Modeling a FCC Riser with the RBF Neural Network

Abstract

A radial basis function(RBF) neural network model has been proposed for predicting production yields and reducing simulation times of a modified special pseudo-components(SPCs) fluid catalytic cracking riser. Subsequently, a modified special pseudo-components riser, served as the source of reliable data for artificial neural network(ANN) model training and testing. The ANN model experimental results show a substantial reduction of computation time and good match compared to special pseudo-components fluid catalytic cracking riser mechanism model. Its potential use for modeling optimize and control of the riser is enormous.