Synchronisation based adaptive parameter identification for permanent magnet synchronous motors

Abstract

High power and high performance applications of permanent magnet synchronous motors (PMSM) are increasing. PMSM models with accurate parameters are required for precise control and fault diagnostic systems. An adaptive synchronisation based parameter identification method for the PMSM with a nonlinear structure is proposed. During the identification, the PMSM's dynamic response is synchronised with another system of similar dynamic structure. The algorithm is different from the commonly known adaptive estimator. A globally converged feedback control is applied to the reference model. Lyapunov function analysis and LaSalle's invariance principle are used to assure performance. To validate the proposed identification approach, both simulation and experiments are performed for the identification of motor parameters in a PMSM servo system. During experiments, the method is further improved to counteract voltage measurement problems brought by high frequency switching devices. The results demonstrate the effectiveness of this approach.