采用径向基神经网络的卫星网络申报趋势分析方法

将径向基函数(RBF)神经网络应用于卫星网络申报趋势分析,构建基于RBF神经网络的趋势量化分析方法,改变当前主要依赖专家经验分析申报趋势的现状,为卫星网络申报趋势的评估提供量化指标.首先,梳理当前卫星网络申报的业务特点;然后,对主流预测方法进行分析,提出基于RBF神经网络的申报趋势分析方法;最后,通过实际申报数据进行算法验证.结果表明:文中方法对卫星网络申报趋势的预测误差总体小于20%,对实际申报工作具有指导意义.

Analysis Method of Satellite Network Declaration Trend Using Radial Basis Function Neural Network

Radial basis function(RBF)neural network is innovatively applied to the trend analysis of satellite network declaration. A quantitative trend analysis method based on RBF neural network is proposed, which provides quantitative indicators and methods for evaluating the trend of satellite network declaration and changes current situation that mainly depends on expert experience to analyze declaration trend. Firstly, the business characteristics of current satellite network declaration are combed. Then, the widely used prediction methods are analyzed, thus trend analysis method of declaration is proposed based on RBF neural network. Finally, the algorithm by actual declaration data is verified. The results show that prediction error of the proposed method is less than 20% for the trend of satellite network declaration, which has practical guiding significance.