Abstract
The need to convert from fossil fuels to renewable energy sources to combat climate change has driven the electrification of transportation. The key components of an electric vehicle powertrain are the propulsion drive and energy storage system. High-performance control techniques for electric drives, such as model predictive control, require accurate machine models to realize their potential, which in turn requires accurate knowledge of the model parameters. In addition, many condition monitoring techniques for electric machines and energy storage systems are based upon an atypical variation of the model or the model’s parameters. Adaptive techniques can be used to accurately determine parameters in real time. However, to be effective these techniques require the input possess a persistency of excitation to guarantee convergence to the true parameters, which in general can conflict with control objectives.
This talk will discuss the application of Simultaneous Identification and Control (SIC) techniques to electric powertrain components. The talk will focus on over-actuated systems, which have more controllable inputs than regulated outputs. This extra degree of freedom can be exploited to accurately track desired reference values while at the same time providing a sufficient signal richness to the inputs to ensure parameter convergence. It will be shown how the Fisher Information Matrix can be used to develop techniques (both online and offline) for crafting inputs that achieve these objectives. The use of these estimated parameters in both control and condition monitoring applications will be discussed. Both simulation and experimental results will be provided to illustrate the benefits of SIC in electric powertrains.
Biography
Dr. Hofmann received his Ph.D. in Electrical Engineering and Computer Science (EECS) from the University of California at Berkeley in 1998. He was then a faculty member of the Department of Electrical Engineering at Penn State University from 1999 to 2010. In 2010 he moved to the EECS Department of the University of Michigan, where he is a Full Professor and currently serves as Associate Chair of Graduate Affairs. Dr. Hofmann’s research area is power electronics, specializing in the design, analysis, and control of electromechanical systems. Specific research interests include propulsion drives for electric and hybrid electric vehicles, multi-physics modelling and simulation of electric machines and power electronics, energy storage systems, and energy harvesting. He is an IEEE Fellow and is currently an Associate Editor for IEEE Transactions on Power Electronics.
