Modelling of an industrial fluid catalytic cracking unit using neural networks

An artificial neural network (ANN) model for determining the steady-state b ehaviour of an industrial Fluid Catalytic Cracking (FCC) unit is presented in this paper. Industrial data from a Creek petroleum refinery were used to develop, train and check the model. FCC is one of the most important oil r efinery processes. Due to its complexity the modelling of the FCC poses a g reat challenge. The proposed model is capable of predicting the volume perc ent of conversion based on six input variables. This work is focused on det ermining the optimum architecture of the ANN, in order to gain good general ization properties. The results show that the ANN is able to accurately pre dict the measured data. The prediction errors in both training and validati on data sets are almost the same, indicating the capabilities of the model to accurately generalize when presented with unseen data. The neural model developed is also compared to an existing non-linear statistical model. The comparison shows that the neural model is superior to the statistical mode l.