Event Review: UoB IEEE Student Branch Event – Webinar: Potential Intersections Between MPC and AI for Electric Drives
About the speaker Issa Hammoud: Issa Hammoud (Student Member, IEEE) received the M.Sc. degree in power engineering with distinction from the Technical University of Munich, Germany, where he is currently a PhD candidate at the Chair of High-Power Converter Systems. He conducted his doctoral research in industry at the Powertrain Mechatronics Control Engineering Excellence Cluster, IAV GmbH, Gifhorn, Germany. He recently started at Robert Bosch GmbH, Stuttgart, Germany, in the field of control of electrical drives. Mr. Hammoud received the best paper awards at the Fifth and Sixth International Symposium on Predictive Control of Electrical Drives and Power Electronics (PRECEDE) in 2019 and 2021, respectively.
The webinar was streamed on the 11th of November from 3-4pm on Zoom. There were 16 attendees. Most attendees were from the University of Birmingham. There were a few additional attendees from other UK institutions..
Contents of talk: the speaker talked about using model predictive control for electric drives. He first outlined the use of first-principle (physics-based) models to achieve this. The shortcomings of first-principle models were highlighted and used to motivate the use of data-based models. Real-world data was used to generate a neural network model. This latter data-based model was found to perform slightly better than the first-principle model. It is also quicker to compute than the first-principle model online, as the majority of its computation is done offline.