Core flooding experiments are always needed to be conducted before the application of Enhanced Oil Production (EOR) in the field, which is a great way to improve the oil recovery. In order to optimize existing resources and obtain the information efficiently, it is necessary to optimize these experiments. As the core flooding experiment are subjected to many uncertainties, the sensitivity of model parameters, in terms of the expected information gain in Bayesian setting, are explored. However, the estimation of such gain, relies on a double-loop integration. Moreover, its numerical integration in multi-dimensional cases is computationally too expensive for core-flooding model. Based on [Long etc. 2013], Laplace approximation for the integration of the posterior pdf was then used to accelerate the estimation of the expected information gains in the model parameters.

 

Longting Mo is a visiting Master student from Nanjing University, China in our group. He has a research interest in uncertainty quantification of groundwater science. He has been working on the optimal information extraction from core flooding experiments during his time at KAUST.

https://b-stochasticnumerics.kaust.edu.sa/pages/people-org/detail/longting-mo