Online estimation of the stator resistances of a PMSM with consideration of magnetic saturation

Precise knowledge of the resistance in each phase of a permanent magnet synchronous machine renders possible the performance optimization of a PMSM drive. It can be applied to torque ripple minimization, fault diagnosis for inverter, and high quality monitoring of e.g. temperature or the aging condition of electrical contacts. This paper provides a method to accurately estimate each individual stator phase resistance of a three-phase PMSM. The resistances are identified using recursive least squares minimization in real time. The estimation algorithm is based on a PMSM model in rotational reference frame. To achieve a high precision estimation in all points of operation the motor model takes account of magnetic saturation and cross-magnetization that are captured by offline estimation. The reliability of the proposed scheme is verified in simulations and test-bench experiments.