Font Size: a A A

Water Quality Monitoring And Early Warning System Of Water Supply Network Based On Hydraulic Model

Posted on:2018-10-29Degree:MasterType:Thesis
Country:ChinaCandidate:H FuFull Text:PDF
GTID:2322330536984390Subject:Architecture and civil engineering
Abstract/Summary:PDF Full Text Request
Water distribution system is an important system for municipal water supply system.However,with the rapid development of urban economy and the continuous expansion of the urban boundary,water distribution system has gradually developed into a complex system with a large number of pipelines.And more and more people pay attention to the problem of water quality in water distribution system.Therefore,the optimal layout of water quality monitoring points and the reverse tracing model of pollution sources have become a hot research direction.In this paper,firstly,the study status of water supply network hydraulic model,water quality model and pollution source traceback model are discussed.Secondly,based on multi factor,the calculation method of node water age is optimized,the optimal layout of conventional water quality monitoring points is realized.The mathematical model and calculation program based on multi factor optimization method are established.The results show that 43.75% of the optimal monitoring points are the same as the actual water quality monitoring points.Based on the simulation of random pollution accident,it is found that 83.3% of the optimal monitoring points are better than the actual monitoring points.According to the theory of fuzzy clustering and the FCM algorithm,the optimal location model of constrained target is established.The method is applied to the hydraulic model of water supply network in M City,and the optimal position of the water quality monitoring points is obtained.The results show that,through the simulation of random pollution accident,there are 70% of the optimal monitoring points are better than the conventional monitoring points.Then,by the EPANET as the simulation engine for water quality,using the method of "virtual partition" on the network,combined with the BP neural network module in the MATLAB to solve the problem of pollution source location in pipe network.According to the different scale and type of the actual water supply network water quality model,the relationship between the number of monitoring points,the size of samples,the number of partitions,and the positioning accuracy of waterpollution in water supply network is studied.The results show that the number of monitoring points and the size of the sample are proportional to the accuracy of the pollution source,while the number of partitions is inversely proportional to the accuracy of the source location.
Keywords/Search Tags:Hydraulic model of water distribution system, The reverse tracing model of pollution sources, FCM algorithm, BP neural network
PDF Full Text Request
Related items