Predicting Valve Failures in the Digital Age of Valve Management

Abstract

Valves are critical and essential parts in an oil & gas (O&G) production plant, which means that if a failure occurs it can potentially great implications on safety and profitability of the plant. The traditional approach in today's oil and gas process industry for verifying valve reliability is simplistic and provides little information regarding valve degradation mechanisms. Therefore, this paper discusses new theories and concepts to help increase the reliability of production & safety critical valves. Focus has historically been targeting on using corrective and scheduled maintenance strategies, rather than using condition and performance monitoring applications as a tool to plan and prioritize maintenance and achieve a higher standard for safety. In the search for a solution to increase the reliability of production and safety critical valves it is reasonable also to look at industry 4.0 and its concepts. The digital revolution which implements concepts such as cyber-physical systems (CPS), industrial internet of things (IIoT) and cloud-based storage. Based on these concepts the paper visits the use of valve diagnostic systems for identifying failure mechanisms of valves subjected to degradation by the application of sensors and small computers used to continuously monitor valve performance and store this data within cloud-based servers. Today we experience an enormous progress in real time monitoring by use of sensors in conjunction with data collectors. There is new technology to process and gather data that supports decision making, such as Valve Diagnostic System (VDS). By taking advantage of new technology there are immense benefits to be had by monitoring valves rather than traditional physical observation and testing. This method has multiple advantages and it can contribute to reducing maintenance cost, down time and increase safety. Multiple unwanted and hazardous events have occurred through the last decades, such as the tragic Piper Alpha disaster (Macleod & Richardson, 2018). This has contributed to drive the industry to improve and adhere to safety regulations and requirements. This makes the sensor technology even more beneficial and opens a wide range of potentials within the maintenance- and risk management genres. Today all in-operation risk calculations are based on reported accidents or unwanted occurrences. This practice gives way to inaccurate reliability calculations that are indisputable and inaccurate. It is a practice where potentially many failures are not reported, understood or even revealed. Experts within Functional Safety Management points to the basis of the calculations with skepticism due to the validity of the data and the fact that the data in itself is not a proactive approach to sound valve management.