基于多目标优化精馏系统综述

综述了精馏系统中多目标的优化问题，分析列举了人工神经网络、正交设计、响应面、遗传算法和粒子群算法在精馏系统多目标优化中的应用，旨在总结精馏系统中的优化算法，寻求最优操作条件的解法，为精馏塔的多目标优化提供参考。结果表明，针对于目前精馏系统的复杂多变性及混合规划问题，这些算法可以很好地对精馏系统进行建模，预测精馏模型，预测值与模拟值拟合较好，有较高的精度。且可用于求解精馏过程中的最优操作条件，降低系统的不可逆性，实现了精馏系统的节能优化，提高了产品质量，降低了能耗。最后指出了多目标优化精馏系统方法的可行性，也表明在实际生产中将多种优化算法相结合进行多步优化的可行性。

A review of optimization rectification systems based on multi-objective

Distillation system is studied in the multi-objective optimization of multiple parameter analysis, analysis of lists of artificial neural network, the orthogonal design, the response surface, genetic algorithm and particle swarm algorithm in distillation system, the application of multi-objective optimization, to summarizes the optimization algorithm of the distillation system and seek the solution to the optimal operating conditions, provide reference for rectifying column of multi-objective optimization. The results show that these algorithms can be used in the actual distillation system, aiming at the complex, variable and mixed programming problems of the current distillation system, the distillation system can be well modeled and predicted. It can be used to solve the optimal operation conditions in the distillation process, reduce the irreversibility of the system, realize the energy saving optimization of the distillation system, improve the product quality and reduce the energy consumption. It is proved that the method of multi-objective optimization distillation system is feasible, and that the method of multi-step optimization distillation system is feasible by combining multiple optimization algorithms in practical production.