Abstract
In order to predict urban water resources utilization more accurately, this paper respectively applied unbiased grey GM (1, 1) model, nonlinear model and combined model of them to calculate water consumption of Beijing from the year 2001 to 2010. Because different cities have different natures, each model used for predicting a specific city may have some disadvantages. Through weighted coefficients, combine the two model unbiased grey model with nonlinear model as weighted combined model. With accuracy testing methods, we concluded the accuracy of each model. The results showed that combination of unbiased grey GM(1,1) model with nonlinear model had highest accuracy than the other single model. The error variance of the predicting result of the unbiased grey GM (1, 1) predicting model was small but the average absolute error of that was large, but the error variance of the result of nonlinear prediction model was large and the average absolute error of this was small. This weighted combination model can integrate advantages of the above two kinds of model and make the result more accurate and reliable, which can be used in the short term and the long term prediction of urban water consumption. Water utilization of Beijing in 2020 was forecasted through the combined model is that the industrial water use is about 1805 million cubic meters and life water use is only 197 million cubic meters.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have