• 9 May 2022:
  2nd GeoDataScience and UQ group research dissemination on-line open day.
  The programme for the day will offer a series of presentations across various aspects of machine learning applications to reservoir uncertainty modelling workflows: seismic interpretation, geological pattern recognition from outcrops, fluvial facies modelling with GANs, dynamic data integration into fracture reservoir modelling and multi-objective optimisation of reservoir production to tackle CO2 emission targets.
  In the afternoon we will present a new HWU JIP initiative on Uncertainty quantification of geomechanically sensitive reservoirs and welcome interested companies to join.

• Presentations:
  • Seismic auto-segmentation with point cloud clustering evaluation & optimisation, ​Quentin Corlay
  • Learning geological patterns from outcrops by using computer vision methods, Athos Nathanail
  • A GAN-based Workflow for 3D Fluvial Facies Modelling​, Chao Sun
  • Integrating geological uncertainty and dynamic data into modelling procedures for fractured reservoirs, Bastian Steffens, PhD overview
  • Well Grouping and Control Optimisation for CO2 Emission Offset in Field Production, Amirsaman Rezaeyan​
  • New JIP: GMUQ – Uncertainty quantification of geomechanically sensitive reservoirs​
    • Use of Artificial Intelligence to Quantify ​Geomechanical Uncertainties ​Induced by Fluid Injection in Subsurface Systems, Bertrand Cuesta
    • Physics-based Machine Learning for Geomechanical Modelling, Farah Rabie

• February 2022:
  “GeoScience Meets DataScience” – Researcher Links workshop to be hosted by Heriot-Watt, sponsored by the British Council.
• February 2021:
  GeoDataScience group Open on-line research dissemination day for industry and academia, over 70 participance from dozens of companies:
  • Turbidite fan interpretation in 3D seismic data by point cloud segmentation using Machine Learning, by Quentin Corlay
  • Machine Learning for sedimentary structure classification, by Athos Nathanail
  • Modeling variations of complex geological concepts with Generative Adversarial Network (GAN) learning from process modelling , by Chao Sun
  • How Generative Networks can ​help improve geological history matching, by Gleb Shishaev
  • A workflow with dynamic screening assisted, automated fractured reservoir modelling, by Bastian Steffens
  • Can agents model hydrocarbon migration? , by Bastian Steffens & Quentin Corlay