Observer Design Based on Fractional-order Model of Permanent Magnet Synchronous Motor

Abstract

Permanent magnet synchronous motor has become the mainstream motor of electric vehicle, elevator and other drive systems due to its high torque density, high efficiency and high reliability. The existing motor model-based fault diagnosis method does not consider the fractional characteristics of the motor, then it is difficult to effectively detect the motor's minor faults. The more accurately mathematical model describes the dynamic relationships of the system, the better performance of model-based fault diagnosis. Effective fault diagnosis relies on effective residual generation, and residual generation relies on the design of observer, so designing an effective state observer to generate residuals is the basis for achieving fault diagnosis. In this paper, two full-dimensional state observers are designed based on integer-order and fractional-order models for permanent magnet synchronous motors with fractional-order characteristics, and it is shown through simulation that more effective residual generation can be obtained based on the fractional-order model.