The Study Of Pollution Event Determination And Pollution Evaluation Of Rainwater Pipe Network

In this paper,the rainwater pipe network of an industrial park in North China is taken as the research object,the pollutant concentration values of three typical pollution events(domestic sewage,industrial wastewater,rainwater retention pollution)in the rainwater pipe network are monitored and investigated,the hydraulic water quality model of the rainwater pipe network is constructed,the pollutant load of the pollution event is input into the model,the pollutant concentration change rule of the rainwater pipe network is simulated,and the pollution event is established Data set.The model of pollution identification and analysis based on limit learning machine is established to judge and identify pollution area and pollution event.Arrange water quality monitoring points in the study area according to the situation of rainwater pipe network,take cod,total phosphorus,ammonia nitrogen,acesulfame and fluoride as monitoring indicators,conduct in-depth analysis on the pollution of rainwater pipe network in the study area;arrange water quality monitoring points in the sewage pipe network,with the same monitoring indicators as rainwater,conduct in-depth analysis on the water quality of sewage in the study area;according to the plant in the study area.In this paper,the concentration range of industrial wastewater was determined by literature search.The rainfall runoff water quality model is established to simulate the process of runoff and concentration as well as the migration of pollutants in the pipe network.The return period of 1-year return period and 3-year return period is set,and the rainfall lasts for 120 minutes.The experimental data are used to check and analyze the underlying surfaces in the model,and to simulate the hydraulic and water quality at the discharge outlet.Three kinds of typical pollution events are added to the model in the form of retention pollution,and the data sets of water quality change of each pollution event are simulated.Based on the method of limit learning machine,the model of pollution event determination of rainwater pipe network is established,and the simulation data set is used to train and identify the model.First of all,determine whether the pollution event occurs,establish the data set according to the concentration of each pollutant when there is no pollution,and use the limit learning machine model to train and identify.The accuracy of the model is about 90%.It can be considered that the model can be used to analyze the pollution.Secondly,the pollution types and locations are determined,and the research situation is divided into four categories,that is,when the recurrence period is one year and three-year return period,corresponding to the high and low concentrations of pollutants in each pollution event.Under this condition,the comparison of four types of pollution shows that the accuracy of P = 1 high value pollution is 79.63%,P = 1low value pollution is 70.37%,P = 3 high value pollution is 81.48%,P = 3 low value pollution is 74.07%.It can be concluded that the order of the discrimination accuracy of various pollution events is: P = 3 high value class > P = 1 high value class > P = 3low value class > P = 1 low value class.The model can accurately identify the types and locations of pollutants,and can identify and analyze pollution events.