Publications
Case studies, papers and articles
Articles
New Workflows Reduce Forecast Cycle Time, Refine Uncertainty
Technology update, JPT July 2006, Information provided by David Millar, Scandpower Petroleum Technology.
Revolutionising History Matching and Uncertainty Assessment
Jens-Petter Nørgård, MEPO, Scandpower Petroleum Technology AS, Geo Expro, Sept 2006.
Optimization technology reduces uncertainty in reservoir models
David Millar and Kjartan Nesse, Scandpower Petroleum Technology, Petroleum Africa, Feb 2006.
Assisted History Matching gives better production forecasts.
David Millar, Scandpower Petroleum Technology, Oil Review Middle East, Nov 2005.
Reducing the cycle time for production forecast
David Millar and Stig Selberg, Scandpower Petroleum Technology White paper, Nov 2005.
Case studies
Enhancing Field Management in Siberia by Quantifying Production Uncertainties
J. David E. Tipping, SPE, Maxim N. Deschenya, TNK-BP, Franz Deimbacher, SPE, & Dmitry Kovyazin, Schlumberger.SPE 101808, SPE Russian Oil and Gas Technical Conference and Exhibition, Moscow, Russia, October 3-6 2006
This paper presents a history matching and uncertainty quantification study for a field in Western Siberia. Within a period of a week 7 different history-matched models each yielding a unique production profile prediction were obtained with MEPO. It was shown that the traditional manual approach would take about 9 man-months to find a single matched model. From the results of the study it was shown that the assisted history matching approach can give major boosts to work productivity. It was also demonstrated that field management decisions should account for uncertainty.
Integration of geologic and dynamic models for History Matching , Medusa Field
J. Lach, K. McMillen and R. Archer, Knowledge Reservoir L.P; , J. Holland and R. DePauw, Murphy E&P Co;, and B. E. Ludvigsen, Scandpower PT."SPE - 95930, Oct 2005
MEPO was used to successfully find multiple history matches in a quick, two week study. Over 1000 simulations were run to evaluate a range of reservoir uncertainties. The T4B reservoir in the Medusa field is a stratigraphically confined syncline. History matching has confirmed the reservoir connectivity across faulting and channel-levee-splay facies boundaries. The forecasts of future production showed a small spread in recovery. However, the oil, gas and water rates were variable which can be significant for development planning in a deepwater setting.
Streamline-based History Matching with Application of Global Optimization Techniques,
R.W. Schulze-Riegert, A. Diab, O. Haase DGMK Spring conference, Celle, Germany 29-30 April 2004.
Case study showing the application of global optimization techniques to full-field streamline simulation model of a heavy-oil field with 675,000 cells, 150 wells and 37 yrs of production. Successfully demonstrates how MEPO and 3DSL can be combined to improve the understanding of large complex fields, and to quantify the uncertainties in production forecasts.
Technical Papers
Modern Techniques for History Matching
Ralf Schulze-Riegert Shawket Ghedan, 9th INTERNATIONAL FORUM ON RESERVOIR SIMULATION December 9 – 13, 2007 Abu Dhabi, United Arab Emirates
This paper presents an introduction to concepts and trends in modern History Matching. Special attention is paid to stochastic optimisation techniques. Uncertainties in reservoir data are summarized and provide the motivation for introducing new workflows in probabilistic forecasting. A distributed computing framework is described which facilitates deploying algorithms for an increasing number of complex simulation problems which require solutions in a brief time.
A Software Component Based Parallel Simulation and Optimisation Environment for Reservoir Simulation, M. Krosche, J.K. Axmann, O. Pajonk, R.W. Schulze-Riegert, O. Haase, DGMK-Tagungsbericht 2005-1, ISBN 3-936418-35-7
This paper presents the technology behind MEPO and the development of next generation parallel optimization tool. Recent research on reservoir simulation has concentrated on alternative optimization methods in addition to gradient type techniques. Because of the long simulation times, parallel computing has become more important, in order to use the availability of cost efficient computing resources effectively. This technology favors optimization techniques that are scalable using parallel computing capabilities. Component based software engineering concepts are applied to develop an open, scalable and extensible system concept.
Combined Global and Local Optimization Techniques Applied to History Matching SPE 79668, R.W. Schulze-Riegert, O. Haase, A. Nekrassov. This paper was prepared for presentation at the SPE Reservoir Simulation Symposium held in Houston, Texas, U.S.A., 3-5 February 2003.
Demonstrates that a combination of global methods such as Evolution Strategy together with a Bayesian approach gives improved convergence when identifying history-matches. Illustrated with two case examples from Germany.
Consistent Parameter Changes in Global Optimization Applied to History Matching
R.W. Schulze-Riegert, O. Haase, B.E. Ludvigsen, A. Nekrassov, Oil Recover Conference, Moscow, Russia, 21-23 May 2003.
Different optimization methods and statistical techniques are applied to a complex real 3-phase reservoir in the North Sea, and a method for consistent update of inter-related model parameters is described. Discusses the introduction of a prior objective term and its effect on stochastic optimization methods.
Evolutionary Algorithms Applied to History Matching of Complex Reservoirs, R.W. Schulze-Riegert, J.K. Axmann, O. Haase, D.T. Rian, Y.-L. You.,SPE Reservoir Evaluation & Engineering (Apr. 2002)
Investigation of the application of optimization methods to the problem of history matching, with a discussion of the advantages of parallel computing for optimization. Illustrated with a North Sea test reservoir. This is an abbreviated version of SPE66393.
Optimization Methods for History Matching of Complex Reservoirs (SPE66393), R.W. Schulze-Riegert, J.K. Axmann, O. Haase, D.T. Rian, Y.-L. You, SPE Reservoir Simulation Symposium, Houston, Texas, 11-14 Feb 2001)
Investigation of the application of optimization methods to the problem of history matching, with a discussion of the advantages of parallel computing for optimization. Illustrated with a North Sea test reservoir. This is the full-length version of the work presented in SPE77301.
Other recent publications on Assisted History Matching and Uncertainty Assessment
Uncertainty Assessment Using Experimental Design: Minagish Oolite Reservoir (SPE91820), W.T.Peake, M. Abadah, and L.Skander [Chevron & Kuwait Oil Company] SPE Reservoir Simulation Symposium, Houston, 31Jan-2 Feb 2005
Describes Chevron and Kuwait Oil Company's experiences using Experimental Design techniques to establish the minimum number of simulation runs required to quantify uncertainty in a large carbonate field. Also describes use of variance and multiple linear regressions to identify the most significant uncertainties, and Monte Carlo simulations to develop P10/P50/P90 forecasts. Although MEPO was not used in this work, the results have been independently verified and reproduced by Scandpower Petroleum Technology using MEPO.
Step Change in Reservoir Simulation Breathes Life into a Mature Oil Field (SPE94940), M.J. Kromah, J.Liou and D.G.MacDonald [BP] SPE LAPEC Conference, Rio de Janeiro, Brazil, 20-23 June 2005.
More complete description of the BP case example from Teak field, where TDRM and AHM methods were used to assess the uncertainties associated with three infill well candidates, and to select the candidate with the highest probability of successfully adding reserves.
Top-Down Reservoir Modelling (SPE89974)
G.J.J. Williams, M. Mansfield, D.G. MacDonald, M.D. Bush [BP] SPE ATCE, Houston, Texas, 29-31 Sept 2004.
Describes BP's proprietary TDRM modeling approach, which has been applied to 18 oil and gas reservoirs and BP claims resulted in an overall 20% increase in NPV for these projects. The TDRM approach includes the use of assisted history matching methods and optimization methods, using the Genetic Algorithm approach, which is also available within MEPO. Illustrated with two example from the Azeri field, Azerbaijan, and Teak field, Trinidad.
A Methodology for History Matching and the Assessment of Uncertainties Associated with Flow Prediction (SPE84465) J.L. Landa and B.Güyagüler [Chevron] SPE ATCE, Denver, 5-8 October 2003.
Describes a framework for the assessment of uncertainties associated with prediction, using an approach based on response surfaces and sensitivity coefficients.
Quantifying Uncertainty in Production Forecasts: Another Look at the PUNQ-S3 Problem (SPE74707) J.W. Barker, M.Cuypers, L. Holden [Total and Norwegian Computing Centre] SPE ATCE, Dallas, Texas, 1-4 Oct 2000.
Comparison of various techniques for quantifying uncertainty, applied to a synthetic reservoir. Demonstrates that relatively small changes in the prior model can have a large impact on the difficulty of finding a history match and on predicted performance.