Research On The Location Of Pipe Burst In Water Supply Network Based On Weighted Least Square Method

Explosion of the water supply pipe network not only causes waste of water resources but also affects the orderly operation of people’s production and life.It is a water supply safety accident that is extremely harmful and has a wide impact.Due to the deep underground water supply network,it is difficult to detect certain tube explosion accidents(such as tube explosions that occurred late at night or in remote areas)by traditional manual inspection and telephon e reporting.Quickly and effectively locating the position of the burst pipe has become a major concern for domestic and foreign water supply industries.With the reduction of monitoring costs and the widespread application of pipe network hydraulic models,the technology of tube explosion based on data and models has attracted widespread attention.Aiming at the location of the blasting tube based on data and model,this paper will conduct research from the following aspects:1.Hydraulic calculation and model construction of water supply network.First introduced the use of hydraulic model,basic theory of hydraulics and modeling steps,then use Water Gems simulation software to calculate the hydraulic adjustment of the actual water supply network,construct the L county pipe network hydraulic model,and finally check the model,the error is 70.3 % Meets modeling requirements.The established hydraulic model provides a reliable model basis for the subsequent blasting tube positioning and exploring the influe nce of model parameters on the blasting tube positioning result.2.Aiming at the problem of locating the pipe explosion in the water supply network,a simulation model of pipe explosion combining the weighted least squares algorithm with the hydraulic model of the pipe network explosion is proposed.By comparing and analyzing the difference between the monitoring value of each monitoring point and the simulated value of the model,the blasting tube simulation model achieves the location of the blasting tub e and shortens the early warning time.Firstly,the hydraulic model of the pipe network in County L is simulated.By continually updating the position and leakage of the pipe explosion to simulate the pipe explosion accident,the pipe explosion simulation model is optimized.It is assumed that the three pipe explosions with the pipe diameter of DN100 and the leakage coefficient K=1,by adding a leakage amount to the pipe network to simulate a burst pipe,the node leakage amount and the residual value of the objective function are calculated,which is consistent with the real situation.The simulation method of the blast tube model can effectively determine the location of the blast tube and the calculation is simple.3.Aiming at the problem that the flow un certainty of the pipe network nodes has an impact on the positioning result of the model pipe bursting,a pipe positioning method based on the false alarm node is proposed.First,establish a model to quantitatively analyze the flow uncertainty of the node s according to the different flow requirements of the pipe network nodes,identify the nodes that are susceptible to the flow uncertainty,compare the correlation between the false alarm nodes and the simulated bursting nodes,and finally use the weighted least squares algorithm to The L county model calculates the residuals,and inversely derives the model burst position.This method focuses on the problem that the locating result of the squib is susceptible to the uncertainty of the node flow.The locatin g performance of the squib of the unified demand model is reflected in the fluctuation of the node false alarm rate of 0.27%～32.77%,and the node false alarm of the distribution demand model The rate fluctuates between 0.36% and 58.02%.4.Aiming at the problem that the existing pipe network monitoring point layout method lacks the division of the pipe explosion area,a pressure monitoring point optimal layout method based on the pipe network explosion area division is proposed.First calculate the pressure difference before and after the simulated blasting event,then use the fuzzy mean clustering algorithm to perform cluster analysis on the fuzzy matrix,and aggregate the pipe network nodes with similar pressure fluctuation characteristics caused by the blasting event into the blasting area,and finally combine Hanoi The leakage rate of each node in the pipeline network is selected as the most representative node as the water pressure monitoring point,k=3 is the most reasonable,the layout of the water pressure monitoring point is optimized,and the model explosion detection performance is improved.This method is more conducive to the detection and positioning of the tube explosion by dividing the tube explosion area in the pipe network and optimizing the layout of the pressure monitoring points.