Author ORCID Identifier

http://0000-0002-5317-9150 Soumava Bhattacharjee

Date of Award

Summer 2021

Publication Type

Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

Keywords

Artificial Intelligence, Electric vehicles, PMSM control, Reinforcement learning, Space vector modulation, Vector control

Supervisor

K.L.V. Iyer

Supervisor

N.C. Kar

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

The demand for electrified vehicles has grown significantly over the last decade causing a shift in the automotive industry from traditional gasoline vehicles to electric vehicles (EVs). With the growing evolution of EVs, high power density, and high efficiency of electric powertrains (e–drive) are of the utmost need to achieve an extended driving range. However, achieving an extended driving range with enhanced e-drive performance is still a bottleneck.

The control algorithm of e–drive plays a vital role in its performance and reliability over time. Artificial intelligence (AI) and machine learning (ML) based intelligent control methods have proven their continued success in fault determination and analysis of motor–drive systems. Considering the potential of intelligent control, this thesis investigates the legacy space vector modulation (SVM) strategy for wide–bandgap (WBG) inverter and conventional current PI controller for permanent magnet synchronous motor (PMSM) control to reduce the switching loss, computation time and enhance transient performance in the available state–of–the-art e–drive systems. The thesis converges on AI– and ML–based control for e–drives to enhance the performance by focusing in reducing switching loss using ANN–based modulation technique for GaN–based inverter and improving transient performance of PMSM by incorporating ML–based parameter independent controller.

Download

Share

COinS