Comparison of rotor position estimation performance in fundamental-model-based sensorless control of PMSM

Abstract

This paper analyzes and compares position estimation performances of fundamental-model-based sensorless control methods considering the effects of inverter nonlinearities and parameter estimation errors. Based on the investigation of four conventional sensorless control algorithms, estimated position error from each algorithm is represented in a quantitative manner. This representation is helpful in understanding how the error influences the position estimation performance. And, it is revealed that steady state position error can be expressed as a unified formula regardless of position estimation methods. The validity of the analysis is verified by computer simulations and experimental results.