Product Yields Prediction of Tehran Refinery Hydrocracking Unit Using Artificial Neural Networks

In this contribution artificial neural network (ann) modeling of the hydrocracking process is presented. the input–output data for the training and simulation phases of the network were obtained from the tehran refinery isomax unit. different network designs were developed and their abilities were compared. backpropagation, elman and rbf networks were used for modeling and simulation of the hydrocracking unit. the residual error (root mean squared difference), correlation coefficient and run time were used as the criteria for judging the best network. the backpropagation model proved to be the best amongst the models considered. the trained networks predicted the yields of products of the isomax unit (diesel, kerosene, light naphtha and heavy naphtha) with good accuracy. the residual error (root mean squared difference) between the model predictions and plant data indicated that the validated model could be reliably used to simulate the isomax unit. a four-lumped kinetic model was also developed and the kinetic parameters were optimized utilizing the plant data. the result of the best ann model was compared to the result of the kinetic model. the root mean square values for the kinetic model were slightly better than the ann model but the ann models are more versatile and more practical tools in such applications as fault diagnosis and pattern recognition.