NONLINEAR DISTURBANCE OBSERVER-ASSISTED FUZZY-NEURAL CONTROL FOR PRECISION SUSPENSION IN AMB SYSTEMS

Abstract

This paper presents a robust controller for the suspension active magnetic bearing system (SAMB), the combination of the sliding mode control (SMC), and a disturbance observer (DOB) is presented. The nonlinear disturbance observer is constructed to estimate and reduce the effect of the outside disturbance and also the variation of inside system parameters. Due to the chattering value will be dealt by the suitable control signal, this paper propose the fuzzy-neural to cover the chattering disadvantage problem. The fuzzy Elman neural network was proposed to approximate the nonlinear term of the chattering values. The fuzzy Elman neural network output is used to regulate the control signal. The Elman neural network is a recurrent neural network structure, which based on the back propagation. The input signal is stored by the cortex layer, which takes in charge of characterize the dynamic phenomenon.  The controller will be chose as two sliding mode controller with two surface value. Which are input of the fuzzy neural network system. The neural network system are included four layer.