Bayes Linear Strategies for Matching Hydrocarbon Reservoir History
- 9 May 1996
- book chapter
- Published by Oxford University Press (OUP)
Abstract
In the oil industry, efficient management and prediction of hydrocarbon production depends crucially on having a good model of the reservoir. Approximate numerical solutions to the model are obtained using a “reservoir simulator”, which takes as input physical descriptions of the reservoir. It is important, therefore, to find settings of the input geology which result in a simulator run with outputs which match as closely as possible to the corresponding reservoir history. This process is termed history matching. Running the simulator with a given input geology may be very time-consuming. A large number of runs is therefore impractical. The natural way to approach the problem is to use the detailed prior knowledge of geologists and reservoir engineers to direct our search. This suggests some kind of Bayesian analysis. A full Bayesian approach is extremely difficult, due to the high dimensionality of the problem. Therefore, we discuss a Bayes linear approach for history matching, which formally incorporates the beliefs of the reservoir engineer, but which only requires a limited specification of aspects of second order beliefs. While we focus on history matching, the methodology that we describe is appropriate in general applications involving sequential design of computer experiments to solve high dimensional inverse problems.This publication has 0 references indexed in Scilit: