一种改进的RBF神经网络在农机总动力预测中的应用

RBF Neural Network Learning Method Based on System Clustering

径向基神经网络是一种局部逼近型神经网络,具有结构简单、训练速度快等优点。为此,分析了传统径向基神经网络采用k-means聚类方法确定网络相关参数中存在的问题,提出了一种基于系统聚类确定基函数中心的方法,利用系统聚类逐个减少类别个数的原理,解决了确定基函数中心时对初始点敏感的问题,同时提出用一个和类间距相关的参数作为聚类终止的条件,确定了基函数的数量。用改进的方法对农机总动力进行预测,结果表明:改进方法在确定网络结构和提高网络性能方面有较好的效果。

Radial basis neural network is a kind of local approximation neural network,which has the advantages of simple structure and fast training speed. This paper analyzes the existing problems in the network-related parameters by k-means clustering method,and proposes a method based on the clustering of the system to determine the basis of the center function. The principle of system clustering is used to reduce the number of categories,which solves the problem of sensitive to the initial point when the center of the basis function is determined. At the same time,it is proposed to determine the number of basis functions by using the condition of a class-related parameter as the condition of clustering termination. The experimental results show that the improved method has a good effect in determining the network structure and improving the network performance.