Refine
Document Type
- Doctoral thesis (12)
- Bachelor thesis (1)
- Conference proceeding (1)
Has Fulltext
- yes (14)
Is part of the Bibliography
- no (14)
Year of publication
Keywords
- Modellierung (14) (remove)
Institute
- FG Aufbereitungstechnik und Sekundärrohstofftechnologie (ehemals) (3)
- FG Thermische Energietechnik (3)
- FG Datenstrukturen und Softwarezuverlässigkeit (1)
- FG Halbleitermaterialien (1)
- FG Ingenieurmathematik und Numerik der Optimierung (1)
- FG Luftchemie und Luftreinhaltung (1)
- FG Rechnernetze und Kommunikationssysteme (1)
- FG Strukturmechanik und Fahrzeugschwingungen (1)
- FG Technische Informatik (1)
Energy efficiency is vital for future low-power electronic applications. This ultra-low power consumption requirement enables the research beyond the conventional charge-based memories. Further, reliability, high scalability, fast switching, CMOS compatibility, high endurance, etc., are some of the characteristics envisaged by the new generation of emerging non-volatile memories (NVMs). A memristor or OxRAM is one among the many emerging NVMs, which can exhibit the aforementioned characteristics, and it has the potential to replace the power-hungry conventional NVMs.
The memristive devices have the advantage of monolithic integration with the CMOS logic, which enables the widening of their application areas. Despite their various advantages, the reliability, forming voltages, and variability of the devices pose a hurdle to their wide commercial usage. Hence, it is crucial to identify these factors and mitigate them. This thesis addresses these issues through fabrication process improvements, electrical characterization techniques, and device-engineering methods.
The improvements in the fabrication processes reduced the pristine state currents of the memristive devices. It impacted the reliability and resistive switching performance of the memristive devices directly. To further improve the performance, the memristive devices are integrated into the 130 nm BiCMOS baseline technology of IHP. Additionally, dedicated test structures are developed to monitor and control the fabrication process steps through in-line electrical characterization.
Further, the forming current and voltage values, along with their dispersions in the 4 kbit memristive arrays, were reduced by utilizing the electrical characterization techniques. Accordingly, the forming operations were performed at high operating temperatures using incremental step pulse and verify algorithm (ISPVA). In contrast to the well-known method of increasing the current compliance (1R) or the gate voltage of the transistor (1T-1R) to increase the conduction filament size, a thin layer of Al2O3 is added. This device engineering technique reduced the variability in both LRS and HRS currents of the memristive devices. Additionally, the conduction filament properties in both states are modeled by using the quantum point contact (QPC) model. Finally, harnessing the intrinsic variability of the memristive devices for neuromorphic computing applications is demonstrated. The reliability of the devices is assessed through endurance and retention characteristics.
1. Memristor-Symposium 2023
(2023)
The 1.Memristor-Symposium was held from 27-28.02.2023 in Bamberg. Young scientists and PhD students presented their research and covered several topics regarding ongoing memristive research in Germany. Several talks on the topic of memristive devices, their testing, modeling and usage were presented and discussed, covering the whole vertical integration from material and divice up to application level.
Recent challenges and novel approaches of wireless communication networks are characterized by high performance requirements on the radio channel and by concepts of self-organization, in-network processing, or resource optimization which ultimately lead to distributed network applications, communication architectures, and radio transceivers.
Associated evaluation is driven by a multitude of ever-increasing requirements that call for multidisciplinary expertise. Depending on the discipline, simulation is one of the most widely used technologies, whereby often abstract assumptions and models do not allow for sufficiently accurate, comparable results.
In contrast, real-world measurements and field-tests can only be performed on actual systems, commonly under non-reproducible conditions.
This thesis establishes with the Split-Protocol-Stack a new type of evaluation method that intends to help closing the gap between purely simulative analyzes and real-world tests.
With the inclusion of real radio hardware and radio channels in the event-based simulation, this central hybrid approach in connection with the Radio-in-the-Loop methodology creates synergies in interdisciplinary fields.
The approach contains analytical discussions, methodological strategies, and practical contributions that are summarized as key elements in the subsequent central considerations and challenges.
With Real-Time-Shift, a pseudo-real-time synchronization approach for parallel simulation and radio channel emulation of communication flows is introduced.
Based on the underlying time compensation scheme, the discrete event simulation is decoupled from real-time constraints when exchanging event messages with real-world wireless hardware.
A physical layer emulation methodology and radio channel interface concept, called Radio-in-the-Loop, is introduced along with two practical realization approaches.
Furthermore, strategic details on radio network planning with an approach to automatic hardware resource allocation are presented for radio channel emulation-capable network testbeds.
The contributions of this work are evaluated using real-world reference measurements, practical application scenarios, and experiments that provide proof of concepts.
By means of an exemplary selected cross-layer optimization scenario, the benefits are practically demonstrated and discussed.
Finally, based on IEEE 802.15.4 as the reference protocol standard for low-power wireless networks, this thesis provides feasibility studies and analysis results using the representative prototype SEmulate for the Split-Protocol-Stack approach.
Over the last twenty years, Petri nets have been increasingly adopted for modelling and simulating biological systems, as they offer an intuitive and graphical approach for this purpose. Their usability convenience comes from the fact that they offer many types of elements to describe systems in a qualitative and quantitative way. Coloured Petri nets are particularly useful to model systems with repeated components in a compact fashion. Our tool Snoopy for modelling and simulating Petri nets is one of the most well-known tools supporting a family of related Petri net classes comprising stochastic, continuous and hybrid Petri nets, and covering uncoloured and coloured Petri nets alike. However, kinetic information of a biological system, i.e. kinetic parameters may be uncertain, due to many reasons, e.g. environmental factors. Besides, coloured Petri nets as they were previously supported in Snoopy suffered from some inconsistencies. Due to these inconsistencies, exploring the model behaviour using different sizes (scaleability) was not feasible. Both challenges call for a new and more powerful approach integrating the modelling of uncertainties together with modelling features supporting repeated structures in a compact and scalable way.
This thesis comprises two major contributions: Firstly, we introduce the definition and present the simulation algorithm for both uncoloured and coloured fuzzy Petri nets, by extending the existing quantitative uncoloured and coloured Petri nets in Snoopy. This includes discretising the uncertain kinetic parameters to crisp values by using sampling strategies. Secondly, we harmonise coloured Petri nets in Snoopy with their uncoloured counterparts and we extend the Snoopy’s coloured Petri nets by all the features, which are supported by the coloured abstract net description language - an exchange format of coloured Petri nets in our PetriNuts tool family.
By performing fuzzy simulation, one can obtain two kinds of output: fuzzy bands of each output variable and their corresponding timed-membership functions. Each fuzzy band describes the uncertainties associated with the input, whereas membership functions give more accurate information about the associated uncertainties. The most important features that we obtain by harmonising coloured Petri nets are to develop scaleable models, by defining scaling factors as constants and unifying the usage of coloured Petri nets with the other tools in our PetriNuts tool family.
Eines der am intensivsten untersuchten Probleme aus dem Bereich der Optimierung stellt das sogenannte Traveling Salesman Problem (TSP, in deutsch: Handlungsreisendenproblem) dar. Die Aufgabe hinter dem Handlungsreisendenproblem besteht darin, für eine gegebene Anzahl an Orten eine Route zu entwickeln, sodass die Gesamtlänge der Route minimal ist. Zudem darf kein Ort, bis auf den ersten, mehrmals besucht werden. Ziel der vorliegenden Bachelorarbeit ist es, statistische Untersuchungen bezüglich der Tourlänge des TSP in Abhängig der Anzahl der besuchten Städte vorzunehmen, wobei der Fokus auf einer kleinen Anzahl von Städten liegt. Dazu wurde das Problem einerseits numerisch gelöst und die Verteilung der Lösungen analysiert. Für Teilaspekte wurden analytische Lösungen berechnet.
Grundlagenuntersuchungen zur Optimierung der technischen Kristallisation von Fettsäuregemischen
(2019)
Trotz der steigenden Nachfrage nach Fettsäuregemischen mit niedrigen gesättigten Anteilen liegen nur relativ wenige Untersuchungen und industrielle Umsetzungen zur Fraktionierung durch Kristallisationsverfahren vor. Die Reduzierung gesättigter Anteilen aus Gemischen gleicher Kettenlängen ist mittels Destillation nicht möglich ist. In dieser Arbeit wurde daher die lösemittelfreie Schmelzkristallisation in Form der Suspensionskristallisation untersucht.
Zur Auslegung und Optimierung von Kristallisationsprozessen ist die Kenntnis der Phasendiagramme und Kristallisationskinetiken von grundlegender Bedeutung.
Daher wurden die thermodynamischen Grundlagen der Kristallisation von Fettsäuregemischen untersucht. Nach der Erarbeitung der Untersuchungs- und Auswertemethoden wurden die zehn binären Systeme aus den fünf Hauptfettsäuren pflanzlicher Fette und Öle mittels DSC analysiert und die Phasendiagramme mit den metastabilen Bereichen sowie Enthalpiediagramme abgeleitet. Durch Modellierung wurden vier ternäre Phasendiagramme konstruiert, wesentliche Charakteristika abgeleitet und in ein Modell zur Berechnung der Phasenumwandlungstemperaturen von Systemen aus den fünf Hauptfettsäuren überführt, welches eine mittlere Abweichung von nur 0,4K und eine Standardabweichung von 0,7K liefert. Zudem konnte erstmals der komplexe Verlauf der Kristallisation von Fettsäuregemischen in den Phasendiagrammen aufgeklärt und beschrieben werden.
Die Kinetik des Kristallwachstums wurde an drei Fettsäuregemischen mit verschiedenen Kühlraten und Zieltemperaturen mittels Polarisationsmikroskopie untersucht. Die eingesetzte Kühlkammer zeigte mit ±0,003 K Temperaturgenauigkeit sehr lineare Temperaturprofile. Die Projektionsfläche der Kristalle wurde, neben den Kristalldichten, aufgrund einer Korrelation zu entsprechenden DSC-Wärmestromkurven als geeigneter Parameter zur Beschreibung des Kristallwachstums erkannt.
Es wurden kinetische Parameter, wie Induktionszeit, Keimbildungsrate, finale Kristalldichte und -fläche sowie die Wachstumsgeschwindigkeiten der Einzelkristalle und des Bulks, ermittelt. Für die zeitliche Entwicklung wurden Modelle für die Berechnung abgeleitet. Zudem erfolgte die Untersuchung der auftretenden Kristallmorphologien, wobei für alle Kristalle eine lanzettförmige Grundstruktur gefunden wurde.
Es wurden Versuche im größeren Maßstab in einem Laborreaktorsystem mit 40ml sowie einem Vibromix- und Wavemixer-System mit jeweils 2l Probenmenge durchgeführt. Zur Überwachung des Kristallwachstums in den Reaktoren wurde ein Laserscanner (FBRM) verwendet.
Versuche mit verschiedenen Kühlraten, Durchmischungsparametern und Zieltemperaturen zeigten den Vorteil einer scherarmen Durchmischung zur Erreichung gut trennbarer Kristallsuspensionen. Die Abtrennbarkeit wurde mittels Druckfiltration und Zentrifugation überprüft, wobei die Druckfiltration mit Ausbeuten an Mutterlauge von bis zu 98% die besten Resultate lieferte.
PEM water electrolysis is a clean technology for hydrogen production. In spite of its many advantages, the costs of the conventional PEM electrolysis cell makes it commercially less competitive vis-à-vis its peers. An alternative cell design has been proposed which has up to a 25 % costs advantage over the conventional cell. In this alternative cell design, the flow channel plate which bears the most costs in the conventional cell design has been replaced with a 3-D Porous Transport Layer (PTL) structure. It has however, been observed that the conventional cell by far out performs the low cost cell at high current density operations, due to increased mass transport limitation in the later. Industrial and commercial hydrogen production efforts are focused towards high current density operation (> 3 A/cm²), so the alternative cell design must be optimized for mass transport limitation.
PEM water electrolysis is a clean technology for hydrogen production. In spite of its many advantages, the costs of the conventional PEM electrolysis cell makes it commercially less competitive vis-à-vis its peers. An alternative cell design has been proposed which has up to a 25 % costs advantage over the conventional cell. In this alternative cell design, the flow channel plate which bears the most costs in the conventional cell design has been replaced with a 3-D Porous Transport Layer (PTL) structure. It has however, been observed that the conventional cell by far out performs the low cost cell at high current density operations, due to increased mass transport limitation in the later. Industrial and commercial hydrogen production efforts are focused towards high current density operation (> 3 A/cm²), so the alternative cell design must be optimized for mass transport limitation.
This work seeks to understand the source of, and to eliminate the mass transport losses in the alternative cell design to get it performing at least as good as the conventional cell at current densities up to 5 A/cm². A 2-D non-isothermal semi-empirical fully-coupled models of both cell designs have been developed and experimentally validated. The developed validated models were then used as tools to simulate and predict the best operating conditions, design parameters and micro-structural properties of the PTL at which the mass transport issues in the alternate cell will be at its minimum, at high current densities. The models are based on a multi-physics approach in which thermodynamic, electrochemical, thermal and mass transport sub-models are coupled and solved numerically, to predict the cell polarization and individual overpotentials, as well as address heat and water management issues. The most unique aspect of this work however, is the development of own semi-empirical equations for predicting the mass transport overpotential imposed by the gas phase (bubbles) at high current densities. For the very first time, calculated polarization curves up to 5 A/cm² have been validated by own experimental data. The results show that, the temperature and pressure, water flowrate and thickness of the PTL are the critical parameters for mitigating mass transport limitation. It was found that, for the size of the cells studied (25 cm² active area each), when both cells are operating at the same temperature of 60 °C, alternative design will have a comparable performance to the conventional designed cell even at 5 A/cm² current density when; the operating pressure is ≥ 5 bar, the feed water flowrate is ≥ 0.024l/min∙cm², PTL porosity is 50 %, PTL pore size is ≥ 11 µm and PTL thickness is 0.5 mm. At these operating, design and micro-structural conditions, the predicted difference between the polarizations of both cells will be only ~10 mV at 5 A/cm² operating current density.
Nach jahrelangem Rückgang des Hanfanbaus in Deutschland konnte in den letzten Jahren wieder eine Zunahme des Hanfanbaus verzeichnet werden. Das bei der Fasergewinnung mit bis zu 70 Masse-% anfallende Nebenprodukt Schäben-Faser-Gemisch muss für eine weitere Verarbeitung und den Einsatz in höherwertigen Anwendungen aufbereitet werden. Ziel war es, dieses für einen wirtschaftlichen Anlagenbetrieb wichtige Nebenprodukt in seine Hauptbestandteile Schäben und Fasern zu separieren und den darin enthaltenen Staub zu entfernen. Das Separieren und damit Reinigen des Gemischs erfolgte nach einem neuartigen und patentierten Prinzip auf Basis einer Paddelschnecke und einer dazu abgestimmten Siebfläche. In verschiedenen Modellen konnten wesentliche Zusammenhänge zwischen Konstruktions-und Betriebsparametern einer Paddelschnecke und dem Abscheideverhalten von Schäben über einem Sieb aufgezeigt werden. Die Ergebnisse der Modellierung zum Massestrom und dem Trennverhalten wurden anhand einer Versuchsmaschine zur Reinigung von Schäben-Faser-Gemischen in einem Aussiebungsverlauf validiert und daraus Konstruktionsvorgaben für einen Prototypen getroffen. Mit der im Ergebnis vorgestellten Konstruktion steht für die hanfverarbeitende Industrie eine Reinigungsmaschine mit minimalem technischem Aufwand, bestehend aus nur einer Prozessstufe und damit verbundene geringe Anschaffungs- sowie Betriebskosten für den Anlagenbetreiber, zur Verfügung.
With the rapid growth of renewable energy sources (RES) in the power generation mix in accordance with the German energy transition policy (‘Energiewende’), fewer baseload coal power plants will be required. Future power generation will be supplied through decentralized power utilities such as off-shore wind parks and also through high operational flexibility of existing conventional coal power units. High operational flexibility means conventional power plants have to increase cyclic operations to cope with feed-ins from variable-RES such as wind and solar.
Unlike medium and peak load power plants that can react quickly to load changes and power ramps, baseload power plants are not suited for such operations. Important technical requirements for flexible operation include among others; frequent start-ups and shut-downs, a minimum downtime, shorter startup time and short operational periods. Baseload coal power plants however do not meet these requirements.
This increased cyclic mode of operation can have severe impacts on vital power plant components such as superheater and reheater tubes resulting in high temperature cyclic oxidation/corrosion especially because these plants were not designed for frequent cyclic operations. To optimize plant operations, minimize material damage and reduce operational and maintenance cost, it is therefore important to understand the oxidation and corrosion risk to plants materials associated with this flexible mode of operation.
In this context, thermochemical modeling in FactSage 6.4ᵀᴹ as well as experimental investigations were carried out. For the experimental investigations, five commercial coal boiler superheater and reheater materials, namely T91, VM12-SHC, TP347-HFG, DMV304 HCu and DMV310 N were exposed for 1000 hours under discontinuous isothermal oxidation conditions and 1000 hours thermo–cyclic oxidation conditions at a metal surface temperature of 650 °C. The synthetic corrosive flue gas consisted of a mixture of CO₂, O₂, SO₂, N₂ and H₂O. The test material samples were partly covered in fly ash to investigate the effect ash deposits on the corrosion and oxidation behavior of the test materials. After exposure metallographic analysis by means of light microscopy and scanning electron microscopy (LOM and SEM–EDS) were carried out to study the oxide morphology and micro–structural properties of the materials. The oxidation kinetics (weight change) results showed significant oxide growth rates (weight gain) under cyclic oxidation conditions especially in the martensitic alloys – T91, VM12-SHC.
Furthermore, metallographic analysis revealed severe oxide spallation in the ash covered sections of these alloys. The austenitic materials (TP374-HFG, DMV310 N) with the exception of DMV304 HCu showed good oxidation behavior with minimal oxide growth both under isothermal and thermal cyclic conditions. However, severe grain boundary attack and internal sulphidation were found in these alloys. DMV310 N showed the best corrosion and oxidation performance. The thermochemical modeling calculations supported the experimental results.
The state of the art in aero engine design and analysis methods is based on mature computer programs, which have been developed during several decades. The classical approach to the preliminary design phase of engine subsystems is to split the complex engineering process into disciplines and subtasks. Different experts manage the time-consuming modelling work. Due to the increasing demand for higher aero engine performance and design cycle time reduction, process integration, accuracy and agility have become key assets of the engineering work-flow.
The intention of this work is to show how multi-disciplinary integration, work-flow automation and CFD-enhanced thermal modelling methods can be used efficiently to support the aero engine preliminary design phase, with focus on the high pressure turbine subsystem. A Java based common design environment for the engine secondary air system, rotors and thermal design disciplines has been developed. This design environment enables the automatic generation of CAD, flow network and thermal models.
The improvements in terms of process agility and model prediction accuracy are demonstrated with the application of the implemented process to a reference high pressure turbine subsystem. For validation purposes, the preliminary design definition of the reference turbine case is reproduced. The automatically generated secondary air system and thermal models can replicate the same level of detail as the previous manual approach. It was found that the CFD-enhanced thermal model improves the prediction accuracy in the preliminary design stage, when no engine test data is available.