伪并行遗传算法在供水管网优化调度中的应用  

Application of Pseudo-Parallel Genetic Algorithm in Optimal Operation on Water Supply Network

作　　者：信昆仑[1] 刘遂庆[1] 陶涛[1] 李树平[1] 

机构地区：[1]同济大学环境科学与工程学院,上海200092

出　　处：《同济大学学报（自然科学版）》2006年第12期1662-1667,共6页Journal of Tongji University:Natural Science

基　　金：国家自然科学青年基金资助项目(50409016);国家"八六三"高技术研究发展计划资助项目(2004AA649410)

摘　　要：讨论了基于微观水力模型的多水源大型供水管网优化调度问题的主要特征,并提出了求解该问题的改进遗传算法.首先,针对决策变量的结构提出二进制-实数混合编码策略;其次,设计并实现了多种群进化的伪并行遗传算法.将本算法与单一群体进化算法同时应用于实际管网的优化调度求解,结果表明,通过个体迁移策略,伪并行遗传算法可以加速优化搜索的进程,显著改善解的质量,并有效节省运行调度费用.Based on a hydraulic simulation model, the main features of operational optimization problem for the multi-source water supply network are discussed. An improved genetic algorithm is then proposed. Firstly, a mixed-coding methodology was adopted according to the characteristics of the decision variables. Secondly, a Pseudo-Parallel GA （PPGA） was designed and implemented. Both simple GA and PPGA were applied to solve an operational optimization problem for a real-life water network. Results show that the evolution process of those populations can be accelerated through the migration of individuals when using PPGA. Compared to SGA, the PPGA proves to be more efficient and able to find much better solutions. And substantial saving on operation costs can also be achieved if the optimal solution is adopted.

关 键 词：供水管网 水力模拟 优化调度 伪并行遗传算法 

分 类 号：TU991.33[建筑科学—市政工程]