应用改进粒子群算法优化管壳式换热网络总成本  

Shell-and-tube heat exchanger network total cost optimization using improved particle swarm algorithm

作　　者：项顺伯[1] 黄燕[2] 

机构地区：[1]广东石油化工学院广东省石化装备故障诊断重点实验室,广东茂名525000 [2]广东石油化工学院化学工程学院,广东茂名525000

出　　处：《黑龙江科技大学学报》2015年第1期115-118,共4页Journal of Heilongjiang University of Science And Technology

基　　金：广东省自然科学基金资助项目(8152500002000003);广东高校石化过程装备故障诊断与信息化控制工程中心开放基金资助项目(512028)

摘　　要：壳式换热网络的优化设计一直是众多学者关注的难题。从热交换总面积出发,对管壳式换热网络各项参数进行建模,得出设计的总成本优化函数。采用基于自适应排斥因子的改进粒子群算法(BARFPSO)进行优化举例。结果表明:BARFPSO能有效地跳出局部最优解,具有计算精度高、有效减小计算复杂度的特点,并能得出最佳的管壳式换热网络各项参数值。优化结果中,换热网络总成本减少了4.3%和1.0%,优化效果明显。This paper is focused specifically on a research into the optimization design of shell-and- tube heat exchanger network, a long-time challenge which has baffled many experts. The research is best obtained by modeling parameters necessary for shell-and-tube heat exchanger network, starting from heat exchange total surface and thereby obtaining the optimization function associated with design total cost. An optimization example resulting from the use of an improved particle swarm algorithm based on adaptive rejection factor（BARFPSO） demonstrates that BARFPSO capable of effectively jumping out of local opti- mum solution, boasts such advantages as a higher computation precision and an effective reduction in computational complexity, thus affording the best parameters required for shell-and-tube heat exchanger network. It follows that this optimization approach yields a 4.3% and 1.0% reduction in the cost of heat exchanger network, thus indicating a significant optimization effect.

关 键 词：改进粒子群算法 管壳式换热网络 自适应排斥因子 总成本优化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]