摘要
针对单独运用BP网络评估舰船装备保障资源保障能力时存在的不足,提出了一种基于改进BP网络的舰船装备保障资源保障能力评估模型。首先构建了保障能力评估的指标体系,在此基础上,采用主成分分析法对原始输入变量进行预处理,选择输入变量的主成分作为神经网络输入,一方面减少了输入变量的维数,消除了各输入变量的相关性;另一方面提高了网络的收敛性和稳定性,同时也简化了网络的结构。然后将遗传优化算法与BP网络结合,利用GA的全局搜索能力优化BP网络的结构参数。最后进行实例计算,结果表明该方法有效克服了纯BP算法局部收敛、泛化能力弱等问题,新模型采用的算法具有收敛速度快,结果精度高的优点,适用于对保障资源保障能力的分析和评估。
Because of the deficiency of support ability evaluation of ship equipment supportability resources by singly using BP neural network,a new assessment model is proposed which is based on an optimized BP neural network.The index system of evaluating support ability of ship equipment supportability resources was brought forward,on the base,inducting PCA to pre-analyze the original multi-objective variables and using the principal components of original variables as the input of network can cut down the dimensions of input variables and eliminate the relativity between input variables,and also improve the convergence speed and stability of network and simplifie network structure.Then bycombine Genetic Algorithm with BP neural network,using GA＇s global search capability the BP network structure parameters are optimized.Finallyt the experiment results show that new algorithm can overcome the local convergence,feebleness of generalization ability and other issues of BP algorithm effectively.The model works with quicker convergence rate and the higher precision,and is suitabl for support ability analysis and evaluation of supportability resources.
出处
《火力与指挥控制》
CSCD
北大核心
2012年第2期65-67,71,共4页
Fire Control & Command Control
基金
总装技术基础课题基金资助项目(4314231428)
关键词
保障资源
保障能力评估
遗传算法
BP网络
主成分分析
supportability resources
support ability evaluation
genetic algorithms
BP network
principal component analysis
