Deadbeat Predictive Current Control Based on a Novel Multi Parameter Identification for the Permanent-Mangent Synchronous Motor

Online parameter identification algorithms allow the motor control system to adjust the control parameters in real application. By using the continuous parameter detection, the motor maintains the optimal performance under different operating conditions. The accuracy of the motor parameters is an important factor affecting the performance of deadbeat current predictive control(DPCC). Traditional identification methods are difficult to balance convergence speed and identification accuracy, which there is an under-ranking problem for the multi-parameter identification of a permanent magnet synchronous motor. In this paper, the sensitivity of the control parameters in DPCC is analysed and an improved hybrid online parameter identification method is proposed. Among the traditional identification methods, the least squares algorithm is fast in identification and the model reference adaptive method has high identification accuracy. The algorithm proposed in this paper combines the advantages of both, which stepwise identifies the model in different time scales, and introduces multiple new hibernation factors to enhance the algorithm's resistance to perturbation. Finally, the effectiveness of the identification method is proved by simulation comparison and experimental verification.