Artificial Intelligence Applications to Power Electronics

Abstract: It has been successfully proven that Artificial Intelligence has more powerful performance for handling some problems than conventional techniques, but lack of computation apparatus, mainly in hardware, had hindered the extensive use of AI. Nonetheless, recent advances in processors/computers is making the use of AI an easier goal to achieve. Current investments in the area of AI across the globe attest to the unquestionable value attributed to AI nowadays. Yet, most of its applications are still restricted to software-based products, such as data modelling and classification type problems. So far, AI applications to power electronics and related areas are still limited, in part because some power electronics researchers do not trust the power performance of AI, and in part because they may not envision opportunities for its application. The goal of this tutorial is to address these issues.
First, it provides an overview of artificial intelligence, of its paradigms (machine learning, search methods, logic-based tools, knowledge-based tools, and probabilistic methods), and the types of problems AI addresses: classification, regression, optimization, and prediction. Second, it presents techniques such as artificial neural network and deep learning, evolution theory-inspired algorithms, fuzzy sets, and rough sets. Third, it explains how these techniques could cope with some power electronics problems, in order to give the audience a glance at potential research opportunities regarding the applying of AI in power electronics. Fourth, it presents a variety of applications of these techniques for different power electronics problems, ranging from optimum design, control, reliability, energy management, to smart grid, and examines successfully implemented examples. Finally, it highlights AI research trends and possible future applications to power electronics. By the end of this tutorial, the attendees will have a good overview of AI, and its potential application to power electronics problems they may be facing in their research.

Joao Pinto was born in Valparaiso, Brazil. He received his EE B.S degree from the Universidade Estadual Paulista, in Brazil, in 1990, his M.S. degree from the Universidade Federal de Uberlândia, Brazil, in 1993, and his Ph.D. degree from The University of Tennessee, Knoxville, in 2001. He is a Faculty Member at the Federal University of Mato Grosso do Sul, Brazil since 1994. He is the founder director of the BATLAB, Artificial Intelligence Applications, Power Electronics and Drives, and Energy Systems. He served for two years as Editor-in-Chief of the Brazilian Power Electronics Transactions, and for two years as the President of the Brazilian Power Electronics Society. He has been IEEE-PELS Administrative Committee member at-large, and IEEE-PELS South America Regional Chair since 2013. He has over 150 published papers in journals and conferences proceedings. His research interests include artificial intelligence applications, data modeling, energy systems, power electronics, and electrical machine drives.

Burak Ozpineci is the group leader of the of the Power Electronics and Electric Machinery Research Group at the Oak Ridge National Laboratory. He received a M.S. and Ph.D. from The University of Tennessee in electrical engineering in 1998 and 2002, respectively, and his B.S. degree from the Orta Dogu Technical University, Ankara, Turkey. He joined the Post-Masters Program with the Power Electronics and Electric Machinery Research Group at the Oak Ridge National Laboratory in 2001 and became a full time staff member at the PEEMRG in 2002. Dr. Ozpineci is the Chair of the IEEE PELS Rectifiers and Inverters Technical Committee and was Transactions Review Chairman of the IEEE Industry Applications Society Industrial Power Converter Committee, and He also has a Joint Faculty Associate Professor position with The University of Tennessee. His research interests include system-level impact of SiC power devices, multilevel inverters, power converters for distributed energy resources and hybrid electric vehicles, and intelligent control applications for power converters.

Raymundo Cordero Garcia received the B.S. degree in electronic engineering from Universidad Católica del Perú, in 2005, the M.Sc. degree in electrical engineering from Universidade Federal de Mato Grosso do Sul, Brazil, in 2009, and the Ph.D. degree in electrical engineering from Universidade Federal de Rio de Janeiro, Brazil, in 2015. He worked in the Didactic Department of FESTO-Perú, in 2006. He was part of the CIEEP research institute of Perú, from 2007 to 2015. Since 2015, he is professor in the Department of Electrical Engineering of the Universidade Federal de Mato Grosso do Sul. His current research interests include motor drives, DC-AC converters, instrumentation, control of nonlinear systems, digital systems, signal processing and artificial intelligence.