Research On Fault Diagnosis Of Heating Pipe Network Based On Simulation Technology

With the rapid development of China’s economy and science and technology,China’s central heating industry has made great progress.With the increase of newly built buildings and the expansion of the scale of heating system,the failure of pipe network has attracted the attention of heating enterprises.The failure of heating pipe network is caused by many reasons,and It often occurs that it is difficult to quickly find the fault location in practical engineering,especially the old pipe network that has been running for many years,the leakage failure of heating pipe network is frequent.Pipe network failure will lead to the decrease of heating effect for heating users and increase the operating cost of heating system.In this thesis,based on the simulation study of heating pipe network in the early stage,the mathematical model of pressure of leakage node of heating pipe network is established,which is transformed into a simulation model by Fortran language,and the resistance coefficient value of pipe segment is added.The fault simulation system of heating pipe network is established by Cybersim simulation software.The experimental platform was reformed by opening a hole in the main pipe section of the hydraulic condition experimental platform.The hole was connected with the short pipe,and the short pipe was connected with the ball valve,and the leakage was controlled by the ball valve.By adjusting the opening of the stop valve on the test table as the plug.Based on the experimental platform,the heating pipe network fault simulation system was built.In the simulation system,the fault was set up and the data was obtained by referring to the fault location and degree of the experimental platform pipe network.Comparing some experimental measurement data with the data simulated by the simulation model,The comparison results show that the experimental data are basically consistent with the simulated data,and the relative error is within 5%,which proves that the fault simulation system of heating pipe network has strong applicability.By setting leakage points and leakage amounts at different positions and plugging points at different positions and different plugging degrees in the fault simulation system of heating pipe network for multi-condition simulation,a large number of operating parameters of heating pipe network under fault conditions are obtained.On the one hand,by analyzing and summarizing the operation parameters of heating pipe network under the fault condition,the partial operation rules of heating pipe network under the fault condition are obtained,so that the fault diagnosis of heating pipe network can be carried out according to the variation rules of the operation parameters of heating pipe network.In this thesis,using the principle of BP neural network,on the other hand,the heating pipe network fault diagnosis model was established,the simulation system to get part of running parameters of the heating pipe network fault conditions as the learning sample and forecast sample,head of section in pressure and flow rate as input layer nodes,to fault section location Numbers for output layer nodes,respectively on two kinds of fault location,Through the operation model,it is found that the accuracy of predicting and locating the two kinds of faults is above 88%,which provides a method for finding the fault location of pipe network in practical engineering.