基于NSGA-Ⅱ的给水管网压力监测点多目标优化布置
|
摘要
针对给水管网压力监测点优化布置过程中采用单目标函数得到的优化结果不能代表多方面综合因素的问题,以监测点压力监测范围及监测点覆盖节点需水量作为目标函数,管网水力连通性、管网节点压力相关性、管网节点水量影响条件作为约束条件建立多目标压力监测点优化布置数学模型,并利用Matlab平台通过非支配排序遗传算法(NSGA-Ⅱ)求解。将该模型应用于某园区实际管网中,并分析不同约束条件的参数取值对模型所得优化结果的影响程度,以提高监测点布设方案的代表性,同时也验证了NSGA-Ⅱ求解多目标给水管网压力监测点优化布置模型的有效性。
Single target function lacks representation of the comprehensive factors in the specification of pressure monitoring points in water distribution network. Using the number of covered nodes and the demand of covered points as target functions,and using hydraulic connectivity,correlation of node pressure,and effect of node demand as constrains,a multi-objective optimization model for the arrangement of pressure monitoring points was developed in Matlab platform using non-dominant sort genetic algorithm (NSGA-Ⅱ). The model was applied to the distribution system in a district to evaluate the impact of various constrains on the optimization outputs,and to enhance the quality of sampling locations. The validity of the optimization model for the pressure monitoring points selection in water distribution network with the use of non-dominant sort genetic algorithm was confirmed.
Single target function lacks representation of the comprehensive factors in the specification of pressure monitoring points in water distribution network. Using the number of covered nodes and the demand of covered points as target functions,and using hydraulic connectivity,correlation of node pressure,and effect of node demand as constrains,a multi-objective optimization model for the arrangement of pressure monitoring points was developed in Matlab platform using non-dominant sort genetic algorithm (NSGA-Ⅱ). The model was applied to the distribution system in a district to evaluate the impact of various constrains on the optimization outputs,and to enhance the quality of sampling locations. The validity of the optimization model for the pressure monitoring points selection in water distribution network with the use of non-dominant sort genetic algorithm was confirmed.
引文
[1]Abo-Monasar A,Al-Zahrani M A.Framework for water quality monitoring system in water distribution networks based on vulnerability and population sensitivity risks[J].Water Sci Technol,2016,17(3):811-824.
[2]金溪,曾小兵,高金良,等.利用遗传算法进行供水管网压力监测点优化布置[J].给水排水,2007,33(增刊):346-349.Jin Xi,Zeng Xiaobing,Gao Jinliang,et al.Optimal placement of pressure monitoring points in water supply network based on genetic algorithm[J].Water&Wastewater Engineering,2007,33(s):346-349(in Chinese).
[3]许刚,张土乔,吕谋.基于灵敏度分析和蚁群算法的管网监测点优化选择[J].中国给水排水,2007,23(11):94-96,101.Xu Gang,Zhang Tuqiao,LüMou.Optimized location of monitoring points for water distribution system based on sensitivity analysis and ant colony algorithm[J].China Water&Wastewater,2007,23(11):94-96,101(in Chinese).
[4]曹宏桂,贠卫国.基于PSO-DE算法的突发水域污染溯源研究[J].中国环境科学,2017,37(10):3807-3812.Cao Honggui,Yun Weiguo.Research of the abrupt waters pollution source based on optimization algorithm of PSO-DE[J].China Environmental Science,2017,37(10):3807-3812(in Chinese).
[5]舒诗湖.基于fm GA的供水管网系统模型自动校核及模型应用[D].哈尔滨:哈尔滨工业大学,2009.Shu Shihu.Water Distribution System Model Automatic Calibration Based on fm GA and Its Applications[D].Harbin:Harbin Institute of Technology,2009(in Chinese).
[6]刘书明,王欢欢,徐鹏,等.多目标大规模供水管网监测点的优化选址[J].清华大学学报:自然科学版,2013,53(1):78-83.Liu Shuming,Wang Huanhuan,Xu Peng,et al.Multiobjective genetic algorithms for optimal monitoring station placement in large water distribution systems[J].Journal of Tsinghua University:Science and Technology,2013,53(1):78-83(in Chinese).
[7]Wu Z Y.Multistep simulation-optimization modeling approach for partitioning water distribution system into district meter areas[J].J Water Resour Plan Manage,2018.DOI:10.1061/(asce)wr.1943-5452.0000927.
[8]Deb K,Pratap A,Agarwal S,et al.A fast and elitist multi-objective genetic algorithm:NSGA-Ⅱ[J].IEEETransactions on Evolutionary Computation,2002,6(2):182-197.
[2]金溪,曾小兵,高金良,等.利用遗传算法进行供水管网压力监测点优化布置[J].给水排水,2007,33(增刊):346-349.Jin Xi,Zeng Xiaobing,Gao Jinliang,et al.Optimal placement of pressure monitoring points in water supply network based on genetic algorithm[J].Water&Wastewater Engineering,2007,33(s):346-349(in Chinese).
[3]许刚,张土乔,吕谋.基于灵敏度分析和蚁群算法的管网监测点优化选择[J].中国给水排水,2007,23(11):94-96,101.Xu Gang,Zhang Tuqiao,LüMou.Optimized location of monitoring points for water distribution system based on sensitivity analysis and ant colony algorithm[J].China Water&Wastewater,2007,23(11):94-96,101(in Chinese).
[4]曹宏桂,贠卫国.基于PSO-DE算法的突发水域污染溯源研究[J].中国环境科学,2017,37(10):3807-3812.Cao Honggui,Yun Weiguo.Research of the abrupt waters pollution source based on optimization algorithm of PSO-DE[J].China Environmental Science,2017,37(10):3807-3812(in Chinese).
[5]舒诗湖.基于fm GA的供水管网系统模型自动校核及模型应用[D].哈尔滨:哈尔滨工业大学,2009.Shu Shihu.Water Distribution System Model Automatic Calibration Based on fm GA and Its Applications[D].Harbin:Harbin Institute of Technology,2009(in Chinese).
[6]刘书明,王欢欢,徐鹏,等.多目标大规模供水管网监测点的优化选址[J].清华大学学报:自然科学版,2013,53(1):78-83.Liu Shuming,Wang Huanhuan,Xu Peng,et al.Multiobjective genetic algorithms for optimal monitoring station placement in large water distribution systems[J].Journal of Tsinghua University:Science and Technology,2013,53(1):78-83(in Chinese).
[7]Wu Z Y.Multistep simulation-optimization modeling approach for partitioning water distribution system into district meter areas[J].J Water Resour Plan Manage,2018.DOI:10.1061/(asce)wr.1943-5452.0000927.
[8]Deb K,Pratap A,Agarwal S,et al.A fast and elitist multi-objective genetic algorithm:NSGA-Ⅱ[J].IEEETransactions on Evolutionary Computation,2002,6(2):182-197.