Fault Diagnosis Of Heat Supply Network Based On BP Neural Network Optimization Algorithm

With the continuous expansion of the scale of the central heating pipe network and the increased use of the number of years,the corresponding faults and breakdowns inevitably occurred.Problem leakage is more frequent in it.This will not only brings security risks to people's production and life in winter,but also wastes energy causing huge economic loss.So Studying and exploring leakage diagnosis methods and detecting leakage fault earlier has substantial economic and social significance.This thesis combined with actual engineering project.First,analysising the heating real-data which setting in heating information management platform.By means of the nonlinear approximation capability of BP artificial neural network,this thesis model and forecast the return temperature of secondary network according to outdoor temperature and the supply temperature of secondary network.The results show that the precision of established BP neural network forecast model basically meets actual requirement.Second,the traditional BP neural network's uncertainty initial weights and thresholds is easily made the neural Salivary gland pitfalls of slow convergence speed and falling into local minima.In order to make up for the defects,this thesis presents a fault diagnosis mode of heating networks based on BP neural network particle swarm optimization.Simulation results by MATLAB verify that the improved algorithms is better,it can effectively accelerate convergence rale and raise the prediction accuracy.Last,designing semi-physical simulation laboratory equipment and a similar experiment is carried out to verify the method of pipeline leakage diagnosis is reasonable for the real-time prediction of the return temperature of secondary network.Use a series of leakage faults experiment,the experimental results indicate that this method for rapid detection diagnosis has certain effectiveness.With increasing central heating network sophistication.The attention for the heating pipeline network leakage diagnosis will also improved.Further enhanced the economy and the security.This method continues to be improved in the field of heating engineering applications.At the same time,researching on heating pipeline leakage diagnosis make a further approach to petroleum,gas and water pipeline's diagnosis.