中文摘要：

改革开放以来,我国农村年用电量受经济和环境的影响,对于波动幅度较大的数据,纯粹的采用单纯灰色预测已经满足不了精度方面的要求,因此文中提出了基于最小二乘法的改进GM（1,1）模型,首先文中介绍了普通GM（1,1）模型建立方法和步骤,接着采用最小二乘法的原理弱化波动较大的数据,加强其规律性从而建立新的GM（1,1）模型,最后通过1999—2008年我国农村年用电量建立新的预测模型,并用2009年的数据对模型进行验证合格,可以用来预测我国未来农村用电量,便于国家对此采取宏观调控政策,结果表明该方法具有更高的精度和可行性,并为其它相关预测提供理论依据。

英文摘要：

Rural electricity consumption in China is influenced by economic and environmental, the influence of data fluctuation is large, the simple using grey prediction model has been unable to accurately predict, so in this paper, based on the least square method is proposed to improve GM（1, 1） model, first of all, this paper introduces the ordinary GM（1, 1）modeling method and the steps, then adopts the principle of least square method to weaken the volatile data, strengthen its regularity to establish a new GM（1, 1） model, at last, through 1999-2008 in China＇s rural electricity consumption, a new prediction model is set up by using 2009 data to verify this model, which can be used to predict the future of rural electricity consumption in our country, facilitate the country took the macroeconomic regulation and control policy, the results show that the method has higher accuracy and feasibility of and provide theoretical basis for other related prediction.