基于胞腔排除双种群遗传算法的泵站优化调度  

Optimal scheduling of water-supply pumping stations based on cell excluding double population genetic algorithm

作　　者：廖莉[1] 张承慧[1] 林家恒[1] 李洪斌[1] 

机构地区：[1]山东大学控制科学与工程学院,山东济南250061

出　　处：《控制理论与应用》2004年第1期63-69,共7页Control Theory & Applications

基　　金：山东省自然科学基金项目(Y2001G01);2003年教育部优秀青年教师资助计划项目.

摘　　要：泵站优化调度是保证整个水工业系统高效运行的关键因素之一.在提出用指数曲线精确拟合水泵性能曲线的基础上,以泵群轴功率最小作为目标函数,以供水指标和水泵高效区为约束条件,建立了泵站优化调度问题的数学模型.结合实际工程水泵切换次数最少的要求,提出了一种工程实用的胞腔排除双种群遗传算法进行模型求解.该算法借助胞腔和胞腔排除得到的有效胞腔作为水泵并联运行的有效组合方案,并在各有效胞腔内采取全局搜索与局部搜索相结合的双种群搜索策略得到各泵负荷分配的优化解,同时给出了符合工程实际的计算机最优调度方案选择方法.实验结果表明:与传统遗传算法相比,胞腔排除双种群遗传算法可快捷、有效地获得泵站最优调度方案和多个有效调度方案.The optimal scheduling of pumping stations is the key to realizing efficient operation of water industry system. An optimal scheduling model of water-supply pumping stations based on approximation of exponent curves was proposed. The shaft horsepower of the pumps was used as the objective function, and both water supply target and high-efficiency area of pumps were taken as the constrained conditions. With the proposed model, solutions based on the cell excluding and variable population genetic algorithm (CEDPGA) were obtained. The CEDPGA not only got valid cells by cell exclude, which were possible parallel schemes and satisfied engineering requirement on lower switching frequency, but also gained optimal load distribution of the working pumps, which has higher precision. The simulation results showed that the CEVPGA is more effective than the conventional GA.

关 键 词：泵站 优化调度 胞腔排除 遗传算法 数学模型 供水行业 

分 类 号：TH3[机械工程—机械制造及自动化]