论文部分内容阅读
为了验证平均秩次法在估计完全删失数据分布函数参数时的有效性,设计了一种明晰平均秩次法适用范围的蒙特卡罗仿真方法。该方法考虑现场故障数据的实际生成过程,以机群为研究对象,通过飞机架数、飞行天数、日利用率动态变化驱动故障样本、删失比变化,获取故障间隔时间的完全删失数据。结合拉依达准则,确定了平均秩次法的计算错误率。仿真结果表明,样本量在9.1,删失比为0.328时,错误率达到28.9%,此条件下,平均秩次法已经不适用。因此,在航空工程实际的可靠性评估中,由于大量高可靠性产品的出现,平均秩次法的适用范围变窄。
In order to verify the validity of the average rank method in estimating the parameters of the complete censored data distribution, a Monte Carlo simulation method is proposed to clarify the scope of the average rank method. The method takes into account the actual process of generating fault data in the field and takes the fleet as the research object. Through the dynamic changes of the number of flight planes, the days of flight and the daily utilization rate, the fault samples are driven and the censored variation is obtained to obtain the complete censored data of the fault interval. Combined with the Raida criterion, the calculation error rate of the average rank method was determined. The simulation results show that when the sample size is 9.1 and the censoring ratio is 0.328, the error rate reaches 28.9%. Under this condition, the average rank method is not applicable. Therefore, in the actual reliability assessment of aeronautical engineering, the application of the average rank method is narrowed due to the appearance of a large number of high-reliability products.