Kohonen neural network and wavelet transform based approach to short-term load forecasting

摘要: Abstract This paper presents Kohonen neural network and wavelet transform (WT) based technique for short-time load forecasting of power systems. Firstly, the historical seasonal data are classified into four patterns using then Daubechies D2, D4 D10 WTs adopted in order to forecast hourly load. The coefficients associated with certain frequency time localisation adjusted conventional multiple regression (MR) method components reconstructed predict final loads through five-scale synthesis technique. outcome study clearly indicates that proposed composite model WT approach can be used as an attractive effective means short-term forecasting.