Refine
Year of publication
Document Type
- Article (713) (remove)
Has Fulltext
- yes (713) (remove)
Keywords
- AG-RESY (42)
- PARO (30)
- SKALP (15)
- Schule (12)
- MINT (11)
- Mathematische Modellierung (11)
- Stadtplanung (9)
- Denkmäler (8)
- HANDFLEX (8)
- Monitoring (8)
Faculty / Organisational entity
- Kaiserslautern - Fachbereich Maschinenbau und Verfahrenstechnik (150)
- Kaiserslautern - Fachbereich Informatik (134)
- Kaiserslautern - Fachbereich Physik (100)
- Kaiserslautern - Fachbereich Mathematik (82)
- Kaiserslautern - Fachbereich Sozialwissenschaften (53)
- Kaiserslautern - Fachbereich Biologie (47)
- Kaiserslautern - Fachbereich Chemie (39)
- Kaiserslautern - Fachbereich Raum- und Umweltplanung (27)
- Kaiserslautern - Fachbereich Elektrotechnik und Informationstechnik (26)
- Kaiserslautern - Fachbereich Bauingenieurwesen (23)
Several governmental organizations all over the world aim for algorithmic accountability of artificial intelligence systems. However, there are few specific proposals on how exactly to achieve it. This article provides an extensive overview of possible transparency and inspectability mechanisms that contribute to accountability for the technical components of an algorithmic decision-making system. Following the different phases of a generic software development process, we identify and discuss several such mechanisms. For each of them, we give an estimate of the cost with respect to time and money that might be associated with that measure.
With the transition of fluid-capillary-based “Lab on a chip 1.0″ concepts in analytical chemistry to “Lab on a chip
2.0″ approaches relying on distinct fluid droplets (“digital microfluidics”, DMF), the need for reliable methods for
droplet actuation has increasingly come into focus. One possible approach is based on “electrowetting on
dielectric” (EWOD). This technique has the disadvantage that any possible desired later positions of the droplets
on the chip have to be defined prior to chip realization because one of the EWOD electrode layers has to be
structured accordingly. “Optoelectrowetting” (OEW) goes a step further in the sense that the later droplet positions
do not have to be known before, and none of the electrode layers has to be structured. Instead, the
electrical parameters of the layer sequence can be altered locally by an impinging (and movable) light spot.
Although some research groups have succeeded in demonstrating OEW actuation of droplets, the optimization of
the relevant parameters of the layer sequence and the droplet – at least half a dozen parameters altogether – is
tedious and not straight-forward. In this contribution, for optimization purposes, the equations governing OEW
are revisited and altered again, e.g., by numerical implementation of the experimentally well-known saturation
of the contact angle change. Additionally, a Nelder-Mead algorithm is applied to find the parameters, on which
the optimization has to focus to maximize contact angle changes and, thus, mechanical forces on the droplets.
The numerical investigation yields diverse results, e.g., the finding that the droplet’s contact area on the
dielectric layer has a strong influence on the contact angle change and the question whether the droplet is pulled
or pushed. Moreover, the interplay between frequency and amplitude of the applied rectangular alternate voltage
is important for optimization.
We report on generation of pulsed broadband terahertz radiation utilizing the inverse spin hall effect in Fe/Pt bilayers on MgO and sapphire substrates. The emitter was optimized with respect to layer thickness, growth parameters, substrates and geometrical arrangement. The experimentally determined optimum layer thicknesses were in qualitative agreement with simulations of the spin current induced in the ferromagnetic layer. Our model takes into account generation of spin polarization, spin diffusion and accumulation in Fe and Pt and electrical as well as optical properties of the bilayer samples. Using the device in a counterintuitive orientation a Si lens was attached to increase the collection efficiency of the emitter. The optimized emitter provided a bandwidth of up to 8 THz which was mainly limited by the low-temperature-grown GaAs (LT-GaAS) photoconductive antenna used as detector and the pulse length of the pump laser. The THz pulse length was as short as 220 fs for a sub 100 fs pulse length of the 800 nm pump laser. Average pump powers as low as 25 mW (at a repetition rate of 75 MHz) have been used for terahertz generation. This and the general performance make the spintronic terahertz emitter compatible with established emitters based on optical rectification in nonlinear crystals.
Individual nitrogen vacancy (NV) color centers in diamond are versatile, spin-based quantum sensors. Coherently controlling the spin of NV centers using microwaves in a typical frequency range between 2.5 and 3.5 GHz is necessary for sensing applications. In this work, we present a stripline-based, planar, Ω-shaped microwave antenna that enables one to reliably manipulate NV spins. We found an optimal antenna design using finite integral simulations. We fabricated our antennas on low-cost, transparent glass substrate. We created highly uniform microwave fields in areas of roughly 400 × 400 μm2 while realizing high Rabi frequencies of up to 10 MHz in an ensemble of NV centers.
The move away from fossil fuels and the diversification of the primary energy sources used are imperative both in terms of mitigating global warming and ensuring the political independence of the Western world. For the industries of agriculture and forestry, it is possible to secure the basic energy supply through their own yield. The use of vegetable oil is a possibility to satisfy the energy requirements for agricultural machines both autonomously and sustainably. Up to now, rapeseed has been the most important plant for oil production in Western Europe. In the EU, rapeseed oil is currently credited with up to 60% fossil CO2 savings compared to conventional diesel fuel. As a result, since 2018, rapeseed oil is no longer considered as biofuel in the EU. However, if cultivation and processing are completely based on renewable energy sources, up to 90% of fossil CO2 emissions can be saved in the future. This also applies to rapeseed oil, which is a by-product of animal feed production. In addition, pure rapeseed oil is chemically unchanged and thus biodegradable, which makes it particularly attractive for use in environmentally sensitive areas.
To increase the attractiveness of rapeseed oil as a fuel for the agricultural industry, a multi-fuel concept for the flexible use of rapeseed oil, diesel fuel and any mixtures of these two fuels would be beneficial, as it minimizes economic risks due to price fluctuations, availability, and taxation. For implementing such a concept, technical adjustments to the propulsion system are necessary. In existing vegetable oil vehicles, cost-intensive additional components are required for diesel particulate filter regeneration. Conventional regeneration via post-injected fuel (which does not participate in combustion) leads to dilution of the engine oil with vegetable oil.
This study elaborates the possibilities of DPF regeneration in vegetable oil operation by internal engine measures without the need for post-injection. This includes strategies for generating exhaust gas temperatures in high-idle operation which are suitable for regeneration. For this purpose, strategies combining throttling and retarded combustion are used. The measures were successfully tested with respect to their effectiveness for DPF regeneration. It could also be proved that no increased engine oil dilution occurs as a result of the regeneration procedure.
For a prospective series application, however, regeneration should also be possible in transient engine operation. For this purpose, the measures developed for high-idle regeneration have been transferred to partial load points to gain insight into their applicability for transient engine operation. In addition, the effect of external EGR on regeneration has been considered. As the previous investigations of high-idle regeneration showed that regeneration is most critical when pure rapeseed oil is used, the studies of regeneration in part-load operation were limited to pure rapeseed oil. The systematic parameter variations carried out during the studies helped to improve the understanding of the system and the mechanisms of regeneration. The results of the investigation show that the exhaust gas temperature can be increased significantly by the measures studied. However, achieving the exhaust temperature required for DPF regeneration remains a challenge for certain operating points.
Due to an excellent ratio of high strength to low density, as well as a strong corrosion resistance, the titanium alloy Ti-6Al-4 V is widely used in industrial applications. However, Ti-6Al-4 V is also a difficult-to-cut material because of its low thermal conductivity and high chemical reactivity, especially at elevated temperatures. As a result, machining Ti-6Al-4 V is characterized by high thermal loads and a rapidly progressing thermo-chemical induced tool wear. An adequate cooling strategy is essential to reduce the thermal load and therefore tool wear. Sub-zero metalworking fluids (MWF) which are applied at liquid state but at supply temperatures below the ambient temperature, offer great potential to significantly reduce the thermal load when machining Ti-6Al-4 V. Within the presented research, systematically varied sub-zero cooling strategies are applied when milling Ti-6Al-4 V. The influences of the supply temperature, as well as the volume flow and the outlet velocity are investigated aiming at a reduction of the thermal loads that occur during milling. The milling experiments were recorded using high-speed cameras in order to characterize the impact of the cooling strategies and resolve the behavior of the MWF. Additionally, the novel sub-zero cooling approach is compared to a cryogenic CO2 cooling strategy. The results show that the optimized sub-zero cooling strategy led to a sufficient reduction of the thermal loads and does outperform the cryogenic cooling even at elevated CO2 mass flows.
Insurance companies and banks regularly have to face stress tests performed by regulatory instances. To model their investment decision problems that includes stress scenarios, we propose the worst-case portfolio approach. Thus, the resulting optimal portfolios are already stress test prone by construction. A central issue of the worst-case portfolio approach is that neither the time nor the order of occurrence of the stress scenarios are known. Even more, there are no probabilistic assumptions regarding the occurrence of the stresses. By defining the relative worst-case loss and introducing the concept of minimum constant portfolio processes, we generalize the traditional concepts of the indifference frontier and the indifference-optimality principle. We prove the existence of a minimum constant portfolio process that is optimal for the multi-stress worst-case problem. As a main result we derive a verification theorem that provides conditions on Lagrange multipliers and nonlinear ordinary differential equations that support the construction of optimal worst-case portfolio strategies. The practical applicability of the verification theorem is demonstrated via numerical solution of various worst-case problems with stresses. There, it is in particular shown that an investor who chooses the worst-case optimal portfolio process may have a preference regarding the order of stresses, but there may also be stress scenarios where he/she is indifferent regarding the order and time of occurrence.
A transition from conventional centralized to hybrid decentralized systems has been increasingly
advised recently due to their capability to enhance the resilience and sustainability of
urban water supply systems. Reusing treated wastewater for non-potable purposes is a promising
opportunity toward the aforementioned resolutions. In this study, we present two optimization
models for integrating reusing systems into existing sewerage systems to bridge the supply–demand
gap in an existing water supply system. In Model-1, the supply–demand gap is bridged by introducing
on-site graywater treatment and reuse, and in Model-2, the gap is bridged by decentralized
wastewater treatment and reuse. The applicability of the proposed models is evaluated using two
test cases: one a proof-of-concept hypothetical network and the other a near realistic network based
on the sewerage network in Chennai, India. The results show that the proposed models outperform
the existing approaches by achieving more than a 20% reduction in the cost of procuring water and
more than a 36% reduction in the demand for freshwater through the implementation of local on-site
graywater reuse for both test cases. These numbers are about 12% and 34% respectively for the
implementation of decentralized wastewater treatment and reuse.
We consider the optimization problem of a large insurance company that wants to maximize the expected utility of its surplus through the optimal control of the proportional reinsurance. In addition, the insurer is exposed to the risk of default of its reinsurer at the worst possible time, a setting that is closely related to a scenario of the Swiss Solvency Test.
In this paper we describe a framework for defining and operationalizing conceptual models of distributed knowledge-based systems which extends published approaches by the notion of ,agents" and multiple task decompositions. The main part deals with techniques underlying our distributed interpreter. We show how a client-server-architecture can be implemented which allows prototyping distributed knowledge-based systems. Further we describe our mechanism which manages task interactions and supports dependency-directed backtracking efficiently.