Fault Detection and Isolation of Tennessee Eastman Process Using Improved RBF Network by Genetic Algorithm

This paper presented Radial Basis Function (RBF) network as a classifier for Fault Detection and Isolation (FDI) and to enhance the model accuracy, a real-coded Genetic Algorithm (GA) is implemented to search for optimal model parameters. Comparison between the performance of FDI by using RBF Neural Network and RBF Neural Network which improved by Genetic algorithm is presented. These techniques are applied to simulated data collected from the Tennessee Eastman Process (TEP) chemical plant simulator that is designed to simulate a wide variety of faults occurring in a chemical plant based on a facility at Eastman chemical. In this study detection and isolation of all faults recorded in the process is investigated