• CN:11-2187/TH
  • ISSN:0577-6686

›› 2009, Vol. 45 ›› Issue (8): 52-57.

• 论文 • 上一篇    下一篇

扫码分享

利用研制阶段试验数据制定测试性验证试验方案新方法

李天梅;邱静;刘冠军   

  1. 国防科技大学机电工程与自动化学院
  • 发布日期:2009-08-15

New Methodology for Determining Testability Integrated Test Scheme with Test Data in the Development Stages

LI Tianmei; QIU Jing; LIU Guanjun   

  1. College of Mechatronics Engineering and Automation, National University of Defense Technology
  • Published:2009-08-15

摘要: 现有的测试性验证试验方案,在一定的风险承受能力下都需要较大的故障样本量。针对这一问题,以故障检测率(Fault detection rate,FDR)为验证指标,提出利用研制阶段试验数据制定测试性验证试验方案的新方法。在传统的FDR的统计检验模型基础上,提出产品研制阶段的成败型试验数据的等效折合方法;考虑Beta分布函数的概率特性,将FDR的置信下限估计转化为Beta分布的置信度分位点,求得FDR的置信概率密度函数;引入差异因子 ,用来反映研制阶段产品的FDR水平和产品验证需达到的FDR水平的差异,求得产品验证需达到的FDR的置信概率密度函数;参考风险的定义,重新确定了测试性验证试验方案。由于充分考虑了研制阶段的试验信息,因此在保证有较好的验证效果的条件下,与传统的验证方案相比,可以明显减少故障样本量,或在故障样本量保持不变的情况下降低双方风险。

关键词: Beta分布, 测试性验证试验, 差异因子, 风险, 故障样本量, 置信概率函数

Abstract: For reducing the amount of fault sample size during the testability demonstration test with specified risk level, and by taking FDR as the target, a new method is presented to make full use of the testability test data acquired in the development stage. Firstly, the conventional statistics test model is given, and then, a method is put forward for processing the data acquired in the development stage. Using the probability of Beta function, the confidence lower limit of FDR is transferred into lower percentile of Beta distribution. A coefficient is introduced to reflect the difference between FDR acquired by development stage data and the required FDR level, finally, the integrated test scheme is determined by the required FDR confidence level distribution function considering the risk level. The integrated test scheme can reduce the necessary fault sample size and then reduce the risk level.

Key words: Beta distribution, Confidence probability function, Difference factor, Fault sample size, Risk level, Testability demonstration test

中图分类号: