Group Maintenance Optimization Of Subsea Xmas Trees With Stochastic Dependency

Dongming Fan,Aibo Zhang,Qiang Feng,Baoping Cai,Yiliu Liu,Yi Ren

DOI: https://doi.org/10.1016/j.ress.2021.107450

IF: 7.247

2021-01-01

Reliability Engineering & System Safety

Abstract:Subsea Xmas trees (XTs) are vital equipment for offshore oil and gas development. Due to a long and continuous operation, components of XTs often become vulnerable subjected to degradation and unexpected failures. Due to the uncertainties of subsea operation and fault tolerance design, current maintenances on heterogeneous components, which are assumed to be independent of each other, perform separately. Only one PM mode (imperfect or perfect) is considered. However, these assumptions impede the application of state-of-the-art research results on the maintenance of this equipment. Therefore, for XTs with stochastic dependency, this study proposes a group maintenance optimization approach that combines maintenance activities to reduce maintenance costs. Reduction factors are introduced to measure the effects of various preventive maintenance (PM) actions, and the optimal component-level PM intervals can be obtained. An improved group strategy can be explored in consideration of stochastic dependency and opportunity maintenance. Utilizing the collaborative particle swarm optimization (CPSO) algorithm, the cost of an optimal group maintenance plan can be minimized while maintaining the availability. The uses and advantages of the proposed group maintenance approach are illustrated by a case study on a Horizon Xmas tree with a 14-component system.