基于灰色预测和Elman神经网络的全国用水量预测

National Water Consumption Forecast Based on Grey Prediction and Elman Neural Network

为解决我国存在的水资源问题,考察各省份的用水状况。从生活、农业、工业和生态4个方面进行全面分析,对各省份2020年的用水量进行预测。采用GM(1,1)模型预测所有用水量影响指标,并使用Elman神经网络拟合了指标与各省份年用水量的映射关系。通过数据检验,可见基于GM(1,1)模型和Elman神经网络的用水量预测模型误差很小。最后,预测结果与国务院发布的水资源"三条红线"中2020年的用水量控制红线进行比较,结果表明:黑龙江、江苏、江西、广西、甘肃、宁夏和新疆7个地区在2020年的用水量预测值超过国家要求。

In order to solve the problem of water resources in China,the water use situation of each province was investigated. A comprehensive analysis of life, agriculture, industry and ecology was car- ried out to forecast the water consumption of the provinces in 2020. The GM （ 1,1 ） model was used to predict all water consumption impact indicators, and the mapping relationship between the indica- tors and the annual water consumption of each province was fitted using Elman neural network. Through data validation, the error of water consumption prediction model based on GM （ 1,1 ） model and Elman neural network is very small. Finally, the prediction results were compared with the ＂Three Red Lines＂ issued by the State Council. The annual water consumption control red line is compared. The forecast results show that the predicted water consumption in Heilongjiang, Jiangsu, Jiangxi, Guangxi, Gansu, Ningxia and Xinjiang will exceed the national requirements in 2020.