Rotor Displacement Self-Sensing Method for Six-Pole Radial Hybrid Magnetic Bearing Using Mixed-Kernel Fuzzy Support Vector Machine
Tiantian Liu, Huangqiu Zhu, Mengyao Wu, Weiyu Zhang
Abstract
In order to solve the problems of larger volume, higher cost and lower reliability caused by displacement sensors in a magnetic bearing system, a self-sensing method using mixed-kernel function fuzzy support vector machine (FSVM) displacement prediction model is proposed. Firstly, The structure and working principle of six-pole radial hybrid magnetic bearing (HMB) are introduced, and its suspension force mathematical model is deduced. Secondly, the FSVM displacement prediction model among control currents and rotor displacements is established, and the performance parameters of FSVM are optimized by genetic algorithm. Finally, a simulation test of the six-pole radial HMB system is constructed, the simulation results show that the performance of the prediction model using the mixed-kernel function FSVM is significantly better than that using other kernel functions, and the predicted values are about 93.35% and 95.5% of the actual values in the x- and y- direction. The anti-disturbance experiments are carried out and the feasibility of the method proposed is verified.