An Improved Displacement Measurement Based on Model Reconstruction for Permanent Magnet Synchronous Motor

Abstract

Permanent magnet synchronous linear motor (PMSLM) and PM synchronous planar motor are important motion devices in the modern precision industry. Their inherent periodic magnetic fields are highly related to motor position. Correspondingly, the magnetic fields can be utilized to determine their own displacement and replace expensive sensors. However, the measurement result of this strategy seriously depends on magnetic field model (MFM) preciseness. To improve measurement accuracy, a novel MFM with Fourier series is proposed in this paper. First, a high-order theoretical model from theoretical derivation is developed. And the analyses show that even harmonics exist in the magnetic field of motor, which are not included in the theoretical model. This model deviation obviously affects measurement accuracy. Hence, the new MFM is reconstructed based on a high-order theoretical model. Comparative experiments with the three different MFMs are carried out to measure the displacement of the PMSLM. In the experiments, a linear hall sensors array containing eight sensors is self-developed to detect motor magnetic field. The root mean square of displacement measurement errors using a reconstruction model, a high-order theoretical model, and a fundamental model is 4.84, 8.42, and $758~\mu \text{m}$ , respectively. The experimental results validate that the measurement result of the proposed reconstruction MFM is better than the other two MFMs.