AGR is delighted to have supported DNV GL
in a six-month research project into investigating the potential of machine learning in the Norwegian Continental Shelf (NCS). The report was launched today at the OG21 Forum in Oslo
The report, commissioned by OG21 (The Research Council of Norway) and with the title ‘OG21 – Study on Machine Learning in the Norwegian Petroleum Industry
’, found that machine learning has the potential to significantly reduce oil and gas well delivery time and green-house gas emissions. Machine learning will also bring down operational spending and accelerate production. Its’ success, however, will depend on whether it is deployed at scale on the NCS and if barriers to adoption, such as machine learning literary and competency, can be overcome.
The study consisted of four Technology Target Areas (TTA’s):
TTA1 – Energy efficiency and Environment
TTA2 – Exploration and increased recovery
TTA3 – Drilling completions and intervention
TTA4 – Production, processing and transport.
AGR was responsible for delivering their expert opinion within exploration and increased recovery (TTA2). Our experts conducted interviews of subject matter expertise in major operating and technology companies in Norway and the US.
AGR found that machine learning has the potential to improve exploration efficiency in seismic processing and interpretation and also in the big data analysis of legacy wells. In terms of increased recovery, the technology is immature but may be important in the analysis of big data in new fields in the future. AGR finds that machine learning developments have the potential to open up new prospective trends, new oil & gas discoveries, improve infill targets, lead to less dry wells, and support enhanced reservoir management.
, CEO of AGR
(pictured), commented: “As the digital transformation of the energy industry continues, it’s clear that machine learning and its broader use of data will have a major role to play in exploration and in seismic interpretation, geological models, well log interpretation and increased oil recovery – to name but a few areas.”
He continues: “We are therefore proud to have worked with DNV GL on this vital project – reaffirming the importance of industry collaboration highlighted in DNV GL’s recommendations. Machine learning can increase volumes, reduce costs, lighten environmental footprints and ensure the continued sustainability of the Norwegian petroleum industry for many years to come.”
Machine learning is an application of artificial intelligence that combines informatics, mathematics and calculation-oriented statistics and provides computer systems with the ability to learn, adapt and improve from empirical data without being explicitly programmed.
The report from DNV GL includes recommendations for how machine learning should be developed, adopted and scaled more quickly through initiatives and joint industry collaboration in areas such as improved machine learning literacy, the sharing of tools and data, and ensuring mechanisms for trusted validation of machine learning solutions. The report highlights the capabilities the Norwegian petroleum industry will need in order to successfully deploy machine learning to add value.
‘OG21 – Study on Machine Learning in the Norwegian petroleum industry’
is available to download here
For more information about the project, please contact:
Ole-Gunnar Tveiten, Advisor Geology, AGR, firstname.lastname@example.org
, +47 95103134
Hans Petter Ellingsen, Principal Consultant, Well & Asset Risk, DNV GL, Hans.Petter.Ellingsen@dnvgl.com