Refine
Year of publication
Document Type
- Conference Proceeding (29)
- Article (reviewed) (19)
- Article (unreviewed) (17)
- Bachelor Thesis (10)
- Master's Thesis (6)
- Doctoral Thesis (5)
- Report (4)
- Letter to Editor (1)
- Study Thesis (1)
Conference Type
- Konferenzartikel (23)
- Konferenz-Abstract (3)
- Konferenz-Poster (1)
- Sonstiges (1)
Keywords
- Deep Leaning (3)
- Deep Learning (3)
- Hochschuldidaktik (3)
- machine learning (3)
- Advanced Footwear Technology (2)
- Alexander von Humboldt (2)
- Artificial Intelligence (2)
- Biomechanics (2)
- Computersicherheit (2)
- Deep learning (2)
Institute
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (31)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (27)
- Fakultät Medien (M) (ab 22.04.2021) (19)
- INES - Institut für nachhaltige Energiesysteme (13)
- Fakultät Wirtschaft (W) (12)
- IMLA - Institute for Machine Learning and Analytics (9)
- ivESK - Institut für verlässliche Embedded Systems und Kommunikationselektronik (4)
- IBMS - Institute for Advanced Biomechanics and Motion Studies (ab 16.11.2022) (3)
- ACI - Affective and Cognitive Institute (2)
- IfTI - Institute for Trade and Innovation (2)
Open Access
- Diamond (92) (remove)
Garbage in, Garbage out: How does ambiguity in data affect state-of-the-art pedestrian detection?
(2024)
This thesis investigates the critical role of data quality in computer vision, particularly in the realm of pedestrian detection. The proliferation of deep learning methods has emphasised the importance of large datasets for model training, while the quality of these datasets is equally crucial. Ambiguity in annotations, arising from factors like mislabelling, inaccurate bounding box geometry and annotator disagreements, poses significant challenges to the reliability and robustness of the pedestrian detection models and their evaluation. This work aims to explore the effects of ambiguous data on model performance with a focus on identifying and separating ambiguous instances, employing an ambiguity measure utilizing annotator estimations of object visibility and identity. Through accurate experimentation and analysis, trade-offs between data cleanliness and representativeness, noise removal and retention of valuable data emerged, elucidating their impact on performance metrics like the log average miss-rate, recall and precision. Furthermore, a strong correlation between ambiguity and occlusion was discovered with higher ambiguity corresponding to greater occlusion prevalence. The EuroCity Persons dataset served as the primary dataset, revealing a significant proportion of ambiguous instances with approximately 8.6% ambiguity in the training dataset and 7.3% in the validation set. Results demonstrated that removing ambiguous data improves the log average miss-rate, particularly by reducing the false positive detections. Augmentation of the training data with samples from neighbouring classes enhanced the recall but diminished precision. Error correction of wrong false positives and false negatives significantly impacts model evaluation results, as evidenced by shifts in the ECP leaderboard rankings. By systematically addressing ambiguity, this thesis lays the foundation for enhancing the reliability of computer vision systems in real-world applications, motivating the prioritisation of developing robust strategies to identify, quantify and address ambiguity.
Die vorliegende Arbeit beschäftigt sich mit der Nutzung von Reinforcement Learning in der Informationsbeschaffungs-Phase eines Penetration Tests. Es werden Kernprobleme in den bisherigen Ansätzen anderer das Thema betreffender wissenschaftlicher Arbeiten analysiert und praktische Lösungsansätze für diese bisherigen Hindernisse vorgestellt und implementiert. Die Arbeit zeigt damit eine beispielhafte Implementierung eines Reinforcement Learning Agenten zur Automatisierung der Informationsbeschaffungs-Phase eines Penetration Tests und stellt Lösungen für existierende Probleme in diesem Bereich dar.
Eingebettet wird diese wissenschaftliche Arbeit in die Anforderungen der Herrenknecht AG hinsichtlich der Absicherung des Tunnelbohrmaschinen-Netzwerks. Dabei werden praktische Ergebnisse des eigen entwickelten Reinforcement Learning Modells im Tunnelbohrmaschinen-Test-Netzwerk der Herrenknecht AG vorgestellt.
With the expansion of IoT devices in many aspects of our life, the security of such systems has become an important challenge. Unlike conventional computer systems, any IoT security solution should consider the constraints of these systems such as computational capability, memory, connectivity, and power consumption limitations. Physical Unclonable Functions (PUFs) with their special characteristics were introduced to satisfy the security needs while respecting the mentioned constraints. They exploit the uncontrollable and reproducible variations of the underlying component for security applications such as identification, authentication, and communication security. Since IoT devices are typically low cost, it is important to reuse existing elements in their hardware (for instance sensors, ADCs, etc.) instead of adding extra costs for the PUF hardware. Micro-electromechanical system (MEMS) devices are widely used in IoT systems as sensors and actuators. In this thesis, a comprehensive study of the potential application of MEMS devices as PUF primitives is provided. MEMS PUF leverages the uncontrollable variations in the parameters of MEMS elements to derive secure keys for cryptographic applications. Experimental and simulation results show that our proposed MEMS PUFs are capable of generating enough entropy for a complex key generation, while their responses show low fluctuations in different environmental conditions.
Keeping in mind that the PUF responses are prone to change in the presence of noise and environmental variations, it is critical to derive reliable keys from the PUF and to use the maximum entropy at the same time. In the second part of this thesis, we elaborate on different key generation schemes and their advantages and drawbacks. We propose the PUF output positioning (POP) and integer linear programming (ILP) methods, which are novel methods for grouping the PUF outputs in order to maximize the extracted entropy. To implement these methods, the key enrollment and key generation algorithms are presented. The proposed methods are then evaluated by applying on the responses of the MEMS PUF, where it can be practically shown that the proposed method outperforms other existing PUF key generation methods.
The final part of this thesis is dedicated to the application of the MEMS PUF as a security solution for IoT systems. We select the mutual authentication of IoT devices and their backend system, and propose two lightweight authentication protocols based on MEMS PUFs. The presented protocols undergo a comprehensive security analysis to show their eligibility to be used in IoT systems. As the result, the output of this thesis is a lightweight security solution based on MEMS PUFs, which introduces a very low overhead on the cost of the hardware.
This report examines exporters’ challenges and possible solutions for public intervention to promote foreign trade. Based on fieldwork conducted in Georgia, we explore which policy approaches can help to stimulate Georgian exports further. Our outcomes show that exporters face substantial barriers such as navigating complex trade regulations, lack of knowledge about target markets, trade finance gaps, as well as new export promotion programs (EPPs) in competitor countries. Other upper-middle-income countries can learn from our results that exporters can significantly benefit from a comprehensive export promotion strategy combined with an ecosystem-based “team” approach. EPPs related to awareness and capacity building in Georgia should be part of this strategy, focusing on challenges such as a lack of knowledge about trade practices and international business skills. Other EPPs must help to mitigate related market failures, as information gathering is costly, and firms have no incentive to share this information with competitors. Furthermore, targeted marketing support and customer matchmaking can answer Georgian exporters’ challenges, such as lack of market access and low sector visibility. Our results also show that public intervention through financial support and risk mitigation is essential for firms with an international orientation. The high-quality, rich outcomes provide significant value for other upper-middle-income countries by exploring the example of Georgia’s contemporary circumstances in an in-depth manner based on extensive interviews and document analysis. Limitations include that our work primarily relies on qualitative data and further research could involve a quantitative study with a diverse range of sectors.
Ultra-low-power passive telemetry systems for industrial and biomedical applications have gained much popularity lately. The reduction of the power consumption and size of the circuits poses critical challenges in ultra-low-power circuit design. Biotelemetry applications like leakage detection in silicone breast implants require low-power-consuming small-size electronics. In this doctoral thesis, the design, simulation, and measurement of a programmable mixed-signal System-on-Chip (SoC) called General Application Passive Sensor Integrated Circuit (GAPSIC) is presented. Owing to the low power consumption, GAPSIC is capable of completely passive operation. Such a batteryless passive system has lower maintenance complexity and is also free from battery-related health hazards. With a die area of 4.92 mm² and a maximum analog power consumption of 592 µW, GAPSIC has one of the best figure-of-merits compared to similar state-of-the-art SoCs. Regarding possible applications, GAPSIC can read out and digitally transmit the signals of resistive sensors for pressure or temperature measurements. Additionally, GAPSIC can measure electrocardiogram (ECG) signals and conductivity.
The design of GAPSIC complies with the International Organization for Standardization (ISO) 15693/NFC (near field communication) 5 standard for radio frequency identification (RFID), corresponding to the frequency range of 13.56 MHz. A passive transponder developed with GAPSIC comprises of an external memory storage and very few other external components, like an antenna and sensors. The passive tag antenna and reader antenna use inductive coupling for communication and energy transfer, which enables passive operation. A passive tag developed with GAPSIC can communicate with an NFC compatible smart device or an ISO 15693 RFID reader. An external memory storage contains the programmable application-specific firmware.
As a mixed-signal SoC, GAPSIC includes both analog and digital circuitries. The analog block of GAPSIC includes a power management unit, an RFID/NFC communication unit, and a sensor readout unit. The digital block includes an integrated 32-bit microcontroller, developed by the Hochschule Offenburg ASIC design center, and digital peripherals. A 16-kilobyte random-access memory and a read-only 16-kilobyte memory constitute the GAPSIC internal memory. For the fabrication of GAPSIC, one poly, six-metal 0.18 µm CMOS process is used.
The design of GAPSIC includes two stages. In the first stage, a standalone RFID/NFC frontend chip with a power management unit, an RFID/NFC communication unit, a clock regenerator unit, and a field detector unit was designed. In the second stage, the rest of the functional blocks were integrated with the blocks of the RFID/NFC frontend chip for the final integration of GAPSIC. To reduce the power consumption, conventional low-power design techniques were applied extensively like multiple power supplies, and the operation of complementary metal-oxide-semiconductor (CMOS) transistors in the sub-threshold region of operation, as well as further innovative circuit designs.
An overvoltage protection circuit, a power rectifier, a bandgap reference circuit, and two low-dropout (LDO) voltage regulators constitute the power management unit of GAPSIC. The overvoltage protection circuit uses a novel method where three stacked transistor pairs shunt the extra voltage. In the power rectifier, four rectifier units are arranged in parallel, which is a unique approach. The four parallel rectifier units provide the optimal choice in terms of voltage drop and the area required.
The communication unit is responsible for RFID/NFC communication and incorporates demodulation and load modulation circuitry. The demodulator circuit comprises of an envelope detector, a high-pass filter, and a comparator. Following a new approach, the bandgap reference circuit itself acts as the load for the envelope detector circuit, which minimizes the circuit complexity and area. For the communication between the reader and the RFID/NFC tag, amplitude-shift keying (ASK) is used to modulate signals, where the smallest modulation index can be as low as 10%. A novel technique involving a comparator with a preset offset voltage effectively demodulates the ASK signal. With an effective die area of 0.7 mm² and power consumption of 107 µW, the standalone RFID/NFC frontend chip has the best figure-of-merits compared to the state-of-the-art frontend chips reported in the relevant literature. A passive RFID/NFC tag developed with the standalone frontend chip, as well as temperature and pressure sensors demonstrate the full passive operational capability of the frontend chip. An NFC reader device using a custom-built Android-based application software reads out the sensor data from the passive tag.
The sensor readout circuit consists of a channel selector with two differential and four single-ended inputs with a programmable-gain instrumentation amplifier. The entire sensor readout part remains deactivated when not in use. The internal memory stores the measured offset voltage of the instrumentation amplifier, where a firmware code removes the offset voltage from the measured sensor signal. A 12-bit successive approximation register (SAR) type analog-to-digital-converter (ADC) based on a charge redistribution architecture converts the measured sensor data to a digital value. The digital peripherals include a serial peripheral interface, four timers, RFID/NFC interfaces, sensor readout unit interfaces, and 12-bit SAR logic.
Two sets of studies with custom-made NFC tag antennas for biomedical applications were conducted to ascertain their compatibility with GAPSIC. The first study involved the link efficiency measurements of NFC tag antennas and an NFC reader antenna with porcine tissue. In a separate experiment, the effect of a ferrite compared to air core on the antenna-coupling factor was investigated. With the ferrite core, the coupling factor increased by four times.
Among the state-of-the-art SoCs published in recent scientific articles, GAPSIC is the only passive programmable SoC with a power management unit, an RFID/NFC communication interface, a sensor readout circuit, a 12-bit SAR ADC, and an integrated 32-bit microcontroller. This doctoral research includes the preliminary study of three passive RFID tags designed with discrete components for biomedical and industrial applications like measurements of temperature, pH, conductivity, and oxygen concentration, along with leakage detection in silicone breast implants. Besides its small size and low power consumption, GAPSIC is suitable for each of the biomedical and industrial applications mentioned above due to the integrated high-performance microcontroller, the robust programmable instrumentation amplifier, and the 12-bit analog-to-digital converter. Furthermore, the simulation and measurement data show that GAPSIC is well suited for the design of a passive tag to monitor arterial blood pressure in patients experiencing Peripheral Artery Disease (PAD), which is proposed in this doctoral thesis as an exemplary application of the developed system.
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.
Vorhofflimmern ist die häufigste tachykarde Herzrhythmusstörung weltweit. Dabei verliert das Herz seinen normofrequenten Sinusrhythmus und schlägt nicht mehr regelmäßig, sondern zu schnell und unregelmäßig. Vorhofflimmern ist normalerweise keine lebensbedrohliche Herzrhythmusstörung, aber es kann zu einem Schlaganfall führen. Die Ursache dieser Herzrhythmusstörung sind die Kreisende bzw. die fokalen Erregungen im linken Atrium, die hauptsächliche aus einer oder mehreren Pulmonalvenen kommen. Die übliche Therapieverfahren des Vorhofflimmerns ist die Pulmonalvenenisolation.
Diese Bachelorthesis beschäftigt sich daher mit der Modellierung unterschiedlicher linksatrialer Fokus-Modelle und intrakardialer Elektrodenkatheter für die Diagnostik und Terminierung von Vorhofflimmern mittels Pulmonalvenenisolation im Offenburger Herzrhythmusmodell nach Schalk, Krämer und Benke, welches in CST
Studio Suite realisiert wurde.
Zu Beginn wurden die verschiedenen linksatrialen fokalen Flimmerquellen modelliert und daraufhin simuliert. Hierbei wurde jeweils eine Simulation mit linksatrialen fokalen Flimmerquellen, die aus einzelnen, dualen oder allen vier Pulmonalvenen kommen, durchgeführt. Es wurde ebenfalls eine weitere Simulation mit Biosignalen (aus der Realität) erstellt. Mit diesen Simulationen konnte nun der elektrische Erregungsablauf sichtbar gemacht werden. Daraufhin wurden die Katheter für die Diagnostik und für die Pulmonalvenenisolation modelliert und in das bestehende Offenburger Herzrhythmusmodell integriert. Bei den Diagnostik-Kathetern handelte es sich um 10-polige Lasso® Katheter, zwei Varianten von PentaRay® NAV eco Katheter und 4-polige Diagnostik-Katheter „OSYPKA FINDER pure®“. Ablationskatheter sind zwei Varianten von Pentaspline Basket pose Katheter und HELIOSTAR™ Ablation Ballon. Abschließend wurden verschiedene Varianten von Isolationsverfahren der Pulmonalvenen modelliert und daraufhin die linksatrialen fokalen Flimmerquellen nach der Isolation der Pulmonalvenen simuliert.
We aim to debate and eventually be able to carefully judge how realistic the following statement of a young computer scientist is: “I would like to become an ethical correctly acting offensive cybersecurity expert”. The objective of this article is not to judge what is good and what is wrong behavior nor to present an overall solution to ethical dilemmas. Instead, the goal is to become aware of the various personal moral dilemmas a security expert may face during his work life. For this, a total of 14 cybersecurity students from HS Offenburg were asked to evaluate several case studies according to different ethical frameworks. The results and particularities are discussed, considering different ethical frameworks. We emphasize, that different ethical frameworks can lead to different preferred actions and that the moral understanding of the frameworks may differ even from student to student.
Artificial intelligence (AI), and in particular machine learning algorithms, are of increasing importance in many application areas but interpretability and understandability as well as responsibility, accountability, and fairness of the algorithms' results, all crucial for increasing the humans' trust into the systems, are still largely missing. Big industrial players, including Google, Microsoft, and Apple, have become aware of this gap and recently published their own guidelines for the use of AI in order to promote fairness, trust, interpretability, and other goals. Interactive visualization is one of the technologies that may help to increase trust in AI systems. During the seminar, we discussed the requirements for trustworthy AI systems as well as the technological possibilities provided by interactive visualizations to increase human trust in AI.
CNN-based deep learning models for disease detection have become popular recently. We compared the binary classification performance of eight prominent deep learning models: DenseNet 121, DenseNet 169, DenseNet 201, EffecientNet b0, EffecientNet lite4, GoogleNet, MobileNet, and ResNet18 for their binary classification performance on combined Pulmonary Chest Xrays dataset. Despite the widespread application in different fields in medical images, there remains a knowledge gap in determining their relative performance when applied to the same dataset, a gap this study aimed to address. The dataset combined Shenzhen, China (CH) and Montgomery, USA (MC) data. We trained our model for binary classification, calculated different parameters of the mentioned models, and compared them. The models were trained to keep in mind all following the same training parameters to maintain a controlled comparison environment. End of the study, we found a distinct difference in performance among the other models when applied to the pulmonary chest Xray image dataset, where DenseNet169 performed with 89.38 percent and MobileNet with 92.2 percent precision.
The COVID19 pandemic, a unique and devastating respiratory disease outbreak, has affected global populations as the disease spreads rapidly. Recent Deep Learning breakthroughs may improve COVID19 prediction and forecasting as a tool of precise and fast detection, however, current methods are still being examined to achieve higher accuracy and precision. This study analyzed the collection contained 8055 CT image samples, 5427 of which were COVID cases and 2628 non COVID. The 9544 Xray samples included 4044 COVID patients and 5500 non COVID cases. The most accurate models are MobileNet V3 (97.872 percent), DenseNet201 (97.567 percent), and GoogleNet Inception V1 (97.643 percent). High accuracy indicates that these models can make many accurate predictions, as well as others, are also high for MobileNetV3 and DenseNet201. An extensive evaluation using accuracy, precision, and recall allows a comprehensive comparison to improve predictive models by combining loss optimization with scalable batch normalization in this study. Our analysis shows that these tactics improve model performance and resilience for advancing COVID19 prediction and detection and shows how Deep Learning can improve disease handling. The methods we suggest would strengthen healthcare systems, policymakers, and researchers to make educated decisions to reduce COVID19 and other contagious diseases.
The use of artificial intelligence continues to impact a broad variety of domains, application areas, and people. However, interpretability, understandability, responsibility, accountability, and fairness of the algorithms' results - all crucial for increasing humans' trust into the systems - are still largely missing. The purpose of this seminar is to understand how these components factor into the holistic view of trust. Further, this seminar seeks to identify design guidelines and best practices for how to build interactive visualization systems to calibrate trust.
With the rising necessity of explainable artificial intelligence (XAI), we see an increase in task-dependent XAI methods on varying abstraction levels. XAI techniques on a global level explain model behavior and on a local level explain sample predictions. We propose a visual analytics workflow to support seamless transitions between global and local explanations, focusing on attributions and counterfactuals on time series classification. In particular, we adapt local XAI techniques (attributions) that are developed for traditional datasets (images, text) to analyze time series classification, a data type that is typically less intelligible to humans. To generate a global overview, we apply local attribution methods to the data, creating explanations for the whole dataset. These explanations are projected onto two dimensions, depicting model behavior trends, strategies, and decision boundaries. To further inspect the model decision-making as well as potential data errors, a what-if analysis facilitates hypothesis generation and verification on both the global and local levels. We constantly collected and incorporated expert user feedback, as well as insights based on their domain knowledge, resulting in a tailored analysis workflow and system that tightly integrates time series transformations into explanations. Lastly, we present three use cases, verifying that our technique enables users to (1)~explore data transformations and feature relevance, (2)~identify model behavior and decision boundaries, as well as, (3)~the reason for misclassifications.
Modern CNNs are learning the weights of vast numbers of convolutional operators. In this paper, we raise the fundamental question if this is actually necessary. We show that even in the extreme case of only randomly initializing and never updating spatial filters, certain CNN architectures can be trained to surpass the accuracy of standard training. By reinterpreting the notion of pointwise ($1\times 1$) convolutions as an operator to learn linear combinations (LC) of frozen (random) spatial filters, we are able to analyze these effects and propose a generic LC convolution block that allows tuning of the linear combination rate. Empirically, we show that this approach not only allows us to reach high test accuracies on CIFAR and ImageNet but also has favorable properties regarding model robustness, generalization, sparsity, and the total number of necessary weights. Additionally, we propose a novel weight sharing mechanism, which allows sharing of a single weight tensor between all spatial convolution layers to massively reduce the number of weights.
Socially assistive robots (SARs) are becoming more prevalent in everyday life, emphasizing the need to make them socially acceptable and aligned with users' expectations. Robots' appearance impacts users' behaviors and attitudes towards them. Therefore, product designers choose visual qualities to give the robot a character and to imply its functionality and personality. In this work, we sought to investigate the effect of cultural differences on Israeli and German designers' perceptions and preferences regarding the suitable visual qualities of SARs in four different contexts: a service robot for an assisted living/retirement residence facility, a medical assistant robot for a hospital environment, a COVID-19 officer robot, and a personal assistant robot for domestic use. Our results indicate that Israeli and German designers share similar perceptions of visual qualities and most of the robotics roles. However, we found differences in the perception of the COVID-19 officer robot's role and, by that, its most suitable visual design. This work indicates that context and culture play a role in users' perceptions and expectations; therefore, they should be taken into account when designing new SARs for diverse contexts.
Recent advances in spiked shoe design, characterized by increased longitudinal stiffness, thicker midsole foams, and reconfigured geometry are considered to improve sprint performance. However, so far there is no empirical data on the effects of advanced spikes technology on maximal sprinting speed (MSS) published yet. Consequently, we assessed MSS via ‘flying 30m’ sprints of 44 trained male (PR: 10.32 s - 12.08 s) and female (PR: 11.56 s - 14.18 s) athletes, wearing both traditional and advanced spikes in a randomized, repeated measures design. The results revealed a statistically significant increase in MSS by 1.21% on average when using advanced spikes technology. Notably, 87% of participants showed improved MSS with the use of advanced spikes. A cluster analysis unveiled that athletes with higher MSS may benefit to a greater extent. However, individual responses varied widely, suggesting the influence of multiple factors that need detailed exploration. Therefore, coaches and athletes are advised to interpret the promising performance enhancements cautiously and evaluate the appropriateness of the advanced spike technology for their athletes critically.
High-tech running shoes and spikes ("super-footwear") are currently being debated in sports. There is direct evidence that distance running super shoes improve running economy; however, it is not well established to which extent world-class performances are affected over the range of track and road running events.
This study examined publicly available performance datasets of annual best track and road performances for evidence of potential systematic performance effects following the introduction of super footwear. The analysis was based on the 100 best performances per year for men and women in outdoor events from 2010 to 2022, provided by the world governing body of athletics (World Athletics).
We found evidence of progressing improvements in track and road running performances after the introduction of super distance running shoes in 2016 and super spike technology in 2019. This evidence is more pronounced for distances longer than 1500 m in women and longer than 5000 m in men. Women seem to benefit more from super footwear in distance running events than men.
While the observational study design limits causal inference, this study provides a database on potential systematic performance effects following the introduction of super shoes/spikes in track and road running events in world-class athletes. Further research is needed to examine the underlying mechanisms and, in particular, potential sex differences in the performance effects of super footwear.
Following the traditional paradigm of convolutional neural networks (CNNs), modern CNNs manage to keep pace with more recent, for example transformer-based, models by not only increasing model depth and width but also the kernel size. This results in large amounts of learnable model parameters that need to be handled during training. While following the convolutional paradigm with the according spatial inductive bias, we question the significance of \emph{learned} convolution filters. In fact, our findings demonstrate that many contemporary CNN architectures can achieve high test accuracies without ever updating randomly initialized (spatial) convolution filters. Instead, simple linear combinations (implemented through efficient 1×1 convolutions) suffice to effectively recombine even random filters into expressive network operators. Furthermore, these combinations of random filters can implicitly regularize the resulting operations, mitigating overfitting and enhancing overall performance and robustness. Conversely, retaining the ability to learn filter updates can impair network performance. Lastly, although we only observe relatively small gains from learning 3×3 convolutions, the learning gains increase proportionally with kernel size, owing to the non-idealities of the independent and identically distributed (\textit{i.i.d.}) nature of default initialization techniques.
We have developed a methodology for the systematic generation of a large image dataset of macerated wood references, which we used to generate image data for nine hardwood genera. This is the basis for a substantial approach to automate, for the first time, the identification of hardwood species in microscopic images of fibrous materials by deep learning. Our methodology includes a flexible pipeline for easy annotation of vessel elements. We compare the performance of different neural network architectures and hyperparameters. Our proposed method performs similarly well to human experts. In the future, this will improve controls on global wood fiber product flows to protect forests.
State-of-the-art models for pixel-wise prediction tasks such as image restoration, image segmentation, or disparity estimation, involve several stages of data resampling, in which the resolution of feature maps is first reduced to aggregate information and then sequentially increased to generate a high-resolution output. Several previous works have investigated the effect of artifacts that are invoked during downsampling and diverse cures have been proposed that facilitate to improve prediction stability and even robustness for image classification. However, equally relevant, artifacts that arise during upsampling have been less discussed. This is significantly relevant as upsampling and downsampling approaches face fundamentally different challenges. While during downsampling, aliases and artifacts can be reduced by blurring feature maps, the emergence of fine details is crucial during upsampling. Blurring is therefore not an option and dedicated operations need to be considered. In this work, we are the first to explore the relevance of context during upsampling by employing convolutional upsampling operations with increasing kernel size while keeping the encoder unchanged. We find that increased kernel sizes can in general improve the prediction stability in tasks such as image restoration or image segmentation, while a block that allows for a combination of small-size kernels for fine details and large-size kernels for artifact removal and increased context yields the best results.
Entity Matching (EM) defines the task of learning to group objects by transferring semantic concepts from example groups (=entities) to unseen data. Despite the general availability of image data in the context of many EM-problems, most currently available EM-algorithms solely rely on (textual) meta data. In this paper, we introduce the first publicly available large-scale dataset for "visual entity matching", based on a production level use case in the retail domain. Using scanned advertisement leaflets, collected over several years from different European retailers, we provide a total of ~786k manually annotated, high resolution product images containing ~18k different individual retail products which are grouped into ~3k entities. The annotation of these product entities is based on a price comparison task, where each entity forms an equivalence class of comparable products. Following on a first baseline evaluation, we show that the proposed "visual entity matching" constitutes a novel learning problem which can not sufficiently be solved using standard image based classification and retrieval algorithms. Instead, novel approaches which allow to transfer example based visual equivalent classes to new data are needed to address the proposed problem. The aim of this paper is to provide a benchmark for such algorithms.
Information about the dataset, evaluation code and download instructions are provided under https://www.retail-786k.org/.
The increasing diffusion of rapidly developing AI technologies led to the idea of the experiment to combine TRIZ-based automated idea generation with the natural language processing tool ChatGPT, using the chatbot to interpret the automatically generated elementary solution principles. The article explores the opportunities and benefits of a novel AI-enhanced approach to teaching systematic innovation, analyses the learning experience, identifies the factors that affect students' innovation and problem-solving performance, and highlights the main difficulties students face, especially in interdisciplinary problems.
Inner Congo
(2023)
This research-creation project, part of the DE\GLOBALIZE artistic research cycle presented at the #IFM2022 Conference, investigates the complexities of Congo violence, care, and colonialism. Drawing on Michel Serres' metaphor of the great estuaries, the study explores the topology of interactive documentaries, blending theory, emotion, and personal experiences. Accessible through the interactive web documentation at http://deglobalize.com, the platform offers a media-archaeological archive for speculative ethnography, enabling the forensic processing of single documents in line with actor-network theory.
Artificial Intelligence (AI) can potentially transform many aspects of modern society in various ways, including automation of tasks, personalization of products and services, diagnosis of diseases and their treatment, transportation, safety, and security in public spaces, etc. Recently, AI technology has been transforming the financial industry, offering new ways to analyse data and automate processes, reduce costs, increase efficiency, and provide more personalized services to customers. However, it also raised important ethical and regulatory questions that need to be addressed by the industry and society as a whole. The aim of the Erasmus+ project Transversal Skills in Applied Artificial Intelligence - TSAAI (KA220-HED - Cooperation Partnerships in higher education) has been to establish a training platform that will incorporate teaching guidelines based on a curriculum covering different areas of application of AI technology. In this work, we will focus on applying AI models in the financial and insurance sectors.
Enhancing engineering creativity with automated formulation of elementary solution principles
(2023)
The paper describes a method for the automated formulation of elementary creative stimuli for product or process design at different levels of abstraction and in different engineering domains. The experimental study evaluates the impact of structured automated idea generation on inventive thinking in engineering design and compares it with previous experimental studies in educational and industrial settings. The outlook highlights the benefits of using automated ideation in the context of AI-assisted invention and innovation.
Training deep neural networks using backpropagation is very memory and computationally intensive. This makes it difficult to run on-device learning or fine-tune neural networks on tiny, embedded devices such as low-power micro-controller units (MCUs). Sparse backpropagation algorithms try to reduce the computational load of on-device learning by training only a subset of the weights and biases. Existing approaches use a static number of weights to train. A poor choice of this so-called backpropagation ratio limits either the computational gain or can lead to severe accuracy losses. In this paper we present TinyProp, the first sparse backpropagation method that dynamically adapts the back-propagation ratio during on-device training for each training step. TinyProp induces a small calculation overhead to sort the elements of the gradient, which does not significantly impact the computational gains. TinyProp works particularly well on fine-tuning trained networks on MCUs, which is a typical use case for embedded applications. For typical datasets from three datasets MNIST, DCASE2020 and CIFAR10, we are 5 times faster compared to non-sparse training with an accuracy loss of on average 1%. On average, TinyProp is 2.9 times faster than existing, static sparse backpropagation algorithms and the accuracy loss is reduced on average by 6 % compared to a typical static setting of the back-propagation ratio.
In der Geschichte »Die Schule« (Originaltitel: ,,The fun they had“) von 1954 beschreibt der russisch-amerikanische Wissenschaftler und Science fiction Autor Isaac Asimov, wie die Schule der Zukunft im Jahr 2157 aussieht – oder genauer: dass es gar keine Schulen mehr gibt. Jedes Kind hat neben seinem Kinderzimmer im Elternhaus einen kleinen Schulraum, in dem es von einem mechanischen Lehrer (einer Maschine mit Bildschirm und einem Schlitz zum Einwerfen der Hausaufgaben) unterrichtet wird. Diese Lehrmaschine ist perfekt auf die Fähigkeiten des einzelnen Kindes eingestellt und kann es optimal beschulen. Nur: Maschinen können kaputt gehen. Die elfjährige Margie wird von ihrem mechanischen Lehrer wieder und wieder in Geographie abgefragt, aber jedes Mal schlechter benotet. Das sieht die Mutter und ruft den Schulinspektor, um den mechanischen Lehrer zu reparieren.
Die Visualisierung von Programmabläufen ist ein zentraler Aspekt für Programmieranfänger, um das Verständnis von Codeabläufen zu erleichtern und den Einstieg in der Softwareentwicklung zu unterstützen. In dieser Masterthesis wird ein speziell auf die Bedürfnisse von Einsteigern zugeschnittenes generisches Framework vorgestellt, wobei der Fokus auf einer einfachen, verständlichen aber auch korrekten Darstellung der Programmausführung liegt. Das Framework integriert das Debugger Adapter Protocol, um den Debugger unterschiedlicher Sprachen ansprechen und verwenden zu können.
In dieser Arbeit werden zunächst die Anforderungen für das generische Framework diskutiert. Anschließend werden bestehende Ansätze zur Visualisierung von Programmabläufen ausführlich untersucht und analysiert. Die Implementierung des Frameworks wird daraufhin detailliert beschrieben, wobei besonderer Wert auf die Erweiterbarkeit unterschiedlicher Sprachen gelegt wird.
Um die Eignung des Frameworks zu evaluieren, werden mehrere Aufgaben aus dem ersten Modul mit der jeweiligen Programmiersprache des Studiengangs Angewandte Informatik der Hochschule Offenburg betrachtet. Die Ergebnisse zeigen, dass das Framework mit den Aufgaben umgehen und diese korrekt und verständlich darstellen kann.
Due to its performance, the field of deep learning has gained a lot of attention, with neural networks succeeding in areas like Computer Vision (CV), Neural Language Processing (NLP), and 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 secondorder 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 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 neural architecture search (NAS) through the framework of group sparsity. Group sparsity is achieved through ℓ2,1-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
Convolutional neural networks (CNN) define the state-of-the-art solution on many perceptual tasks. However, current CNN approaches largely remain vulnerable against adversarial perturbations of the input that have been crafted specifically to fool the system while being quasi-imperceptible to the human eye. In recent years, various approaches have been proposed to defend CNNs against such attacks, for example by model hardening or by adding explicit defence mechanisms. Thereby, a small “detector” is included in the network and trained on the binary classification task of distinguishing genuine data from data containing adversarial perturbations. In this work, we propose a simple and light-weight detector, which leverages recent findings on the relation between networks’ local intrinsic dimensionality (LID) and adversarial attacks. Based on a re-interpretation of the LID measure and several simple adaptations, we surpass the state-of-the-art on adversarial detection by a significant m argin and reach almost perfect results in terms of F1-score for several networks and datasets. Sources available at: https://github.com/adverML/multiLID
In this paper, we describe a first publicly available fine-grained product recognition dataset based on leaflet images. Using advertisement leaflets, collected over several years from different European retailers, we provide a total of 41.6k manually annotated product images in 832 classes. Further, we investigate three different approaches for this fine-grained product classification task, Classification by Image, by Text, as well as by Image and Text. The approach "Classification by Text" uses the text extracted directly from the leaflet product images. We show, that the combination of image and text as input improves the classification of visual difficult to distinguish products. The final model leads to an accuracy of 96.4% with a Top-3 score of 99.2%. We release our code at https://github.com/ladwigd/Leaflet-Product-Classification.
The mathematical representations of data in the Spherical Harmonic (SH) domain has recently regained increasing interest in the machine learning community. This technical report gives an in-depth introduction to the theoretical foundation and practical implementation of SH representations, summarizing works on rotation invariant and equivariant features, as well as convolutions and exact correlations of signals on spheres. In extension, these methods are then generalized from scalar SH representations to Vectorial Harmonics (VH), providing the same capabilities for 3d vector fields on spheres.
Background
Internal tibial loading is influenced by modifiable factors with implications for the risk of stress injury. Runners encounter varied surface steepness (gradients) when running outdoors and may adapt their speed according to the gradient. This study aimed to quantify tibial bending moments and stress at the anterior and posterior peripheries when running at different speeds on surfaces of different gradients.
Methods
Twenty recreational runners ran on a treadmill at 3 different speeds (2.5 m/s, 3.0 m/s, and 3.5 m/s) and gradients (level: 0%; uphill: +5%, +10%, and +15%; downhill: –5%, –10%, and –15%). Force and marker data were collected synchronously throughout. Bending moments were estimated at the distal third centroid of the tibia about the medial–lateral axis by ensuring static equilibrium at each 1% of stance. Stress was derived from bending moments at the anterior and posterior peripheries by modeling the tibia as a hollow ellipse. Two-way repeated-measures analysis of variance were conducted using both functional and discrete statistical analyses.
Results
There were significant main effects for running speed and gradient on peak bending moments and peak anterior and posterior stress. Higher running speeds resulted in greater tibial loading. Running uphill at +10% and +15% resulted in greater tibial loading than level running. Running downhill at –10% and –15% resulted in reduced tibial loading compared to level running. There was no difference between +5% or –5% and level running.
Conclusion
Running at faster speeds and uphill on gradients ≥+10% increased internal tibial loading, whereas slower running and downhill running on gradients ≥–10% reduced internal loading. Adapting running speed according to the gradient could be a protective mechanism, providing runners with a strategy to minimize the risk of tibial stress injuries.
Selbsttests in Lernmanagementsystemen (LMS) ermöglichen es Studierenden, den eigenen Lernfortschritt einzuschätzen. Im Gegensatz zur Einreichung und Korrektur vollständig ausformulierter Aufgabenlösungen nutzen LMS überwiegend die Eingabe der Lösung im Antwort-Auswahl-Verfahren (Single-Choice). Nach didaktischen Ansatz „Physik durch Informatik“ geben die Lernenden stattdessen ihre Aufgabenlösungen in einer Programmiersprache ins LMS ein, was eine automatisierte Rückmeldung erleichtert und das Erreichen einer höheren Kompetenzstufe fördert. Es wurden zehn LMS-Selbsttests erstellt, bei denen die Lösungen zu einer Lehrbuch-Aufgabenstellung jeweils durch Eingabe in einer Programmiersprache und von einer Kontrollgruppe im Antwort-Auswahl-Verfahren abgefragt wurden. Ergebnisse aus dem ersten Einsatz dieser Selbsttests für die Lehrveranstaltung Physik im Studiengang Biotechnologie werden vorgestellt.
Künstliche Intelligenz (KI) durchdringt unser Leben immer stärker. Studierende werden im Alltag und an Hochschulen zunehmend mit KI-Anwendungen konfrontiert. An der Hochschule Offenburg werden deshalb KI-bezogene Lehrangebote curricular verankert, um Studierende im Erwerb von KI-Kompetenz zu unterstützen.
Der Beitrag stellt ein Konzept für die Entwicklung von Lehrveranstaltungen nach der Idee des pädagogischen Makings zur Förderung von KI-Kompetenz in der Hochschullehre vor. Konkretisiert wird das Konzept anhand eines Moduls zum Thema Chatbots, dessen Lehrinhalte interdisziplinär aus verschiedenen Perspektiven ausgearbeitet werden.
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.
The variable refrigerant flow system is one of the best heating, ventilation, and air conditioning systems (HVAC) thanks to its ability to provide thermal comfort inside buildings. But, at the same time, these systems are considered one of the most energy-consuming systems in the building sector. Thus, it is crucial to well size the system according to the building’s cooling and heating needs and the indoor temperature fluctuations. Although many researchers have studied the optimization of the building energy performance considering heating or cooling needs, using air handling units, radiant floor heating, and direct expansion valves, few studies have considered the use of multi-objective optimization using only the thermostat setpoints of VRF systems for both cooling and heating needs. Thus, the main aim of this study is to conduct a sensitivity analysis and a multi-objective optimization strategy for a residential building containing a variable refrigerant flow system, to evaluate the effect of the building performance on energy consumption and improve the building energy efficiency. The numerical model was based on the EnergyPlus, jEPlus, and jEPlus+EA simulation engines. The approach used in this paper has allowed us to reach significant quantitative energy saving by varying the cooling and heating setpoints and scheduling scenarios. It should be stressed that this approach could be applied to several HVAC systems to reduce energy-building consumption.
In den letzten Jahren haben Recommender Systeme zunehmend an Bedeutung gewonnen. Diese Systeme sind meist für Bereiche des E-Commerce konzipiert und berücksichtigen oftmals nicht den aktuellen Kontext der nutzenden Person. Recommender Systeme können allerdings nicht nur im E-Commerce zum Einsatz kommen, sondern finden ihren Anwendungszweck auch im Gesundheitswesen. Ziel dieser Bachelorarbeit ist es, ein Recommender System zu entwickeln, das den aktuellen Kontext der nutzenden Person (Chatverlauf, demografische Daten) besser berücksichtigen kann. Dazu befasst sich diese Arbeit mit der Konzeption und prototypischen Umsetzung eines kontextsensitiven Recommender Systems für einen bereits existierenden Chatbot aus dem Gesundheitswesen. Das in dieser Arbeit konzipierte und entwickelte Recommender System soll Mitarbeitende aus dem Gesundheits- und Sozialwesen entlasten und ihnen hilfreiche sowie thematisch sinnvolle Informationen zur Verfügung stellen. Basierend auf festgelegten Anforderungen wurde ein Konzept für das Recommender System entwickelt und zu Teilen als Prototyp umgesetzt. Abschließend wurde der Prototyp im Hinblick auf die Anforderungen evaluiert. Zudem fand eine technische Evaluation und eine Evaluation mithilfe von Anwendenden statt, welche den implementierten Prototypen bereits existierenden Systemen gegenüberstellte. Die von dem Prototyp empfohlenen Textausschnitte erzielten in der Evaluation mit nutzenden Personen eine thematisch signifikant höhere Übereinstimmung mit den Chatdaten.
In den letzten Jahren entstand ein regelrechter Hype um das Thema Kryptowährungen und sie sind in Gesellschaft, Politik und Wirtschaft kaum noch wegzudenken. Trotz der hohen Volatilität und des Risikos für Investoren werden Kryptowährungen teilweise auch als eine Alternative für herkömmliche Währungen angesehen. Daher stellt sich die Frage, ob die Einstellung zu Kryptowährungen und auch das Investitionsverhalten auf einer sorgfältigen Auseinandersetzung von Argumenten basiert. Dafür wurde die Forschungsfrage: „Basieren Kaufverhalten und Einstellung zu Kryptowährungen eher auf der intensiven Auseinandersetzung mit tatsächlichen Argumenten oder auf oberflächlichen Reizen?“ aufgestellt. Zur Untersuchung dieser wurde eine Umfrage mit 283 Teilnehmenden durchgeführt. Auf Basis theoretischer Überlegungen zur Einstellungsforschung durch das Elaboration-Likelihood-Model wurde ein Fragebogen entworfen, der den Einfluss der Elaboration auf die Einstellung und das Kaufverhalten empirisch messbar machen sollte. Durch eine Kausalanalyse mittels Strukturgleichungsmodell konnte ein teilweise signifikanter Einfluss von elaborationsbestimmenden Größen auf Einstellung und Kaufverhalten festgestellt werden. Eine Überprüfung der Gütekriterien des Fragebogens mittels explorativer und konfirmatorischer Faktoranalyse ergab in Hinsicht auf Reliabilität und Validität jedoch keine zufriedenstellenden Ergebnisse. Die Ergebnisse des Kausalmodells sollten deswegen mit Vorsicht betrachtet werden. In weiterführenden Forschungen könnte die Struktur der durch den Fragebogen erhobenen Konstrukte für Elaboration und Einstellung überarbeitet werden, um eine bessere Reliabilität und Validität zu erreichen und somit genauere Aussagen über die eigentlichen Beziehungen der Konstrukte treffen zu können.
Public export credits and trade insurance require a global framework of institutions, rules and regulations to avoid subsidies and a race to the bottom. The extensive modernisation of the Arrangement on Officially Supported Export Credits (Arrangement) of the Organisation for Economic Co-operation and Development intends to re-level the playing field. This Practitioner Commentary describes the demand for adequate government interventions, considers the need for the reform and discusses key aspects of the new Arrangement. We argue that there is a breakthrough in several important areas such as tenors, repayment terms and green finance. However, we also find that the modernisation falls short in areas such as the interplay between different rulebooks, pre-shipment instruments' regulations and climate action.
Additive manufacturing enables the production of lightweight and resilient components with extensive design freedom. In the low-cost sector, material extrusion (e.g. Fused Deposition Modeling - FDM) has been the main method used to date. Thus, robust 3D printers and inexpensive 3D materials (polymer filaments) can be used. However, the printing times for FDM are very long and the quality of the dimensions and surfaces is limited. Recently, new processes from the field of Vat polymerization have entered the market. For example, masked stereolithography (mSLA) offers a significant improvement in component quality and build speed through the use of resins and large-area curing at still reasonable costs. Currently, there is only limited knowledge available on the optimal design of components using this young process. In this contribution, design guidelines are developed to determine the possibilities and limitations of mSLA from a design point of view. For this purpose, a number of test geometries are designed and investigated to obtain systematic insights into important design features, such as wall thickness, grooves and holes. In addition, typical problems in additive manufacturing, such as the design of overhangs and fits or the hollowing of components, are investigated. The evaluation of practical 3D printing tests thus provides important parameters that can be transferred to design guidelines of components for additive manufacturing using mSLA.
eLetter zum Artikel "Condiciones neuropsi-quiátricas y probable causa de muerte de Maurice Ravel" von Gómez-Carvajal AM, Botero-Meneses JS, Palacios-Espinosa X und Palacios-Sánchez L., veröffentlicht in Iatreia 35(3), Seite 341-8 (DOI: https://doi.org/10.17533/udea.iatreia.154).
A balcony photovoltaic (PV) system, also known as a micro-PV system, is a small PV system consisting of one or two solar modules with an output of 100–600 Wp and a corresponding inverter that uses standard plugs to feed the renewable energy into the house grid. In the present study we demonstrate the integration of a commercial lithium-ion battery into a commercial micro-PV system. We firstly show simulations over one year with one second time resolution which we use to assess the influence of battery and PV size on self-consumption, self-sufficiency and the annual cost savings. We then develop and operate experimental setups using two different architectures for integrating the battery into the micro-PV system. In the passive hybrid architecture, the battery is in parallel electrical connection to the PV module. In the active hybrid architecture, an additional DC-DC converter is used. Both architectures include measures to avoid maximum power point tracking of the battery by the module inverter. Resulting PV/battery/inverter systems with 300 Wp PV and 555 Wh battery were tested in continuous operation over three days under real solar irradiance conditions. Both architectures were able to maintain stable operation and demonstrate the shift of PV energy from the day into the night. System efficiencies were observed comparable to a reference system without battery. This study therefore demonstrates the feasibility of both active and passive coupling architectures.
Electrochemical pressure impedance spectroscopy (EPIS) is an emerging tool for the diagnosis of polymer electrolyte membrane fuel cells (PEMFC). It is based on analyzing the frequency response of the cell voltage with respect to an excitation of the gas-phase pressure. Several experimental studies in the past decade have shown the complexity of EPIS signals, and so far there is no agreement on the interpretation of EPIS features. The present study contributes to shed light into the physicochemical origin of EPIS features, by using a combination of pseudo-two-dimensional modeling and analytical interpretation. Using static simulations, the contributions of cathode equilibrium potential, cathode overpotential, and membrane resistance on the quasi-static EPIS response are quantified. Using model reduction, the EPIS responses of individual dynamic processes are predicted and compared to the response of the full model. We show that the EPIS signal of the PEMFC studied here is dominated by the humidifier. The signal is further analyzed by using transfer functions between various internal cell states and the outlet pressure excitation. We show that the EPIS response of the humidifier is caused by an oscillating oxygen molar fraction due to an oscillating mass flow rate.
Electrochemical pressure impedance spectroscopy (EPIS) has received the attention of researchers as a method to study mass transport processes in polymer electrolyte mem-brane fuel cells (PEMFC). It is based on analyzing the cell voltage response to a harmonic excitation of the gas phase pressure in the frequency domain. Several experiments with a single-cell fuel cell have shown that the spectra contain information in the frequency range typical for mass transport processes and are sensitive to specific operating condi-tions and structural fuel cell parameters. To further benefit from the observed features, it is essential to identify why they occur, which to date has not yet been accomplished. The aim of the present work, therefore, is to identify causal links between internal processes and the corresponding EPIS features.
To this end, the study follows a model-based approach, which allows the analysis of inter-nal states that are not experimentally accessible. The PEMFC model is a pseudo-2D model, which connects the mass transport along the gas channel with the mass transport through the membrane electrode assembly. A modeling novelty is the consideration of the gas vol-ume inside the humidifier upstream the fuel cell inlet, which proves to be crucial for the reproduction of EPIS. The PEMFC model is parametrized to a 100 cm² single cell of the French project partner, who provided the experimental EPIS results reproduced and in-terpreted in the present study.
The simulated EPIS results show a good agreement with the experiments at current den-sities ≤ 0.4 A cm–2, where they allow a further analysis of the observed features. At the lowest excitation frequency of 1 mHz, the dynamic cell voltage response approaches the static pressure-voltage response. In the simulated frequency range between 1 mHz – 100 Hz, the cell voltage oscillation is found to strongly correlate with the partial pressure oscillation of oxygen, whereas the influence of the water pressure is limited to the low frequency region.
The two prominent EPIS features, namely the strong increase of the cell voltage oscillation and the increase of phase shift with frequency, can be traced back via the oxygen pressure to the oscillation of the inlet flow rate. The phenomenon of the oscillating inlet flow rate is a consequence of the pressure change of the gas phase inside the humidifier and in-creases with frequency. This important finding enables the interpretation of experimen-tally observed EPIS trends for a variation of operational and structural fuel cell parame-ters by tracing them back to the influence of the oscillating inlet flow rate.
The separate simulation of the time-dependent processes of the PEMFC model through model reduction shows their individual influence on EPIS. The sluggish process of the wa-ter uptake by the membrane is visible below 0.1 Hz, while the charge and discharge of the double layer becomes visible above 1 Hz. The gas transport through the gas diffusion layer is only visible above 100 Hz. The simulation of the gas transport through the gas channel
without consideration of the humidifier becomes visible above 1 Hz. With consideration of the humidifier the gas transport through the gas channel is visible throughout the fre-quency range. The strong similarity of the spectra considering the humidifier with the spectra of the full model setup shows the dominant influence of the humidifier on EPIS.
A promising observation is the change in the amplitude relationship between the cell volt-age and the oxygen partial pressure oscillation as a function of the oxygen concentration in the catalyst layer. At a frequency where the influence of oxygen pressure on the cell voltage is dominant, for example at 1 Hz, the amplitude of the cell voltage oscillation could be used to indirectly measure the oxygen concentration in the catalyst layer.