (VAEs) and foster the understanding
of these models. As the name suggests, VAEs are based on standard autoencoders (AEs)
and therefore used to perform dimensionality reduction of data. This is achieved by a
bottleneck structure within the hidden layers of the ANN. From a data input the encoder,
that is the part up to the bottleneck, produces a low dimensional representation. The
decoder, the part from the bottleneck to the output, uses this representation to reconstruct
the input. The model is learned by minimizing the error from the reconstruction.
In our point of view, the most remarkable property and, hence, also a central topic
in this thesis is the auto-pruning property of VAEs. Simply speaking, the auto-pruning
is preventing the VAE with thousands of parameters from overfitting. However, such a
desirable property comes with the risk for the model of learning nothing at all. In this
thesis, we look at VAEs and the auto-pruning from two different angles and our main
contributions to research are the following:
(i) We find an analytic explanation of the auto-pruning. We do so, by leveraging the
framework of generalized linear models (GLMs). As a result, we are able to explain
training results of VAEs before conducting the actual training.
(ii) We construct a time dependent VAE and show the effects of the auto-pruning in
this model. As a result, we are able to model financial data sequences and estimate
the value-at-risk (VaR) of associated portfolios. Our results show that we surpass
the standard benchmarks for VaR estimation.
Analyzing Centrality Indices in Complex Networks: an Approach Using Fuzzy Aggregation Operators
(2018)
The identification of entities that play an important role in a system is one of the fundamental analyses being performed in network studies. This topic is mainly related to centrality indices, which quantify node centrality with respect to several properties in the represented network. The nodes identified in such an analysis are called central nodes. Although centrality indices are very useful for these analyses, there exist several challenges regarding which one fits best
for a network. In addition, if the usage of only one index for determining central
nodes leads to under- or overestimation of the importance of nodes and is
insufficient for finding important nodes, then the question is how multiple indices
can be used in conjunction in such an evaluation. Thus, in this thesis an approach is proposed that includes multiple indices of nodes, each indicating
an aspect of importance, in the respective evaluation and where all the aspects of a node’s centrality are analyzed in an explorative manner. To achieve this
aim, the proposed idea uses fuzzy operators, including a parameter for generating different types of aggregations over multiple indices. In addition, several preprocessing methods for normalization of those values are proposed and discussed. We investigate whether the choice of different decisions regarding the
aggregation of the values changes the ranking of the nodes or not. It is revealed that (1) there are nodes that remain stable among the top-ranking nodes, which
makes them the most central nodes, and there are nodes that remain stable
among the bottom-ranking nodes, which makes them the least central nodes; and (2) there are nodes that show high sensitivity to the choice of normalization
methods and/or aggregations. We explain both cases and the reasons why the nodes’ rankings are stable or sensitive to the corresponding choices in various networks, such as social networks, communication networks, and air transportation networks.
Image restoration and enhancement methods that respect important features such as edges play a fundamental role in digital image processing. In the last decades a large
variety of methods have been proposed. Nevertheless, the correct restoration and
preservation of, e.g., sharp corners, crossings or texture in images is still a challenge, in particular in the presence of severe distortions. Moreover, in the context of image denoising many methods are designed for the removal of additive Gaussian noise and their adaptation for other types of noise occurring in practice requires usually additional efforts.
The aim of this thesis is to contribute to these topics and to develop and analyze new
methods for restoring images corrupted by different types of noise:
First, we present variational models and diffusion methods which are particularly well
suited for the restoration of sharp corners and X junctions in images corrupted by
strong additive Gaussian noise. For their deduction we present and analyze different
tensor based methods for locally estimating orientations in images and show how to
successfully incorporate the obtained information in the denoising process. The advantageous
properties of the obtained methods are shown theoretically as well as by
numerical experiments. Moreover, the potential of the proposed methods is demonstrated
for applications beyond image denoising.
Afterwards, we focus on variational methods for the restoration of images corrupted
by Poisson and multiplicative Gamma noise. Here, different methods from the literature
are compared and the surprising equivalence between a standard model for
the removal of Poisson noise and a recently introduced approach for multiplicative
Gamma noise is proven. Since this Poisson model has not been considered for multiplicative
Gamma noise before, we investigate its properties further for more general
regularizers including also nonlocal ones. Moreover, an efficient algorithm for solving
the involved minimization problems is proposed, which can also handle an additional
linear transformation of the data. The good performance of this algorithm is demonstrated
experimentally and different examples with images corrupted by Poisson and
multiplicative Gamma noise are presented.
In the final part of this thesis new nonlocal filters for images corrupted by multiplicative
noise are presented. These filters are deduced in a weighted maximum likelihood
estimation framework and for the definition of the involved weights a new similarity measure for the comparison of data corrupted by multiplicative noise is applied. The
advantageous properties of the new measure are demonstrated theoretically and by
numerical examples. Besides, denoising results for images corrupted by multiplicative
Gamma and Rayleigh noise show the very good performance of the new filters.
In this work we study and investigate the minimum width annulus problem (MWAP), the circle center location or circle location problem (CLP) and the point center location or point location problem (PLP) on Rectilinear and Chebyshev planes as well as in networks. The relations between the problems have served as a basis for finding of elegant solution, algorithms for both new and well known problems. So, MWAP was formulated and investigated in Rectilinear space. In contrast to Euclidean metric, MWAP and PLP have at least one common optimal point. Therefore, MWAP on Rectilinear plane was solved in linear time with the help of PLP. Hence, the solution sequence was PLP-->MWAP. It was shown, that MWAP and CLP are equivalent. Thus, CLP can be also solved in linear time. The obtained results were analysed and transfered to Chebyshev metric. After that, the notions of circle, sphere and annulus in networks were introduced. It should be noted that the notion of a circle in a network is different from the notion of a cycle. An O(mn) time algorithm for solution of MWAP was constructed and implemented. The algorithm is based on the fact that the middle point of an edge represents an optimal solution of a local minimum width annulus on this edge. The resulting complexity is better than the complexity O(mn+n^2logn) in unweighted case of the fastest known algorithm for minimizing of the range function, which is mathematically equivalent to MWAP. MWAP in unweighted undirected networks was extended to the MWAP on subsets and to the restricted MWAP. Resulting problems were analysed and solved. Also the p–minimum width annulus problem was formulated and explored. This problem is NP–hard. However, the p–MWAP has been solved in polynomial O(m^2n^3p) time with a natural assumption, that each minimum width annulus covers all vertexes of a network having distances to the central point of annulus less than or equal to the radius of its outer circle. In contrast to the planar case MWAP in undirected unweighted networks have appeared to be a root problem among considered problems. During investigation of properties of circles in networks it was shown that the difference between planar and network circles is significant. This leads to the nonequivalence of CLP and MWAP in the general case. However, MWAP was effectively used in solution procedures for CLP giving the sequence MWAP-->CLP. The complexity of the developed and implemented algorithm is of order O(m^2n^2). It is important to mention that CLP in networks has been formulated for the first time in this work and differs from the well–studied location of cycles in networks. We have constructed an O(mn+n^2logn) algorithm for well–known PLP. The complexity of this algorithm is not worse than the complexity of the currently best algorithms. But the concept of the solution procedure is new – we use MWAP in order to solve PLP building the opposite to the planar case solution sequence MWAP-->PLP and this method has the following advantages: First, the lower bounds LB obtained in the solution procedure are proved to be in any case better than the strongest Halpern’s lower bound. Second, the developed algorithm is so simple that it can be easily applied to complex networks manually. Third, the empirical complexity of the algorithm is equal to O(mn). MWAP was extended to and explored in directed unweighted and weighted networks. The complexity bound O(n^2) of the developed algorithm for finding of the center of a minimum width annulus in the unweighted case does not depend on the number of edges in a network, because the problems can be solved in the order PLP-->MWAP. In the weighted case computational time is of order O(mn^2).
Thermoplastic polymer-polymer composites consist of a polymeric matrix and a
polymeric reinforcement. The combination of these materials offers outstanding
mechanical properties at lower weight than standard fiber reinforced materials.
Furthermore, when both polymeric components originate from the same family or,
ideally, from the same polymer, their sustainability degree is higher than standard
fiber reinforced composites.
A challenge of polymer-polymer composites is the subsequent processing of their
semi-finished materials by heating techniques. Since the fibers are made of meltable
thermoplastic, the reinforcing fiber structure might be lost during the heating process.
Hence, the mechanical properties of an overheated polymer-polymer composite
would decline, and finally, they would be even lower than the neat matrix. A decrease
of process temperature to manage the heating challenge is not reasonable since the
cycle time would be increased at the same time. Therefore, this work pursues the
adaption of a fast and selective heating method on the use with polymer-polymer
composites. Inductively activatable particles, so-called susceptors, were distributed in
the matrix to evoke a local heating in the matrix when being exposed to an
alternating magnetic field. In this way, the energy input to the fibers is limited.
The experimental series revealed the induction particle heating effect to be mainly
related to susceptor material, susceptor fraction, susceptor distribution as well as
magnetic field strength, coupling distance, and heating time. A proper heating was
achieved with ferromagnetic particles at a filler content of only 5 wt-% in HDPE as
well as with its respective polymer fiber reinforced composites. The study included
the analysis of susceptor impact on mechanical and thermal matrix properties as well
as a degradation evaluation. The susceptors were identified to have only a marginal
impact on matrix properties. Furthermore, a semi-empiric simulation of the particle
induction heating was applied, which served for the investigation of intrinsic melting
processes.
The achieved results, the experimental as well as the analytic study, were
successfully adapted to a thermoforming process with a polymer-polymer material,
which had been preheated by means of particle induction.
The overall goal of the work is to simulate rarefied flows inside geometries with moving boundaries. The behavior of a rarefied flow is characterized through the Knudsen number \(Kn\), which can be very small (\(Kn < 0.01\) continuum flow) or larger (\(Kn > 1\) molecular flow). The transition region (\(0.01 < Kn < 1\)) is referred to as the transition flow regime.
Continuum flows are mainly simulated by using commercial CFD methods, which are used to solve the Euler equations. In the case of molecular flows one uses statistical methods, such as the Direct Simulation Monte Carlo (DSMC) method. In the transition region Euler equations are not adequate to model gas flows. Because of the rapid increase of particle collisions the DSMC method tends to fail, as well
Therefore, we develop a deterministic method, which is suitable to simulate problems of rarefied gases for any Knudsen number and is appropriate to simulate flows inside geometries with moving boundaries. Thus, the method we use is the Finite Pointset Method (FPM), which is a mesh-free numerical method developed at the ITWM Kaiserslautern and is mainly used to solve fluid dynamical problems.
More precisely, we develop a method in the FPM framework to solve the BGK model equation, which is a simplification of the Boltzmann equation. This equation is mainly used to describe rarefied flows.
The FPM based method is implemented for one and two dimensional physical and velocity space and different ranges of the Knudsen number. Numerical examples are shown for problems with moving boundaries. It is seen, that our method is superior to regular grid methods with respect to the implementation of boundary conditions. Furthermore, our results are comparable to reference solutions gained through CFD- and DSMC methods, respectevly.
In this thesis, we focus on the application of the Heath-Platen (HP) estimator in option
pricing. In particular, we extend the approach of the HP estimator for pricing path dependent
options under the Heston model. The theoretical background of the estimator
was first introduced by Heath and Platen [32]. The HP estimator was originally interpreted
as a control variate technique and an application for European vanilla options was
presented in [32]. For European vanilla options, the HP estimator provided a considerable
amount of variance reduction. Thus, applying the technique for path dependent options
under the Heston model is the main contribution of this thesis.
The first part of the thesis deals with the implementation of the HP estimator for pricing
one-sided knockout barrier options. The main difficulty for the implementation of the HP
estimator is located in the determination of the first hitting time of the barrier. To test the
efficiency of the HP estimator we conduct numerical tests with regard to various aspects.
We provide a comparison among the crude Monte Carlo estimation, the crude control
variate technique and the HP estimator for all types of barrier options. Furthermore, we
present the numerical results for at the money, in the money and out of the money barrier
options. As numerical results imply, the HP estimator performs superior among others
for pricing one-sided knockout barrier options under the Heston model.
Another contribution of this thesis is the application of the HP estimator in pricing bond
options under the Cox-Ingersoll-Ross (CIR) model and the Fong-Vasicek (FV) model. As
suggested in the original paper of Heath and Platen [32], the HP estimator has a wide
range of applicability for derivative pricing. Therefore, transferring the structure of the
HP estimator for pricing bond options is a promising contribution. As the approximating
Vasicek process does not seem to be as good as the deterministic volatility process in the
Heston setting, the performance of the HP estimator in the CIR model is only relatively
good. However, for the FV model the variance reduction provided by the HP estimator is
again considerable.
Finally, the numerical result concerning the weak convergence rate of the HP estimator
for pricing European vanilla options in the Heston model is presented. As supported by
numerical analysis, the HP estimator has weak convergence of order almost 1.
Hardware devices fabricated with recent process technology are intrinsically
more susceptible to faults than before. Resilience against hardware faults is,
therefore, a major concern for safety-critical embedded systems and has been
addressed in several standards. These standards demand a systematic and
thorough safety evaluation, especially for the highest safety levels. However,
any attempt to cover all faults for all theoretically possible scenarios that a sys-
tem might be used in can easily lead to excessive costs. Instead, an application-
dependent approach should be taken: strategies for test and fault resilience
must target only those faults that can actually have an effect in the situations
in which the hardware is being used.
In order to provide the data for such safety evaluations, we propose scalable
and formal methods to analyse the effects of hardware faults on hardware/soft-
ware systems across three abstraction levels where we:
(1) perform a fault effect analysis at instruction set architecture level by em-
ploying fault injection into a hardware-dependent software model called
program netlist,
(2) use the results from the program netlist analysis to perform a deductive
analysis to determine “application-redundant” faults at the gate level by
exploiting standard combinational test pattern generation,
(3) use the results from the program netlist analysis to perform an inductive
analysis to identify all faults of a given fault list that can have an effect
on selected objects of the high-level software, such as specified safety
functions, by employing Abstract Interpretation.
These methods aid in the certification process for the higher safety levels
by (a) providing formal guarantees that certain faults can be ignored and (b)
pointing to those faults which need to be detected in order to ensure product
safety.
We consider transient and permanent faults corrupting data in program-
visible hardware registers and model them using the single-event upset and
stuck-at fault models, respectively.
Scalability of our approaches results from combining an analysis at the ma-
chine and hardware level with separate analyses on gate level and C level
source code, as well as, exploiting certain properties that are characteristic for
embedded systems software. We demonstrate the effectiveness and scalability
of each method on industry-oriented software, including a software system
with about 138 k lines of C code.
This thesis discusses several applications of computational topology to the visualization
of scalar fields. Scalar field data come from different measurements and simulations. The
intrinsic properties of this kind of data, which make the visualization of it to a complicated
task, are the large size and presence of noise. Computational topology is a powerful tool
for automatic feature extraction, which allows the user to interpret the information contained
in the dataset in a more efficient way. Utilizing it one can make the main purpose of
scientific visualization, namely extracting knowledge from data, a more convenient task.
Volume rendering is a class of methods designed for realistic visual representation of 3D
scalar fields. It is used in a wide range of applications with different data size, noise
rate and requirements on interactivity and flexibility. At the moment there is no known
technique which can meet the needs of every application domain, therefore development
of methods solving specific problems is required. One of such algorithms, designed for
rendering of noisy data with high frequencies is presented in the first part of this thesis.
The method works with multidimensional transfer functions and is especially suited for
functions exhibiting sharp features. Compared with known methods the presented algorithm
achieves better visual quality with a faster performance in presence of mentioned
features. An improvement on the method utilizing a topological theory, Morse theory, and
a topological construct, Morse-Smale complex, is also presented in this part of the thesis.
The improvement allows for performance speedup at a little precomputation and memory
cost.
The usage of topological methods for feature extraction on a real world dataset often
results in a very large feature space which easily leads to information overflow. Topology
simplification is designed to reduce the number of features and allow a domain expert
to concentrate on the most important ones. In the terms of Morse theory features are
represented by critical points. An importance measure which is usually used for removing
critical points is called homological persistence. Critical points are cancelled pairwise
according to their homological persistence value. In the presence of outlier-like noise
homological persistence has a clear drawback: the outliers get a high importance value
assigned and therefore are not being removed. In the second part of this thesis a new
importance measure is presented which is especially suited for data with outliers. This
importance measure is called scale space persistence. The algorithm for the computation
of this measure is based on the scale space theory known from the area of computer
vision. The development of a critical point in scale space gives information about its
spacial extent, therefore outliers can be distinguished from other critical points. The usage
of the presented importance measure is demonstrated on a real world application, crater
identification on a surface of Mars.
The third part of this work presents a system for general interactive topology analysis
and exploration. The development of such a system is motivated by the fact that topological
methods are often considered to be complicated and hard to understand, because
application of topology for visualization requires deep understanding of the mathematical
background behind it. A domain expert exploring the data using topology for feature
extraction needs an intuitive way to manipulate the exploration process. The presented
system is based on an intuitive notion of a scene graph, where the user can choose and
place the component blocks to achieve an individual result. This way the domain expert
can extract more knowledge from given data independent on the application domain. The
tool gives the possibility for calculation and simplification of the underlying topological
structure, Morse-Smale complex, and also the visualization of parts of it. The system also
includes a simple generic query language to acquire different structures of the topological
structure at different levels of hierarchy.
The fourth part of this dissertation is concentrated on an application of computational
geometry for quality assessment of a triangulated surface. Quality assessment of a triangulation
is called surface interrogation and is aimed for revealing intrinsic irregularities
of a surface. Curvature and continuity are the properties required to design a visually
pleasing geometric object. For example, a surface of a manufactured body usually should
be convex without bumps of wiggles. Conventional rendering methods hide the regions
of interest because of smoothing or interpolation. Two new methods which are presented
here: curvature estimation using local fitting with B´ezier patches and computation of reflection
lines for visual representation of continuity, are specially designed for assessment
problems. The examples and comparisons presented in this part of the thesis prove the
benefits of the introduced algorithms. The methods are also well suited for concurrent visualization
of the results from simulation and surface interrogation to reveal the possible
intrinsic relationship between them.
Photonic crystals are inhomogeneous dielectric media with periodic variation of the refractive index. A photonic crystal gives us new tools for the manipulation of photons and thus has received great interests in a variety of fields. Photonic crystals are expected to be used in novel optical devices such as thresholdless laser diodes, single-mode light emitting diodes, small waveguides with low-loss sharp bends, small prisms, and small integrated optical circuits. They can be operated in some aspects as "left handed materials" which are capable of focusing transmitted waves into a sub-wavelength spot due to negative refraction. The thesis is focused on the applications of photonic crystals in communications and optical imaging: • Photonic crystal structures for potential dispersion management in optical telecommunication systems • 2D non-uniform photonic crystal waveguides with a square lattice for wide-angle beam refocusing using negative refraction • 2D non-uniform photonic crystal slabs with triangular lattice for all-angle beam refocusing • Compact phase-shifted band-pass transmission filter based on photonic crystals
Streams and their adjacent terrestrial ecosystem are tightly linked via the flux of organisms and matter. Emergent aquatic insects can be an important food source for riparian predators like bats, birds, spiders, and lizards. Information about the quality, quantity and phenology of emergent aquatic insects is necessary to estimate how riparian predators can benefit from them as food source. Though intensive agriculture is a globally dominant land use, little is known about how agricultural land use affects the quantity, quality as well as phenology of emergent aquatic insects. Typically, emergent aquatic insects contain more long-chain polyunsaturated fatty acids (PUFA) than terrestrial insects. Especially long-chain PUFA, were shown to enhance growth and immune response of spiders and birds.
In chapter 2, the PUFA transfer to spiders and the effect of food sources differing in their PUFA profiles on spiders was examined in outdoor microcosms under environmentally realistic conditions (i.e., normal weather conditions, possibility to construct orb webs as in their natural habitat). The environmental context determined how PUFA can affect the spiders. For instance, besides PUFA profiles of food sources, environmental variables like the temperature were important for the growth and body condition of spiders.
In the third chapter, the effect of agricultural land use on the quantity in terms of biomass as well as abundance, phenology and composition of emergent aquatic insects was assessed. Previous studies were limited to seasons or single time points, which hampered determining annual biomass export and shifts in phenology. Therefore, emergent aquatic insects were sampled continuously over the primary emergence period of one year and environmental variables associated with agricultural land use were monitored. The biomass and abundance in total were higher (61 – 68 and 79 – 86%, respectively) in agricultural than forested sites. In addition to that, a turn-over of emergent aquatic insect assemblages and a shift in phenology of aquatic insects was identified. In agricultural sites, 71% families of aquatic insects emerged earlier than in forested sites. Pesticide toxicity was associated with different aquatic insect order biomass and abundances. During the same experiment spiders were sampled in spring, summer, and autumn. Additionally, the fatty acid (FA) content of the spiders and emergent aquatic insects was determined. These results are presented in chapter 4. The FA export via emergent aquatic insects was higher (26 – 29%) in forested than agricultural sites, which indicated a reduced quality of aquatic insects as food source for riparian predators in agricultural sites. The FA profiles of mayflies, flies and caddisflies differed between land-use types, but not for spiders. Shading and pool habitats were the most important environmental variables for the FA profiles, though environmental variables explained only little variation in FA profiles. Overall, the quantity, quality and phenology of emergent aquatic insects differed between land-use types, which can affect population dynamics in the adjacent terrestrial ecosystem. Our results can be used in modeling food-web dynamics or meta-ecosystems to improve understanding of linked ecosystems.
Adult emerging aquatic insects can transfer micropollutants, accumulated during their aquatic development, from aquatic to terrestrial ecosystems. This process depends on both contaminant- and organism-specific properties and processes. The transfer of contaminants can result in the dietary exposure of terrestrial insectivores at the aquatic-terrestrial ecosystem boundary. It is, however, unknown whether this route of contaminant transfer is relevant for current-use pesticides, despite their ubiquity in freshwater ecosystems globally. Furthermore, empirical investigation of pesticides in terrestrial insectivores which consume emerging aquatic insects (e.g. riparian spiders) is lacking. In the present work, two laboratory batch-scale studies and a field study were conducted to investigate the transfer of current-use pesticides by emerging aquatic insects and the dietary exposure of riparian spiders preying on emerging insects. In the two laboratory studies, larvae of the model organism, Chironomus riparius, were exposed, either chronically to seven fungicides and two herbicides, or acutely (24-hours) to three individual insecticides during their development. The pesticides were all small organic molecules, selected to cover a low to moderate lipophilicity range (logKow 1.2 – 4.7). Exposure took place at three environmentally relevant concentrations for the fungicides and herbicides (1.2 – 2.5, 17.5 – 35.0 or 50.0 – 100.0 ng/mL) and two for the insecticides (0.1 and either 4 or 16 ng/mL). Eight of the nine fungicides and herbicides, as well as one of the three insecticides were detected in the adult insects after metamorphosis. Concentrations of the pesticides decreased over metamorphosis. However, the transfer of individual pesticides was not well predicted using published models which are based on contaminant lipophilicity andwere developed using other contaminant classes. In the present work, pesticide-specific differences in bioaccumulation by the larvae, retention through metamorphosis and sex-specific bioamplification and elimination over the course of the terrestrial life stage were observed. The neonicotinoid, thiacloprid, was the only insecticide retained by the emerging insects, due to its slow elimination by the larvae. Thiacloprid also decreased insect emergence success. An approximate 30 % higher survival to emergence in the low exposure level (0.1 ng/mL), however, resulted in a relatively higher insecticide flux, from the aquatic to the terrestrial environment compared to the higher exposure (4 ng/mL). For the field study, a method for the analysis of 82 current-use pesticides by high-performance liquid chromatography tandem to triple quadrupole mass spectrometry using small volumes (30 mg) of insect material was validated and applied to samples of emerging insects and Tetragnatha spp. spiders which were collected from stream sites impacted by agricultural activities. Emerging aquatic insects from three orders (Diptera, Ephemeroptera and Trichoptera) contained 27 pesticides whereas 49 pesticides were found in the aquatic environment (water, sediment and aquatic leaf litter). This included mixtures of up to four neonicotinoid insecticides in the insects, with concentrations up to 12300 times greater than were found in the water. Furthermore, the web-building riparian spiders contained 29 pesticides, generally at low concentrations, however concentrations of three neonicotinoids and one herbicide were biomagnified compared to the emerging insects. The three studies included in this thesis thus reveal that the aquatic-terrestrial transfer of current-use pesticides occurs, even at very low environmentally relevant exposure concentrations. Furthermore, new knowledge was generated on the diverse interactions between current-use pesticides and organisms over their entire lifecycles, affecting the propensities for individual pesticides to be transferred via insect emergence. A wide range of pesticides were found to be dietarily bioavailable to riparian spiders, and likely many other riparian insectivores. The neonicotinoid insecticides stood out for their potential to negatively impact adjacent terrestrial food webs through negative impacts on aquatic insect emergence (i.e. biomass flux), while still having a high propensity to be transferred by emerging insects and bioaccumulated in riparian spiders.
The proliferation of sensors in everyday devices – especially in smartphones – has led to crowd sensing becoming an important technique in many urban applications ranging from noise pollution mapping or road condition monitoring to tracking the spreading of diseases. However, in order to establish integrated crowd sensing environments on a large scale, some open issues need to be tackled first. On a high level, this thesis concentrates on dealing with two of those key issues: (1) efficiently collecting and processing large amounts of sensor data from smartphones in a scalable manner and (2) extracting abstract data models from those collected data sets thereby enabling the development of complex smart city services based on the extracted knowledge.
Going more into detail, the first main contribution of this thesis is the development of methods and architectures to facilitate simple and efficient deployments, scalability and adaptability of crowd sensing applications in a broad range of scenarios while at the same time enabling the integration of incentivation mechanisms for the participating general public. During an evaluation within a complex, large-scale environment it is shown that real-world deployments of the proposed data recording architecture are in fact feasible. The second major contribution of this thesis is the development of a novel methodology for using the recorded data to extract abstract data models which are representing the inherent core characteristics of the source data correctly. Finally – and in order to bring together the results of the thesis – it is demonstrated how the proposed architecture and the modeling method can be used to implement a complex smart city service by employing a data driven development approach.
The intensive use of pesticides is one of the main causes for global arthropod decline which can subsequently affect ecosystem services such as pollination, natural pest control, and soil fertility and cascade to higher trophic levels including bats and birds. However, agriculture in large parts is strongly dependent on pesticides, and viticulture in particular is one of the major consumers of fungicides. Fungus-resistant grape varieties offer a very good opportunity to reduce fungicide applications by more than 80 % while maintaining healthy grapes. Here, the effects of fungicide reduction on arthropods and natural pest control were investigated on the one hand in a long- term study in an experimental vineyard and on the other hand in 32 commercially managed vineyards in southwestern Germany. In both designs, fungicide reduction resulted in mostly positive effects on arthropods and natural pest control. Particularly beneficial arthropods such as predatory mites and spiders were promoted by reduced fungicide applications. Contrastingly, potential vineyard pests such as phytophagous mites and leafhoppers decreased under fungicide reduction. Fungus-resistant grape varieties are thus a promising approach to foster resilient agroecosystems and a more sustainable viticulture.
The Wilkie model is a stochastic asset model, developed by A.D. Wilkie in 1984 with a purpose to explore the behaviour of investment factors of insurers within the United Kingdom. Even so, there is still no analysis that studies the Wilkie model in a portfolio optimization framework thus far. Originally, the Wilkie model is considering a discrete-time horizon and we apply the concept of Wilkie model to develop a suitable ARIMA model for Malaysian data by using Box-Jenkins methodology. We obtained the estimated parameters for each sub model within the Wilkie model that suits the case of Malaysia, and permits us to analyse the result based on statistics and economics view. We then tend to review the continuous time case which was initially introduced by Terence Chan in 1998. The continuous-time Wilkie model inspired is then being employed to develop the wealth equation of a portfolio that consists of a bond and a stock. We are interested in building portfolios based on three well-known trading strategies, a self-financing strategy, a constant growth optimal strategy as well as a buy-and-hold strategy. In dealing with the portfolio optimization problems, we use the stochastic control technique consisting of the maximization problem itself, the Hamilton-Jacobi-equation, the solution to the Hamilton-Jacobi-equation and finally the verification theorem. In finding the optimal portfolio, we obtained the specific solution of the Hamilton-Jacobi-equation and proved the solution via the verification theorem. For a simple buy-and-hold strategy, we use the mean-variance analysis to solve the portfolio optimization problem.
In this thesis, we have dealt with two modeling approaches of the credit risk, namely the structural (firm value) and the reduced form. In the former one, the firm value is modeled by a stochastic process and the first hitting time of this stochastic process to a given boundary defines the default time of the firm. In the existing literature, the stochastic process, triggering the firm value, has been generally chosen as a diffusion process. Therefore, on one hand it is possible to obtain closed form solutions for the pricing problems of credit derivatives and on the other hand the optimal capital structure of a firm can be analysed by obtaining closed form solutions of firm's corporate securities such as; equity value, debt value and total firm value, see Leland(1994). We have extended this approach by modeling the firm value as a jump-diffusion process. The choice of the jump-diffusion process was a crucial step to obtain closed form solutions for corporate securities. As a result, we have chosen a jump-diffusion process with double exponentially distributed jump heights, which enabled us to analyse the effects of jump on the optimal capital structure of a firm. In the second part of the thesis, by following the reduced form models, we have assumed that the default is triggered by the first jump of a Cox process. Further, by following Schönbucher(2005), we have modeled the forward default intensity of a firm as a geometric Brownian motion and derived pricing formulas for credit default swap options in a more general setup than the ones in Schönbucher(2005).
Phycobilisomes (PBS) are the major light-harvesting complexes for the majority of cyanobacteria
and allow these organisms to absorb in the so-called green gap. They consist of smaller units called
phycobiliproteins (PBPs), which are composed of an α- and a β-subunit with covalently bound
linear tetrapyrroles (phycobilins). The latter are attached to the apo-PBPs by phycobiliprotein
lyases. Interestingly, cyanobacteria of the genus Prochlorococcus lack complete PBS and instead
use prochlorophyte chlorophyll-binding proteins (Pcbs), which effectively utilize the energy of the
blue light region. The low-light-adapted (LL) strain Prochlorococcus marinus SS120 has a single
PBP, phycoerythrin-III (PE-III). It has been postulated that PE-III is chromophorylated with the
phycobilins phycourobilin (PUB) and phycoerythrobilin (PEB) in a 3:1 ratio. Thereby, the function
of PE-III remains unclear so far, so that light-gathering function and also photoreceptor function
are discussed.
The main goal of this work was to characterize the assembly of PE-III and thus the function of the
six putative phycobiliprotein lyases of P. marinus SS120. Previous work found that the individual
lyases could not be produced in soluble form, so we switched to a dual pDuet™ plasmid system in
E. coli, which was successfully established. Investigation of the binding of PEB to Apo-PE
revealed that the CpeS lyase specifically chromophorylated Cys82 with 3Z-PEB. Unfortunately,
additional chromophorylation could not be observed using the pDuet system. Therefore, in a
second part of the work, the entire PE gene cluster from P. marinus SS120 was to be introduced
into E. coli and expressed. Although the gene cluster was successfully transcribed within E. coli,
no translation was observed, possibly due to incompatible translation initiation between
Prochlorococcus and E. coli. The introduction of a mini PE cluster (CpeAB) into the
cyanobacterium Synechococcus sp. PCC 7002 was also successfully performed, in which case
production of CpeB but not CpeA from Prochlorococcus was detected. Recombinant CpeB was
also detected together with intrinsic PBP in Synechococcussp. 7002, indicating structural similarity
and incorporation into PBS in Synechococcus sp. 7002. Overall, the obtained results suggest that a
cyanobacterial host is a good option for the studies on the assembly of PE-III from P. marinus and,
based on this, future work could aim at generating an artificial operon using synthetic biology to
achieve efficient translation of all genes.
The usage of sensors in modern technical systems and consumer products is in a rapid increase. This advancement can be characterized by two major factors, namely, the mass introduction of consumer oriented sensing devices to the market and the sheer amount of sensor data being generated. These characteristics raise subsequent challenges regarding both the consumer sensing devices' reliability and the management and utilization of the generated sensor data. This thesis addresses these challenges through two main contributions. It presents a novel framework that leverages sentiment analysis techniques in order to assess the quality of consumer sensing devices. It also couples semantic technologies with big data technologies to present a new optimized approach for realization and management of semantic sensor data, hence providing a robust means of integration, analysis, and reuse of the generated data. The thesis also presents several applications that show the potential of the contributions in real-life scenarios.
Due to the broad range, growing feature set and fast release pace of new sensor-based products, evaluating these products is very challenging as standard product testing is not practical. As an alternative, an end-to-end aspect-based sentiment summarizer pipeline for evaluation of consumer sensing devices is presented. The pipeline uses product reviews to extract the sentiment at the aspect level and includes several components namely, product name extractor, aspects extractor and a lexicon-based sentiment extractor which handles multiple sentiment analysis challenges such as sentiment shifters, negations, and comparative sentences among others. The proposed summarizer's components generally outperform the state-of-the-art approaches. As a use case, features of the market leading fitness trackers are evaluated and a dynamic visual summarizer is presented to display the evaluation results and to provide personalized product recommendations for potential customers.
The increased usage of sensing devices in the consumer market is accompanied with increased deployment of sensors in various other fields such as industry, agriculture, and energy production systems. This necessitates using efficient and scalable methods for storing and processing of sensor data. Coupling big data technologies with semantic techniques not only helps to achieve the desired storage and processing goals, but also facilitates data integration, data analysis, and the utilization of data in unforeseen future applications through preserving the data generation context. This thesis proposes an efficient and scalable solution for semantification, storage and processing of raw sensor data through ontological modelling of sensor data and a novel encoding scheme that harnesses the split between the statements of the conceptual model of an ontology (TBox) and the individual facts (ABox) along with in-memory processing capabilities of modern big data systems. A sample use case is further introduced where a smartphone is deployed in a transportation bus to collect various sensor data which is then utilized in detecting street anomalies.
In addition to the aforementioned contributions, and to highlight the potential use cases of sensor data publicly available, a recommender system is developed using running route data, used for proximity-based retrieval, to provide personalized suggestions for new routes considering the runner's performance, visual and nature of route preferences.
This thesis aims at enhancing the integration of sensing devices in daily life applications through facilitating the public acquisition of consumer sensing devices. It also aims at achieving better integration and processing of sensor data in order to enable new potential usage scenarios of the raw generated data.
The Symbol Grounding Problem (SGP) is one of the first attempts to proposed a hypothesis about mapping abstract concepts and the real world. For example, the concept "ball" can be represented by an object with a round shape (visual modality) and phonemes /b/ /a/ /l/ (audio modality).
This thesis is inspired by the association learning presented in infant development.
Newborns can associate visual and audio modalities of the same concept that are presented at the same time for vocabulary acquisition task.
The goal of this thesis is to develop a novel framework that combines the constraints of the Symbol Grounding Problem and Neural Networks in a simplified scenario of association learning in infants. The first motivation is that the network output can be considered as numerical symbolic features because the attributes of input samples are already embedded. The second motivation is the association between two samples is predefined before training via the same vectorial representation. This thesis proposes to associate two samples and the vectorial representation during training. Two scenarios are considered: sample pair association and sequence pair association.
Three main contributions are presented in this work.
The first contribution is a novel Symbolic Association Model based on two parallel MLPs.
The association task is defined by learning that two instances that represent one concept.
Moreover, a novel training algorithm is defined by matching the output vectors of the MLPs with a statistical distribution for obtaining the relationship between concepts and vectorial representations.
The second contribution is a novel Symbolic Association Model based on two parallel LSTM networks that are trained on weakly labeled sequences.
The definition of association task is extended to learn that two sequences represent the same series of concepts.
This model uses a training algorithm that is similar to MLP-based approach.
The last contribution is a Classless Association.
The association task is defined by learning based on the relationship of two samples that represents the same unknown concept.
In summary, the contributions of this thesis are to extend Artificial Intelligence and Cognitive Computation research with a new constraint that is cognitive motivated. Moreover, two training algorithms with a new constraint are proposed for two cases: single and sequence associations. Besides, a new training rule with no-labels with promising results is proposed.