论文部分内容阅读
针对小样本条件下系统可靠性评估的特点,以及传统Bayes方法在进行可靠性数据分析时存在的不足,提出了一种基于改进Bayes-Bootstrap的系统可靠性仿真评估方法。首先,对基于Bayes方法进行系统可靠性评估时易发生的现场试验数据“湮灭”问题进行了分析;其次,针对传统Bayes-Bootstrap方法的局限性,提出了一种基于经验函数修正的Bayes-Bootstrap数据抽样算法;最后,结合算例对方法的有效性进行了验证。结果表明:该仿真评估算法具有较高的评估精度,有效降低了传统Bayes方法进行可靠性评估的主观性和处理过程的复杂性。
In view of the characteristics of system reliability evaluation under small sample conditions and the shortcomings of traditional Bayesian method in reliability data analysis, a system reliability simulation evaluation method based on improved Bayes-Bootstrap is proposed. First of all, the field test data “annihilation” problem which is easy to occur when evaluating the system reliability based on the Bayes method is analyzed. Secondly, in view of the limitation of the traditional Bayes-Bootstrap method, a modified Bayesian method based on the empirical function -Bootstrap data sampling algorithm; Finally, the effectiveness of the method is verified with the example. The results show that the simulation evaluation algorithm has high evaluation accuracy and effectively reduces the subjectivity of the reliability evaluation of the traditional Bayesian method and the complexity of the processing.