换热网络多目标综合优化算法研究进展

摘要

For the sustainable development in resources and energy,the designers should not only consider economy,but also flexibility,reliability,operability and environmental impact in the synthesis of heat exchanger networks(HENs). Multi-objective synthesis of HENs has got preliminary progress and drawn great attention. This paper illustrates the necessity of multi-objective synthesis of HENs and summarizes the research progress on multi-objective synthesis of HENs. The summary and comparison of the algorithms for solving multi-objective optimization problems were mainly focused. Application in multi-objective synthesis of HENs was reviewed. Research shows that traditional multi-objective algorithms are less suitable for solving the problems of complex superstructure. However, multi-objective evolutionary algorithms can solve multi-objective problems better in the synthesis of HENs. Non-dominated sorting genetic algorithm(NSGA-Ⅱ) is one of the most popular and effective applied algorithms. It was proposed that based on superstructure,establishing multi-objective models which involve economy,flexibility,reliability,operability and environmental impact and then present the decision makers with Pareto optimum solutions is the future of HEN synthesis.%资源和能源的可持续发展使得换热网络综合不仅要考虑经济性，同时要满足柔性、可靠性、可操作性和环境影响度等指标的要求。目前，换热网络多目标综合的研究有了初步进展并引起了广泛关注。本文阐述了进行换热网络多目标综合的必要性并总结了相关研究。重点对常用的多目标优化算法作了总结和对比，综述了其在换热网络多目标优化设计中的应用进展。研究表明，传统多目标算法越来越无法满足复杂模型的求解，而多目标进化算法可以很好地求解换热网络综合多目标优化问题，其中 NSGA-Ⅱ算法是目前应用最广的有效算法。提出尝试 NSGA-Ⅱ等多目标进化算法，基于超结构建立包括经济性、柔性、可靠性、可操作性和环境影响度等在内的换热网络多目标综合模型，给出Pareto最优解集合供决策者选择是未来的研究方向。