Improvement of control characteristics of interior permanent-magnet synchronous motor for electric vehicle

Abstract

This paper describes a method to improve the control characteristics of an interior permanent-magnet synchronous motor for an electric vehicle (EV). At the motor for an electric vehicle, d- and q-axes inductances are varied to a great extent. Therefore, variation characteristics for d- and q-axes inductances are analyzed as a function of both current magnitude and current phase angle. A new vector control algorithm is proposed that makes use of the learning capability of a neural network to implement d- and q-axes inductances to be varied according to current magnitude and also current phase angle. Also, a distribution law of torque reference to improve the driving characteristics of the two motor-driven EV is proposed. Advanced performance is verified through numerical methods and experiments.