DOI: 10.1016/J.ENERGY.2008.09.020
关键词:
摘要: Short-term load forecast (STLF) is a key issue for operation of both regulated power systems and electricity markets. In spite all performed research in this area, there still an essential need more accurate robust methods. paper, new hybrid method proposed purpose, composed wavelet transform (WT), neural network (NN) evolutionary algorithm (EA). Hourly time series usually consists global smooth trends sharp local variations, i.e. low- high-frequency components. WT can efficiently decompose the into its Each component predicted by combination NN EA then inverse hourly obtained. The examined on three practical compared with some most recent STLF