基于改进灰色模型GM(1,1)的生活用水量预测研究

Prediction of Domestic Water Consumption Based on Improved grey model GM(1,1)

生活用水量的预测对水资源规划、决策及管理起到重要作用。分别采用传统灰色预测模型和背景值改进的灰色预测模型,以河南省居民2012年~2018年生活用水量为例,检验两种模型的拟合情况。研究结果表明:两种模型的预测精度等级均为优,但改进后模型的平均相对误差及后验方差比值均低于传统模型,这表明改进模型比传统模型具有更高的预测精度,能够更好地预测河南省居民生活用水量。最后利用改进后的灰色预测模型对2019年~2021年河南省居民生活用水量进行预测,分别为43.37亿m^3、45.42亿m^3、47.56亿m^3。

Water consumption prediction plays an important role in water resources planning, decision-making and management. With domestic water of Henan Province in 2012~2018 as an example, the traditional grey model and the background value-improved gray model are tested in the paper. The results show that the prediction precision levels of both models are "excellent", and the average relative error and posterior error ratio of the improved model are lower than the traditional one, which indicates that the improved model has higher prediction precision, and we can arrive at the conclusion that the improved model can predict the domestic water consumption in Henan Province better. Finally, this paper uses the improved model to predict domestic water consumption in Henan Province from 2019 to 2021, respectively, are 4.337, 4.542,4.756 billion cubic meters.