| Petrochemical industry is the pillar industry of economic development all over the world.Petrochemical safety equipment is the key device to protect the safe operation of equipment.Once the safety equipment fails,it cannot guarantee the safe operation of the device,resulting in leakage and other problems.The consequences are unimaginable.The lifetime data related to safety equipment is the key data to help equipment engineers grasp the reliability status of safety equipment and specify the maintenance plan rationally.Therefore,the reliability and maintainability RM data related to plant safety equipment and its operation are the basic data for large data applications such as lifetime parameter estimation.Whether the life parameters are accurate or not will directly affect the engineers’ judgment of the reliability status of safety equipment,which will easily lead to the absence of equipment maintenance strategy and affect the mechanical integrity of safety equipment.Therefore,collecting reliability and maintainability data of safety equipment on site and using reliability analysis method to analyze lifetime data can reflect the real reliability status of safety equipment in factories.It is of great significance to prevent and reduce equipment failure and reduce the frequency of fire and explosion accidents.In view of the above problems,the main research contents of this paper are as follows:(1)Firstly,based on ISO14224-2016 standard,according to the operation characteristics of petrochemical industry,this paper designs the data collection process of reliability and maintainability of safety equipment in petrochemical industry.Then,the specific implementation process of collecting reliability and maintenance data of safety equipment in petrochemical industry is introduced in detail.According to the current situation of censored data,the identification and optimization design of key collection items of reliability and maintainability data are implemented.Finally,taking the relief valve as an example,the RM data collection table is compiled,and the data content that each part needs to collect is presented in detail.(2)To compensate for the inaccuracy of the median method in evaluating the life time parameters of interval censored data and right censored data,a quantile filling algorithm is introduced to preprocess the censored data.By setting failure detection strategy,this paper defines the way to discover the failure of relief valve,and distinguishes the interval censored data and the right censored data by using symbolic markers.Then,the censored data is transformed into virtual complete data with the help of quantile probability P,and the maximum likelihood estimation is incorporated into the iteration process to realize the combination of the quantile filling algorithm and the maximum likelihood estimation method.(3)Taking the emergency relief valve of the wax oil hydrogenation unit as an example,a numerical example is given to evaluate the performance of the quantile filling algorithm.In this paper,we use MATLAB tools to obtain mixed censored data of small samples including interval censored data and right censored data,and use Weibull++ reliability analysis software as an auxiliary tool to illustrate the advantages of this algorithm by setting relevant comparison content.The results of the corresponding case studies show that the algorithm in this paper shows better stability and accuracy when dealing with small sample data;the algorithm has good stability in small sample data and can converge after finite iterations;the quantile filling algorithm has higher accuracy than the median algorithm,and the algorithm will not be affected by single type of data.The accuracy of life distribution parameters obtained by this algorithm is 8.1% higher than that by the median method.This helps to formulate effective maintenance strategies to ensure the mechanical integrity of safety equipment and the safe operation of petrochemical plants. |
