Diffusion behavior study of model diesel components in polymer membranes based on neural network for pattern recognition

A neural network for a pattern recognition model is developed for the first time to predict the diffusion behavior of the model diesel components (dibenzothiophene and quinolone) in three different membranes of polyvinyl alcohol, polyvinyl chloride and polymethyl acrylate. The simulation results show that the excellent performance target parameter optimization area can be obtained and the effective desulfurization and denitrification agent can be found. Compared with the advanced molecular dynamic simulation method and verified by adsorption experiments, the simulation values are in good agreement with the experimental data and molecular dynamic simulation data. The results reveal that the polyvinyl chloride membrane can improve the diffusion selectivity of dibenzothiophene and it is selected as the most effective desulfurization agent, while the polyvinyl alcohol membrane is selected as the most effective denitrification agent to remove the nitrogen compounds. Development time and effort of screening desulfurization agent and denitrification agent tests are also reduced because the neural network for the pattern recognition model provides ready-made decisions. Therefore, the neural network for pattern recognition is a prospect and practicable theoretical method to research the diffusion behavior of model diesel components in polymer membranes.