时间序列与神经网络法相结合的短期风速预测

Short-Term Wind Speed Forecasting Combing Time Series and Neural Network Method

利用时间序列-神经网络法研究了短期风速预测。该方法用时间序列模型来选择神经网络的输入变量,选用多层反向传播(back propagation,BP)神经网络和广义回归神经网络(generalized regression neural network,GRNN)分别对采样时间间隔为10min、20min和30min的风速序列进行预测。结果表明,时间序列结合GRNN的方法精度更高,具有一定的实用价值。

By use of time series and neural network the short-term wind speed forecasting is researched in which the time series model is used to select the input variables and multi-layer back propagation neural network and generalized regression neural network are used to conduct forecasting. The wind speed series are forecasted by sampling time interval of 10min, 20min and 30min respectively. Forecasting results show that the method integrating time series with generalized regression neural network possesses higher accuracy and is of a certain practical value.