Prediction of Urban Water Consumption Based on Residual Modified GM(1,1)Grey Dynamic Model Group
[Purposes]This paper aims to explore the application of residual modified GM(1,1)grey dy-namic model group in urban water consumption prediction.[Methods]On the basis of the traditional single GM(1,1)gray model,the logarithmic curve was used to fit and correct the residual results,and the traditional single GM(1,1)gray model and the residual modified GM(1,1)gray dynamic model group were constructed,the actual urban water consumption data of Jining City from 2013 to 2022 were used for case verification and prediction analysis.[Findings]Compared with the traditional single GM(1,1)gray model,the fitting accuracy and prediction accuracy of the traditional single GM(1,1)gray model group were significantly improved,and the fitting relative error of the traditional single GM(1,1)gray model was 0.22%～4.22%,the prediction relative error was 0.91%～2.74%,the prediction average relative error was 1.62%,the prediction mean square error was 0.143,and the residual modified GM(1,1)0 gray dynamic model group fitting relative error was 0.02%～3.09%,the relative error of prediction is 0.75%～2.42%,the average relative error of prediction is 1.32%,and the mean square error of prediction is 0.102.[Con-clusions]Through the prediction results of Jining's water consumption from 2023 to 2030,it can be seen that Jining's future water consumption is on a downward trend,which is consistent with Jining's strict implementation of the strictest water resources management system,rigid constraints on water-related in-dicators,and the improvement of water use efficiency.And the residual modified GM(1,1)gray dynamic model group can meet the prediction of urban water consumption.
residual correctiongrey dynamic model groupwater consumption forecastingJining City
NETL
NSTL
万方数据
