Research on Denoising PMSM Resolver Signals Based on Morphological Filter

Abstract

The resolver is often used to measure rotor positions in space vector control of permanent magnet synchronous motor (PMSM) due to many advantages. Its angle decoding scheme can be divided into hardware decoding and software decoding. Hardware decoding features high cost, while traditional software decoding is low precision. In order to decrease its application cost and promote its decoding precision, a software decoding method has been proposed based on mathematical morphology. The morphological opening and closing algorithm has been used to suppress the positive and negative pulse noise of signal and improve the precision of angle decoding. The MATLAB simulation shows that this method can eliminate the noise pollution effectively and improve its operation speed.