Robust Electrical and Mechanical Parameter Identification of Low-Power PMSMs

Abstract

Parameter identification is an important task in the field of motor control and final stage inspection. Over the years, the demand for robust parameter identification increased especially in the field of highly miniaturized and integrated PMSMs. Nevertheless, current measurements on such kind of motors are characterized by small signal-to-noise ratios, thus limiting the performance of identification algorithms. In this paper, a RLS algorithm is combined with a first order Sliding Mode Differentiator (SMD) used for derivative estimation. This allows the algorithm to estimate the electrical and mechanical parameters of a PMSM while rejecting noise, PWM switching components or quantization effects. Experimental results on a PMSM without and with SMD are shown and compared in terms of estimation performance.