Refine
Year of publication
- 2023 (65) (remove)
Document Type
- Master's Thesis (65) (remove)
Has Fulltext
- yes (65)
Is part of the Bibliography
- no (65) (remove)
Keywords
- IT-Sicherheit (5)
- Computersicherheit (3)
- Digitaler Zwilling (3)
- Effizienz (3)
- Internet der Dinge (3)
- Künstliche Intelligenz (3)
- Maschinelles Lernen (3)
- Change Management (2)
- Cloud Computing (2)
- Deep learning (2)
Institute
- Fakultät Medien (M) (ab 22.04.2021) (29)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (16)
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (14)
- Fakultät Wirtschaft (W) (10)
- INES - Institut für nachhaltige Energiesysteme (2)
- ivESK - Institut für verlässliche Embedded Systems und Kommunikationselektronik (2)
- IDEeP - Institute for Digital Engineering and Production (ab 12.10.2022) (1)
- IUAS - Institute for Unmanned Aerial Systems (1)
Open Access
- Closed (51)
- Closed Access (9)
- Diamond (5)
- Open Access (5)
Linux and Linux-based operating systems have been gaining more popularity among the general users and among developers. Many big enterprises and large companies are using Linux for servers that host their websites, some even require their developers to have knowledge about Linux OS. Even in embedded systems one can find many Linux-based OS that run them. With its increasing popularity, one can deduce the need to secure such a system that many personnel rely on, be it to protect the data that it stores or to protect the integrity of the system itself, or even to protect the availability of the services it offers. Many researchers and Linux enthusiasts have been coming up with various ways to secure Linux OS, however new vulnerabilities and new bugs are always found, by malicious attackers, with every update or change, which calls for the need of more ways to secure these systems.
This Thesis explores the possibility and feasibility of another way to secure Linux OS, specifically securing the terminal of such OS, by altering the commands of the terminal, getting in the way of attackers that have gained terminal access and delaying, giving more time for the response teams and for forensics to stop the attack, minimize the damage, restore operations, and to identify collect and store evidence of the cyber-attack. This research will discuss the advantages and disadvantages of various security measures and compare and contrast with the method suggested in this research.
This research is significant because it paints a better picture of what the state of the art of Linux and Linux-based operating systems security looks like, and it addresses the concerns of security enthusiasts, while exploring new uncharted area of security that have been looked at as a not so significant part of protecting the OSes out of concern of the various limitations and problems it entails. This research will address these concerns while exploring few ways to solve them, as well as addressing the ideal areas and situations in which the proposed method can be used, and when would such method be more of a burden than help if used.
AI-based Ground Penetrating Radar Signal Processing for Thickness Estimation of Subsurface Layers
(2023)
This thesis focuses on the estimation of subsurface layer thickness using Ground Penetrating Radar (GPR) A-scan and B-scan data through the application of neural networks. The objective is to develop accurate models capable of estimating the thickness of up to two subsurface layers.
Two different approaches are explored for processing the A-scan data. In the first approach, A-scans are compressed using Principal Component Analysis (PCA), and a regression feedforward neural network is employed to estimate the layers’ thicknesses. The second approach utilizes a regression one-dimensional Convolutional Neural Network (1-D CNN) for the same purpose. Comparative analysis reveals that the second approach yields superior results in terms of accuracy.
Subsequently, the proposed 1-D CNN architecture is adapted and evaluated for Step Frequency Continuous Wave (SFCW) radar, expanding its applicability to this type of radar system. The effectiveness of the proposed network in estimating subsurface layer thickness for SFCW radar is demonstrated.
Furthermore, the thesis investigates the utilization of GPR B-scan images as input data for subsurface layer thickness estimation. A regression CNN is employed for this purpose, although the results achieved are not as promising as those obtained with the 1-D CNN using A-scan data. This disparity is attributed to the limited availability of B-scan data, as B-scan generation is a resource-intensive process.
The objective of this project is to enhance the operations of a micro-enterprise that deals with food ingredients. The emphasis is on streamlining procedures and executing effective tactics. By utilizing tools like SWOT analysis, evaluations, and strategy development, the company's strengths, weaknesses, opportunities, and threats were assessed. The company developed business-level and functional-level strategies to expedite growth and attain objectives based on the findings. Moreover, precise suggestions were given to minimize the quantity of SKUs and optimize operations. The work highlighted the significance of developing a process map for streamlining operations, boosting efficiency, and elevating customer contentment. Through the implementation of said recommendations and strategies, the company can strategically position itself for success within the highly competitive food ingredients industry.
The progress in machine learning has led to advanced deep neural networks. These networks are widely used in computer vision tasks and safety-critical applications. The automotive industry, in particular, has experienced a significant transformation with the integration of deep learning techniques and neural networks. This integration contributes to the realization of autonomous driving systems. Object detection is a crucial element in autonomous driving. It contributes to vehicular safety and operational efficiency. This technology allows vehicles to perceive and identify their surroundings. It detects objects like pedestrians, vehicles, road signs, and obstacles. Object detection has evolved from being a conceptual necessity to an integral part of advanced driver assistance systems (ADAS) and the foundation of autonomous driving technologies. These advancements enable vehicles to make real-time decisions based on their understanding of the environment, improving safety and driving experiences. However, the increasing reliance on deep neural networks for object detection and autonomous driving has brought attention to potential vulnerabilities within these systems. Recent research has highlighted the susceptibility of these systems to adversarial attacks. Adversarial attacks are well-designed inputs that exploit weaknesses in the deep learning models underlying object detection. Successful attacks can cause misclassifications and critical errors, posing a significant threat to the functionality and safety of autonomous vehicles. With the rapid development of object detection systems, the vulnerability to adversarial attacks has become a major concern. These attacks manipulate inputs to deceive the target system, significantly compromising the reliability and safety of autonomous vehicles. In this study, we focus on analyzing adversarial attacks on state-of-the-art object detection models. We create adversarial examples to test the models’ robustness. We also check if the attacks work on a different object detection model meant for similar tasks. Additionally, we extensively evaluate recent defense mechanisms to see how effective they are in protecting deep neural networks (DNNs) from adversarial attacks and provide a comprehensive overview of the most commonly used defense strategies against adversarial attacks, highlighting how they can be implemented practically in real-world situations.
Much of the research in the field of audio-based machine learning has focused on recreating human speech via feature extraction and imitation, known as deepfakes. The current state of affairs has prompted a look into other areas, such as the recognition of recording devices, and potentially speakers, by only analysing sound files. Segregation and feature extraction are at the core of this approach.
This research focuses on determining whether a recorded sound can reveal the recording device with which it was captured. Each specific microphone manufacturer and model, among other characteristics and imperfections, can have subtle but compounding effects on the results, whether it be differences in noise, or the recording tempo and sensitivity of the microphone while recording. By studying these slight perturbations, it was found to be possible to distinguish between microphones based on the sounds they recorded.
After the recording, pre-processing, and feature extraction phases we completed, the prepared data was fed into several different machine learning algorithms, with results ranging from 70% to 100% accuracy, showing Multi-Layer Perceptron and Logistic Regression to be the most effective for this type of task.
This was further extended to be able to tell the difference between two microphones of the same make and model. Achieving the identification of identical models of a microphone suggests that the small deviations in their manufacturing process are enough of a factor to uniquely distinguish them and potentially target individuals using them. This however does not take into account any form of compression applied to the sound files, as that may alter or degrade some or most of the distinguishing features that are necessary for this experiment.
Building on top of prior research in the area, such as by Das et al. in in which different acoustic features were explored and assessed on their ability to be used to uniquely fingerprint smartphones, more concrete results along with the methodology by which they were achieved are published in this project’s publicly accessible code repository.
Total Cost of Ownership (TCO) is a key tool to have a complete understanding of the costs associated with an investment, as it allows to analyze not only the initial acquisition costs, but also the long-term costs related to operation, maintenance, depreciation, and other factors. In the context of the cement industry, TCO is especially important due to the complexity of the production processes and the wide variety of components and machinery involved in the process.
For this reason, a TCO analysis for the cement industry has been conducted in this study, with the objective of showing the different components of the cost of production. This analysis will allow the reader to gain knowledge about these costs, in the industrial model will be to make informed decisions on the adoption of technologies and practices that will allow them to reduce costs in the long run and improve their operational efficiency.
In particular, this study pursues to give visibility to technologies and practices that enable the reduction of carbon emissions in cement production, thus contributing to the sustainability of industry and the protection of the environment. By being at the forefront of sustainability issues, the cement industry can contribute to the achievement of environmentally friendly technologies and enable the development of people and industry.
The Oxyfuel technology has been selected as a carbon capture solution for the cement industry due to its practical application, low costs, and practical adaptation to non-capture processes. The adoption of this technology allows for a significant reduction in CO2 emissions, which is a crucial factor in achieving sustainability in the cement manufacturing process.
Carbon capture storage technologies represent a high investment, although these technologies increase the cost of production, the application of Oxyfuel technology is one of the most economically viable as the cheapest technology per capture according to the comparison. However, this price increase is a technical advantage as the carbon capture efficiency of this technology reaches 90%. This level of efficiency leads to a decrease in taxes for the generation of CO2 emissions, making the cement manufacturing process sustainable.
Cloud computing is a combination of technologies, including grid computing and distributed computing, that use the Internet as a network for service delivery. Organizations can select the price and service models that best accommodate their demands and financial restrictions. Cloud service providers choose the pricing model for their cloud services, taking the size, usage, user, infrastructure, and service size into account. Thus, cloud computing’s economic and business advantages are driving firms to shift more applications to the cloud, boosting future development. It enlarges the possibilities of current IT systems.
Over the past several years, the ”cloud computing” industry has exploded in popularity, going from a promising business concept to one of the fastest expanding areas of the IT sector. Most enterprises are hosting or installing web services in a cloud architecture for management simplicity and improved availability. Virtual environments are applied to accomplish multi-tenancy in the cloud. A vulnerability in a cloud computing environment poses a direct threat to the users’ privacy and security. In our digital age, the user has many identities. At all levels, access rights and digital identities must be regulated and controlled.
Identity and access management(IAM) are the process of managing identities and regulating access privileges. It is considered as a front-line soldier of IT security. It is the goal of identity and access management systems to protect an organization’s assets by limiting access to just those who need it and in the appropriate cases. It is required for all businesses with thousands of users and is the best practice for ensuring user access control. It identifies, authenticates, and authorizes people to access an organization’s resources. This, in turn, enhances access management efficiency. Authentication, authorization, data protection, and accountability are just a few of the areas in which cloud-based web services have security issues. These features come under identity and access management.
The implementation of identity and access management(IAM) is essential for any business. It’s becoming more and more business-centric, so we need more than technical know-how to succeed. Organizations may save money on identity management and, more crucially, become much nimbler in their support of new business initiatives if they have developed sophisticated IAM capabilities. We used these features of identity and access management to validate the robustness of the cloud computing environment with a comparison of traditional identity and access management.
Viralität auf TikTok
(2023)
Die Social Media Plattform TikTok erfreut sich spätestens seit der Corona-Pandemie einer immer größer werdenden Gemeinschaft. Mittlerweile verfügt die App über mehr als 20 Millionen Nutzer:innen - alleine in Deutschland. Virale Videos sprießen förmlich aus dem Boden. Diese Masterarbeit beschäftig sich mit der Frage, welche Faktoren der Viralität zu Grunde liegen und ob man die Viralität maßgeblich beeinflussen kann. Dies erfolgt mittels theoretischer Grundlagen, einer quantitativen Nutzerumfrage und Experteninterivews mit erfolgreichen deutschen Creatorn. Abschließend werden Videos für TikTok konzipiert und analysiert.
Die rasante Digitalisierung verändert die Hochschule Offenburg nachhaltig. Jedes Semester entstehen zahlreiche akademische Arbeiten und Prüfungsdaten, die von hoher Bedeutung für die Qualitätssicherung und den Bildungsprozess sind. Bisher fehlte jedoch eine effiziente Lösung zur Archivierung dieser Daten. Meine Masterarbeit präsentiert ein neues Archivierungskonzept, das eine zentrale, digitale Plattform schafft. Diese ermöglicht es Lehrenden und Studierenden, leicht auf ihre Daten zuzugreifen, sei es für Projektarbeiten, Seminarleistungen oder Prüfungen. Ich stelle nicht nur das Konzept vor, sondern auch die Umsetzung eines Prototyps auf Intrexx, einer Low-Code-Entwicklungsplattform. Mein Ziel ist es, ein Handbuch für zukünftige Entwickler zu hinterlassen, um zur digitalen Transformation der Hochschule beizutragen und die Bildungsprozesse zu optimieren.
Go ist eine 2009 veröffentlichte Programmiersprache mit einem statischen Typsystem. Seit Version 1.18 sind auch Generics ein Teil der Sprache. Deren Übersetzung wurde im de facto Standard-Compiler mittels Monomorphisierung umgesetzt. Diese bringt neben einigen Vorteilen auch Nachteile mit sich. Aus diesem Grund beschäftigt sich diese Arbeit mit einer alternativen Übersetzungsstrategie für Generics in Go und implementiert diese in einem neuen Compiler für Featherweight Generic Go, einem Subset von Go. Zum Schluss steht damit ein nahezu funktionierender Compiler, welcher schließlich Racket-Code ausgibt. Eine Evaluierung der Performanz der Übersetzungsstrategie ist allerdings noch ausstehend.
Empfehlungssysteme sind auf E-Commerce-Webseiten omnipräsent und unterstützen die Nutzer an bestimmten Touchpoints beim Onlineshopping, indem sie auf Produkte aufmerksam machen. In den meisten Anwendungsfällen werden Produkte empfohlen, die den Interessen der Nutzer entsprechen oder einen Warenkorb komplettieren sollen. Während klassische Empfehlungssysteme die Nutzer meistens zu Beginn oder zum Ende einer Produktannäherung erreichen, finden sie dazwischen selten Anwendung. Dabei könnten Empfehlungssysteme bereits in die Navigations-Journey der Nutzer integriert werden und so maßgeblich zur Produktauffindbarkeit beitragen. Trotz der Tatsache, dass Empfehlungssysteme maßgeblich zum Geschäftserfolg im Online-Handel beitragen, ist ihre Integration in die Navigation von E-Commerce-Plattformen noch wenig erforscht.
Ziel dieser Arbeit ist es daher zu ergründen, wie ein Empfehlungssystem in der lokalen Navigation von E-Commerce-Plattformen ausgestaltet sein sollte. Dabei soll der Prototyp eines Empfehlungssystems in der lokalen Navigation konzipiert werden, indem einerseits echte Warenkorbdaten einer E-Commerce-Plattform und andererseits Nutzerpräferenzen untersucht werden. Die Untersuchung umfasst die algorithmische Datenverarbeitung von Warenkorbdaten und die Durchführung einer quantitativen Befragung. Darüber hinaus wurde domänenspezifisches Wissen über die Verfahren, Algorithmen, Designkriterien und weitere Erfolgsfaktoren recherchiert.
Die Ergebnisse deuten darauf hin, dass die alleinige Verarbeitung von Warenkorbdaten nicht ausreichend für ein navigatorisches Empfehlungssystem ist, obwohl sich die Alternating Least Squares Matrixfaktorisierung als geeignetes Verfahren herausgestellt hat. Des Weiteren können Empfehlungssysteme in der lokalen Navigation sowohl einen positiven als auch einen negativen Einfluss auf die Nutzerzufriedenheit haben. Aus den Ergebnissen konnte ein detaillierter Prototyp konzipiert und vorgestellt werden. Bei diesem Prototyp handelt es sich um ein Session-basiertes Empfehlungssystem, das den Onlineshopping-Kontext der Nutzer ermitteln kann. Auf dieser Basis kann das System Produktkategorien in Echtzeit empfehlen, die zu diesem Kontext komplementär sind oder einen Social Proof abbilden.
Ziel dieser Arbeit ist die Modellierung und Nutzung eines digitalen Zwillings am Beispiel eines realen Tiny-Houses. Dazu werden für die Komponenten der technischen Gebäudeausrüstung Wärmepumpe, thermische Speicher, thermoaktives Bauteilsystem, PVT-Kollektoren und Batterie als Grey-Box-Modelle modelliert und in der Python-Umgebung umgesetzt. In dieser Arbeit wird auf die physikalischen Hintergründe und mathematischen Formulierungen für jede Komponente eingegangen.
In einem automatisierten Programm werden die digitalen Komponenten mit Messwerten aus dem realen Anlagenbetrieb verknüpft. Dieses Skript wird zu Fehlererkennung verwendet. Dabei konnte ein fehlerhafter Betrieb der PVT-Kollektoren bewiesen werden.
Die Verknüpfung der einzelnen Komponenten zum digitalen Zwilling wird zur Betriebsoptimierung verwendet. Hierzu wird die Simulation des Ausgangszustands mit verschiedenen Optimierungsstrategien verglichen. Dabei konnte der Anlagenbetrieb hinsichtlich Komfortbedingungen und Energieeffizienz deutlich optimiert werden. Die finale Optimierungsstrategie basiert auf einereinfachen Wettervorhersage.
Mit der Modellierung und Nutzung eines digitalen Zwillings trägt diese Arbeit dazu bei, innovative Lösungen für die zukünftige Entwicklung und Gestaltung von Gebäuden sowie die Optimierung bereits bestehender Gebäude mithilfe digitaler Zwillinge voranzutreiben.
It is generally agreed that the development and deployment of an important amount of IoT devices throughout the world has revolutionized our lives in a way that we can rely on these devices to complete certain tasks that may have not been possible just years ago which also brought a new level of convenience and value to our lives.
This technology is allowing us in a smart home environment to remotely control doors, windows, and fridges, purchase online, stream music easily with the use of voice assistants such as Amazon Echo Alexa, also close a garage door from anywhere in the world to cite some examples as this technology has added value to several domains ranging from household environments, cites, industries by exchanging and transferring data between these devices and customers. Many of these devices’ sensors, collect and share information in real-time which enables us to make important business decisions.
However, these devices pose some risks and also some security and privacy challenges that need to be addressed to reach their full potential or be considered to be secure. That is why, comprehensive risk analysis techniques are essential to enhance the security posture of IoT devices as they can help evaluate the robustness and reliability towards potential susceptibility to risks, and vulnerabilities that IoT devices in a smart home setting might possess.
This approach relies on the basis of ISO/IEC 27005 methodology and risk matrix method to highlight the level of risks, impact, and likelihood that an IoT device in smart home settings can have, map the related vulnerability, threats and risks and propose the necessary mitigation strategies or countermeasures that can be taken to secure a device and therefore satisfying some security principles. Around 30 risks were identified on Amazon Echo and the related IoT system using the methodology. A detailed list of countermeasures is proposed as a result of the risk analysis. These results, in turn, can be used to elevate the security posture of the device.
Die akustischen Emissionen von Wasserinstallationsgeräten, insbesondere während der nächtlichen Ruhezeit in angrenzenden Räumen, können erhebliche Störungen verursachen. Die primäre Ursache dieser Belästigungen liegt in der spezifischen Funktionsweise von Wasserarmaturen, welche somit als Hauptquelle der Lärmemissionen identifiziert werden können. Hydrodynamische und -akustische Simulationen erweisen sich in diesem Zusammenhang als wertvolle Werkzeuge zur Identifikation von kritischen Bereichen, welche für die Schallabstrahlung verantwortlich sind.
In dieser vorliegenden Arbeit werden die hydrodynamischen und -akustischen Eigenschaften eines Installationsgeräuschnormals mithilfe numerischer Methoden untersucht, wobei gleichzeitig ein kommerzielles Berechnungswerkzeug zur Vorhersage von akustischen Schallfeldern in Bezug auf ein hybrides Verfahren evaluiert und validiert wird. Das Installationsgeräuschnormal dient dabei als Testfall, um die Leistungsfähigkeit des hybriden Akustikmodells in Verbindung mit einem RANS- sowie RANS/LES-Ansatz zu überprüfen. Es werden verschiedene Varianten des Installationsgeräuschnormals untersucht und mit experimentellen Messdaten validiert.
Die hydrodynamischen Simulationsergebnisse der verschiedenen Varianten zeigen eine hervorragende Übereinstimmung mit den experimentellen Daten bei der Verwendung eines RANS/LES-Ansatzes. Der Einsatz eines solchen Hybrid-Modells zur Turbulenzmodellierung hat gezeigt, dass eine verbesserte Wiedergabe von zeitabhängigen Phänomenen in der Strömung erreicht werden kann, welche in direktem Zusammenhang mit der Vorhersage der Schallausbreitung stehen und somit eine bedeutende Auswirkung darauf haben.
Die akustische Untersuchung der verschiedenen Varianten des Installationsgeräuschnormals zeigt, dass die Schallabstrahlung ausschließlich in breitbandigen Frequenzspektren erfolgt. Im Rahmen des Vergleichs mit den Messdaten lässt sich erkennen, dass eine akustische Analyse auf Basis eines hybriden Modells in Verbindung mit dem Akustikmodell vielversprechende Ergebnisse bezüglich der Vorhersage des Schallfelds liefern kann und dabei im Vergleich zu einfachen RANS-Modellen Vorteile aufweist.
As the population grows, so does the amount of biowaste. As demand for energy grows, biogas is a promising solution to the problem. Lignocellulosic materials are challenged of slow degradability due to the presence of polymers such as cellulose, lignin and hemicellulose. There are several pretreatment methods available to enhance the degradability of such materials, including enzymatic pretreatment. In this pretreatment, there are few parameters that can influence the results, the most important being the enzyme to solid ratio and the solid to liquid ratio. During this project, experiments were conducted to determine the optimal conditions for those two factors. It was discovered that a solid to liquid ratio of 31 g of buffer per 1 gram of organic dry matter produced the highest reducing sugar release in flasks when combined with 34 mg of protein per 1 gram of organic dry mass. Additionally, another experiment was carried out to investigate the impact of enzymatic pretreatment on biogas production using artificial biowaste as a substrate. Artificial biowaste produced 577,9 NL/kg oDM, while enzymatically pretreated biowaste produced 639,3 NL/kg oDM. This resulted in a 10,6% rise in cumulative biogas production compared to its use without enzymatic pretreatment. By the conclusion of the investigation, specific cumulative dry methane yields of 364,7 NL/kg oDM and 426,3 NL/kg oDM were obtained from artificial biowaste without and with enzymatic pretreatment, respectively. This resulted in a methane production boost of 16,9%. Additionally in case of the reactors with enzymatically pretreated substrate kinetic constant was lower more than double, where maximum volume of biogas increased, comparing to the reactors without enzymatic pretreatment.
One of the most critical areas of research and expansion has been exploiting new technologies in supply chain risk management. One example of this is the use of Digital Twins. The performance of physical systems can be analyzed and simulated using digital twins, virtual versions of these systems that use real-time data, and sophisticated algorithms. Inside the supply chain risk management field, digital twins present a one-of-a-kind opportunity to improve an organization's ability to anticipate, address, and react to the possibility of problems within the supply chain.
The objective of this study is to identify and assess the advantages that accrue to supply chain risk management as a result of Digital Twins' adoption into the system, as well as to identify the challenges associated with achieving those benefits. In the context of supply chain risk management, a thorough literature study is conducted to analyze the essential traits and capabilities of digital twins and how these qualities lead to enhanced risk management methods. This study investigates the essential properties and capacities of digital twins. In addition, the state of digital twin technology and its applications in supply chain risk management are evaluated, and prospective areas for further study and development are highlighted.
The primary purpose of this investigation is to provide a comprehensive and in-depth analysis of the digital twins' role in supply chain risk management through the utilization of digital twins, as well as to highlight the potential benefits and challenges associated with the implementation of digital twins. The research was carried out based on the existing body of written material and the replies of 27 individuals who had previous experience making use of digital twins and took part in an online questionnaire.
The results of this study will be relevant to a diverse group of stakeholders, including specialists in risk management and researchers, amongst others.
Eine reine Passwortauthentifizierung, wie sie im Hochschulumfeld eingesetzt wird, bringt Sicherheitsrisiken mit sich. Ziel dieser Arbeit ist es zu analysieren, wie die Sicherheit in der Praxis mittels einer Zwei-Faktor-Authentifizierung erhöht werden kann und in welcher Weise die Einführung eines Single-Sign-On Konzept zum Erreichen dieses Ziels beiträgt. Es konnte gezeigt werden, dass eine Shibboleth IdP Installation als zentraler Zugangspunkt für Single-Sign-On sowie eine daran angebundene privacyIDEA-Instanz die Umsetzung der Zwei-Faktor-Authentifizierung für einige aber nicht alle Dienste ermöglicht.
Licht war für die Menschheit schon immer ein Hilfsmittel zur Orientierung. Das Zusammenspiel zwischen hellen und schattierten Oberflächen macht eine räumliche Wahrnehmung erst möglich. Die Lokalisierung von Lichtquellen bietet darüber hinaus für zahlreiche Anwendungsfelder, wie beispielsweise Augmented Reality, ein großes Potential.
Das Ziel der vorliegenden Arbeit war es, ein neuronales Netzwerk zu entwickeln, welches mit Hilfe eines selbst generierten, synthetischen Datensatzes eine Lichtsetzung parametrisiert. Dafür wurden State-of-the-Art Netzwerke aus der digitalen Bildverarbeitung eingesetzt.
Zu Beginn der Arbeit mussten die Eigenschaften der Lichtsetzung extrahiert werden. Eine weitere fundamentale Anforderung war die Aufbereitung des Wissens von Deep Learning.
Für die Generierung des synthetischen Datensatzes wurde eigens ein Framework entwickelt, welches auf der Blender Engine basiert.
Anschließend wurden die generierten Bilder und Metadaten in einem abgewandelten VGG16- und ResNet50-Netz trainiert, validiert und evaluiert.
Eine gewonnene Erkenntnis ist, dass sich künstlich generierte Daten eignen um ein neuronales Netz zu trainieren. Des Weiteren konnte gezeigt werden, dass sich mit Hilfe von Deep Learning Lichtsetzungsparameter extrahieren lassen.
Eine weiterführende Forschungsaufgabe könnte mit dem vorgeschlagenen Ansatzdie Lichtinszenierung von Augmented Reality Anwendungen verbessern.
Decarbonisation Strategies in Energy Systems Modelling: APV and e-tractors as Flexibility Assets
(2023)
This work presents an analysis of the impact of introducing Agrophotovoltaic technologies and electric tractors into Germany’s energy system. Agrophotovoltaics involves installing photovoltaic systems in agricultural areas, allowing for dual usage of the land for both energy generation and food production. Electric tractors, which are agricultural machinery powered by electric motors, can also function as energy storage units, providing flexibility to the grid. The analysis includes a sensitivity study to understand how the availability of agricultural land influences Agrophotovoltaic investments, followed by the examination of various scenarios that involve converting diesel tractors to electric tractors. These scenarios are based on the current CO2 emission reduction targets set by the German Government, aiming for a 65% reduction below 1990 levels by 2030 and achieving zero emissions by 2045. The results indicate that approximately 3% of available agricultural land is necessary to establish a viable energy mix in Germany. Furthermore, the expansion of electric tractors tends to reduce the overall system costs and enhances the energy-cost-efficiency of Agrophotovoltaic investments.