Estimating the Value of Information Using Closed Loop Reservoir Management of Capacitance Resistive Models

Abstract

The optimal goal of reservoir management is to achieve the highest possible recovery factor for the lowest possible cost. Changes in injection control have the potential to cause significant gain in net present value. In order to set the injection control that gives the highest possible net present value, it is important to have detailed reservoir description. Gathering information is expensive and it is difficult to determine its worth.

A value-of-information (VOI) analysis can be helpful to identify when and if information should be gathered. This is a powerful method that can be used to identify and exclude alternatives in a decision context.

%This project uses a previous proposed methodology for estimating VOI.

One of the two purposes in this thesis, is to evaluate the usability of a closed loop reservoir management (CLRM) structure using capacitance resistive models (CRM) to estimate VOI.

A CLRM structure for CRM was implemented in MATLAB. This structure used the ensemble Kalman filter (EnKF) for history matching and the ensemble optimization (EnOpt) for optimizing injection.

The EnKF successfully identified the reservoir flow pattern in reservoir models, whilst the EnOpt successfully increased the objective value after new information became available.

The CLRM with the CRMs were shown to be suitable to estimate VOI, although the slow computer speed limits the usability for more complex reservoir models.

The other purpose of this thesis is to provide information on the CRM, oil fractional flow models, the EnKF, the EnOpt and the CLRM, to ease the way for future research students.