论文部分内容阅读
分布参数估计是可靠性数据分析的手段,用于研究产品可靠性的变化规律及评估产品的可靠性水平。使用寿命试验中产品可靠度所服从的β分布评价可靠度估计值的合理程度,认为可靠度估计值在β分布中概率密度较大的更合理,给出了一种评价可靠度函数估计好坏的β似然函数,探讨了使用该函数进行分布参数估计的β似然估计方法,通过仿真验证了该方法在指数分布和威布尔分布下的适用性,并给出了应用实例。本参数估计方法理论依据充分,适用于各种分布类型,估计结果合理可信。极大似然估计方法以样本在待估分布中的概率密度作为评价准则,与之不同,本方法以累积发生比例的估计值作为评价准则,更适用于可靠性、生存率等关注事件累积发生比例的场合。
Distributed parameter estimation is the means of reliability data analysis, used to study the variation of product reliability and evaluate the reliability of products. In the life test, the β distribution obeying the product reliability is used to evaluate the reasonableness of the reliability estimation. It is considered that the reliability estimation value is more reasonable in the β distribution with a larger probability density, and an evaluation reliability function is good or bad The β likelihood function of this method is used to estimate the β parameters of distributed parameters. The simulation results show the applicability of this method under the exponential distribution and Weibull distribution, and the application examples are given. This parameter estimation method is based on sufficient theory, suitable for all kinds of distribution types, the estimation result is reasonable and credible. In contrast, the maximum likelihood estimation method uses the probability density of samples in the distribution to be estimated as the evaluation criteria. In contrast, this method uses the estimated value of cumulative occurrence ratio as the evaluation criterion, and is more suitable for the cumulative occurrence of concerns such as reliability and survival rate Proportion of the occasion.