Adaptive neural network control of magnetically suspended reaction-wheel with high accuracy

Aiming at sensor noises and non-linear response of magnetically suspended reaction-wheel (MSRW), an adaptive neural network control algorithm was investigated. This control algorithm could realize the stabilization of MSRW, approach complicated non-linearity functions, implement the noises' restraint and the non-linear compensate of MSRW. Simulation results show that the proposed method can eliminate the noises and restrain the non-linearity of MSRW effectively. MSRW can realize high accuracy control by the adaptive neural network method.