正态分布定时无失效数据可靠性分析方法

Method of reliability analysis for time truncated zero-failure data based on normal distribution

定时无失效是指产品试验到预定的时间未发生失效的情况,定时无失效数据的处理是当前工程中亟需解决的问题.为此,提出一种正态分布定时无失效数据可靠性分析方法,该方法能够根据定时无失效数据对产品进行高置信水平的可靠性评定.分别针对标准差或标准差上限已知的情况进行了详细讨论,给出了可靠度和使用寿命的单侧置信下限.并结合某型号卫星推力器脉冲点火试验得到的定时无失效数据,进行了分析计算,结果表明本方法既可以充分利用定时无失效数据,又能最大化发挥对数正态分布置信下限较高的优点,且计算简单,便于工程应用.

The term＇ time truncated zero-failure＇means the product under experiment does not fail at the end of the experiment, and this is a pressing problem in engineering to solve problems related to time truncated zero-failure data. In this paper, a method of reliability analysis for time truncated zero-failure data based on normal distribution was presented, providing reliability assessment of a product with high confidence using the information from time truncated zero-failure data. The one-side lower confidence limits of liability and life were calculated in this paper when the standard deviation or its higher limit was known. Then, time truncated time zero-failure data from some satellite thrust tests with normal distribution, were use of b the analyzed. The results indicate that the method in this paper can make full information from these data and the higher one-side lower confidence limits of lognormal distribution. The method provided in this paper can be easily realized for the benefit of engineering application.