针对干气提浓装置能耗较高的现状,对最新的中冷油闪蒸工艺进行了研究。采用Aspen Plus软件进行流程模拟,使用改进的遗传算法(NSGA-Ⅱ),以年总费用(TAC)、CO₂排放量(E_carbon)和碳二回收率(R_C2)为目标函数,通过罚函数法转化为无约束问题,对中冷油闪蒸工艺进行多目标优化,获得了Pareto前沿。统计后发现,半贫液与贫液质量比的变异系数仅为2.37%,可以使用平均值1.95来代表。最后使用优劣解距离法(TOPSIS)选取最优点进行对比,优化结果显示,相比于浅冷油吸收工艺,中冷油闪蒸工艺的R_C2上升3.09%,TAC下降43.75%,E_carbon减少41.77%。结果表明,中冷油闪蒸工艺在各方面性能均有大幅提升,且基于NSGA-Ⅱ算法的多目标优化方法能够发现更多的有益性结论。

In view of the high energy consumption in dry gas concentration unit,the latest middle-cool oil flash process is studied.Aspen Plus software is utilized for process simulation,and an improved genetic algorithm (NSGA-Ⅱ) is used.Taking the annual total cost (TAC),CO₂ emissions (E_carbon) and C₂ recovery (R_C2) as the objective functions,the penalty function method is used to transform the problem into an unconstrained problem.A multi-objective optimization is conducted on the middle-cool oil flash process to obtain a Pareto front.After statistics,it is found that the coefficient of variation for the mass ratio of semi lean liquid to lean liquid is only 2.37%,which can be represented by an average value of 1.95.Finally,the TOPSIS method is employed to select the best solution.The optimization results show that compared with the shallow-cool oil absorption process,the R_C2 of the middle-cool oil flash process increases by 3.09%,the TAC decreases by 43.75%,and the E_carbon decreases by 41.77%.It is shown that the performance of middle-cool oil flash process shows a great improvement in all properties,and the multi-objective optimization method based on NSGA-Ⅱalgorithm can help to find more beneficial conclusions.

贾继龙(1998-),男,硕士生。