Parameter optimization of Neural Network Model Based on Simulated Moving Bed

In recent years Simulated Moving Bed (SMB) chromatography separation technology has been used in petroleum chemicals, food processing and fine chemicals. In this paper, Bayesian regularization BP neural network algorithm is used to fit the input and output data of mechanism model of Simulated Moving Bed. Genetic algorithm and particle swarm algorithm are used to optimize the operating parameters and to find out the best concentration. The results are put into the mechanistic model to verify the point. The simulation results show that the error is small and these methods can solve the problem that the parameter optimization time is too long.