Huai Wang is currently Professor at the Center of Reliable Power Electronics (CORPE) at Aalborg University, Denmark. His research addresses the fundamental challenges in modeling and validation of power electronic component failure mechanisms, and application issues in system-level predictability, condition monitoring, circuit architecture, and robustness design. He also leads a project on light-AI for cognitive power electronics. His team collaborates with various industry companies across the value chain, from power electronic materials, components to systems. Prof. Wang lectures three short-term Industrial/PhD courses on Reliability of Power Electronic Systems, Design FMEA in Power Electronics, and Capacitors in Power Electronics Applications at Aalborg University. He has contributed more than 120 journal papers and co-edited a book on the Reliability of Power Electronic Converter Systems in 2015. He has given 25 tutorials at leading power electronics conferences (e.g., PCIM Europe, APEC, ECCE, etc.) and more than 80 invited talks. Prof. Wang received his PhD degree from the City University of Hong Kong, Hong Kong, China, and B. E. degree from the Huazhong University of Science and Technology, Wuhan, China. He was a short-term visiting scientist with the Massachusetts Institute of Technology (MIT), USA, and ETH Zurich, Switzerland. He was with the ABB Corporate Research Center, Baden, Switzerland, in 2009. Dr. Wang received the Richard M. Bass Outstanding Young Power Electronics Engineer Award from the IEEE Power Electronics Society in 2016 for the contribution to reliability of power electronic converter systems. He serves as General Chair of IEEE IFEC 2020 and Associate Editor of IEEE Transactions on Power Electronics.
Shuai Zhao is currently a postdoctoral researcher with the Center of Reliable Power Electronics (CORPE), Department of Energy Technology, Aalborg University, Denmark. He received the BE (Hons), ME, and Ph.D. degrees in information and communication engineering from Northwestern Polytechnical University, Xi'an, China, in 2011, 2014, and 2018, respectively. From Sep. 2014 to Sep. 2016, he was a visiting Ph.D. student with the Department of Mechanical and Industrial Engineering at the University of Toronto, Toronto, ON, Canada, with a scholarship from the China Scholarship Council (CSC). In Aug. 2018, he was a visiting scholar with the Power Electronics and Drives Laboratory, Department of Electrical and Computer Science at the University of Texas at Dallas, Richardson, TX, USA. His research interests include system informatics, intelligent condition monitoring, diagnostics & prognostics, and tailored AI tools for power electronic systems.
AI-Assisted Condition and Health Monitoring in Power Electronics
Abstract – Artificial Intelligence (AI) plays an increasing role in solving the optimization, regression, and classification problems in condition and health monitoring of power electronic converters. The outcomes from the condition and health monitoring are essential to operation optimization and predictive maintenance. This tutorial will cover three sub-topics: 1) introduction to condition and health monitoring in power electronics and its engineering problems; 2) introduction to AI methods and the motivation to apply them for the associated engineering problems; 3) case studies in parameter estimations, early failure prediction, and remaining useful lifetime prediction for power electronic applications.
Asst. Prof. Miroslav Vasić
Universidad Politecnicade Madrid
Center for Industrial Electronics
Miroslav Vasić was born in Serbia in 1981. He received the B.S. degree from the School of Electrical Engineering, University of Belgrade, Belgrade, Serbia, in 2005. Since then he has been working at Centro de Electrónica Industrial at ETSII (UPM) where he received his M.S. in 2007 and his Ph.D. degree in 2010. He has been working as assistant professor at UPM since 2015.
His research interest includes application of power converters and their optimization. In the recent years great part of his research activities has been related to the research of new semiconductor devices based on GaN and their impact on power electronics.
Miroslav Vasić has published more than 70 peer-reviewed technical papers at conferences and in IEEE journals. In 2012 he received the Semikron Innovation Award for the teamwork on “RF Power Amplifier with Increased Efficiency and Bandwidth.” In 2015 he received a medal from Spanish Royal Academy of Engineering as a recognition of his research trajectory and in 2016 he received UPM Research Projection Award for the best young researcher at Universidad Politécnica de Madrid.
Miroslav actively serves as a reviewer in several IEEE journals such as IEEE Transactions on Power Electronic and IEEE Transactions on Industrial Electronics and as an Associated Editor in IEEE Journal of Emerging and Selected Topics in Power Electronics and IEEE Transactions on Vehicular Technology. Since 2021 he acts as the Vice-chair of the IEEE PELS TC 10- Design Methodologies.
Luis Gómez Navajas was born in Jaraíz de la Vera, Spain, in 1997. He received the bachelor degree in industrial engineering from the Universidad Politecnica de Madrid, Madrid, Spain, in 2019. He is currently studying the double Master in industrial engineering and industrial electronics in the Universidad Politecnica de Madrid, Madrid, Spain.
Since 2018, he has been a research student at the Centro de Electrónica Industrial (CEI), Madrid. His current research interests include high frequency converters with carrier phase-shifted modulation and compact GaN based design to drive capacitive loads.
Javier Galindos Vicente Received the B.S. degree in Industrial Engineering with a minor in Control and Industrial Electronics from ETSII-UPM, Madrid, Spain in 2020. He has done his Bachelor Thesis researching on the field of Power Electronics with GaN devices. (Spanish patent under evaluation). He is currently studying an MSc in Digital Manufacturing at EIT Digital (UPM-TalTech). At the same time, he is involved in a research project with AIRBUS D&S and UPM to characterize failure mechanism of GaN devices. His main interests include power electronics and AI.
Design Challenges for high-performance GaN based converters in multi-MHz applications
Abstract – Nowadays Silicon (Si) semiconductors present high reliability and maturity, however, the limits in terms of power density, operation temperature, and switching frequency are close to being reached. Gallium Nitride (GaN) power devices promise superior operation at higher junction temperatures, in harsh conditions such as the space, and superior conduction and switching properties than traditional Si technology. Although the theoretical electrical properties of GaN devices are far superior to those of Silicon, all their benefits have not been exploited yet. In order to fully empower the emerging GaN based power electronics applications and unleash the full potential of GaN devices, it is necessary to fully understand what is behind (or inside) a GaN HEMT, as well as their design and reliability challenges.
We were expecting to move the switching frequencies to MHz range, but it looks like that it is more complex than expected. What are the design challenges if you want to design a 20 MHz inverter? How the single event failures influence the health of your GaN converter? How to proceed with further application development? These are the questions in front of us, and this Tutorial will try to respond them. It will cover two fundamental aspects, device theory and practical design issues, starting with a comprehensive introduction to GaN devices and their applications, to provide the attendees with the fundamental knowledge of GaN devices advantages, as well as basic know-how related to practical GaN based circuit design. Afterwards we will present the applications which can clearly benefit from GaN employment, clearly identifying the problems that we have seen during the development.
The tutorial is based on the results published by the speakers and other researchers in the field, as well as on our latest designs and accomplished experimental results.