Short-Term Load Forecasting Using Immune Wavelet Network
WU Hong-xiao,HOU Zhi-jian(School of Electric Information and Electrical Engineering,Shanghai Jiaotong University,Shanghai 200030,China)
The IWN is presented to overcome the weakness of BP neural network(BPNN) for short-term load forecasting.In the design of the IWN,a novel algorithm based on the symbiotic evolutionary and the immune programming method(SEIPM) is proposed. The SEIPM can automatically adjust the number and parameters of hidden neurons. The proposed IWN model has been implemented using actual data and compared with the traditional BPNN method.The forecasting results demonstrate that the IWN method possesses superior forecast precision than the BPNN method does.
【CateGory Index】： TM715