Hydraulic Station Simulation On Water Supply Network And Study On The Identification Of Pipe Burst Signal

Water supply network is one of the important infrastructure of the city.In recent years,the water supply industry in various countries developed rapidly.With the expanding of the water supply network,the phenomenon of pipe burst is also increasing.It is obvious that the research of pipe burst has become one of the research hotspots in all fields.After analyzing the current research situation,the author found that there are still blanks in the signal recognition of pipe burst and the accuracy of current methods are still limited.Therefore,this paper presents an effective method to detect pipe burst signal by CUSUM algorithm.Firstly,the Water GEMS software is used to construct the hydraulic model of the pipe network.In recent years,the application of the hydraulic model in the design,operation and scheduling of water supply network has become more and more widely.In order to solve the difficult of data acquisition,use the Water GEMS software to build the hydraulic model of the example network.Secondly,Combined with the facts of the example network and the research fruit of overseas and domestic scholars.Analysised the water using characteristic curve by Time Series Analysis Method and Monte Carlo simulation method.On this foundation,simulated the normal operation of the water supply network.Then,the leakage model and the calculation formula of the leakage area of previous research are analyzed.According to the free flow theory of small holes and considering the obstacles of the soil,calculated the leakage model used in this paper.On this foundation,simulated the pipes burst.Finally,CUSUM algorithm can making a clear response to the smaller abnormal,has filtering function and the advantage of simple and sensitive.Therefore,in order to improve the recognition accuracy,the CUSUM algorithm is creatively applied to the pipe burst signal recognition.The application research of CUSUM includes the following contents:CUSUM algorithm analysis,data stability analysis,data normal distribution test,data normal conversion and instance's appliy.When analyze the stability of the data,time series is constitute by water consumption of a single moment,so the influence of normal operating fluctuation can be reduced.By this way,ensured the stability of data and at the same time also improved the recognition accuracy.Finally,the example results show that using CUSUM algorithm to deal with those data can effectively detect the burst signal in the water distribution system and avoid the false alarm.The methods mentioned in this paper can offer reference for the research of water supply network.