城市用水量预测中的多变量灰色预测模型

当观测资料的数据量少,而又存在多个相互影响或关联的变量时,常用的灰色预测模型GM(1,1)不能全面考虑多个变量。采用自适应MGM(1,n)模型———多变量灰色预测模型,较好地解决了这一问题。MGM(1,n)模型是GM(1,1)模型在n元多变量情况下的推广,通过联立求解n个n元微分方程,使模型中的参数能够反映实际工程或社会系统中多个变量间相互影响、相互制约的关系。以杭州市用水量统计资料为研究对象,运用灰色理论建立MGM(1,n)模型,获得了较好的预测效果。

Application of grey multi-variable forecasting model in predicting urban water consumption

The general grey model GM (1,1) cannot take multi variables into account in forecast problems with several variables relating with each other and few observation data available. In this paper, a self-adapting MGM (1,n) model, which is a grey multi-variable forecasting model, is introduced to solve this problem. The MGM (1,n) model is an extension of GM (1,1) in the case of n variables. Through the application of the grey MGM (1,n) model to statistical data of water consumption in Hangzhou City, satisfying results are obtained. The results show that the proposed method is feasible and effective for application.