Virtual Reference-Based Fuzzy Noncascade Speed Control for PMSM Systems With Unmatched Disturbances and Current Constraints

Abstract

This paper aims to develop a non-cascade speed control approach for permanent magnet synchronous motor (PMSM) systems based on the Takagi–Sugeno (T–S) fuzzy model. In non-cascade control structures, the lack of firsthand current references imposes a tricky challenge in dealing with current constraints from the circuit and unmatched disturbances caused by load torques. To this end, a virtual reference-based fuzzy non-cascade speed control scheme is proposed for PMSM systems. First, unmatched load torques are estimated by a fuzzy disturbance observer. And then, a novel virtual reference planner is constructed to inject the estimated unmatched disturbance into the control channel as feedback compensation. This improves anti-disturbance performance, while keeping the current re-sponse within the feasible domain and reducing speed overshoot. Furthermore, the non-cascade speed regulation problem is con-verted into a system stabilization problem and addressed by the composite control strategy based on the non-parallel distributed compensation. Finally, closed-loop stability with reduced con-servatism is ensured through the fuzzy Lyapunov functions and improved slack variables. Simulations and experiments show that the proposed scheme achieves excellent speed response without overcurrent even under mismatched load torque variations.