A novel method of short-term load forecasting based on wavelet coefficients and BP neural network is proposed in this paper. The method of forecasting of load sequences has been replaced by the method of forecasting of wavelet coefficients. The wavelet coefficients on different scales are forecasted by BP neural networks respectively after wavelet detail coefficients have been dealt with by layer soft threshold. The new method combining wavelet coefficients with BP neural network is introduced in detail in this paper and the example about the method is given as well.
Published in:
Power and Energy Engineering Conference, 2009. APPEEC 2009. Asia-Pacific
Date of Conference: 27-31 March 2009