A Robust Deadbeat Predictive Current Control Method for IPMSM

Deadbeat predictive current control (DPCC) demonstrates excellent dynamic performance. However, in practical applications, its effectiveness is degraded by parameter mismatches and inverter nonlinearities. Among the various improvement methods addressed for these issues, incremental model-based DPCC (I-DPCC) achieves zero static current error with a low computational burden but suffers from instability under parameter variation, especially when applied to interior permanent magnet synchronous motors (IPMSMs). In this paper, a robust I-DPCC (RI-DPCC) combining feedforward control is proposed for IPMSM, with an adjustable stable operation range that can be extended to twice the actual inductance or even larger. To further improve the robustness of dynamic performance, an inductance correction method is introduced to track the variation of inductance during dynamic processes. Thus, the current commands can be well tracked even when significant inductance variation occurs. With sufficient voltage margin, the dynamic processes under mismatched inductance can be shortened to four control periods. Finally, experimental results validate the effectiveness of the proposed method.