Research On Water Use Quota And Water Demand Forecasting Of Different Areas Of The City In Shaanxi

The prediction of water requirement plays a key role in the design of urban drainage.Accurately predict the water requirement is significant to the reasonable design of watersupply system and the construction of water facilities. However, the selection of water quotaand the prediction of water requirement affect the result of the prediction on a large scale.There are many standards make rules about waterquota, but lack ofpertinence. Meanwhile, themethod of quota in the prediction of water requirement tends to be higher in practical projects.Thus, it must have differences in Shaanxi province for its regional character, such as the south,north and central. In order to develop the design of water supply rational and effectively, thispaper worked on the water quota and water demand forecasting methodin different regions ofshaanxi, then proposed a suggestion value of waterquota and a recommended method topredict the water requirement.This paper mainly used Two times average method and the probability of budget methodin Statistical analysis method, and the Subjective probability method (PERT) in Experiencemethod to calculate the residents living water quota and per capita water consumption quotain different regions of shaanxi, according to relevant standards and considering the futuredevelopment trend, proposeda suggestion value of waterquota, also analyzed its rationality ineconomics, climate and water resources.This paper used Multivariate regression analysis method, Grey GM（1，1）model, DGMmodel and Verhulst model to calculate the maximum daily water consumption and the annualwater consumption. When used Multivariate regression analysis method, factors of tap waterrate,the total water consumption population, per million population water distribution pipelength, all personnel labor productivity and comprehensive water price were taken asindependents, then make model to get the equation, the overall effect of regression andSignificant of regression were detected using multiple correlation coefficient R value and Ftest. While using Grey prediction method, based on the actual value of water used, modelingcalculated the model equations and the simulation values, analysis of the actual values andsimulation values, calculated the mean absolute percentage error of MAPE and a posteriorierror ratio C to accuracy test. Comparedthe test results, it showed that the grey GM (1,1)modelis more suitable for the maximum daily water demand forecasting andmultiple linearregression modelfor annual water demand predictionin southern area of Shaanxi Province;Guanzhong region water demand prediction is favorable whenusing the grey GM (1,1) model in the near future,while theprediction of long-term water demand prefer to use the multiplelinear regression model; multiple linear regression model is more suitable for NorthernShaanxi area. Using the recommended model, the water demand of every region in2015and2020were predicted.