用于风电功率预测的RPCL优化神经网络模型

为了提高风电功率预测的准确度,提出了一种基于对手竞争惩罚学习算法( rival penalized competitive learning,RPCL)优化径向基函数( radial basis function,RBF)神经网络的风电功率预测模型。首先通过RPCL确定网络隐含层神经元数目以及中心点初始值,然后由K均值聚类法确定隐含层神经元的中心点和宽度,最后通过最小均值算法确定隐含层神经元与输出层神经元之间的权值。仿真结果表明：此优化模型相较于传统RBF网络具有更高的准确性。

Rival Penalized Competitive Learning-based Neural Network Model for Wind Power Forecasting

For increasing the accuracy of wind power forecasting, a rival penalized competitive learning-based radial basis function ( RBF) neural network model was presented. Firstly the number of neural network hidden-layer-nodes and its initial center values were determined by rival penalized competitive learning. And then the width of RBF and the center values of network were identified accurately through K-means clustering. At last, appropriate weights of network were estimated by least mean square. The forecasting result shows that the presented model can lead to more accurate forecasting compared with the traditional neural network.