PUMA publications for /https://puma.uni-kassel.de/PUMA RSS feed for /2024-04-10T22:53:48+02:00An Examination of Organic Computing Strategies in Design Optimizationhttps://puma.uni-kassel.de/bibtex/2a8bae3d276fcc8c617284a15017b0f58/04068750040687502024-04-09T12:07:03+02:00imported itegpub isac-www DesignOptimization DeepLearning ActiveLearning GenerativeLearning <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Jens Decke" itemprop="url" href="/author/Jens%20Decke"><span itemprop="name">J. Decke</span></a></span>. </span><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"><em><span itemprop="name">Organic Computing -- Doctoral Dissertation Colloquium 2023</span>, </em><em><span itemprop="publisher">kassel university press</span>, </em></span>(<em><span>2023<meta content="2023" itemprop="datePublished"/></span></em>)Tue Apr 09 12:07:03 CEST 2024Organic Computing -- Doctoral Dissertation Colloquium 202341--54An Examination of Organic Computing Strategies in Design Optimization2023imported itegpub isac-www DesignOptimization DeepLearning ActiveLearning GenerativeLearning In the field of design optimization, numerical equation solvers,
commonly used for finite element analysis and computational fluid dy-
namics, impose significant computational demands. The iterative nature
of design optimization, involving numerous simulations, magnifies re-
source requirements in terms of time, energy, costs, and greenhouse gas
emissions. Accelerating this process has been a long-standing research
challenge, driven by the potential for resource savings and the ability
to tackle increasingly complex problems. Recently, organic computing
methods have gained prominence as promising approaches to address
this challenge. This study aims to bridge the gap by integrating tech-
niques from deep learning, active learning, and generative learning into
the field of design optimization. The goal is to accelerate the design op-
timization process while addressing specific challenges such as exploring
large and complex design spaces and evaluating the advantages and con-
straints of these methods. This research has the potential to significantly
impact the industrial use of design optimization by providing faster and
more efficient tools.Machine Learning Supported Optimisation and Experimental Evaluation of Electrical Motors for Small Urban Passenger Vehicleshttps://puma.uni-kassel.de/bibtex/2f68d78692b7048293d759eb4527731be/04068750040687502024-04-09T12:01:11+02:00imported itegpub isac-www electrical_traction_machines sustainable_mobility data-driven_models machine_learning multi-objective_optimisation experimental_evaluation deep_learning_techniques test_benches <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Diego Botache" itemprop="url" href="/author/Diego%20Botache"><span itemprop="name">D. Botache</span></a></span>. </span><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"><em><span itemprop="name">Organic Computing -- Doctoral Dissertation Colloquium 2023</span>, </em><em><span itemprop="publisher">kassel university press</span>, </em></span>(<em><span>2023<meta content="2023" itemprop="datePublished"/></span></em>)Tue Apr 09 12:01:11 CEST 2024Organic Computing -- Doctoral Dissertation Colloquium 202314--26Machine Learning Supported Optimisation and Experimental Evaluation of Electrical Motors for Small Urban Passenger Vehicles2023imported itegpub isac-www electrical_traction_machines sustainable_mobility data-driven_models machine_learning multi-objective_optimisation experimental_evaluation deep_learning_techniques test_benches The increasing environmental pollution and rising global energy demands require the rapid development of novel and sustainable traction machines from urban transport up to large-scale transport vehicles. The optimisation of the development cycle of electrical traction machines plays a vital role in sustainable mobility. Still, developers have to face the difficulty of analysing time-consuming multiphysics simulations to assess the performance of many prototype variants. Accelerating this process involves implementing machine learning-supported strategies that can effectively acquire results and predictions about the performance and efficiency of prototypes in an iterative optimisation process and detect possible issues during the experimental procedure. The following research project streamlines the development cycle by identifying and presenting multiple strategies for applying and evaluating machine learning and deep learning techniques in different stages and areas in the development cycle including self-optimisation and self-adaptation capabilities. By implementing these strategies, the optimisation process can be further improved and become more efficient and robust, reducing the time and cost required for testing and validation. Therefore, this proposal addresses two application areas. The first area consists of the supported machine learning motor performance prediction at the topology design and optimisation stage of structural components. The second area corresponds to the machine learning-supported experimental evaluation process of prototypes at desired test benches. Enhancing Multi-Objective Optimization through Machine Learning-Supported Multiphysics Simulationhttps://puma.uni-kassel.de/bibtex/2730ae1996394beab35bc8f865eadaf67/04068750040687502024-04-09T11:57:48+02:00imported itegpub isac-www Electric_Motors Multiobjective_Optimisation Surrogate-Modelling Deep-Learning Explainable_Artificial_Intelligence <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Diego Botache" itemprop="url" href="/author/Diego%20Botache"><span itemprop="name">D. Botache</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Jens Decke" itemprop="url" href="/author/Jens%20Decke"><span itemprop="name">J. Decke</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Winfried Ripken" itemprop="url" href="/author/Winfried%20Ripken"><span itemprop="name">W. Ripken</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Abhinay Dornipati" itemprop="url" href="/author/Abhinay%20Dornipati"><span itemprop="name">A. Dornipati</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Franz Götz-Hahn" itemprop="url" href="/author/Franz%20G%c3%b6tz-Hahn"><span itemprop="name">F. Götz-Hahn</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Mohamed Ayeb" itemprop="url" href="/author/Mohamed%20Ayeb"><span itemprop="name">M. Ayeb</span></a></span>, and <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Bernhard Sick" itemprop="url" href="/author/Bernhard%20Sick"><span itemprop="name">B. Sick</span></a></span>. </span><span itemtype="http://schema.org/PublicationIssue" itemscope="itemscope" itemprop="isPartOf"><span itemtype="http://schema.org/Periodical" itemscope="itemscope" itemprop="isPartOf"><span itemprop="name"><em>arXiv e-prints</em></span></span> </span>(<em><span>2024<meta content="2024" itemprop="datePublished"/></span></em>)Tue Apr 09 11:57:48 CEST 2024arXiv e-printsarXiv:2309.13179v2Enhancing Multi-Objective Optimization through Machine Learning-Supported Multiphysics Simulation2024imported itegpub isac-www Electric_Motors Multiobjective_Optimisation Surrogate-Modelling Deep-Learning Explainable_Artificial_Intelligence This paper presents a methodological framework for training, self-optimising, and self-organising surrogate models to approximate and speed up multiobjective optimisation of technical systems based on multiphysics simulations. At the hand of two real-world datasets, we illustrate that surrogate models can be trained on relatively small amounts of data to approximate the underlying simulations accurately. Including explainable AI techniques allow for highlighting feature relevancy or dependencies and supporting the possible extension of the used datasets. One of the datasets was created for this paper and is made publicly available for the broader scientific community. Extensive experiments combine four machine learning and deep learning algorithms with an evolutionary optimisation algorithm. The performance of the combined training and optimisation pipeline is evaluated by verifying the generated Pareto-optimal results using the ground truth simulations. The results from our pipeline and a comprehensive evaluation strategy show the potential for efficiently acquiring solution candidates in multiobjective optimisation tasks by reducing the number of simulations and conserving a higher prediction accuracy, i.e., with a MAPE score under 5% for one of the presented use cases.DADO – Low-Cost Query Strategies for Deep Active Design Optimizationhttps://puma.uni-kassel.de/bibtex/2e6ecfa5a1489519a2c4c3b9f08ceb2ee/04068750040687502024-03-21T17:06:48+01:00imported itegpub isac-www Self-Optimization Self-Supervised-Learning Design-Optimization Active-Learning Numerical-Simulation <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Jens Decke" itemprop="url" href="/author/Jens%20Decke"><span itemprop="name">J. Decke</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Christian Gruhl" itemprop="url" href="/author/Christian%20Gruhl"><span itemprop="name">C. Gruhl</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Lukas Rauch" itemprop="url" href="/author/Lukas%20Rauch"><span itemprop="name">L. Rauch</span></a></span>, and <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Bernhard Sick" itemprop="url" href="/author/Bernhard%20Sick"><span itemprop="name">B. Sick</span></a></span>. </span><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"><em><span itemprop="name">International Conference on Machine Learning and Applications (ICMLA)</span>, </em></span><em>page <span itemprop="pagination">1611--1618</span>. </em><em><span itemprop="publisher">IEEE</span>, </em>(<em><span>2023<meta content="2023" itemprop="datePublished"/></span></em>)Thu Mar 21 17:06:48 CET 2024International Conference on Machine Learning and Applications (ICMLA)1611--1618DADO – Low-Cost Query Strategies for Deep Active Design Optimization2023imported itegpub isac-www Self-Optimization Self-Supervised-Learning Design-Optimization Active-Learning Numerical-Simulation In this work, we apply deep active learning to the field of design optimization to reduce the number of computationally expensive numerical simulations widely used in industry and engineering.
We are interested in optimizing the design of structural components, where a set of parameters describes the shape. If we can predict the performance based on these parameters and consider only the promising candidates for simulation, there is an enormous potential for saving computing power.
We present two query strategies for self-optimization to reduce the computational cost in multi-objective design optimization problems. Our proposed methodology provides an intuitive approach that is easy to apply, offers significant improvements over random sampling, and circumvents the need for uncertainty estimation.
We evaluate our strategies on a large dataset from the domain of fluid dynamics and introduce two new evaluation metrics to determine the model's performance.
Findings from our evaluation highlights the effectiveness of our query strategies in accelerating design optimization.
Furthermore, the introduced method is easily transferable to other self-optimization problems in industry and engineering.Towards Few-Shot Time Series Anomaly Detection with Temporal Attention and Dynamic Thresholdinghttps://puma.uni-kassel.de/bibtex/2a4a29acb67656f837ca6e532fc88958d/04068750040687502024-03-21T17:06:17+01:00imported itegpub isac-www few-shot learning anomaly detection temporal attention dynamic thresholding <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Chandana Priya Nivarthi" itemprop="url" href="/author/Chandana%20Priya%20Nivarthi"><span itemprop="name">C. Nivarthi</span></a></span>, and <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Bernhard Sick" itemprop="url" href="/author/Bernhard%20Sick"><span itemprop="name">B. Sick</span></a></span>. </span><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"><em><span itemprop="name">International Conference on Machine Learning and Applications (ICMLA)</span>, </em></span><em>page <span itemprop="pagination">1444--1450</span>. </em><em><span itemprop="publisher">IEEE</span>, </em>(<em><span>2023<meta content="2023" itemprop="datePublished"/></span></em>)Thu Mar 21 17:06:17 CET 2024International Conference on Machine Learning and Applications (ICMLA)1444--1450Towards Few-Shot Time Series Anomaly Detection with Temporal Attention and Dynamic Thresholding2023imported itegpub isac-www few-shot learning anomaly detection temporal attention dynamic thresholding Anomaly detection plays a pivotal role in diverse realworld applications such as cybersecurity, fault detection, network
monitoring, predictive maintenance, and highly automated driving. However, obtaining labeled anomalous data can be a formidable
challenge, especially when anomalies exhibit temporal evolution. This paper introduces LATAM (Long short-term memory Autoencoder with Temporal Attention Mechanism) for few-shot anomaly detection, with the aim of enhancing detection performance in scenarios with limited labeled anomaly data. LATAM effectively captures temporal dependencies and emphasizes significant patterns in multivariate time series data. In our investigation, we
comprehensively evaluate LATAM against other anomaly detection models, particularly assessing its capability in few-shot learning
scenarios where we have minimal examples from the normal class and none from the anomalous class in the training data. Our
experimental results, derived from real-world photovoltaic inverter data, highlight LATAM’s superiority, showcasing a substantial
27% mean F1 score improvement, even when trained on a mere two-week dataset. Furthermore, LATAM demonstrates remarkable
results on the open-source SWaT dataset, achieving a 12% boost in accuracy with only two days of training data. Moreover, we
introduce a simple yet effective dynamic thresholding mechanism, further enhancing the anomaly detection capabilities of LATAM.
This underscores LATAM’s efficacy in addressing the challenges posed by limited labeled anomalies in practical scenarios and it
proves valuable for downstream tasks involving temporal representation and time series prediction, extending its utility beyond
anomaly detection applications.