Inverse Control of Servo System Based on Neural Network

In this paper neural network is used for servo system design. It is proved that a class of nonlinear servo system can be described by its inverse dynamic model. We use open-loop and closed-loop schemes for feedforward multilayer network learning. Alopex algorithm is used to train inverse dynamic model of the plant. At the end of the paper,the design of servo system on-line control based on Feedback Error Learning is presented. Simulation results show that the method is effective in identification and control of servo system.