Sensorless Control of Permanent Magnet Synchronous Motor Based on Double- Cost-Function Finite-Position-Set Phase-Locked Loop

Abstract

Finite-position-set phase-locked loop (FPS-PLL) is a research hotspot in the sensorless control of permanent magnet synchronous motor (PMSM) due to its excellent dynamic response and no parameters tuning, but it takes the high computational burden and may suffer from a rotor position error of 180°. To solve the above problems, a double-cost-function FPS-PLL (DCF-FPS-PLL) is proposed in this article, where two different cost functions are defined to decide the iteration direction and guarantee the convergence of the Newton iteration method. The estimation accuracy of the rotor position is theoretically 4.73e-6 rad for the DCF-FPS-PLL without considering the error caused by the observer while only three iterations are needed, which reduces the large amount of computation without sacrificing the accuracy. The experimental results verify the advantages of the DCF-FPS-PLL.