Research On Leakage Fault Diagnosis Of Heating Pipe Network Based On ANFIS

As the scale of central heating continues to expand,the quality of heat supply network management is increasingly valued by heating companies.In recent years,pipe network leaks frequently occur,which has a great impact on the reliability of the heating system.Since the leakage of the thermal network is a sudden event,the time and location of the leak can be found irregularly,and the pipeline is laid underground.This brings great difficulties to the maintenance and repair of the thermal network,which seriously affects the thermal network.The use of normal operations management and hot users.Therefore,this paper aims to provide a practical and effective method for the diagnosis of leakage faults in heating networks.First,a mathematical model of leakage hydraulic conditions in heating network was established based on the spatial network topology.Due to lack of actual data of fault conditions,the model was used for data simulation,and then the model's simulation results were processed to obtain leakage fault training and test data sets.Based on adaptive neuro-fuzzy inference system(ANFIS),a grid fault diagnosis model for heating network is established by using grid method and subtraction clustering algorithm in the toolbox.Both methods can be used in combination with actual operating data to diagnose leaking pipe sections,locating leak locations,and predicting leaks.A corresponding valve closing plan can be given based on the diagnostic results.Through analysis and test results,it is found that in the diagnosis of leaky pipe segments,although both methods can accurately diagnose the leaked pipe segment,the grid method has a significantly higher diagnostic accuracy.In the leak location and leak prediction,the grid method The average relative error of the prediction results is also better than the subtraction clustering algorithm,and the prediction effect is better.Finally,conduct an actual engineering case analysis.Taking the heat pipe network of the first station of Yantai Qingquan No.1 as an engineering example,combined with the actual operation data of the pipe network,the grid fault diagnosis model of the heating network based on the grid method in the adaptive neuro-fuzzy inference system proposed in this paper was tested.The test results show that the method proposed in this paper has a good prediction effect and can meet the actual fault diagnosis of the heating network,providing theoretical guidance for practical projects.