Font Size: a A A

Research On Accurate Simulation Of Leakage Of Pipe Network By Improved FAVAD Model

Posted on:2019-12-23Degree:MasterType:Thesis
Country:ChinaCandidate:T L ShiFull Text:PDF
GTID:2382330566996651Subject:Municipal engineering
Abstract/Summary:PDF Full Text Request
Physical leakage is an important part of the leakage of water supply network.How to quantify the physical leakage of regional cities is of great significance to the prediction of the leakage of the pipe network in the whole region and the subsequent leakage control.The existing leakage theory of the pipe network has some defects.The aim of leakage is to further improve and enrich the leakage theory of water supply network from the single leak point theory,and use the new point leakage model to predict the leakage of the pipe network.In the leakage process,the relationship between the leakage area and the water head is taken into account in the FAVAD model,and the leakage calculation is more accurate.Based on the theoretical basis of the traditional FAVAD model,this paper first analyses the fluid solid coupling of different types of leaks,and studies the change law of the different types of leaks under the pressure conditions,and obtains the change law of pressure,stress and strain at the leakage of various forms of failure under different pressure.It is found that the change of the leakage of the circular and circumferential cracks is in line with the traditional FAVAD leakage model.The leakage area of the axial crack is influenced by the stress concentration,and is directly proportional to the square of the water pressure.Based on this,it provides a theoretical basis for subsequent leakage experiments.Through the related pipeline leakage test,the leakage law of single leak point pipeline is explored.From the basic information of the two basic information of the leakage form and the size of the leakage,the basic information of the three pipelines,the diameter of the pipe,the pipe material and the water pressure of the pipeline,the effects of the leakage information and the basic pipeline information on the leakage flow are studied.On the basis of the factors,the parameters in the FAVAD model are optimized,and the specific expression of the m value under different failure forms in the FAVAD leakage formula is obtained,and the expression form of the FAVAD leakage formula is refined.Based on a large number of FAVAD leakage tests,the GA-BP neural network is used to predict the m value of the FAVAD model,which greatly simplifies the computational complexity of the model and has a good fitting precision.After optimization,the prediction accuracy of the leakage prediction value can be as high as 99%,and the leakage loss prediction model is combined with the EPANET software.A new model is used to predict leakage flow in the leakage area.By verifying the pipe network,the prediction error of the improved model is reduced from 18.7% to 12.1% for the leakage area of 1,and the prediction error of the leakage region is reduced from 13.8% to 10.2% for the leakage region 3.The traditional pressure leakage model is mostly based on the exponential leakage model.Although the exponential leakage model can predict the leakage of the area,the basic principle ignores the influence of the basic parameters of the pipeline.It has some limitations.The new model is based on the FAVAD leakage formula,and the leakage information,pipeline information and pipeline are considered.The effect of water pressure on leakage is more accurate and practical.
Keywords/Search Tags:Fluid solid coupling, FAVAD model, Neural network, Leakage prediction, Genetic algorithm
PDF Full Text Request
Related items