Institute of GeoEnergy Engineering, School of Energy, Geoscience, Infrastructure and Society
AutoWell:Automated Well Monitoring and Control, 2022-2025 PETROMAKS2 project supported by Research Council of Norway, ConocoPhillips, Sumitomo, Wintershall Dea and Aker BP. R&D partners: NORCE, University of Stavanger (UiS) and Heriot-Watt University (HWU)
GMUQ: Uncertainty quantification for the simulation of reservoir geomechanics and flow using machine learning, JIP 2021-2025, Dr Dan Arnold (PI), Dr Helen Lewis (co-PI), Prof V. Demyanov (co-PI), Dr Saeed Ghanbari (co-PI)
Rock Flow Dynamics research and academic support, 2019-2023 Rock Flow Dynamics supports GeoDataScience group research, individual MSc research project and other academic activities wider across the Institute of GeoEnergy Engineering by providing educational licenses for tNavigator Integrated Solutions for reservoir simulation and geomodelling.
Robust uncertainty quantification for green hydrocarbon production in carbonate reservoirs, 2019-2023 Qatar National Research Fund
Improve Local History Match Quality with geologically realistic model update based on AI ensemble data analytics, 2020-2021
Optimisation of mature reservoir development under geological uncertainty using machine learning and distributed HPC, January 2018 – September 2019.
The review of predictive methodologies to estimate reservoirs characteristics by applying Machine learning techniques, Oct 2019 – Feb 2020
Feasibility, Ensemble Based method for History Matching – phase III, September 2017 – September 2018
Uncertainty Quantification JIP, 2002 – 2016 founded and led by Prof Mike Christie, phases I-IV, 1999 – 2016, sponsored by 9 companies. The JIP investigated the best ways to history match reservoir models and from those history matches produce accurate estimates of risk and uncertainty in predictions of future oil production/reserves. Phase I: Uncertainty and Upscaling was a Joint Industry Project which started in March 2002. It contains two separate but linked modules – one on uncertainty and one on upscaling. The aim of the uncertainty module of the uncertainty and upscaling project is to develop tools for practical engineering use that will enable practical deployment of uncertainty tools. The specific goals of the project were: (i) to develop practical methods for generating multiple history matched reservoir models; (ii) to develop ways of incorporating prior geological data in uncertain history matches and forward predictions; (iii) to generate tools and methods for predicting reservoir performance uncertainty; (iv) to develop methods for quantifying the value of additional data. Sponsors: Anadarko, BG plc, BP, ConocoPhillips, DTI, JNOC, Landmark, Shell, Statoil. Phase II: The approach developed was to construct multiple models matching known reservoir performance data (rates, pressures etc) which are also consistent with prior geological beliefs. Because our knowledge of the reservoir properties is from a limited set of sample points, there will always be multiple reservoir models consistent with our knowledge, and our reservoir model predictions will be uncertain. Sponsors: BG plc, BP, Chevron, ConocoPhillips, DTI, ENI, JNOC, Norsk Hydro, Shell, Transform Phase III: The phase III research advanced in developing new methods for geologically realistic model update using machine learning to integrate prior geological knowledge. Sponsors: BG plc, BP, ConocoPhillips, JOGMEC Phase IV: The focus of the Phase IV research shifted to optimisation under geological uncertainty and more intensive use of data driven machine learning methods to describe realistic geological uncertainty. Sponsors: BG plc, E.ON, RFD