| 引用本文: | 张 平,潘学萍,薛文超.基于小波分解模糊灰色聚类和BP神经网络的短期负荷预测[J].电力自动化设备,2012,32(11): |
| ZHANG Ping,PAN Xueping,XUE Wenchao.Short-term load forecasting based on wavelet decomposition，fuzzy gray correlation clustering and BP neural network[J].Electric Power Automation Equipment,2012,32(11): |
|
| 摘要: |
| 提出基于小波分解、模糊灰色聚类和BP神经网络的短期负荷预测方法。通过小波分解将负荷序列分解成低频分量和高频分量，找出负荷各频率分量的规律；通过模糊灰色关联聚类方法选取待预测日的负荷相似日；针对不同频段负荷的规律采用相对应的神经网络模型进行负荷预测，获得不同频段的待预测日负荷各分量，将各分量的预测结果叠加得到负荷预测值。采用所提方法对某地区 2010 年实际负荷进行预测，并与已有的负荷预测方法比较，结果表明所提方法可提高负荷预测的精度。 |
| 关键词: 小波分析 模糊灰色关联聚类 神经网络 负荷预测 有效性指标 预测 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Short-term load forecasting based on wavelet decomposition，fuzzy gray correlation clustering and BP neural network |
|
ZHANG Ping, PAN Xueping, XUE Wenchao
|
|
College of Energy and Electrical Engineering，Hohai University，Nanjing 210098，China
|
| Abstract: |
| A short-term load forecasting method is proposed based on the wavelet decomposition，fuzzy gray correlation clustering and BP neural network. The wavelet decomposition is applied to decompose the load series into the low-frequency and high-frequency components to find the law of each load component. The fuzzy gray correlation clustering is applied to select the days with the load similar to the day to be forecasted. Corresponding neural network model is used to forecast the load for each component. The forecasted load is the superposition of all forecasted component loads. The proposed method is applied to forecast the load of an actual region for three days in 2010 and the results are compared with those by other existing forecasting methods，which shows that the forecasting accuracy is increased. |
| Key words: wavelet analysis fuzzy gray correlation clustering neural networks load forecasting efficiency index forecasting |
