Citation

Peng X, Wang H, Liu J, Yu X. China Saf. Sci. J. 2019; 29(2): 94-99.

Copyright

(Copyright © 2019, China Occupational Safety and Health Association, Publisher Gai Xue bao)

DOI

10.16265/j.cnki.issn1003-3033.2019.02.016

PMID

unavailable

Abstract

In order to compensate for the shortcomings of the mid-point method in evaluating the lifetime distribution parameters of interval censored data, and make it easy to use in engineering practice, a quantile-filling algorithm was worked out, which can convert the interval censored data into virtual complete data, and the obtained virtual complete data will be used in MLE to obtain parameter estimation of the probability distribution of equipment lifetime. In the case of the emergency relief valve of the wax oil hydrogenation unit, the use and performance of the algorithm were verified by comparing the estimation results of the 50 - 100 data volume. The results show that the algorithm has good stability under small sample, that it can converge within 5 times of iteration, the accuracy of life distribution parameters obtained by this algorithm is 7. 5% higher than that by the mid-point method, which helps to develop effective maintenance strategy to ensure the mechanical integrity of the safety equipment, and then ensure the safe operation of the petrochemical plant. © 2019 China Safety Science Journal. All rights reserved.

Language: zh