作者: Qi Di Zhao , Yang Yu , Meng Meng Jia
DOI: 10.4028/WWW.SCIENTIFIC.NET/AMM.662.259
关键词: Engineering 、 Control theory 、 Artificial intelligence 、 Convergence (routing) 、 Wind speed 、 Artificial neural network 、 Wind power generation 、 Ant colony optimization algorithms 、 Backpropagation 、 Term (time)
摘要: To improve the short-term wind speed forecasting accuracy of farms, a prediction model based on back propagation (BP) neural network combining ant colony algorithm is built to predict speed. The input variables BP predictive are historical speeds, temperature, and air pressure. Ant used optimize weights bias networks. Using optimization future 1h speed, simulation results show that proposed method offers advantages high precision fast convergence in contrast with network.