基于自适应Elman神经网络的短期风电功率预测

Short-term wind power prediction based on adaptive Elman neural network

针对BP神经网络动态性能的不足、适应性较差的问题,提出了基于自适应Elman神经网络的短期风电功率预测模型。通过对比不同隐含层数的Elman预测模型的预测误差,选取最小误差的隐含层数作为自适应Elman预测模型的隐含层数;根据不同的训练集和预测集的输入,自动调节Elman隐含层节点数,实现隐含层节点数的自适应,寻求具有最佳隐含层节点数的预测模型,提高了风电功率预测精度。

With an aim at the problems of shortage of the BP neural network dynamic performance and low-adaptability,a short-term wind power prediction model based on the adaptive Elman neural network is proposed in this paper.Through the comparison of prediction errors in the number of different hidden layers of Elman prediction model,the number of the minimum error hidden layers is selected as the number of hidden layers in the adaptive Elman prediction model.In accordance with the inputs of different training set and prediction set,the number of Elman hidden layer nodes can be automatically regulated so as to realize the adaptation of number of hidden layer nodes and to seek for the prediction model with the best number of hidden layer nodes,where-by improving the prediction accuracy of wind power.