Enhanced Position Sensorless Control Using Bilinear Recursive Least Squares Adaptive Filter for Interior Permanent Magnet Synchronous Motor

TLDR: A bilinear recursive least squares (BRLS) adaptive filter is proposed and integrated into a sliding-mode position observer to suppress the dominant harmonic components in the estimated back EMF and as a result, the accuracy of the estimated rotor position can be greatly improved.

Abstract: In the back electromotive force (EMF)-based sensorless control of interior permanent magnet synchronous motor (IPMSM), the inverter nonlinearity and flux linkage spatial harmonics will possibly give rise to (6 k ± 1)th harmonics in the estimated back EMF, especially the fifth and seventh harmonics. Those harmonics will consequently introduce (6 k )th harmonic ripples to the estimated rotor position, especially the sixth harmonic component. In order to solve this problem, a bilinear recursive least squares (BRLS) adaptive filter is proposed and integrated into a sliding-mode position observer to suppress the dominant harmonic components in the estimated back EMF and as a result, the accuracy of the estimated rotor position can be greatly improved. A unique feature of the BRLS adaptive filter is its ability to track and suppress the specified harmonic components in different steady state and dynamic operational conditions. The proposed method can compensate for harmonic ripples caused by the inverter nonlinearity and machine spatial harmonics at the same time; this method is also robust to machine parameter variation, and the BRLS algorithm itself is machine parameter independent. The implementation of the proposed BRLS filter in the sensorless control of IPMSM is explained in details in this paper. The enhanced drive performances using the BRLS filter have been thoroughly validated in different steady state and dynamic operational conditions on a 1.5-kW IPMSM sensorless drive.