Investigation of MTPA Parameter Dependency in MPC‐Based IPMSM Drives and an Experimental Parameter Robust MTPA Strategy

ABSTRACT Model predictive control (MPC) has become a widely adopted control technique for interior permanent magnet synchronous machine (IPMSM) drives due to its fast dynamic response and simple control structure. Several studies exist in the literature aiming to improve the robustness of MPCs against parameter variations. However, the impact of parameter variations on maximum torque per ampere (MTPA) performance in MPC‐based IPMSM drives has not been comprehensively investigated. Existing robust MPC studies generally employ conventional MTPA strategies or control, which may significantly degrade torque control accuracy under electrical parameter variations. This study investigates the influence of MTPA parameter dependency on both model predictive current control (MPCC) and model predictive torque control (MPTC) based IPMSM drives. Conventional MTPA and cost function based MTPA (CFBMTPA) techniques are comparatively evaluated under variations in , , and parameters. Experimental results demonstrate that parameter dependency associated with MTPA generation can affect torque accuracy more critically than the inherent model dependency of MPC. To address this limitation, an experimental parameter robust MTPA strategy based on experimentally derived trigonometric current mapping is proposed. The proposed approach generates optimal −dq axis current references without requiring machine parameter information or weighting factor tuning. The proposed strategy has been experimentally validated using a 4.1‐kW IPMSM. The experimental results demonstrate improved torque tracking capability and enhanced robustness against electrical parameter variations compared with conventional MTPA and CFBMTPA approaches.