Abstract: To address the problem that the lifetime data of petrochemical safety devices belong to the typical multiple timing censored leading to the difficulty of reliability estimate of the devices,a reliability estimation method of Filled Least Squares Method (FLSM) combined with improved Bootstrap method was proposed.A representative Weibull distribution was fitted and a point estimate of the equipment reliability was obtained using the FLSM.The sample size was expanded using the improved Bootstrap method to obtain the confidence lower bound of the reliability.As a result of the Monte Carlo simulation,the proposed FLSM outperformed the traditional method in terms of accuracy and applicability,and had excellent robustness.The application to the emergency de-pressure valves of the hydrocracker,the equipment maintenance time was determined to be 12,929 hours,and the confidence lower bound obtained by the improved Bootstrap method suggested an equipment maintenance time of 6,969 hours to ensure the smooth operation of petrochemical safety equipment.

Key words: petrochemical safety devices, multiple timing censored data, Monte Carlo simulation, parameter estimation, reliability

摘要： 针对多重定时截尾类型的石化安全设备寿命数据和由此导致的可靠度评估挑战,提出一种填充最小二乘法结合改进Bootstrap法的可靠度估计方法。采用代表性的Weibull分布拟合,并利用填充最小二乘法获得设备可靠度的点估计,再利用改进Bootstrap法扩大样本量,得到可靠度的置信下限。通过蒙特卡洛模拟实验结果表明,所提出方法在精确性和适用性上均优于传统的方法且具有优良的稳健性。最后,以加氢裂化装置中紧急泄压阀为例进行数值计算,确定设备维护时间为12 929 h,改进Bootstrap法获得的置信下限估计为6 969 h,确保石化安全设备平稳运行。

关键词: 石化安全设备, 多重定时截尾数据, 蒙特卡洛模拟, 参数估计, 可靠度