An Enhanced Deadbeat Predictive Current Control for High-Speed PMSM Drives

Deadbeat predictive current control (DBPCC) has gained popularity in high-speed motor drives account for its rapid dynamic response. However, the accuracy of motor modeling and parameters seriously impacts the control performance of DBPCC, resulting in static tracking errors and even leading to instability. In this paper, we introduce an inductance identification algorithm based on disturbance term observed by extended state observer(ESO) to enhance the system’s performance. The proposed method eliminates the need for introducing supplementary recognition algorithms. Instead, it leverages ESO to extract inductance changes from disturbance terms observed at specific times, updating it in the digital controller. Consequently, it prevents the degradation of the control system caused by inductance mismatch, achieving error-free current tracking in dq-axis. Both simulation and analytical results validate the effectiveness of the proposal.