基于熵值法的组合模型用电量预测方法研究

电力系统的中长期负荷预测是配电网规划的基础,对实现电网的安全经济运行起着重要作用。以年度用电量预测作为研究的对象,年度用电量预测采用4种主要方法,即分别按照年度、季度、月度和行业用电量预测得到对应年用电量预测值,在此基础上再按其发展序列预测结合起来,建立了一种线性组合预测模型。并采用熵值法对组合模型的权系数进行求解,实证分析表明该模型使预测精度得到了明显提高,具有良好的预测效果。

A Combination Model Based on Entropy Method for Annual Electricity Consumption Forecasting

Tile medium-term and long-term load forecasting is the foundation of the distribution network planning. The precise load forecasting can improve the reliability and economy of the power grid. In this paper, with the annual electricity consumption taken as the research object, the forecasted value of the annual electricity consumption is obtained by the following four means： the annual, quarterly, monthly, and industry electricity consumption. A linear combination load forecasting model is established by its developing sequence forecasting. The entropy method is adopted to obtain the weight coefficient of the combination model, and analysis of the example result shows that this new model significantly improves the forecasting precision.