CAJ | 학술논문

针对当前测试性评估方法中,经典方法无法利用历史测试性试验信息,且在小样本量下,评估结论置信度低、风险大的问题,提出了一种基于研制阶段试验数据的复杂装备测试性评估模型。在对经典评估方法进行建模与分析的基础上,运用Bayes理论,建立了综合利用研制阶段历史试验信息和现场试验数据的Bayes测试性评估模型;该模型结合验前信息与现场信息的相容性给出了一种混合验前分布,并利用拟合优度检验确定继承因子。最后开展了案例应用研究,结果表明,在相同的现场试验条件下,该模型能给出较高置信度的测试性评估结论,比经典评估方法更合理。
침대당전측시성평고방법중,경전방법무법이용역사측시성시험신식,차재소양본량하,평고결론치신도저、풍험대적문제,제출료일충기우연제계단시험수거적복잡장비측시성평고모형。재대경전평고방법진행건모여분석적기출상,운용Bayes이론,건립료종합이용연제계단역사시험신식화현장시험수거적Bayes측시성평고모형;해모형결합험전신식여현장신식적상용성급출료일충혼합험전분포,병이용의합우도검험학정계승인자。최후개전료안례응용연구,결과표명,재상동적현장시험조건하,해모형능급출교고치신도적측시성평고결론,비경전평고방법경합리。
In view of the problems that the classical statistic method can not make use of the historical test information and produce the evaluation conclusion with low confidence level and high risk under the con- dition of small sample, a new testability evaluation method based on test data in development stages was pro- posed herein. Firstly, after modeling and analysis of the classical evaluation method, a Bayes evaluation mod- el for testability of complex equipment based on fusing the historical samples and the current samples were es- tablished by using Bayes theory. The paper presented a new mixed Beta prior distribution after introducing the concept of prior data compatibility, and determined the inheritance factor by using goodness of fit between the historical test informations and the current informations. Finally, the fault detection rate of an example was validated by means of such a model. The results show that this method can produce the evaluation con- clusion with high confidence level in the same test condition and show that this method is more rational than the classical statistical evaluation method.