Multi-Objective Optimization of Gas Fractionation Unit Based on SUMT and NSGA-II Algorithm

This study provide insights into the multi-objective optimization problem of gas fractionation unit. The production data are used to fit a multiple linear regression model to identify the gas fractionation unit model parameters and to establish its multi-objective optimization model,using the energy consumption product output and recovery as the objective function. The non-dominant sorting genetic algorithm( NSGA-II) and penalty function algorithm are used to solve the multi-objective optimization model and to obtain the Pareto optimal solution sets. The optimization results show that the proposed method can increase the production and decrease energy consumption and emission of propylene by optimizing the operational conditions,and the improved NSGA-II algorithm has better feasibility validity and versatility than NSGA-II and NSGA algorithm. This optimal method can guide the operational of the gas fractionation unit.