论文部分内容阅读
针对现代空袭兵器的特点,建立地空导弹人机环系统的作战效能评估指标体系,并采用BP神经网络方法对其作战效能进行分析评估。该神经网络输入层节点分别对应评估指标体系最底层的指标,隐层节点数在训练过程中可视网络的泛化能力调整,输出节点数则分别对应于效能评估等级。神经元输入与输出之间转移函数选取S型函数。评估步骤包括:训练样本的选取、样本的归一化处理、神经网络的训练、示例分析。
According to the characteristics of modern air raid weapon, this paper establishes the evaluation index system of operational effectiveness of surface-to-air missile man-machine loop system, and uses BP neural network method to analyze and evaluate its operational effectiveness. The input nodes of the neural network correspond to the lowest index of the evaluation index system and the generalization ability of the visual network in the hidden layer during training. The output nodes correspond to the performance evaluation levels respectively. Neuron input and output transfer function between the S-function selection. Evaluation steps include: training sample selection, sample normalization, neural network training, sample analysis.