摘要
针对弹药试验数据样本量少时无法直接采用经典统计学方法的问题,介绍了弹药贮存可靠度评估的传统方法和现有自助扩容方法,并针对其不足提出了改进自助扩容算法,增加了自助样本量信息、保证了抽样样本与原始样本特征的一致性、降低了异常值的影响,并对所提算法利用Matlab软件仿真验证.结果表明,所提算法能有效进行贮存可靠度评估,具有显著优越性.该结论可用于弹药小样本数据贮存可靠度评估,为弹药贮存管理提供理论支撑.
Aiming at the issue which cannot use classical statistical method directly while ammo test data has less samples, this paper introduces the traditional method and existing Bootstrap enlargement method to the estimation of ammo storage reliability. And it comes up with the modified algorithm for the disadvantages of these methods, which increases the information of auto samples, ensures the feature consistency between sampling samples and original samples, reduces the impact of ＂abnormal value＂. Moreover, it uses Matlab software to simulating verify the proposed algorithm. The result indicates that the proposed algorithm can effectively proceed the estimation of storage reliability and it has remarkable superiority. The conclusion can be applied to the estimation of ammo storage reliability for small sample data and it provides the theoretical support for the management of ammo storage.
作者
陈佳欣
李青
郭红霞
CHEN Jia-xin;LI Qing;GUO Hong-xia(Graduate School,Engineering University of PAP,Xi＇an 710086,China)
出处
《数学的实践与认识》
北大核心
2018年第21期127-133,共7页
Mathematics in Practice and Theory
基金
陕西省自然科学基金(2018JQ6019)
博士后创新人才支持计划基金(BX201700104)
关键词
弹药
自助扩容
贮存可靠度
仿真验证
ammo
bootstrap enlargement
storage reliability
simulation verify
