Kaiserslautern - Fachbereich Maschinenbau und Verfahrenstechnik
Refine
Year of publication
Document Type
- Doctoral Thesis (370)
- Article (150)
- Periodical Part (29)
- Preprint (17)
- Habilitation (9)
- Diploma Thesis (3)
- Conference Proceeding (1)
- Periodical (1)
- Study Thesis (1)
- Working Paper (1)
Keywords
- Phasengleichgewicht (12)
- Finite-Elemente-Methode (11)
- finite element method (11)
- Flüssig-Flüssig-Extraktion (9)
- Tribologie (8)
- Messtechnik (6)
- Modellierung (6)
- Faser-Kunststoff-Verbunde (5)
- Kontinuumsmechanik (5)
- Polymere (5)
Faculty / Organisational entity
Diese Dissertation erläutert die Umsetzung eines RAMI 4.0 konformen Marktplatz in der spanenden Bearbeitung. Ziel ist es einen Lösungsansatz zu definieren, in dem firmenübergreifende Prozessketten für kleine Losgrößen automatisiert identifiziert werden und die Fertigung eines individuellen Produktes realisiert wird. Die Extraktion von Produktinformationen, die Fertigung eines individualisierten Produktes sowie die Beschreibung der Informationen in den Verwaltungsschalen wird validiert. Vor allem stellt sich als Herausforderung für die Zukunft heraus, eine gemeinsame Semantik für die Beschreibung von Capabilities zu definieren. Diese würde ermöglichen, dass ein Matching zwischen proprietären Produktinformationen und Skills möglich wird.
This thesis outlines the development of thermoplastic-graphite based plate heat exchangers from material screening to operation including performance evaluation and fouling investi-gations. Polypropylene and polyphenylene sulfide as matrix and graphite as filler were cho-sen as feedstock materials, as they possess a low density and excellent corrosion resistance at a comparatively low price.
For the purpose of material screening, custom-made polymer composite plates with a plate thickness of 1-2 mm and a filler content of up to 80 wt.% were investigated for their thermal and mechanical suitability with regard to their use in plate heat exchangers. Three-point flexural tests show that the loading of polypropylene with graphite leads to mechanical prop-erties that allow the composites to be applied as corrugated heat exchanger plates. The simu-lated maximum overpressure is greater than 7 bar, depending on the wall thickness. The thermal conductivity of the composites was increased by a factor of 12.5 compared to pure polypropylene, resulting in thermal conductivities of up to 2.74 W/mK.
The fabrication of the developed corrugated heat exchanger plates, with a thickness between 0.85 mm and 2.5 mm and a heat transfer surface area of 11.13·10-3 m² was carried out via processes that can be automized, namely extrusion and embossing. With the manufactured plate heat exchanger, overall heat transfer coefficients are determined over a wide range of operating conditions (Re = 200 - 1600), which are used to validate a plate heat exchanger model and consequently to compare the composites with conventional materials. The em-bossing, which seems to result in a shift of the internal graphite structure, leads to a further improvement of the thermal conductivity by 7-20 %, in addition to the impact of the filler. With low plate thicknesses, overall heat transfer coefficients of up to 1850 W/m²K could be obtained. Considering the low density of the manufactured thermal plates, this ensures com-parable performance with metallic materials over a wide range of process conditions (Re = 200 - 4000).
The fouling kinetics and amount of calcium sulfate and calcium carbonate, respectively, on different polypropylene/graphite composites in a flat plate heat exchanger and the developed chevron type plate heat exchanger are determined and compared to the reference material stainless steel. For a straight evaluation of the fouling susceptibility of the materials the for-mation of bubbles on the materials is considered by optical imaging or excluded by a degas-ser. The results are interpreted using surface free energy and roughness of the surfaces. The results show that if bubble formation is avoided, the polymer composites have a very low fouling tendency compared to stainless steel, which is attributed to the low surface free ener-gies of approximately 25 mN/m. This is particularly the case when turbulent flows are pre-sent, as is in plate heat exchangers or when sandblasted specimen are used. Sandblasting also continues to increase heat transfer compared to untreated samples by increasing thermal conductivity and creating local turbulences. Depending on the test conditions, the fouling resistance formed on the stainless steel surface is an order of magnitude greater than on the flat plate polymer composites. In addition, the fouling layers adhere only weakly to the com-posites, which indicates an easy cleaning in place after the formation of deposits. The fouling investigations in the plate heat exchanger reveal sensitivity to calcium sulfate fouling, how-ever, CFD simulations indicate that this is due to flow maldistribution and not the actual pol-ymer composite materials.
Cyber-physische Produktionssysteme (CPPS) ermöglichen die Herstellung kundenindividueller Produkte in kleinen Losgrößen durch Nutzung aktueller Entwicklungen der Informations- und Kommunikationstechnologien. Im Materialfluss in CPPS ist jedoch aufgrund unterschiedlicher physikalischer Eigenschaften der Fördergüter und dynamischer Prozesszuweisungen die Gefahr physikalisch bedingter Störungen erhöht. Diese Arbeit untersucht die Nutzung von Physiksimulation als Basis eines Digitalen Zwillings von Fördermitteln, um diesen Herausforderungen zu begegnen. Das Ziel besteht darin, durch die Simulation der physikalischen Phänomene einzelner Materialflussprozesse die negativen Einflüsse von Störungen zu verringern und somit die Leistungsfähigkeit des Produktionssystems zu erhöhen. Hierzu findet zunächst eine konzeptionelle Entwicklung des Digitalen Zwillings statt, die eine Analyse der beteiligten Systeme, eine Anforderungsdefinition, eine Festlegung von Aufbau- und Ablaufstruktur, sowie eine Formalisierung der einzelnen Funktionsbestandteile umfasst. Im Anschluss wird der Digitale Zwilling softwaretechnisch implementiert, mit einem exemplarischen Fördermittel vernetzt und prototypisch in Betrieb genommen. Die Ergebnisse zeigen die Eignung der Physiksimulation für den beschriebenen Zweck und die Wirksamkeit des Einsatzes auf Produktionssystemebene, indem Materialflussprozesse beschleunigt durchgeführt, überwacht und im Falle von Störungen nachträglich simulativ untersucht werden können.
In contrast to motorbike tyres, whose friction during cornering has to be as high as possible, the desired effect in skiing is the opposite, that of low friction. The reduced friction between skis and ice or snow is made possible by a film of meltwater that forms as a function of friction power. To support this friction mechanism, skis are waxed with different waxes in both hobby and professional sports, depending on a variety of conditions. Waxes with fluorine additives show best performance in most conditions, corresponding to the lowest friction coefficients. However, for health and environmental reasons, the International Ski Federation (FIS) and the Biathlon Un-ion (IBU) have imposed a complete ban on fluorine additives at all FIS races and IBU events with effect from the 2023/2024 season. As a result, wax manufacturers are required to develop and extensively test fluorine-free waxes in order to remain competitive.
Traditional tests take place either indoors or outdoors in the field. Athletes, who complete a particular distance and whose time is measured, also note the impres-sions that the prepared skis provide to the skiers. The time and cost involved in nu-merous individual tests is a drawback, and the presence of only a single type of snow in the hall or field, air resistance, changing environmental conditions and var-iations in the athlete's movement, limit the depth of information. For the need of re-ducing the time-consuming procedure of indoor and outdoor tests, a tribometer of-fers a solution where friction measurements can be performed on a laboratory scale. Due to the consistent adjustable conditions such as temperature, speed and load applied to the friction partners, scientific studies can be carried out with reduced dis-turbance variables. At present, the tribometric results of laboratory instruments for predicting friction values do not translate into application in practice. The reasons for this are the compromises that have to be made in the design of the tribometers.
This work reviews the existing tribometers for their operating conditions and con-firms the need for a scientific method of characterising different waxes. In order to fill the gap between friction results obtained in laboratory tests which cannot yet be used in the selection of waxes, and traditional field tests, this thesis is dedicated to the methodical design and manufacture of a linear tribometer capable of measuring friction between a ski base made of UHMWPE (ultra high molecular weight polyeth-ylene) and an ice sample. The tribometer provides for the first time results that allow differentiating be-tween different modified waxes with regard to their running performance. Friction-influencing factors such as speed, temperature and the surface pressure below the ski base can be adjusted within the range relevant for ski sports. Furthermore, the laboratory-scale test stand, which is located in a cold chamber, is capable of ac-commodating not only typical ski jumping base lengths and widths, but also cross-country and alpine ski bases. To verify the tribometer, a ski base is treated with three waxes of different fluorine content and measured comparatively. With a minimum of 95% confidence, the friction differences between the tested waxes depending on their fluorine content is validated and proven at the end of this work.
Infobrief FBK 71/24
(2024)
Production, purification and analysis of novel peptide antibiotics from terrestrial cyanobacteria
(2024)
Cyanobacteria are a known source for bioactive compounds, of which several also show antibiotic activity. In regard to the growing number of multi-resistant pathogens, the search for novel antibiotic substances is of great importance and unexploited sources should be explored. So, this thesis initially dealt with the identification of productive strains, especially within the group of the terrestrial cyanobacteria, which are less well studied than marine and freshwater strains. Amongst these, Chroococcidiopsis cubana, an extremely desiccation and radiation tolerant, unicellular cyanobacterium was found to produce an extracellular antimicrobial metabolite effective against the Gram-positive indicator bacterium Micrococcus luteus as well as the pathogenic yeast Candida auris. However, as the sole identification of a productive cyanobacterium is not sufficient for further analysis and a future production scale-up, the second part of this thesis targeted the identification of compound synthesis prerequisites. As a result, a limitation of nitrogen was shown to be the production trigger, a finding that was used for the establishment of a continuous production system. The increased compound formation was then used for purification and analysis steps. As a second approach, in silico identified bacteriocin gene clusters from C. cubana were cloned and heterologously expressed in Escherichia coli. By this, the bacteriocin B135CC was identified as a strong bacteriolytic agent, active predominantly against the Gram-positive strains Staphylococcus aureus and Mycobacterium phlei. The peptide showed no cytotoxic effects against mouse neuroblastoma (N2a-) cells and a high temperature tolerance up to 60 °C. In order to facilitate the whole project, two standard protocols, specifically adapted for the work with cyanobacteria, were established. First, a method for a quick and easy in vivo vitality estimation of phototrophic cells and second, an approach for a high throughput determination of nitrate concentrations in microalgal cultures. Both methods greatly helped to proceed the main objectives of this work, the first one by simplifying the development of suitable cryopreservation protocols for individual cyanobacteria strains and the second one by accelerating the determination of the optimal nitrate concentration for the production of the antimicrobial compound from C. cubana. In the course of this cultivation optimization, the ability of cyanobacteria to utilize organic carbon sources for an accelerated cell growth was examined in greater detail. It could be shown that C. cubana reaches significantly higher growth rates when mixotrophically cultivated with fructose or glucose. Interestingly, this effect was even further enhanced when light intensity was decreased. Under these low-light conditions, phototrophically cultivated C. cubana cells showed a clearly decreased cell growth. This effect might be extremely useful for a quick and economic preparation of precultures.
Industrial robots are vital in automation technology, but their limitations become evident in applications requiring high path accuracy. This research focuses on improving the dynamic path accuracy of industrial robots by integrating additional sensor technology and employing intelligent feed-forward control. Specifically, the inclusion of secondary encoder sensors enables explicit measurement and compensation of robot gear deformations. Three types of model-based feed-forward controllers, namely physics-based, data-based, and hybrid, are developed to effectively counteract dynamic effects.
Firstly, a physics-based feed-forward control method is proposed, explicitly modeling joint deformations, hydraulic weight compensation, and other relevant features. Nonlinear friction parameters are accurately identified using a globally optimized design of experiments. The resulting physics-based model is fully continuously differentiable, facilitating its transformation into a code-optimized flatness-based feed-forward control.
Secondly, a data-based feed-forward control approach is introduced, leveraging a continuous-time neural network. The continuous-time approach demonstrates enhanced model generalization capabilities even with limited data. Furthermore, a time domain normalization method is introduced, significantly improving numerical properties by concurrently normalizing measurement timelines, robot states, and state derivatives. Based on previous work, a method ensuring input-to-state and global-asymptotic stability is presented, employing a Lyapunov function. Model stability is enforced already during training using constrained optimization techniques. Moreover, the data-based methods are evaluated on public benchmarks, extending its applicability beyond the field of robotics.
Both the physics-based and data-based models are combined into a hybrid model. Comparative analysis of the three models reveals that the continuous-time neural network yields the highest model accuracy, while the physics-based model delivers the best safety properties. The effectiveness of all three models is experimentally validated using an industrial robot.
Die Interaktion zwischen Prozess, Werkzeug, Spindel und Maschine kann die erreichbare Bearbeitungsgenauigkeit spanender Bearbeitungsverfahren beeinflussen. Bei der spanenden Mikrobearbeitung sind die Größen- und Kraftverhältnisse zwischen Span, Werkzeug und Werkzeugmaschine im Vergleich zur spanenden Bearbeitung mit Werkzeuggrößen über einem Millimeter jedoch grundlegend unterschiedlich. Aufgrund dessen können dort gewonnene Erkenntnisse nicht ohne Weiteres für die spanende Mikrobearbeitung adaptiert werden. So gilt es für die spanende Mikrobearbeitung gesondert zu identifizieren, welche Effekte und Faktoren die erreichbare Bearbeitungsgenauigkeit beeinflussen. Die veränderten Größenverhältnisse, Eingriffsverhältnisse und eingesetzten Maschinenkomponenten erschweren jedoch eine experimentelle Untersuchung. Eine simulationsgestützte Analyse des Prozesses und der Maschinenkomponenten kann deshalb maßgeblich dazu beitragen, die Interaktion zwischen Prozess, Werkzeug, Spindel und Maschine bei der spanenden Mikrobearbeitung zu verstehen.
Diese Arbeit präsentiert simulationsgestützte Methoden zur Analyse der Interaktion zwischen Prozess, Werkzeug, Spindel und Maschine bei der spanenden Mikrobearbeitung. Darauf aufbauend werden die Interaktion zwischen Spindelwelle und Elektromotor sowie die Interaktion zwischen Prozess, Werkzeug, Spindel und Maschine für das Mikrofräsen und Mikroschleifen untersucht. Zwischen der Spindelwelle und dem Elektromotor kann keine Interaktion identifiziert werden. Stattdessen liegt ein nicht vernachlässigbarer unidirektionaler Einfluss des Elektromotors auf die Spindelwelle vor. Ebenso konnte eine unidirektionale Beeinflussung des Werkzeugs durch die Werkzeugspindel ermittelt werden. Zwischen dem Prozess und dem Werkzeug kommt es zu einer Interaktion. Jedoch beschränkt sich diese Interaktion auf das Werkzeug, das heißt, die Spindelwelle wird nicht vom Werkzeug beeinflusst. Insgesamt zeigt sich, dass bei der spanenden Mikrobearbeitung nicht nur die Auftrennung der Werkzeugmaschine und des Spindel-Werkzeug-Systems zweckmäßig ist, sondern dass auch das Werkzeug und die Werkzeugspindel als separate Aspekte betrachtet werden müssen.
Die funktionale Wechselwirkung zwischen geometrischen Oberflächeneigenschaften und dem daraus resultierenden Haftreibwert wird in der vorliegenden Arbeit anhand von mechanisch bearbeiteten Stahloberflächen untersucht. Dabei wird der Fokus neben einer umfangreichen Analyse der Einflussfaktoren auf die Oberflächencharakterisierung gelegt. Basierend auf Drückversuchen und der Untersuchung der Oberflächendeformation wird eine Methode zur funktional relevanten Beschreibung der Oberfläche entwickelt. Die am Haftreibwert beteiligten Oberflächenanteile sind durch die Parameter Inselanzahl, projizierte Durchschnittsoberfläche und Durchschnittsmaterialvolumen beschrieben. Diese Kenngrößen fließen in eine mathema-tische Berechnung eines theoretischen Haftreibwertes ein. Es werden der theoretisch errech-nete und der aus einer statistischen Versuchsreihe ermittelte Haftreibwert miteinander vergli-chen. Statistische Untersuchungen sowie die Aufstellung eines Messunsicherheitsbilanz stüt-zen die Forschungsergebnisse. Damit leistet diese Arbeit nicht nur einen Beitrag zur funktion-sorientierten Oberflächenbeschreibung, sondern auch zur methodischen Korrelations-/ Re-gressionsanalyse und zur Integration geometrischer Oberflächenparameter in Haftreibwert-untersuchungen.
In recent years, the automotive industry has shifted from purely combustion engine-driven vehicles towards hybridization due to the introduction of CO2 emission legislation. Hybrid powertrains also represent an important pillar and starting point in the journey towards zero-emission and full electrification. Fulfilling the most recent emission standards requires efficient control strategies for the engine, capable of real-time operation. Model accuracy is one of the main parameters which directly influence the performance of such control strategies. Specific methodologies developed in the past, such as physically- or phenomenologically-based approaches, have already facilitated the modeling of the combustion engine. Even though these models can accurately predict emissions in steady state conditions, their performance during transient engine operation is time-consuming and still not sufficiently reliable. The major contribution of the current work is to clarify and apply the recent advancements in data-driven modeling techniques, especially in time series forecasting with feedforward neural networks (FFNNs) and long short-term memory networks (LSTMs), to address the limitations mentioned above and to compare the different approaches. The quantity and quality of data are significant challenges for data-driven modeling. This paper studies the modeling of gasoline engine emissions using FFNNs and LSTMs. The data quantity and quality requirements are studied based on a portable emission measurement system (PEMS), measuring at 1 Hz, and additional analyses on an engine test bench with a HiL setup, providing the possibility of increasing the measurement frequency with more sophisticated devices by a factor of five. Subsequently, the training and validation of the FFNNs and LSTMs are outlined, and finally, the model accuracy is discussed.