A joint project involving researchers from the University of Aberdeen is using artificial intelligence (AI) to unlock the "treasure trove" of raw data collected by the oil and gas industry in order to maximise economic recovery.
The Intelligent Data Quality Improver (IDQI) project is an initiative involving the University and co-funders the Oil & Gas Innovation Centre (OGIC), software R&D company HyperDAP Ltd, and The Data Lab – Scotland’s innovation centre for data and AI.
The project aims to develop algorithms that will bridge the gap between the vast amounts of big data available to the oil and gas industry - predominantly in exploration and production - and its ability to analyse, interpret and manage it effectively.
Professor Wamberto Vasconcelos, from the University’s Department of Computing Science, said: “Data is potentially among the most valuable assets created and owned by a business, but without proper interpretation it has very little value.
“The oil and gas industry alone produces 2.5 quintillion bytes of data each day, but only 1 per cent of this data is analysed, which is a missed opportunity in terms of asset value maximisation and new field discovery.
“The IDQI aims to address this issue, using Distributed Optical Sensing data to develop algorithms capable of performing automated analyses on digital exploration and production datasets.
The oil and gas industry alone produces 2.5 quintillion bytes of data each day, but only 1 per cent of this data is analysed, which is a missed opportunity in terms of asset value maximisation and new field discovery" Professor Wamberto Vasconcelos
“This means we can extract and interpret most of the hidden information in a matter of minutes using, among others, a range of AI techniques such as machine learning, fuzzy logic and rule-based reasoning. This has the potential to unlock a vast treasure trove of data that is not currently exploited.”
Nicoletta Compatangelo, Managing Director of HyperDAP Ltd, added: “There is a substantial amount of context-related information hidden inside datasets, which adds value well beyond what data themselves convey on the surface.
“Once extracted and made explicit through data analysis and interpretation, this information can be formally measured against the quality of those datasets – the more information, the higher the corresponding dataset quality.
“At present, very little analysis and interpretation is performed in several specialist areas of oil and gas exploration and production. For example, distributed optical sensing data from wells is constantly acquired but almost invariably stored without in-depth analysis, as it can take weeks for a human expert to analyse one single day of data records.
“IDQI has the potential to revolutionise the way that this raw data is interpreted, providing greater speed, accuracy, and ultimately providing insights that can be of real benefit to operators."
Author: Robert Turbyne