Neural Network Based Predictive Control of a Heat Exchanger Nonlinear Process

In this paper, a neural network based predictive controller is designed to govern the dynamics of a heat exchanger pilot plant. Heat exchanger is a highly nonlinear process; therefore, a nonlinear prediction method can be a better match in a predictive control strategy. Advantages of neural networks for the process modeling are studied and a neural network based predictor is designed, trained and tested as a part of the predictive controller. The dynamics of the plant is identified using a Multi Layer Perceptron (MLP) neural network. The predictive control strategy based on the neural network model of the plant is applied then to achieve set point tracking of the output of the palnt. The performance of the proposed controller is compared with that of Generalized Predictive Control (GPC) through simulation studies. Obtained results demonstrate the effectiveness and superiority of the proposed approach..