基于协整-格兰杰因果检验和季节分解的中期负荷预测

Medium-term Load Forecasting Based on Cointegration-Granger Causality Test and Seasonal Decomposition

近年来,随着国民经济的转型,中国的经济结构发生了较大的变化,仅仅依靠电力负荷历史数据进行负荷电量预测会造成较大的误差。为解决传统负荷预测方法对于经济、气象等因素考虑不足的问题,提出了一种可以计及经济与气象等因素影响的中期负荷电量预测方法。首先利用季节分解将历史月度用电量分解为长期趋势及循环分量、季节分量以及不规则分量;并以计量经济学中的协整检验以及格兰杰因果检验分析经济因素与用电量长期趋势及循环分量的关系,确定影响该部分电量预测的关键性指标;基于电量、气象以及经济数据,对各个分量利用支持向量机分别进行预测并综合得到月度电量总量预测值;最后通过算例分析了方法的有效性与可行性。

In recent years,with the transformation of national economy,great changes have taken place in the economic structure of China.The prediction based on the historical data of electric power load will cause great error.In order to solve the problem which traditional load forecasting method is not enough for economic and meteorological factors,a forecasting method for medium-term load is proposed.This method can consider the influence of economy,climate and other factors.First,using seasonal decomposition,the monthly electricity consumption of history is decomposed into long-term and cycle component,seasonal component and irregular component,and the relationship between economic factors and long-term trend and cyclic components of electricity consumption is analyzed by cointegration test and Granger causality test in econometrics.The key indexes to influence the prediction of electric quantity is determined.Each component is predicted by support vector machine(SVM)based on electricity,meteorology and economic data,and the monthly total quantity of electricity is predicted.Finally,the effectiveness and feasibility of the method are illustrated by an example.