Please wait a minute...
浙江大学学报(工学版)
JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE)
    
Improved K average spatial clustering method for nodes of water distribution system
LIU Jing qing1, GUO Dong jin1, YE Ping2
1.Department of Civil Engineering, Zhejiang University, Hangzhou 310027 ,China;2.Jia Yuan Water Supply and Sewerage Company,Jiaxing 314000,China
Download:   PDF(726KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

The adaptive elitist genetic algorithm was introducted to optimize the choice of the initial cluster centers. The informationg entropy was combined to objectively determine the attributes’ weights, which can improve K-average spatial clustering for nodes of water distribution system.The case study prove that adaptive elitist genetic algorithm K average spatial clustering has obvious advantages in clustering accuracy, stability, elapsed time and weight choice. Relative to traditional K-average spatial clustering, the average value and standard deviation of inner-class distance resulting from adaptive elitist genetic algorithm K-average spatial clustering respectively decrease from 6.92＼1.06 to 4.39 ＼0.  The clustering accuracy and stability can be  apparently improved. Elapsed time of genetic algorithm K-average spatial clustering, adaptive elitist genetic algorithm K average spatial clustering and simulated annealing genetic algorithm K average spatial clustering were respectively 342,123,383 s, Elapsed time of adaptive elitist genetic algorithm K average spatial clustering is the shortest among the three methods；Network topology shows that the information entropy can objectively determine attributes’ weights, and the result is more reasonable．

Published: 01 November 2015
CLC:  TU 990.3  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

LIU Jing qing, GUO Dong jin, YE Ping. Improved K average spatial clustering method for nodes of water distribution system. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2015, 49(11): 2128-2134.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008 973X.2015.11.013     OR     http://www.zjujournals.com/eng/Y2015/V49/I11/2128


改进的给水管网节点K均值空间聚类

利用自适应精英保留遗传算法优选初始聚类中心,信息熵确定属性权重改进K均值空间聚类,应用于给水管网节点聚类.实例验证表明,改进的K均值空间聚类方法在聚类精度、稳定性、耗时、权重计算方面具有明显的优越性：相对于传统的K均值空间聚类,自适应精英保留策略遗传算法和信息熵确定权重的K均值空间聚类得到的类内距离均值和标准差分别由6.92＼1.06下降至4.39＼0,聚类精度和稳定性均有较大程度提高；普通遗传算法、自适应精英保留策略遗传算法和模拟退火遗传算法优化的K均值空间聚类消耗时间分别为342、123、383 s,自适应精英保留策略遗传算法消耗时间最短；管网拓扑图表明,信息熵权重能客观计算属性权重,结果更加合理.
［1］ 王旭冕,黄廷林,刘勇,等.给水管网水质分区聚类分析中的指标三步筛选法［J］.西安建筑大学学报：自然科学版.2009,41（5）:709-714．
WANG Xu mian,HUANG Ting lin,LIU Yong, et al. The three step screening method of indexes for cluster analysis in water supply system‘s subregion ［J］. Journal of Xi’an University of Architeture & Technology， 2009, 41（5）:709-714．
［2］ 李雅洁,王景成,赵平伟.聚类算法在给水管网节点选择中的应用［J］.计算机工程,2010,36（8）：245-250.
LI Ya jie, WANG Jing cheng, ZHAO Ping wei.Application of clustering algorithm in choosing nodes from water supply network ［J］. Computer Engineering,2010,36（8）：245-250.
［3］ 程广河, 任绪才, 张让勇.城市给水管网水量漏失监测机制研究［J］. 给水排水,2007,33（zl）:355-357．
CHENG Guang he, REN Xu cai, ZHANG Rang yong.Monitoring mechanisms for leakage of urban water distribution system ［J］. Water & Wastewater Engineering, 2007,33（zl）:355-357．
［4］PERELMAN L,  OSTFELD A. Water distribution systems simplificationsthrough clustering ［J］. Journal of Water Resources Planning andManagement，2012,18（3）：218-228.
［5］黄璐, 李树平, 周巍巍,等.基于原有测压点的给水管网压力监测点优化布置研究［J］. 给水排水.2013,39（2）:119-122．
HUANG Lu, LI Shu ping, ZHOU Wei wei.Study on optimized layout of pressure monitoring points in water supply network based on existing pressure monitoring points ［J］. Water & Wastewater Engineering, 2013,39（2）:119-122．
［6］贾新强. 基于趋势面分析的城市给水管网系统漏损区判定的研究［D］.太原：太原理工大学. 2007:25-28.
JIA Xin qang.Research on network leakage region determination based on the analysis method of trend surface ［D］.Taiyuan:Taiyuan University of Technology. 2007:2528．
［7］ 邰明明.基于kriging 的给水管网节点压力插值研究［J］.中国建筑金属结构.2013,（6）:210-211．
Tai Ming ming.Study on node pressure interpolation based on kriging ［J］ .China Construction Metal Structure, 2013（6）:210-211．
［8］ 戴一明. 基于Kmeans聚类法的给水管网水压监测点优化布置［J］.建筑安全.2014,29（8）：74-77
DAI Yi ming.Optimal placement of water distribution system pressure monitors based on k means clustering ［J］.Construction Safety,2014,29（8）: 74-77．
［9］李光强, 邓敏, 程涛,等.一种基于双重距离的空间聚类方法［J］测绘学报.2008,37（4）：482-483．
LI Guang qian,DENG Min,CHENG Tao.et al. A dual distance based spatial clustering method ［J］. Acta Geodaetica et Cartographica Sinica,2008,37（4）：482-483．
［10］李新运,郑新奇,闫弘文.坐标与属性一体化的空间聚类方法研究［J］.地理与地理信息科学.2004,20（03）：38-40．
LI Xin  yun,ZHENG Xin  gi,YAN Hong  wen.On spatial clustering of combination of coordinate and attribute ［J］.Geography and Geo Information Science,2004,20（03）：38-40．
［11］武兆慧,张桂娟,刘希玉.基于模拟退火遗传算法的聚类分析［J］.计算机应用研究.2005,22（12）:24-26．
WU Zhao hui,ZHANG Gui juan,LIU Xi yu.Clustering based on simulated annealing genetic algorithm ［J］. Application Research of Computers,2005,22（12）:24-26．
［12］桂云苗,朱金福.一种用信息熵确定聚类权重的方法［J］.统计与决策.2005 （16）:29-30．
GUI Yun miao,ZHU Jin fu.A method of determining clustering weights with entropy information ［J］.Statistics and Decision .2005 （16）:29-30．
［13］ZHANG Jun, CHUNG H S H, HU B J.Adaptive probabilities of crossover and mutation in genetic algorithms ［J］.IEEE Trans Syst Man and Cybernetics,1994,24（4）：656-667．
［14］魏书堤,姜小奇.一种利用信息熵确定属性权重的模糊单因素评价方法［J］.计算机工程与科学.2010,7（32）:93-94．
WEI Shu di,JIANG Xiao qi.A single factor fuzzy evaluation method of using information entropy to determine the property weights ［J］.Computer Engineering and Science,2010,7（32）:93-94．
［15］朱红灿,陈能华.粗糙集条件信息熵权重确定方法的改进［J］.统计与决策.2011（08）：154-156.
ZHU Hong can,CHEN Neng hua.Improved rough set information entropy method for weight determining ［J］.Statistics and Decision,2011（08）：154-156．
［16］刘冀,王本德.基于组合权重的模糊可变模型及在防洪风险评价中应用［J］.大连理工大学学报.2009,49（02）:272-275
LIU Ji,WANG Ben de. Variable fuzzy model based on combined weights and its application to risk assessment for flood control engineering ［J］.Journal of Dalian University of Technology,2009,49（02）:272-275.
No related articles found!