Variable-structure control for a linear synchronous motor using a recurrent fuzzy neural network

Abstract

A variable-structure controller, using a recurrent fuzzy neural network (RFNN) to control the mover position of a permanent-magnet linear synchronous motor (PMLSM) servo drive is developed. A variable-structure adaptive (VSA) controller is first adopted to control the mover position of the PMLSM, where a simple adaptive algorithm is utilised to estimate the uncertainty bounds. Then, to further improve the rate of convergence of the estimation, a variable-structure controller using an RFNN is investigated, in which the RFNN is utilised to estimate the real-time lumped uncertainty. Simulated and experimental results show that the proposed variable-structure controller using an RFNN provides high-performance dynamic characteristics and is robust with regard to plant-parameter variations and external disturbance. Furthermore, comparing with the VSA controller, a smaller control effort is required and the chattering phenomenon is reduced by the proposed variable-structure controller using an RFNN.