Active control for the base motion of active magnetic bearings based on RBF neural network

ZHAO Hong-kai, JIANG Ke-jian
(School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China)

Abstract: Aiming at the problem that active magnetic bearing system installed on a moving carrier was often subject to excitation interference from its bearing foundation, the rotor acceleration, rotor dynamic displacement and stator dynamic displacement of magnetic bearings under basic excitation interference were studied. Through the dynamic model of singledegreeoffreedom magnetic bearing system under basic excitation, a PID control algorithm based on RBF neural network was proposed, the vibration characteristic of the magnetic bearing under nonrandom and random excitation was analyzed as well. Numerical simulation was established by MATLAB to analyze the control performance of magnetic bearing. The results indicate that the method can effectively suppress the vibration of the magnetic bearing rotor on the moving carrier, the adaptive adjustment of the control parameters of the AMB control system on the moving carrier is realized, and the bearing rotor can keep stable suspension under the disturbance of the basic excitation.

Key words: active magnetic bearings(AMB); base motion; neural network; active control