Research On The Numerical Model Of TOC And COD_Mn In The Water Supply Network In The North Of China

In recent years,the water quality of drinking water has been more and more attention.With the development of industrial technology,surface water was polluted by different levels,the water of the organic matter concentration increased,more threat to the safety of drinking water.The treated water will still contain some organic matter,the water in the water supply pipe network transmission process,the presence of organic matter will lead to the increase of bacteria in the water and disinfection by-products,scaling phenomenon can also be generated,increasing the risk of tube burst.With northeast J city as the research object,this study on the characterization of organic matter in water supply network of two water quality indexes of TOC and COD_Mn were studied,after obtaining a large number of measured data,on the basis of analysis of pipeline network of water supply pipe network the common water quality indicators evaluation analysis,TOC,COD_Mn multiple linear regression model and BP neural network model are established,which provide a theoretical basis for the control of organic matter in water supply network.First,in this study,on the basis of reading a large amount of relevant literature,has carried on the on-the-spot investigation to J city water supply pipe network,combined with the monitoring point arrangement principle eventually identified at the end of the entrance,the way of water supply pipe and 6 monitoring stations,for 20 consecutive days cycle for sampling and determination of drinking water.Second,the data obtained through the analysis of this research,to study the change of the indexes of water supply network,and analyzes the correlation between the indexes,providing data basis for the establishment of the model and theoretical basis.Studies have shown that the index with the increase of distance water,residual chlorine,TOC and COD_Mn,overall present attenuation change,turbidity and water age,on the whole,showing a growth change,UV₂₅₄ change regularity of poor,other indicators and TOC and COD_Mn all have certain relevance,residual chlorine and the two of the most significant correlation.Finally,used a large amount of measured data obtained from this study established the TOC,COD_Mn,multivariate linear regression model and BP neural network model,and test of the two kinds of model.Results show that the BP neural network model for the TOC and COD_Mn prediction effect significantly better than multiple linear regression model,the model of optimization can be used for practical prediction in the future.