Study On The Leakage Risk Of Urban Water Supply Network

Leakage accident of urban water supply network occurs regularly, because thenetwork is susceptible to the threats of municipal, vandalism and nature accident dueto its own vulnerabilities. Generally, leakage accident will affect water quality andcause huge economic loss. The water supply network is facing leakage risk caused bymultiple threats, so it is necessary to study the leakage risk assessment and riskcontrol. In this paper, taking the pipe networks in the main urban area of a city inNorth China as research object, a method of leakage risk assessment and a model ofleakage risk control were proposed by using the systems science method and theory.Firstly, the leakage threats that urban water supply network facing were studied.The threat set was established and the occurrence probability of each threat was thenobtained by two groups of experts scoring method. The uncertainty analysis of experts’scores was conducted, and the occurrence probability of the leakage threats waseventually calculated using the Monte Carlo simulation. The reasonability andfeasibility of this method was proved by a case study.Secondly, a method of leakage vulnerability assessment of the urban watersupply network was proposed. Two-level evaluation index system was set up on thebasis of analyzing previous studies, and vulnerability of each index was calculatedusing statistical data related to leakage. The weights of the first and second gradeindexes were obtained respectively using Analytic Hierarchy Process and FactorAnalysis Method. And the uncertainty of leakage vulnerability assessment is studiedby Fuzzy Comprehensive Evaluation. The accuracy of the methods above was provedby a case study.Thirdly, the assessment and control of leakage risk were quantitativelyresearched. The definition of leakage consequence was proposed and its value wascalculated. Leakage risk was quantitatively assessed and its critical value wascalculated using a formula. Based on the risk assessment, a model of leakage riskcontrol was established using multivariate nonlinear regression method. The value ofleakage risk can be controlled within a certain range by changing the value ofindependent variables in the model. At last, the pipe network in the main urban area ofa city in North China was taken as an example to verify the feasibility of the above methods.