An open-source approach to Integrated Operations
MetadataVis full innførsel
This thesis is based on exploring the topic of Integrated Operations (IO) and software development in the oil and gas industry. The philosophy of IO and how it impacted the oil and gas industry in the last decade is investigated. The term was introduced in the early 2000s, and builds on an old idea; allow engineers to collaborate in cross-disciplinary teams, and enable them to access all data, independent of their geographical location. IO is a philosophy to solve the problems of tomorrow concerning the governance, work processes, technology, and people. The investigation show that engineers and data have become increasingly intertwined as rising amount of data have become available, and the communication between offshore and on-shore facilities have significantly improved. Through the last decade, companies of all industries have boosted their attention on information acquisition and data management. This increase in focus for information technology is often explained with the term digitalization. The idea of digitalization and Integrated Operations share multiple aspects; IO could be interpreted as a holistic approach which includes digitalization as the enabling dimension. The implementation of Integrated Operations has seen a number successes and a comparable number of failures. Challenges associated with the deployment of IO related projects are often caused by the lack of adequate technological solutions.Innovation is the driving force behind change; it improves efficiency and increases the abilities of engineers. In many domains of the oil and gas industry innovation frequently happen, in other areas they do not. Reservoir engineering is an example of the latter. Typical reservoir engineering workflows involve simulating fluid behaviour in a reservoir and Eclipse is a tool. The simulator has been a workhorse of the industry for more than two decades, and have enabled large-scale field development and an increase in value realization. Schlumberger uses a proprietary model to distribute Eclipse, a secure system to manage, but might also hamper innovation. This thesis treats the topics of software development models and distribution licenses. An alternative approach to the proprietary development model is to use an open-source model, which is much harder to sustain but stimulates innovation. The Open Porous Media (OPM) Initiative is currently developing an alternative open-source reservoir simulator named Flow. Flow offers unique opportunities and may offer Eclipse competition in the future. In a benchmark study conducted on the Norne full-scale reservoir model, the results show that the open-source alternative, Flow, delivers both accurate and consistent results compared to the reference simulator, Eclipse. The performance shows that in a single-core environment, Eclipse slightly outperforms Flow concerning speed, while the opposite applies to a dual-core environment. These results give an indication of potential that may materialise when an open-source framework and mindset is applied to a domain previously dominated by a proprietary software solution. The structure of open-source software development increases the pace at which features and the core capabilities are developed. Open-source software is rarely built from scratch, rather, it is most commonly built on other projects with the same development model. This structure creates co-dependencies between multiple projects and allows the scope of one project to be maintained at the core functionality. Flow applies a modular toolbox for solving partial differential equations (PDEs) with grid-based methods, called DUNE (DUNE, 2017). DUNE is an independent project that continuously evolves and optimizes its capabilities, which through a co-dependency contributes to a better framework for Flow and improved run-times for simulations. Among the biggest challenges in an open-source development model is to establish a sustainable business model and create a reputation as a reliable alternative. In a project like OPM, it is vital to find partners to expand, develop and verify. This involves a great deal of risk for OPM, but not necessarily for those who would partake in such an endeavour. OPM offers a unique opportunity for oil and gas companies to secure their ability to operate. Giving greater leverage towards existing suppliers and avoid a vendor lock-in situation. As of today, OPM does not deliver all the tools that are needed for a field development. It does, however, give an insight into what the future may bring regarding software. An open-source alternative allows for full transparency regarding how the software operates and thereby could increase the trust by the engineers that utilize the software. This increased trust could in the future increase the amount of automation in typical workflows and transition engineers from doing iterative tasks to value adding activities. A fundamental part of the open-source software is its openness towards modifications. This would allow an operator to modify the simulator to fit the needs of individual assets. Also, support for custom scripts and third-party programs could have native support and deliver additional information and increase the functionality.