Monthly Load Forecasting Model Considering Seasonal Decomposition and Calendar Effects

摘要：

Monthly load forecasting plays an important role in power system planning and marketing. Seasonal and calendar effects are major factors resulting in forecasting error, however, in most of the common monthly load forecasting models, they are always ignored, therefore reduces the accuracy of forecasting models. Regarding seasonal decomposition method which is maturity used in economic data processing, a new method considering seasonal decomposition and calendar effects regression is presented. By adopting approaches of moving average and least-squares regression, monthly electricity consumption curve is decomposed into four components, which are product of trend component, seasonal component, calendar effects component, and irregular component. Corresponding models are chosen to estimate different components according to their characteristics, obtaining sub-forecasting results. The final synthetical forecasting result is the combination of all subforecasting results. A case study using real load data is presented to illustrate the validity of the proposed method.

关键词： Load forecasting Seasonal decomposition Seasonal effects Calendar effects

作者: Ning Zhang Chongqing Kang Qing Xia

作者单位: State Key Lab of Power Systems，Dept. of Electrical Engineering ， Tsinghua University ， Haidian District ， Beijing 100084，China

会议类型: 国际会议

会议名称: The International Conference on Electrical Engineering 2009(2009 电机工程国际会议)

会议地点: 沈阳

会议语种:英文

页码: 1-5

在线出版日期: 2009-07-05（万方平台首次上网日期，不代表论文的发表时间）