Neural Network Control for PPDCV Clinker Cooler System

Pilot Proportional Directional Control Valve (PPDCV) is used to control the clinker cooler actuator position for Cement Industries. PID controllers are satisfying control for the actuator position control with transient response and reduce steady state error. Uncertainty parameter cause unsmooth response small oscillation in actuator position. This work explains a neural network control for (PPDCV) system with uncertainty parameter, reduces the disturbance, and removes the oscillation in the response. The paper is focused on the possibilities of applying trained artificial neural networks for creating the system inverse models that are used to design inverse control algorithm for non-linear dynamic system. The simulation experiment is performed using Matlab and Simulink to show the response of these controllers.