基于动态最近邻聚类算法的RBF神经网络及其在MH-Ni电池容量预测中的应用

RBF Neural Networks Based on Dynamic Nearest Neighbor-Clustering Algorithm and Its Application in Prediction of MH-Ni Battery Capacity

基于RBF神经网络的设计难点提出了一种动态确定隐含层节点数及数据中心的新方法,即动态最近邻聚类算法,消除了现有算法中人为因素对数据中心的影响。所设计的神经网络具有最少的隐含层节点数,结构简单,提高了网络学习训练速度,基于动态RBF神经网络建立了MH-Ni电池容量预测模型,通过仿真,取得了理想的结果,为MH-Ni电池容量预测提供了新方法。

The method of controlling the RBFNN data centers of the hidden layer is raised in this article based on the feature of the RBFNN. It is the dynamic nearest neighbor-Clustering Algorithm. This algorithm eliminates the factitious factor affecting the choice of data centers in existing algorithms. The RBFNN has the least nodes and high studying speed. A prediction model of MH-Ni battery is cited and simulated with the dynamic RBFNN. The ideal result is made, and a novel method is presented for prediction of MH-Ni battery capacity.