Tracking Control Strategy of PMSLM with A Novel Observer-based Compensator and A RBFNN-based Controller

Abstract

This paper is devoted to a high-precision tracking control strategy of permanent magnet synchronous linear motor (PMSLM). Firstly, the field-oriented control model of the PMSLM is established to calculate the electromagnetic thrust. To reconstruct the system states and reject the lump disturbance, a novel observer-based compensator is proposed, by taking the basic ideas of the frequency-domain disturbance observer and the time-domain one (the extended state observer in active disturbance rejection control, ADRC). A radial-basis-function neural network (RBFNN) controller with accurate approximation capability is utilized to tracking the desired motion trajectory. Contrasted to the RBFNN’s parameters self-evolved, a simple and unique parameters tuning method is derived to guarantee the compensator performance. All the proposed algorithms are implemented in a rapid control prototype (RCP) real-time simulation platform and the simulation and experiment results validate the rightness of theoretical analysis and the feasibility of the proposed methods.