Research On Intelligent Leakage Analysis And Positioning Technology Of Urban Water Supply Network

Network leakage is a common phenomenon in the water supply industry. Allcountries in the world regard the leakage control of the water supply network as animportant subject to study. In order to resolve the problem radically, the reason andmechanism of the network leakage should be analysised, and effective technology tocut and control the leakage should be researched and developed, so as to provideimportant basis for putting forward scientific and reasonable meathods of leakagecontrol.According to the present problem in the research on leakage location analysisand control management of water supply network, the researches are carried out inthis article as follows: cause analysis and simulation experiment of urban watersupply network leakage, pressure variation characteristics analysis based on leakagevolume change, inverse calculation of leakage location and optimization based ongenetic algorithm, and leakage state identification and location of water supplynetwork based on the BP neural network. Based on the above researches, theintelligent leakage analysis and positioning technology can be applied in practiceperfectly.Most of the studies on the relationship between leakage and pressure of watersupply network are based on the traditional orifice flow calculation formula, in whichthe influence of external environment in the actual leakage has not been taken intoaccount. Therefore there are limitations in current research. This paper studies on thesynergistic effect of the soil outside of the orifice and pipe on leakage. Throughclassifying the proportion of the two influencing factors, the functional relationshipsbetween network leakage and pipe head under different states are studied, and theanalysis model is built correspondingly.The simulation experimental platform of water supply network is established.Furthermore, the pressure variation characteristic under leakage state in water supplynetwork is studied. The hydraulic simulation model of the platform is built by EPANET. According to the real-time monitoring data of network leakage andpressure, the simulation experiments of the pressure variation characteristic in watersupply network are done based on the leakage volume change, in order to master thepressure variation law in water supply network under the condition of one or moreleakage points, and provide important theoretical basis for the implementation ofwater supply network location technology.Based on the researches above, using the hydraulic simulation model, the watersupply network inverse problem model of leakage location is set up based on thegenetic algorithm, which takes the leakage location and volume in the network as thevariables, and takes the minimum difference between monitoring and simulationvalues of the pressure monitoring points as the objective. The rationality andeffectiveness of the leakage location method are verified though experiments underthree leakage conditions.In order to improve the precision of the inverse problem leakage location model,the leakage state identification and location technology of water supply network is putforward based on the BP neural network. Through the sensitivity analysis of theleakage model, the accuracy and practicability of the technology are improved on thebasis of on-line monitoring data. In addition, the optimization results are verified bymeans of experiments. The researches above provide strong technical support fordiagnosing leakage or tube explosion timely and reducing the leakage volumeeffectively in water supply network.