Application of Artificial Neural Networks to Downhole Fluid Analysis
- Peter S. Hegeman (Schlumberger) | Chengli Dong (Schlumberger) | Nikos Varotsis (Tech. U. of Crete) | Vassilis Gaganis (Tech. U. of Crete)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- February 2009
- Document Type
- Journal Paper
- 8 - 13
- 2009. Society of Petroleum Engineers
- 1 in the last 30 days
- 1,328 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
Reservoir characterization and asset management require comprehensive information about formation fluids. Obtaining this information at all stages of the exploration and development cycle is essential for field planning and operation. Traditionally, fluid information has been obtained by capturing samples and then measuring the pressure/volume/temperature (PVT) properties in a laboratory. More recently, downhole fluid analysis (DFA) during formation testing has provided real-time fluid information. However, the extreme conditions of the downhole environment limit the DFA-tool measurements to only a small subset of the fluid properties provided by a laboratory. Nevertheless, these tools are valuable in predicting other PVT properties from the measured data. These predictions can be used in real time to optimize the sampling program, to help evaluate completion decisions, and to understand flow-assurance issues.
The petroleum industry has devoted much effort to developing computational methods to model phase behavior. Two approaches are prevalent—simple correlations and equation-of-state (EOS) models. However, in recent years, artificial-neural-network (ANN) technology has been applied successfully to many petroleum-engineering problems, including the prediction of PVT behavior. ANN technology can recognize patterns in data, adjust dynamically to changes, infer general rules from specific cases, and accept a large number of input variables. An ANN architecture can allow for continuous improvement by expanding the training database with new data.
In this paper, we present the application of ANN technology to DFA. We demonstrate this with an ANN model that uses the DFA-tool measurements of fluid composition as input and produces predictions of gas/oil ratio (GOR), a key PVT property used in real time to monitor a formation-tester sampling job. The ANN also provides an uncertainty estimation of its outputs as a quality-assurance indicator. We compare ANN results with those from the algorithms used by DFA tools.
|File Size||465 KB||Number of Pages||6|
Al-Kaabi, A.U. and Lee, W.J. 1993. Using Artificial Neural Networks ToIdentify the Well Test Interpretation Model. SPE FormEval8 (3): 233-240; Trans, AIME, 295.SPE-20332-PA. DOI: 10.2118/20332-PA.
Al-Marhoun, M.A. and Osman, E.A. 2002. Using Artificial Neural Networks toDevelop New PVT Correlations for Saudi Crude Oils. Paper SPE 78592presented at the Abu Dhabi International Petroleum Exhibition and Conference,Abu Dhabi, UAE, 13-16 October. DOI: 10.2118/78592-MS.
Ayoub, M.A., Raja, D.M., and Al-Marhoun, M.A. 2007. Evaluation of Below Bubble PointViscosity Correlations and Construction of a New Neural Network Model.Paper SPE 108439 presented at the Asia Pacific Oil and Gas Conference andExhibition, Jakarta, 30 October-1 November. DOI: 10.2118/108439-MS.
Betancourt, S.S., Fujisawa, G., Mullins, O.C., Eriksen, K.O., Dong, C., Pop,J., and Carnegie, A. 2004. Exploration Applications of DownholeMeasurement of Crude Oil Composition and Fluorescence. Paper SPE 87011presented at the SPE Asia Pacific Conference on Integrated Modelling for AssetManagement, Kuala Lumpur, 29-30 March. DOI: 10.2118/87011-MS.
Dashevskiy, D., Dubinsky, V., and Macpherson, J.D. 1999. Application of Neural Networks forPredictive Control in Drilling Dynamics. Paper SPE 56442 presented at theSPE Annual Technical Conference and Exhibition, Houston, 3-6 October. DOI:10.2118/56442-MS.
Del Campo, C., Dong, C., Vasques, R., Hegeman, P., and Yamate, T. 2006. Advancesin Fluid Sampling With Formation Testers for Offshore Exploration. PaperOTC 18201 presented at the Offshore Technology Conference, Houston, 1-4May.
Dong, C., Hegeman, P.S., Carnegie, A., and Elshahawi, H. 2006. Downhole Measurement of Methane andGOR Content in Formation Fluid Samples. SPE Res Eval &Eng9 (1): 7-14. SPE-81481-PA. DOI: 10.2118/81481-PA.
Dong, C., O'Keefe, M., Elshahawi, H., Hashem, M., Williams, S., Stensland,D., Norge, E. et al. 2007. NewDownhole Fluid Analyzer Tool for Improved Reservoir Characterization. PaperSPE 108566 presented at Offshore Europe, Aberdeen, 4-7 September. DOI:10.2118/108566-MS.
Elshahawi, H., Hashem, M., Dong, C., Hegeman, P., Mullins, O.C., Fujisawa,G., and Betancourt, S. 2004. In-situ Characterization of FormationFluid Samples--Case Studies. Paper SPE 90932 presented at the SPE AnnualTechnical Conference and Exhibition, Houston, 26-29 September. DOI:10.2118/90932-MS.
Fujisawa, G., Betancourt, S.S., Mullins, O.C., Torgersen, T., O'Keefe, M.,Terabayashi, T., Dong, C., and Eriksen, K.O. 2008. Hydrocarbon Compositional GradientRevealed by In-Situ Optical Spectroscopy. SPE Res Eval &Eng11 (2): 233-237. SPE-89704-PA. DOI: 10.2118/89704-PA.
Fujisawa, G., van Agthoven, M.A., Jenet, F., Rabbito, P., and Mullins, O.C.2002. Near-InfraredCompositional Analysis of Gas and Condensate Reservoir Fluids at ElevatedPressures and Temperatures. Applied Spectroscopy 56 (12):1615-1620. DOI: 10.1366/000370202321116101.
Gaganis, V. and Varotsis, N. 2005. Sensitivity Controlled Neural Networksfor the Prediction of Thermodynamic Properties of Fluids. Paper presented atthe 1st IC-EpsMso, Athens, 6-9 July.
Garrouch, A. and Smaoui, N.H. 1998. An Artificial Neural Network Modelfor Estimating Tight Gas Sand Permeability. Paper SPE 39703 presented atthe SPE Asia Pacific Conference on Integrated Modelling for Asset Management,Kuala Lumpur, 23-24 March. DOI: 10.2118/39703-MS.
Gharbi, R.B. and Elsharkawy, A.M. 1997. Universal Neural Network Based Modelfor Estimating the PVT Properties of Crude Oil Systems. Paper SPE 38099presented at the SPE Asia Pacific Oil and Gas Conference and Exhibition, KualaLumpur, 14-16 April. DOI: 10.2118/38099-MS.
Haykin, S. 1999. Neural Networks, a Comprehensive Foundation. UpperSaddle River, New Jersey: Prentice-Hall.
Hornik, K., Stinchcombe, M., and White, H. 1989. Multilayer FeedforwardNeural Networks are Universal Approximators. Neural Networks2 (5): 359-366. DOI:10.1016/0893-6080(89)90020-8.
Mohaghegh, S. 2000. Virtual-Intelligence Applications in Petroleum Engineering: Part 1--NeuralNetworks. J. Pet Tech52 (9): 64-73. SPE-58046-PA. DOI:10.2118/58046-PA.
Mullins, O.C., Schroer, J., and Beck, G.F. 2000. Real-time quantification ofOBM filtrate contamination during open hole wireline sampling by opticalspectroscopy. Trans., SPWLA Annual Logging Symposium, Dallas, 4-7 June,Paper SS, 1-10.
Nocedal, J. and Wright, S.J. 1999. Numerical Optimization. New YorkCity: Springer.
O'Keefe, M., Eriksen, K.O., Williams, S., Stensland, D., and Vasques, R.2008. Focused Sampling ofReservoir Fluids Achieves Undetectable Levels of Contamination. SPE ResEval & Eng11 (2): 205-218. SPE-101084-PA. DOI:10.2118/101084-PA.
Osman, E.A., Abdel-Wahhab, O.A., and Al-Marhoun, M.A. 2001. Prediction of Oil PVT PropertiesUsing Neural Networks. Paper SPE 68233 presented at the SPE Middle East OilShow, Bahrain, 17-20 March. DOI: 10.2118/68233-MS.
Smits, A.R., Fincher, D.V., Nishida, K., Mullins, O.C., Schroeder, R.J., andYamate, T. 1995. In-Situ OpticalFluid Analysis as an Aid to Wireline Formation Sampling. SPE FormEval10 (2): 91-98. SPE-26496-PA. DOI: 10.2118/26496-PA.
Varotsis, N., Gaganis, V., and Nighswander, J. 2002. Quality Assurance Tool for PVTSimulator Predictions. SPE Res Eval & Eng5 (6):499-506. SPE-81751-PA. DOI: 10.2118/81751-PA.
Varotsis, N., Gaganis, V., Nighswander, J., and Guieze, P. 1999. A Novel Non-Iterative Method for thePrediction of the PVT Behavior of Reservoir Fluids. Paper SPE 56745presented at the SPE Annual Technical Conference and Exhibition, Houston, 3-6October. DOI: 10.2118/56745-MS.