Technical health, remaining useful life and life extension of subsea equipment

Abstract

This PhD thesis proposes concepts, tools and processes for analyses of technical health and remaining useful life (RUL) for decision-making regarding life extension of subsea equipment. The thesis is the outcome of the Integrated Operations (IO) programme, hosted at the Norwegian University of Science and Technology, Trondheim. The mission of the programme is to develop new methods and tools, which can be embedded in improved work processes in the oil companies and enhanced products and services from the suppliers. Program 3, subsection 3.1, of the IO, has the mandate to create a generic framework or a toolbox for remaining useful life prediction. This work clarifies the meaning of the useful life and the RUL of equipment. It then discusses the terms useful life and the life termination criteria. It explains the weaknesses of using the mean remaining life and the ‘technical health index’ for determination of the RUL. It defines the concept of technical health, presents the factors that influence the technical health and shows the relation between technical health and RUL. This work provides a process to determine a technical condition vector by bringing in the idea of multi-parameter monitoring. It presents technical health as a process of assessment of the elements of the technical condition vector, resulting in the formulation of a technical health picture. The work clarifies the terms tacit knowledge and explicit knowledge in the context of technical health. It explains that the technical health is a tacit knowledge and that all the factors influencing RUL are in tacit form. The knowledge needs to be converted into explicit form. The work explains the meaning of community of practice (CoP), and how this can be used to achieve the knowledge conversion. The work presents various failure modes and mechanisms that are commonly observed in the subsea industry and recommends suitable modelling approaches based on several references from top-side (oil and gas), nuclear and avionics industries. It evaluates the suitability of Bayesian Belief Network (BBN) as a modelling tool for technical health assessment in the subsea oil and gas industry. The work explains how to utilize expert judgment, knowledge conversion and knowledge sharing for technical health assessment and for model selection. A new approach, combining elements from storytelling and root cause analysis, to address knowledge conversion and knowledge sharing is presented. A flexible RUL prediction process is suggested that is adaptable to both objective (classical, frequentist) and subjective (Bayesian) interpretations of probability depending on the availability of data and suggestions from experts. Condition based maintenance (CBM) for RUL prediction is not yet a practice in the subsea oil and gas industry because the real time data acquisition process is not established. The PhD thesis also describes the work done by various standardization groups to ensure interoperability among sensors and interfaces i.e., making them smart. In particular, it informs on how these divergent standards complement each other and how they can work together to get real time operational data. The main contributions of the thesis are: (1) Clarification of the terms technical condition and technical health, (2) Explanation of the relationship between technical condition, technical health, RUL and decision-making related life extension, and (3) Presentation of a step-wise process for determination of technical health and RUL that can be used in the subsea oil and gas industry.