@article{65099, author = {{Weber, Daniel and Schmies, Dominik and Lange, Jarren H. and Schenke, Maximilian and Wallscheid, Oliver}}, issn = {{2169-3536}}, journal = {{IEEE Access}}, pages = {{38517--38535}}, publisher = {{Institute of Electrical and Electronics Engineers (IEEE)}}, title = {{{Optimal Control of Voltage-Forming Grid Inverters by Model Predictive Control and Reinforcement Learning}}}, doi = {{10.1109/access.2026.3670948}}, volume = {{14}}, year = {{2026}}, } @article{65098, author = {{Weber, Daniel and Lange, Jarren and Wallscheid, Oliver}}, issn = {{2687-9735}}, journal = {{IEEE Journal of Emerging and Selected Topics in Industrial Electronics}}, pages = {{1--12}}, publisher = {{Institute of Electrical and Electronics Engineers (IEEE)}}, title = {{{Reinforcement Learning-Based Control of Voltage-Forming Grid Inverters With Arbitrary Loads}}}, doi = {{10.1109/jestie.2026.3654784}}, year = {{2026}}, } @article{65253, author = {{Abdelwanis, Ali Hassan Ali and Haucke-Korber, Barnabas and Jakobeit, Darius and Kirchgässner, Wilhelm and Meyer, Marvin and Schenke, Maximilian and Vater, Hendrik and Wallscheid, Oliver and Weber, Daniel}}, issn = {{2577-3569}}, journal = {{Journal of Open Source Education}}, number = {{97}}, publisher = {{The Open Journal}}, title = {{{Reinforcement Learning: A Comprehensive Open-Source Course}}}, doi = {{10.21105/jose.00306}}, volume = {{9}}, year = {{2026}}, } @article{59805, abstract = {{The LLC converter achieves the highest efficiency in resonant operation. Conventionally, the input DC-link voltage is controlled to operate the LLC converter at resonance for the given operating point. However, the DC-link capacitor voltage shows a low-frequency voltage ripple (typically the second harmonic of grid frequency) in cascaded converters so that the LLC has to adapt its switching frequency within the grid period. Conventionally, the LLC converter operates 50% of the time above the resonant frequency of 40 kHz and 50% below resonance. Both operating conditions cause additional losses. However, experimental measurements indicate that the below-resonance operation causes significantly higher losses than above-resonance operation due to much higher primary and secondary transformer currents. It is better to increase the DC-link voltage by 30% of the peak-to-peak low-frequency voltage ripple to mostly avoid below-resonance operation (i.e., from 650 V to 680 V in this case). With the proposed control, the LLC converter operates about 75% of time over resonance and only 25% of time below resonance. The overall efficiency increases from 97.66% to 97.7% for the average operating point with an 80% load current. This corresponds to a 2% total loss reduction. Finally, the peak resonance capacitor voltage decreases from 910 V to 790 V (−13%).}}, author = {{Unruh, Roland and Böcker, Joachim and Schafmeister, Frank}}, issn = {{2079-9292}}, journal = {{Electronics}}, keywords = {{adaptive DC-link voltage, cascaded H-bridge, resonant operation, Full-Bridge Converter, loss minimization, LLC Resonant Converter, peak capacitor voltage reduction}}, number = {{8}}, publisher = {{MDPI AG}}, title = {{{Adaptive DC-Link Voltage Control for 22 kW, 40 kHz LLC Resonant Converter Considering Low-Frequency Voltage Ripple}}}, doi = {{10.3390/electronics14081517}}, volume = {{14}}, year = {{2025}}, } @inproceedings{63157, abstract = {{Three-phase cascaded H-bridge converters (CHBs) in star configuration require reliable current controllers to evenly charge the module DC-link capacitors. Conventionally, a current control in dq-coordinates is utilized. At steady state, the resulting calculated reference arm voltages are sinusoidal, have identical amplitudes and show a phase shift of 120 degree to each other. For balanced grid inductors, the resulting grid currents also have the same amplitude. However, own simulations show that unbalanced grid inductors always lead to different grid current amplitudes (4% difference in this case). As a result, the averaged charging module powers differ and the peak DC-link capacitor voltage rises as well. In the first step, an adaptation of an existing zero-sequence voltage injection is proposed. For balanced grid inductors, it converges to the 3rd harmonic voltage injection which can reduce the peak-to-peak DC-link voltage ripple up by to 50% and balances the power between the phases. However, unbalanced grid inductors still lead to the same unbalanced grid currents of 4%. Therefore, a new method with 4 integrators based on linear regression is proposed to achieve sinusoidal grid currents for unbalanced inductors. The proposed method has a similar transient dynamic as the conventional dq control, but balances the grid currents nearly ideally. Simulation results of a 1MW cascaded H bridge and a scaled-down prototype verify the proposed method.}}, author = {{Unruh, Roland and Böcker, Joachim and Schafmeister, Frank}}, booktitle = {{2025 Energy Conversion Congress & Expo Europe (ECCE Europe)}}, keywords = {{Cascaded H-Bridge, Current Control, dq Transformation, Linear Regression, Unbalanced Inductors}}, location = {{Birmingham, United Kingdom}}, publisher = {{IEEE}}, title = {{{Three-Phase Instantaneous Current Controller for Unbalanced Grid Inductors Without DQ Transform for Cascaded H-Bridge Converters}}}, doi = {{10.1109/ecce-europe62795.2025.11238538}}, year = {{2025}}, } @inproceedings{60746, author = {{Jakobeit, Darius and Peña López, Mario and Schenke, Maximilian and Haucke-Korber, Barnabas and Wallscheid, Oliver}}, booktitle = {{2025 IEEE International Electric Machines & Drives Conference (IEMDC)}}, publisher = {{IEEE}}, title = {{{Structural Optimization of Meta-Reinforcement Learning-based Finite-Control-Set Direct Torque Control of Permanent Magnet Synchronous Motors}}}, doi = {{10.1109/iemdc60492.2025.11061179}}, year = {{2025}}, } @inproceedings{60745, author = {{Haucke-Korber, Barnabas and Aung, Nyi Nyi and Schenke, Maximilian and Peña López, Mario and Jakobeit, Darius and Wallscheid, Oliver}}, booktitle = {{2025 IEEE International Electric Machines & Drives Conference (IEMDC)}}, publisher = {{IEEE}}, title = {{{Reinforcement Learning-based Direct Torque Control of Externally Excited Synchronous Motors: a Proof of Concept}}}, doi = {{10.1109/iemdc60492.2025.11061093}}, year = {{2025}}, } @inproceedings{60744, author = {{Peña López, Mario and Schenke, Maximilian and Jakobeit, Darius and Haucke-Korber, Barnabas and Wallscheid, Oliver}}, booktitle = {{2025 IEEE International Electric Machines & Drives Conference (IEMDC)}}, publisher = {{IEEE}}, title = {{{Reinforcement Learning Control of Three-Level Converter Permanent Magnet Synchronous Machine Drives}}}, doi = {{10.1109/iemdc60492.2025.11061032}}, year = {{2025}}, } @article{63498, author = {{Kirchgässner, Wilhelm and Förster, Nikolas and Piepenbrock, Till and Schweins, Oliver and Wallscheid, Oliver}}, journal = {{IEEE Transactions on Power Electronics}}, keywords = {{Mathematical models, Estimation, Data models, Convolutional neural networks, Accuracy, Magnetic hysteresis, Magnetic cores, Temperature measurement, Magnetic domains, Temperature distribution, Convolutional neural network (CNN), machine learning (ML), magnetics}}, number = {{2}}, pages = {{3326--3335}}, title = {{{HARDCORE: H-Field and Power Loss Estimation for Arbitrary Waveforms With Residual, Dilated Convolutional Neural Networks in Ferrite Cores}}}, doi = {{10.1109/TPEL.2024.3488174}}, volume = {{40}}, year = {{2025}}, } @inproceedings{63496, author = {{Foerster, Nikolas and Urbaneck, Daniel and Schenke, Maximilian and Ebers, Anastacia and Schoenlau, Nicolas and Wallscheid, Oliver and Schafmeister, Frank}}, booktitle = {{PCIM Conference 2025; International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management}}, pages = {{2881--2890}}, title = {{{Improving the Usability of Calorimetric Measuring Chambers for Reliable Thermal Measurements}}}, doi = {{10.30420/566541386}}, year = {{2025}}, } @inproceedings{65097, author = {{Weber, Daniel and Lange, Jarren and Wallscheid, Oliver}}, booktitle = {{2025 IEEE Kiel PowerTech}}, publisher = {{IEEE}}, title = {{{Safe Reinforcement Learning-based Control for a Voltage Source Inverter Operating in an Unbalanced Grid}}}, doi = {{10.1109/powertech59965.2025.11180230}}, year = {{2025}}, } @inproceedings{54355, author = {{Urbaneck, Daniel and Wiegard, Jan and Schafmeister, Frank}}, booktitle = {{PCIM Europe 2024; IEEE International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management}}, location = {{Nuremberg}}, title = {{{Analysis of Inverter Operation Modes of an IGBT-Based ZCS LLC Converter for a 2 kW Automotive On-Board DC-DC}}}, year = {{2024}}, } @inproceedings{54353, author = {{Piepenbrock, Till and Keuck, Lukas and Schachten, Sebastian and Böcker, Joachim and Schafmeister, Frank}}, booktitle = {{PCIM Europe 2024; IEEE International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management}}, location = {{Nuremberg}}, title = {{{Study on Sample Geometries for Ferrite Characterisation in the MHz-Range}}}, year = {{2024}}, } @inproceedings{54354, author = {{Förster, Nikolas and Urbaneck, Daniel and Kohlhepp, Benedikt and Kübrich, Daniel and Schafmeister, Frank}}, booktitle = {{PCIM Europe 2024; IEEE International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management}}, location = {{Nuremberg}}, title = {{{Pitfalls and their Avoidability in the Double-Pulse Test}}}, year = {{2024}}, } @article{53309, author = {{Hölsch, Lukas and Brosch, Anian and Steckel, Richard and Braun, Tristan and Wendel, Sebastian and Böcker, Joachim and Wallscheid, Oliver}}, issn = {{0885-8969}}, journal = {{IEEE Transactions on Energy Conversion}}, keywords = {{Electrical and Electronic Engineering, Energy Engineering and Power Technology}}, pages = {{1--12}}, publisher = {{Institute of Electrical and Electronics Engineers (IEEE)}}, title = {{{Insights and Challenges of Co-Simulation-Based Optimal Pulse Pattern Evaluation for Electric Drives}}}, doi = {{10.1109/tec.2024.3374962}}, year = {{2024}}, } @inproceedings{54357, author = {{Piepenbrock, Till and Schafmeister, Frank and Böcker, Joachim}}, booktitle = {{SPEEDAM 2024; 27th International Symposium on Power Electronics, Electrical Drives, Automation and Motion}}, location = {{Ischia, near Naples, Italy}}, title = {{{FEM Modelling of Dimensional-Resonant Inductors for LLC Converters in MHz Range}}}, doi = {{10.1109/SPEEDAM61530.2024.10609111}}, year = {{2024}}, } @inproceedings{54356, abstract = {{Although there are numerous design and control methodologies for the LLC resonant converter, they often do not consider decentralized control strategies to operate them as isolated DC-DC converters within a cascaded H-bridge. The total output power of all LLC converters must be constant to supply a load such as a wa- ter electrolyzer. However, each individual LLC converter can vary its output power as long as the total output power remains constant. This opens new possibilities in increasing the system efficiency and robustness. Usually, the DC-link voltage of each module capacitor shows a 2nd harmonic voltage ripple. However, the total stored energy in all DC-link capacitors is constant within a grid period for a balanced three-phase system. By controlling each LLC converter’s output power locally to be proportional to the energy stored in its DC-link capacitor, modules with a lower instantaneous DC-link voltage transfer less power to the load than modules with a higher DC-link voltage. As a result, a higher efficiency, voltage gain and lower peak resonant capacitor voltage can be achieved with the same components. The 22.2kW experimental prototype of the LLC converter reaches an efficiency of over 97% at resonance which is similar to the precalculated value.}}, author = {{Unruh, Roland and Böcker, Joachim and Schafmeister, Frank}}, booktitle = {{ECCE Europe 2024; IEEE Energy Conversion Congress & Exposition Europe}}, isbn = {{979-8-3503-6444-6}}, keywords = {{Cascaded H-Bridge, Converter Losses, Decentralized Control, Full-Bridge Converter, LLC Resonant Converter}}, location = {{Darmstadt, Germany}}, publisher = {{IEEE}}, title = {{{Experimentally Verified 22 kW, 40 kHz LLC Resonant Converter Design with new Control for a 1 MW Cascaded H-Bridge Converter}}}, doi = {{https://doi.org/10.1109/ECCEEurope62508.2024.10751954}}, year = {{2024}}, } @inproceedings{58648, author = {{Unruh, Roland and Böcker, Joachim and Schafmeister, Frank}}, booktitle = {{Proceedings of the Energy Conversion Congress & Expo (ECCE Europe)}}, location = {{Darmstadt}}, publisher = {{IEEE}}, title = {{{Experimentally Verified 22 kW, 40 kHz LLC Resonant Converter Design with new Control for a 1 MW Cascaded H-Bridge Converter}}}, doi = {{10.1109/ECCEEurope62508.2024.10751954}}, year = {{2024}}, } @phdthesis{58756, abstract = {{Der Permanentmagnet-Synchronmotor (PMSM) ist aufgrund seiner hohen Leistungs- und Drehmomentdichte bezogen auf Volumen und Gewicht ein häufig verwendeter Traktionsmotor in Automobilanwendungen. Jene Charakteristika werden jedoch maßgeblich durch Temperaturhöchstwerte begrenzt. Hinzu kommt, dass die Temperatur wichtiger Rotorkomponenten nicht wirtschaftlich messbar ist. Temperaturschätzverfahren wie modellbasierte Ansätze sind potentiell in der Lage, das Problem der fehlenden Temperaturinformation zu relativieren, ohne zusätzliche Geräte zu erfordern. Diese Arbeit stellt ein Portfolio von thermischen Modellen aus dem Bereich des maschinellen Lernens zusammen. Die Untersuchung basiert auf einem PMSM-Datensatz, der auf einem Prüfstand aufgezeichnet wurde. Neben dem durchschnittlichen Schätzfehler diktiert die erforderliche Anzahl von Modellparametern zahlreiche Auslegungsentscheidungen. Der gesamte Entwurfsprozess eines Modells aus dem maschinellen Lernen wird beleuchtet und für verschiedene lineare, sowie baumbasierte Modelle; vorschiebende, rekurrente und faltende neuronale Netze als auch für verschiedene hybride Modellierungsansätze durchgeführt. Desweiteren wird der hybride Modellierungsansatz über thermische neuronale Netze besonders hervorgehoben. Sie setzen sich aus neuronalen Netzen und einem thermischen Ersatzschaltbild zusammen und wurden erstmals vom Autor dieser Arbeit veröffentlicht. Schließlich wird ein von Experten entworfenes, datengetriebenes thermisches Netz mit konzentrierten Parametern über verschiedene Algorithmen optimiert und als Stand der Technik herangezogen.}}, author = {{Kirchgässner, Wilhelm}}, publisher = {{LibreCat University}}, title = {{{Data-driven thermal modeling of a permanent magnet synchronous motor with machine learning}}}, doi = {{10.17619/UNIPB/1-2068}}, year = {{2024}}, } @phdthesis{58757, abstract = {{On-bord DC-DC-Konverter sind das Bindeglied zwischen der Traktionsbatterie und der Hilfsbatterie und versorgen wichtige Komponenten des Elektrofahrzeugs. Diese Arbeit adressiert den weiten Spannungsbereich des Wandlers, der eine Folge der variierenden Spannungen der Batterien ist. Als potentielle Topologien werden der LLC Resonanzwandler, der aktiv geklemmte Flusswandler und der isolierte Vollbrücken-Konverter untersucht.Zunächst wird hierbei der LLC untersucht und verschiedene Modulationstechniken zur Abdeckung des weiten Spannungsbereichs gegenübergestellt, um zu zeigen, dass die Frequenzverdoppler-Modulation und die alternierende Phasenverschiebungsmodulation die maximale Temperatur der Halbleiter deutlich senken. Zum Wechsel zwischen Voll- und Halbbrückenmodulation wird eine Modulationstechnik vorgeschlagen, welche den transienten Magnetisierungsfluss um über 70 % respektive des konventionellen Konzept senkt. Für den aktiv geklemmten Flusswandler wird ein verbessertes Modell vorgestellt, das die Blockierspannung sehr genau modelliert. Zudem wird eine Snubber-Schaltung vorgeschlagen, welche die sekundärseitige transiente Blockierspannung deutlich reduziert. Für den isolierten Vollbrücken-Konverter werden hart- und weichschaltende Modulationstechniken analysiert und eine hartschaltende Frequenz-Verdoppler-Modulationstechnik vorgeschlagen, welche die maximale Schaltertemperatur deutlich reduziert und eine Modulationstechnik mit Beschaltung vorgestellt, um zwischen dem Voll- und Halbbrückenmodus zu wechseln. Die zuvor erarbeiteten Konverter werden unter Anwendung einer vorgestellten Designmethodik verglichen und messtechnisch evaluiert.}}, author = {{Rehlaender, Philipp}}, publisher = {{LibreCat University}}, title = {{{Single-stage DC-DC converters for a wide input & output voltage range}}}, doi = {{10.17619/UNIPB/1-2148}}, year = {{2024}}, }