Research On The Identification Of Leakage Signals And The Method Of Regional Location In Urban Water Supply Networks

The distribution of water resources in China is uneven,fresh water resources are scarce,water environment pollution is serious,and the leakage in urban water supply system is also serious,which has not been effectively solved.Under the condition of large-scale urbanization,how to ensure the safety and reliability of urban water supply system,reduce the leakage rate of water supply,and improve the efficiency of water supply has become a scientific problem of sustainable development of social economy,and also the goal of study of scientific researchers.In order to improve the operation safety and leakage management level of urban water supply system,the regional positioning model of leakage identification in urban water supply network is studied in this papers.Firstly,the hydraulic model of urban pipe network is constructed.The accurate hydraulic model is the basis of optimal layout of pressure monitoring points,leakage signal recognition and leakage area location.This paper takes the establishment of a city’s pipe network hydraulic model as an example.The calibration results show that the percentage of the pressure error range ≤ 4m is 100%,and the percentage of the error range ≤ 2m is 94%;the flow error shows that the relative error values of the two water treatment plants are all below2.5%,so the hydraulic model can be used in the pipe network simulation calculation.Secondly,according to the current situation of water supply network operation,the optimal layout method of pressure monitoring points is proposed.The influence coefficient matrix of network node pressure is constructed firstly,and then the fuzzy c-means clustering algorithm of particle swarm optimization(PSO)is used to solve the matrix to get the optimal location of monitoring points.Particle swarm optimization avoids the disadvantage that FCM clustering algorithm is easy to fall into local optimum when the initial value is not properly selected,and improves the accuracy of covering nodes of pressure monitoring points.Combined with the example analysis,the results show that the pressure monitoring points are reasonably arranged and evenly distributed in the pipe network.At the same time,the new pressure monitoring points arrangement method is proposed,which lays the foundation for the identification and location of pipeline network leakage.Thirdly,by collecting the historical data of pressure value of SCADA monitoring system of water supply network,and constructing the time series prediction model of PSO to optimize LSSVM,the prediction model is used to predict the pressure value of pressure monitoring point at the next time,and SPSS software is used to analyze and calculate the threshold value of each pressure monitoring point at each time.The residual value is obtained by the difference between the real-time remote monitoring point pressure value and the predicted value.When the residual value is within the threshold value of the pressure monitoring point,it belongs to the normal working condition;once the residual value exceeds the threshold value,it belongs to the abnormal working condition,i.e.leakage event or pipe explosion accident.Combined with the model analysis,the results show that the prediction accuracy of the improved time series prediction model is high,and the thresholds of each pressure monitoring point can be determined to identify the leakage accident of the pipe network.Finally,two kinds of leakage area positioning methods are proposed,one is to calculate the residual value of each pressure monitoring point in the water supply network.When leakage occurs,the position of the monitoring point whose residual value changes to a large extent constitutes the leakage area;the other is to divide the water supply network into different areas and establish the leakage sensitive area of the network based on the hydraulic correlation between pressure monitoring equipment and node leakage.In this paper,the hydraulic model of the study area is established,and the effect locating the leakage area is analyzed.The results show that the minimum distance between actual leakage point and the monitoring point is about 300 m,while the maximum distance between actual leakage point and the monitoring point is about 620m;when the value of the three leakage response thresholds is 97%,the number of covered nodes is the most,the division of sensitive areas is the most accurate,and the leakage point is about 200 m away from the presusure monitoring point,which indicated the leakage positioning effect is obvious,and can reduce the leakage detection time and improve the leakage detection efficiency.