基于GMDH的中长期电力负荷组合预测模型

组合负荷预测模型能够充分利用数据信息,有效降低预测风险、改善预测效果,在中长期负荷预测中获得了广泛应用。而目前的组合预测模型实质大都为单一预测模型的加权平均,没有能够充分发挥综合预测的优势.应用数据分组处理方法（GMDH）进行组合预测,在充分考虑各单一模型特点和预测效果的基础上,形成多元非线性组合预测模型,自动从数据中挖掘出重要信息,克服了传统组合预测模型建模中的主观因素影响,可以改善预测精度。并将该预测模型应用于实际电网,计算结果表明该模型有效提高了预测精度,适用于中长期负荷预测.

Study on combination forecasting model for mid-long term power load based on GMDH

The combination forecasting model has been widely used in mid-long term load forecasting,which can reduce the forecasting risk and then improve the forecasting effect.Many combining forecast models are the weighted average results of some single forecasting models,so advantages of the combining forecast model have not been discovered yet.The group method of data handling(GMDH) was applied to form the multiple non-linear combination forecasting model in this paper.The important information was automatically extracted from data,and non-linear mapping model was automatically generated.The forecasting accuracy was improved.The results in practical operation demonstrated that the proposed method can overcome the disadvantage of traditional combination forecasting model.It is a good way for mid-long term load forecasting.