Safe State Estimation With Disturbance Rejection in PMSM Systems

This paper investigates the disturbance rejection safe state estimation problem of the permanent magnet synchronous motor (PMSM) with load torque disturbance and safe current specification via a new moving horizon extended state estimation (MHESE) approach. To be specific, the state extension technique is first introduced to deal with the undesirable impact of load torque in PMSM to significantly improve disturbance rejection capability of traditional moving horizon estimation (MHE). By redesigning the cost function that takes into account the influence of the extended state, the new MHESE is developed for disturbance rejection safe state estimation in PMSM. Under MHESE, the safe current specification (e.g., overcurrent protection) is explicitly taking into account, which substantially improves state estimation accuracy in the presence of load torque disturbance. The disturbance rejection safe state estimation problem of PMSMs is transformed a quadratic programming problem with inequality constraints that will be recursively solved in a fixed length moving horizon window. The proposed approach is validated by experimental studies. The results show that MHESE provides much better dynamic estimation performance compared with the traditional MHE in the presence of load torque disturbance.