Title: Active steering PMSM speed control with wavelet neural network

Authors: Mingzhu Xu; Zhaohan Huo; Shaohua Li; Li Jiang

Addresses: State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, School of Mechanical Engineering, Shijiazhuang Tiedao University, Shijiazhuang, 050043, China ' School of Mechanical Engineering, Shijiazhuang Tiedao University, Shijiazhuang, 050043, China ' State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, Shijiazhuang Tiedao University, Shijiazhuang, 050043, China ' Department of Electronic, Electrical and Systems Engineering, University of Birmingham, Birmingham, B15 2TT, UK

Abstract: In view of the high requirements of active steering system on the noise, vibration and weight of the motor, this paper chooses permanent magnet synchronous motor as the active power steering motor. A speed control strategy of permanent magnet synchronous motor based on wavelet neural network (WNN) proportional integral differential controller (PID) is proposed. According to the change of system operation parameters, a three-layer feedforward artificial neural network is used to update the PID parameters online training based on gradient descent correction error method. Wavelet neural network and incremental PID are used to construct the speed loop controller. A control system based on DSP28335 is designed for experimental verification. The results show that the proposed control strategy has better static and dynamic performance, which can provide a new choice for small motor control of vehicles.

Keywords: active steering motor; permanent magnet synchronous motor; WNN; wavelet neural network; incremental PID; vehicle electronic control system; driver assistance function; auxiliary drive; active steering; power steering; PMSM control experiment.

DOI: 10.1504/IJVD.2020.113912

International Journal of Vehicle Design, 2020 Vol.82 No.1/2/3/4, pp.64 - 74

Received: 18 Jan 2020
Accepted: 03 May 2020

Published online: 01 Apr 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article