Robust Petri Recurrent-Fuzzy-Neural-Network Sliding-Mode Control for Micro-PMSM Servo Drive System

This paper proposes an intelligent hybrid control system (IHCS) for identification and control of micro-permanent-magnet synchronous motor (micro-PMSM) servo drive to achieve high precision tracking performance. Based on the principle of computed torque control (CTC), a position tracking controller is designed and analyzed. Moreover, to relax the requirement of the lumped uncertainty, an IHCS is proposed. The IHCS is composed of a Petri recurrent-fuzzy-neural-network controller (PRFNNC), PRFNN identifier (PRFNNI) and a sliding-mode controller (SMC). The PRFNNC is used as the main tracking controller to mimic the CTC law and the SMC is designed with adaptive bound estimation algorithm to compensate for the approximation error between the PRFNNC and the CTC. The PRFNNI is used to provide the sensitivity information of the micro-PMSM servo drive system to the PRFNNC. The on-line learning algorithms of the PRFNNC and PRFNNI are derived using Lyapunov stability analysis. In addition, the online adaptive control laws of the SMC are derived based on the Lyapunov stability analysis, so that the stability of the system can be guaranteed. The position tracking performance is significantly improved using the proposed IHCS and robustness to external disturbances can be obtained as well. A computer simulation is developed to validate the effectiveness of the proposed IHCS. The simulation results confirm that the IHCS grants robust performance and precise response regardless of load disturbances and micro-PMSM parameters uncertainties. Key-Words: Computed torque control, intelligent control, Lyapunov stability theorem, micro-permanentmagnet synchronous motor (micro-PMSM), Petri net (PN), recurrent-fuzzy-neural-network.