基于决策级数据融合的可靠性综合验证方法

可靠性验证试验是检验产品的可靠性水平是否达到要求而进行的试验,是一项重要的可靠性活动.价格昂贵或者可靠性水平较高的设备,按照经典的抽样方案进行试验,时间和费用几乎是不能承受的.为了解决上述实际问题,针对成功率的抽样检验方案,利用证据理论方法,研究了基于研制试验数据、可靠性评估结果和专家信息的基本可信度分配函数的构建方法,并给出了基于决策级数据融合的可靠性综合验证试验模型.同时,对可靠性综合验证模型的性质进行了讨论,确定了可靠性综合验证试验方案的存在性,并提出了方案的选取原则.模型可以广泛地利用研制信息,并可以统一给出基于不同分布类型和不同信息的可靠性综合验证试验方案.通过示例表明该方法可以利用各种研制信息,达到减少可靠性验证试验样本的目的.

Reliability integrated compliance testing plans based on decision making level data fusion

The reliability compliance test is the test inspected reliability level against the requirements, and is the important reliability activities. It is a formidable task for some expensive or high reliability products according to classic reliability compliance testing plans to demonstrate reliability. Based on these problems, the basic probability assignment was built using varieties of development information, including development data, reliability assessment result and expert experience. And the reliability integrated compliance testing model was given for success ratio applying the evidence theory. In the meantime, the character of this reliability integrated compliance testing model was researched. The existence of the integrated compliance test plan was evidenced, and the choosing principle was discussed. This model can avail diversiform information, and the integrated compliance test plan serving difference distribution and difference information can derive form this model. The target to reduce the samples by the use of multifarious development information was illustrated by instance.