基于多元线性回归模型的电力负荷预测研究

电力负荷预测是电力系统规划和电网运行的重要内容、前提和基础。科学、准确的电力需求预测对电力工业的健康发展，乃至对整个国民经济的发展均有着十分重要的意义。针对我国1995年至2008年人口、GDP和全社会用电量的历史数据，基于多元线性回归分析进行中、长期电力负荷预测，多元线性回归模型通过变量GDP和人口进行全社会用电量的电力负荷定量预测。结果表明模型的有效性，为电力负荷预测的滚动修正，进而为电力负荷控制和预测提供科学依据。

Research on the prediction of power load based on Multiple linear regression model

The prediction of power load is important content, prerequisite and basis of the planning of power system and running of power grid. Scientific and accurate prediction is of great importance to the healthy development of power industry and even the entire national economy. In this paper a prediction for interim and long term power load was presented on the basis of multi factor line regression analysis. The model in this paper opplied the variant GDP and population to predict the total social electricity consumption quantitatively, history datums on the popula- tion, GDP and total social electricity consumption were utlized in this model, the span of datums was from 1995 to 2008. The result showed that the model is effective to the rolling rectifying of power load prediction and further can provide foundation for the controlling and prediction of power load.