Abstract
The wish to better know the condition of equipment and thus be able to plan and perform maintenance before a critical failure occurs has existed for many years. Early leak detection of liquid and gas from topside process piping is a critical task for economical and safety reasons. For an operator, there is a need to accurately detect, and localize leaks to timely recover them cost-effectively, or the most viable solution to be able to prevent leaks to happen. The latter named prognostic maintenance, condition-based maintenance or predictive maintenance represents a well-known method where the goal is to ensure preventive measures are taken in due time where leak may happen in a piping system.
There is a rapid development in sensor technologies, algorithm development, data storage systems, data transfer, data processing and computing power. The robustness and reliability of electronic components is constantly improving while the cost is decreasing. To ‘unhide’ hidden failures through condition knowledge and thus reduce inspection and unnecessary repairs and testing efforts on critical (e.g. safety) equipment is a compelling need for the Energy sector today. Digitalization is currently boosting this technology both within model-based prognostics, data-driven prognostics, and hybrid solutions. To make use of this development, classical understanding of fatigue, corrosion and moisture penetration on piping systems must be combined with several IT-related disciplines. The paper will discuss the latest state-of-the art related to sensor monitoring systems with the aim to provide sound recommendations related to fatigue, corrosion and moisture penetration on piping system for topside process piping.
