Self-sensing and characteristic of magnetic bearing based on least squares method

HU Xiong-xin, XU Fang, TAN Da-peng
(Key Laboratory of E&M，Ministry of Education & Zhejiang Province,
Zhejiang University of Technology, Hangzhou 310014, China)

Abstract: Aiming at the complex structure and the lower precision of the self-sensing active magnetic bearings (AMBs), the inductance calculation method based on the least square method, the rotor displacement estimation method, the amplitude demodulation method and the direct current measurement method are studied. Firstly, a method for direct estimation of rotor displacement and the model of position control system based on the least square method was proposed. Secondly, based on TMS320F2812 and flexure hinge micro displacement structure, a single degree of freedom magnetic bearing experimental platform was built to test the estimated displacement and error of the self-sensing method. Finally, the position controller of a 1-DOF AMB was implemented in the real-time rig which consists of the digital signal processor TMS320F2812 and micro-position platform based on flexible hinge. The results indicate that the relative error is about 1.39% as the simulation result of absolution error is -3.49 μm, and the relative error in test is achieved about 0.96% as the simulation absolution error is 2.41μm within position range of 0~250 μm. The displacement estimation method does not only improve the estimation accuracy, but also works well when the structure of the self-sensing system is simplified.

Key words: self-sensing; magnetic bearing; position estimation; least squares method (LSM)