Implementation of Neural Network Control in a Nonlinear Plant Using MATLAB

In modern control theory, there are several variations to different controller designs. The same can be said for Neural Network (NN) Controllers. The goal of this paper is to implement a variant of NN controllers called the Predictive Neural Network controller for a nonlinear plant using MATLAB. The controller will not only be used to determine the performance of the plant but also model future inputs of the system by using the data it has collected. The data will undergo training to create a predictive model of the system. The predicted inputs can then be used to optimize the performance of the system. The NN controller was implemented on a nonlinear plant model and simulated using Simulink which is available in MATLAB using the Deep Learning Toolbox. Another motivation for this paper is to gain a better understanding of the applications of predictive neural networks in control systems.