Analysis of Data From the Barnett Shale Using Conventional Statistical and Virtual Intelligence Techniques
- Obadare Awoleke (Texas A&M University) | Robert Lane (Texas A&M University)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- October 2011
- Document Type
- Journal Paper
- 544 - 556
- 2011. Society of Petroleum Engineers
- 1.6 Drilling Operations, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 5.8.2 Shale Gas, 5.7.2 Recovery Factors
- Water loading, Analysis, Water production, Shale gas, Competitive learning network
- 1 in the last 30 days
- 1,194 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
A Barnett shale water-production data set from approximately 11,000 completions was analyzed using conventional statistical techniques. Additionally, a water/hydrocarbon ratio and first-derivative diagnostic-plot technique developed elsewhere for conventional reservoirs was extended to analyze Barnett shale water-production mechanisms. To determine hidden structure in well and production data, self-organizing maps and the k-means algorithm were used to identify clusters in data. A competitive-learning-based network was used to predict the potential for continuous water production from a new well, and a feed-forward neural network was used to predict average water production for wells drilled in Denton and Parker Counties, Texas, of the Barnett shale.
Using conventional techniques, we concluded that for wells of the same completion type, location is more important than time of completion or hydraulic-fracturing strategy. Liquid loading has potential to affect vertical more than horizontal wells. Different features were observed in the spreadsheet diagnostic plots for wells in the Barnett shale, and we made a subjective interpretation of these features. We find that 15% of the horizontal and vertical wells drilled in Denton County have a load-water-recovery factor greater than unity. Also, 15 and 35% of the horizontal and vertical wells drilled, respectively, in Parker County have a load-recovery factor greater than unity.
The use of both self-organizing maps and the k-means algorithm showed that the data set is divided into two main clusters. The physical properties of these clusters are unknown but interpreted to represent wells with high water throughput and those with low water throughput. Expected misclassification error for the competitive-learning-based tool was approximately 10% for a data set containing both vertical and horizontal wells. The average prediction error for the neural-network tool varied between 10 and 26%, depending on well type and location.
Results from this work can be used to mitigate risk of water problems in new Barnett shale wells and predict water issues in other shale plays. Engineers are provided a tool to predict potential for water production in new wells. The method used to develop this tool can be used to solve similar challenges in new and existing shale plays.
|File Size||1 MB||Number of Pages||13|
Awoleke, O.O. 2009. Analysis of Data from the Barnett Shale withConventional Statistical and Virtual Intelligence Techniques. MS thesis, TexasA&M University, College Station, Texas, USA.
Chan, K.S. 1995. Water Control Diagnostic Plots. Paper SPE 30775 presentedat the SPE Annual Technical Conference and Exhibition, Dallas, 22-25 October.http://dx.doi.org/10.2118/30775-MS.
Cybenko, G. 1988. Continuous valued neural networks with two hidden layersare sufficient. Technical Report, Department of Computer Science, TuftsUniversity, Medford, Massachusetts, USA.
Cybenko, G. 1989. Approximations by superposition of a sigmoidal function.Mathematics of Control, Signals, and Systems (MCSS) 2 (4):303-314. http://dx.doi.org/10.1007/BF02551274.
Friedel, T., Mtchedlishvili, G., Behr, A., Voigt, H.-D., and Frieder, H.2007. Comparative Analysis of Damage Mechanisms in Fractured Gas Wells. PaperSPE 107662 presented at the European Formation Damage Conference, Scheveningen,The Netherlands, 30 May-1 June. http://dx.doi.org/10.2118/107662-MS.
Haykin, S. 2005. Neural Networks, a Comprehensive Foundation, secondedition, ninth reprint. Delhi, India: Prentice-Hall of India.
Holditch, S.A. 1979. Factors Affecting Water Blocking and Gas Flow FromHydraulically Fractured Gas Wells. J Pet Technol 31(12): 1515-1524. SPE-7561-PA. http://dx.doi.org/10.2118/7561-PA.
Hornik, K., Stinchcombe, M., and White, H. 1989. Multilayer FeedforwardNeural Networks are Universal Approximators. Neural Networks 2 (5): 359-366. http://dx.doi.org/10.1016/0893-6080(89)90020-8.
Iqbal, G.M. 1988. Cleanup of water based liquids from hydraulicallyfractured gas wells. PhD dissertation, The University of Oklahoma, Norman,Oklahoma, USA.
Martineau, D.F. 2007. History of the Newark East Field and the Barnett shaleas a gas reservoir. AAPG Bulletin 91 (4): 399-403. http://dx.doi.org/10.1306/intro910407.
Mitchell, T. 1997. Machine Learning, international edition.Singapore: McGraw-Hill.
Mohaghegh, S. 2000. Virtual-Intelligence Applications in PetroleumEngineering: Part 1--Neural Networks. J Pet Technol 52 (9):64-73. SPE-58046-MS. http://dx.doi.org/10.2118/58046-MS.
Montgomery, K.T., Holditch, S.A., and Berthelot, J.M. 1990. Effects ofFracture Fluid Invasion on Cleanup Behavior and Pressure Buildup Analysis.Paper SPE 20643 presented at the SPE Annual Technical Conference andExhibition, New Orleans, 23-26 September. http://dx.doi.org/10.2118/20643-MS.
Ozkan, K., Ohaeri, U., and Raghavan, R. 1987. Unsteady Flow to a WellProduced at a Constant Pressure in a Fractured Reservoir. SPE Form Eval 2 (2): 186-200. SPE-9902-PA. http://dx.doi.org/10.2118/9902-PA.
Pollastro, R.M., Jarvie, D.M., Hill, R.J., and Adams, C.W. 2007. GeologicFramework of the Missippian Barnett Shale, Barnett-Paleozoic total petroleumsystem, Bend Arch-Fort Worth Basin, Texas. AAPG Bulletin 91(4): 405-436. http://dx.doi.org/10.1306/10300606008.
Seright, R.S. 1997. Improved Methods for Water Shutoff. Annual Report,Contract No. DE-AC22-94PC91008, Subcontract No. G4S60330, New Mexico Instituteof Mining and Technology, Socorro, New Mexico, USA (November 1997).
Shelley, B. and Harris, P.C. 2009. Data Mining Identifies Production Driversin a Complex High-Temperature Gas Reservoir. SPE Prod & Oper 24 (1): 74-80. SPE-106463-PA. http://dx.doi.org/10.2118/106463-PA.
Shelley, B., Johnson, B.J., Fielder, E.O., Heinze, J.R., and Werline, J.R.2008. Data Analysis of Barnett Shale Completions. SPE J. 13(3): 366-374. SPE-100674-PA. http://dx.doi.org/10.2118/100674-PA.
Soliman, M.Y. and Hunt, J.L. 1985. Effect of Fracturing Fluid and ItsCleanup on Well Performance. Paper SPE 14514 presented at the SPE EasternRegional Meeting, Morgantown, West Virginia, USA, 6-8 November. http://dx.doi.org/10.2118/14514-MS.
Tannich, J.D. 1975. Liquid Removal from Hydraulically Fractured Gas Wells.J Pet Technol 27 (11): 1309-1317. SPE-5113-PA. http://dx.doi.org/10.2118/5113-PA.
Turner, R.G., Hubbard, M.G., and Dukler, A.E. 1969. Analysis and Predictionof Minimum Flowrate for the Continuous Removal of Liquids from Gas Wells. JPet Technol 21 (11): 1475-1482; Trans., AIME, 246.SPE-2198-PA. http://dx.doi.org/10.2118/2198-PA.
Walton, I.C. and Erkal, A. 2009. Evaluation of Shale Gas Productivity Usinga New Asymptotic Model. Paper No. 120577 presented at the SPEOKC Production andOperations Symposium, Oklahoma City, Oklahoma, USA, 4-8 April.