Full text loading...
-
Channelized Reservoirs Characterization Using Ensemble Smoother with Selective Measurement Data
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 76th EAGE Conference and Exhibition 2014, Jun 2014, Volume 2014, p.1 - 5
Abstract
Reservoir characterization is one of the most important works for predicting future performances and making development plan. Ensemble Kalman filter (EnKF) uses recursive update whenever observed data are obtained. Ensemble smoother (ES) does not require multiple updates because it assimilates all available data at a time. ES has advantage on simulation time over EnKF, but it sometimes gives instable results on history matching. In this paper, we propose the concept of selective use of measurement data for ES. We have compared two cases. One case uses all data which are oil production and water cut from 20 to 900 days. The other case uses data selectively such as only oil production before water breakthrough and water cut after water breakthrough. For 2D channelized reservoirs, ES with all data case shows overshooting and filter divergence problems. However, ES with selective data case controls the two problems and characterizes main channel distribution properly. It keeps bimodal distribution of the channel fields better than EnKF. Furthermore, ES with selective data provides proper estimation of reservoir performances with uncertainty and enables us to make a rational decision. It requires low simulation cost which is only 2.2% of that of EnKF for 45 times updates.