FCTM Esope

S2.14- Life-Cycle Decision Optimization: Managing Plant-Wide Thickness Data for Fixed-Equipment and Piping Subject to Corrosion

7 oct. 2021 | 15:00 - 15:30

Studio Photo

Description

In the pressure vessels and piping community, internal corrosion of all sorts (i.e. both general corrosion mechanisms such as CO2 corrosion and local corrosion mechanisms such as microbiologically induced corrosion) continue to wreak havoc on fixed-equipment (FE) and piping and provide ongoing struggles for inspection planning and risk management. What type of corrosion is lurking in the circuit, what is its expected coverage area, is the piping system properly circuitized, when should inspections take place, where should inspection efforts be focused, what NDT inspection techniques should be used, and when gathering inspection data, how should it be managed to optimize the life-cycle of the plant? All of these questions are addressed by implementing a proper life-cycle management (LCM) program. Preparation of such a program can be overwhelming and may require a diverse team of engineering experts with extensive field experience. However, with recent advances in technology, such an effort doesn’t need to be as daunting as it seems. Artificial Intelligence Bayesian Decision Networks (AI-BDN) are probabilistic computational algorithms that utilize multiple, often disparate, sources of knowledge to make more informed day-to-day decisions. Compilation of AI-BDNs, for life-cycle decision optimization in industrial applications, requires cloud-based HPC methods and algorithms. Incorporating such a tool into daily operations promotes pro-active decision-making and risk management to minimize the likelihood of a potentially catastrophic, unexpected failure event. The outcome is complete plant awareness (i.e. improved confidence throughout the facility; knowing that the site has a grasp on the corrosion that is occurring in each system). The concepts demonstrated herein are directly applicable to other failure modes (e.g. creep, fatigue, fracture, et cetera) throughout all aging energy industries and may be used to properly manage plant-wide FE/piping subject to damage.

Présentée par