Identification and control of nonlinear processes using neural network

In this paper, a dynamic model of the magnet levitation nonlinear process is identified as a neural network. The accuracy of the model is tested and verified even if the observed input/output data contains noisy components. Three layers neural network controller is proposed and developed in order to track the set point and regulate against disturbance. The response of the proposed neural controller is tested and verified. Simulation results show the power of neural network to model and control nonlinear processes.