Minimum-Delay Deadbeat Predictive Current Control Strategy With Model Mismatch Compensation for SPMSM Drives

Deadbeat predictive current control (DPCC) scheme has been widely applied in permanent magnet synchronous motor (PMSM) drive systems due to its superior dynamic response and current tracking performance. However, the control performance of conventional DPCC is susceptible to inherent one-step control delays and motor parameter mismatches. To address these challenges, a minimum-delay DPCC (MD-DPCC) strategy with model mismatch compensation is proposed in this paper. Differing from conventional method that relies on the average stator voltage equations of the motor, the proposed strategy constructs the transient current ripple prediction model by minimizing the delay between the sampling instant and system update instant. This prediction model ensures high performance and robustness even when affected by nonlinear disturbances such as model parameter mismatches. Furthermore, the mathematical expression of this transient model at various sampling instants is derived and discussed in detail. To further enhance steady-state accuracy, an improved feedback correction loop is introduced, which leverages the high-precision current prediction to effectively eliminate steady-state errors. Finally, experimental results conducted on a SPMSM drive platform based on STM32H742 validate that the proposed MD-DPCC strategy achieves fast and precise current tracking as well as strong robustness to model parameter variations.