摘要
换热网络优化属于混合整数非线性规划问题,具有严重的非凸非线性特征,是过程系统优化的重点和难点。文章介绍了强制进化随机游走算法(RWCE)的基本原理,以有固定投资,计算规模相对较大的H13C7换热网络算例为例,从计算效率和精度等方面综合考虑算法在不同参数设置下对结构进化和连续变量优化进程的影响,提出了对应策略。经证明,参数的组合优化设置有助于提高算法的效能,取得较好的优化结果。
The heat exchanger network synthesis is primarily formulated as a mixed integer nonlinear programming problem, with serious non-convex and non-linear feature, which is the focus and hard problem in the process system. In this paper, the basic principle of RWCE algorithm is introduced. From the prospective of computational efficiency and accuracy, one large-scale case, H13C7 with fixed cost, is used to test the control parameters of algorithm to find characteristics of structure evolution and continuous variable. Then, a strategy is proposed to improve the global search ability. It is proved that the combination optimization of parameters is helpful to improve the performance of the algorithm and get better results.
出处
《工程热物理学报》
EI
CAS
CSCD
北大核心
2018年第2期404-411,共8页
Journal of Engineering Thermophysics
基金
上海市科委部分地方院校能力建设计划(No.16060502600)
国家自然科学基金资助项目(No.51176125)
沪江基金研究基地专项(No.D14001)
关键词
换热网络优化
强制进化随机游走算法
参数分析
heat exchanger network synthesis
a random algorithm with compulsive evolution
parameter analysis