Efficiency Optimization for Permanent-Magnet Synchronous Machines using Model-Referenced Square-wave Extremum Seeking Control

We propose a novel optimization control strategy that integrates model-based optimization and square-wave extremum seeking control to optimize the efficiency of permanent-magnet synchronous machines using direct torque control (PMSM-DTC). Ordinary efficiency-optimization methods tend to require a precise mathematical model of the PMSM for estimating the optimal stator flux. Model-based optimization can rapidly shift the optimal parameters when operating conditions change. However, deviation in the stator flux estimation will occur if variation of the internal parameters is considered. The strategy proposed in this paper allows rapid switching of optimal parameters, and uses square-wave extremum seeking control to offset the deviation in the stator flux estimation quickly. The latter control scheme, ensures that the system always runs near the optimal point.