Extreme learning machine based phase angle control for stator-doubly-fed doubly salient motor for electric vehicles

This paper develops a novel advanced angle control scheme for the stator-doubly-fed doubly salient (SDFDS) motor for electric vehicles (EVs) based on the extreme learning machine (ELM) so as to satisfy the requirement of EVs. The SDFDS motor runs with constant torque below the base speed and with constant power by field weakening over the base speed. To achieve high torque at low speed for cranking and widen speed operation range fro cruising, phase angle of armature current must be advanced. Hence phase angle control is the key factor. As a new learning algorithm for single-hidden layer feed-forward neural networks (SLFNs), the extreme learning machine (ELM) can solve the nonlinear relationships among phase angle, torque and speed. Thus phase angle control based on extreme learning machine is presented in this paper, in which the experimental data is applied to train the SLFNs in off-line way and afterwards, the trained data is applied to control the motor on-line. The experimental results verify the effectiveness of the developed control scheme.