Kjersti Solberg Eikrem

Optimisation, data assimilation, seismic inversion, wind energy, batteries.

The object of the SURF project is to develop efficient history-matching methods for an improved understanding of the properties of the subsurface reservoirs. Improvements in subsurface understanding are critical for efficient reduction in energy usage, reduction in CO2 emissions, and improved recovery of resources.

The process of adjusting the parameters in large reservoir models to simultaneously match data and quantify uncertainty in forecasts is difficult and costly. In this project, methods will be developed to ensure that the history matching methods are adapted to the complex types of uncertainty observed in real field models where multiple geologic scenarios are often used to quantify uncertainty. The project will ensure that the properties of the model/data relationship are accounted for in the selection of the iteration step size through developments in an analogous problem of adaptive time-stepping for the numerical solution of stochastic differential equations. In most applications, not all history-matched models are equally good at predicting future behavior. The weighting of models for the predictability of decision-targeted forecasts will be an outcome of the project. The methods developed will be used to evaluate the potential for the reduction of CO2 emissions resulting from the production operation of a field on the Norwegian Continental Shelf.

Acknowledgements

This is the official webpage of the project SURF (Subsurface Understanding for Robust emissions Forecasting), funded by Equinor, TotalEnergies, and Petrobras.

Resources