基于最小维修的石化安全设备区间截尾寿命数据分析

Interval censored life data of petrochemical safety equipment based on minimal maintenance

在石化安全关键设备寿命截尾数据的可靠性评估过程中,为避免传统可靠性评估方法未考虑维修对故障间隔时间的影响,以及中值填充法误差较大的缺点,本文提出了一种逆变换区间截尾数据填充方法。首先,在可修系统基于最小维修的假设下,通过非齐次泊松过程描述系统的故障趋势,利用逆变换填充得到监测区间的缺失数据;随后根据蒙特卡罗期望最大化算法,获得填充数据的参数估计值。通过蜡油加氢装置紧急泄压阀的案例分析,该算法能够很好地融合已有数据,得到的形状参数估计值偏差与中值填充法相比缩小4%以上,从而验证了该算法的有效性;这有助于安全关键设备的定量化完整性管理,保障石化装置平稳运行。

For the censored life data of critical petrochemical safety equipment,the traditional reliability assessment methods do not consider the effect of maintenance on time between failures.In order to avoid this shortcoming and the large error of median padding algorithm,an inverse transform interval censored data filling algorithm was proposed.Firstly,based on the assumption that the repairable system was in the minimal maintenance,the non-homogeneous Poisson process to describe the system’s failure trend was used,and the missing data of the monitoring interval using the inverse transformation filling intensity function were obtained.Then,according to the Monte Carlo expectation maximization algorithm,the parameter estimates of the filled data were obtained.The case analysis of the emergency pressure relief valve of the wax oil hydrogenation device showed that the algorithm can well integrate the existing data.Comparing with the median padding algorithm,this calculated deviation of the shape parameter estimates was reduced more than 4%,which proved the effectiveness of this algorithm.It helped the quantitative integrity management of the critical safety equipment and guaranteed the smooth operation of the petrochemical equipment.