PREDICTIVE NEURO-FUZZY CONTROLLER FOR MULTIVARIATE PROCESS-CONTROL

A self-learning fuzzy controller with a neural estimator was designed for predictive process control, and it was applied to snack food fryin g process control. Main features of the designed controller are that i t can be applied to plants having nonlinear dynamics, and its structur e can be easily extended to multivariable systems. The neural estimato r, composed of a time delay multilayer perceptron with output feedback , was structured to model the dynamics of a frying process and to pred ict the actual plant output affected by controller output after the ti me lag. The neuro-fuzzy controller was composed of a multilayer feedfo rward network, and it was trained using the backpropagation algorithm. The neuro-fuzzy controller trained with only two data sets performed the control task successfully, and it showed that the controller had t he robustness for the control task starting from the untrained initial conditions by computer simulation.