Loss Minimization Control of High-Speed SPMSM Based on Multiparameter Identification

Compared with maximum torque per ampere (MTPA) control, loss minimization control (LMC) offers enhanced efficiency for high-speed surface-mounted permanent magnet synchronous motors (SPMSMs), yet its performance is limited by motor parameter accuracy. To address this challenge, a multiparameter identification-based LMC strategy is proposed. Primarily, the influence of parameter variations on the electrical loss and optimal reference current is analyzed. Then, offline identification is employed to obtain stator resistance and synchronous inductance, which are compiled into a lookup table. Based on these parameters, a stepwise online identification method integrating model reference adaptive system with enhanced recursive total least squares is proposed to estimate iron loss resistance and permanent magnet flux linkage. Subsequently, the integration of offline and online identified parameters into optimal current calculations substantially improves computational accuracy, thereby optimizing the loss reduction effect of LMC strategy. Finally, comparative experiments are carried out on a high-speed SPMSM-driven magnetic levitation rotor spinning machine. The results show that the improved LMC scheme improves the prototype efficiency by 3.80 and 1.24 percentage points at the rated speed, compared with the MTPA control and the traditional LMC method, respectively.