Kristian Fossum

Data assimilation and machine learning with a focus on reservoir applications

Background

The Data Assimilation & Optimization Group at NORCE-Energy is one of the world’s foremost research communities on data assimilation (DA), the AI-adjacent discipline of optimally combining simulation models (digital twins) and observational data for improved predictions and decisions. The Group has developed digital methods for reservoir management for over 25 years. These methods have led to a paradigm shift in the petroleum industry worldwide and are currently in commercial use. Recently, we published our state-of-the-art methods in open source on GitHub. The methodology is also used globally in oceanography and meteorology and has potential far beyond today’s use for many other operational systems.

Focus

  • Uncertainty quantification
  • Decision support
  • Combination of physical models and data-driven models
  • Rapid prototyping
  • Improved understanding

Problems

  • Big hybrid models and big data
  • Operational systems
  • Handling uncertainties in both models and data

Goal

Join an AI center to improve uncertainty quantification and explainability for a variety of applied operational systems, such as:

  • Healthcare diagnostics
  • Enhancing autonomous vehicle safety
  • Optimizing agricultural yields
  • Increasing energy efficiency
  • Advancing manufacturing processes

Contact

The vision for Data Assimilation in the AI Era