基于混合训练神经网络的武器系统效能评估  

A Method of Evaluating Effectiveness of Military System Based on Combined-training Neutral Network

作　　者：陈有伟[1] 李为民[1] 

机构地区：[1]空军工程大学导弹学院,陕西三原713800

出　　处：《计算机工程与应用》2003年第8期95-97,115,共4页Computer Engineering and Applications

基　　金：高等学校骨干教师资助计划基金项目(编号:GG-110-90039-1004)

摘　　要：提出了一种将遗传算法和神经网络相结合的混合训练神经网络模型,分析了其基本思想方法、具体实现过程,以及在武器效能评估中的具体应用。实例分析表明,该模型用于武器系统效能评估可较好地克服人为因素和模糊随机性的影响,使评估结果更为合理可信。The paper presents a combined-training neutral network model by combining neutral network with genetic algorithm,analyzes its basic ideal and method and its actual course of realizing,then applies it in evaluating effective-ness of military system.Through analyzing with example,the results show that using this model in evaluating effectiveness of military system can greatly surmount the influence of artificial factor and fuzzy randomness,make evaluating result more reasonable and credible.

关 键 词：遗传算法 BP神经网络 武器系统效能评估 

分 类 号：TP389.1[自动化与计算机技术—计算机系统结构]