Leakage Rate Model of Urban Water Supply Networks Using Principal Component Regression Analysis  

作　　者：Zhiguang Niu Chong Wang Ying Zhang Xiaoting Wei Xili Gao 

机构地区：[1]School of Environmental Science and Engineering,Tianjin University,Tianjin 300072,China [2]Key Laboratory of Pollution Processes and Environmental Criteria of Ministry of Education,College of Environmental Science and Engineering,Nankai University,Tianjin 300071,China [3]Binhai Industrial Technology Research Institute of Zhejiang University,Tianjin 300450,China [4]Tianjin Urban Construction Design Institute,Tianjin 300122,China

出　　处：《Transactions of Tianjin University》2018年第2期172-181,共10页天津大学学报（英文版）

基　　金：supported by the Ministry of Science and Technology of China (No.2014ZX07203-009);the Fundamental Research Funds for the Central Universities;the Program for New Century Excellent Talents at the University of China

摘　　要：To analyze the factors affecting the leakage rate of water distribution system, we built a macroscopic "leakage rate–leakage factors"(LRLF) model. In this model, we consider the pipe attributes(quality, diameter,age), maintenance cost, valve replacement cost, and annual average pressure. Based on variable selection and principal component analysis results, we extracted three main principle components—the pipe attribute principal component(PAPC), operation management principal component, and water pressure principal component. Of these, we found PAPC to have the most influence. Using principal component regression, we established an LRLF model with no detectable serial correlations. The adjusted R2 and RMSE values of the model were 0.717 and 2.067, respectively.This model represents a potentially useful tool for controlling leakage rate from the macroscopic viewpoint.To analyze the factors affecting the leakage rate of water distribution system, we built a macroscopic "leakage rate–leakage factors"(LRLF) model. In this model, we consider the pipe attributes(quality, diameter,age), maintenance cost, valve replacement cost, and annual average pressure. Based on variable selection and principal component analysis results, we extracted three main principle components—the pipe attribute principal component(PAPC), operation management principal component, and water pressure principal component. Of these, we found PAPC to have the most influence. Using principal component regression, we established an LRLF model with no detectable serial correlations. The adjusted R2 and RMSE values of the model were 0.717 and 2.067, respectively.This model represents a potentially useful tool for controlling leakage rate from the macroscopic viewpoint.

关 键 词：Water DISTRIBUTION system LEAKAGE RATE LEAKAGE influencing FACTOR QUANTITATIVE model Principal COMPONENT regression 

分 类 号：G64[文化科学—高等教育学]