Shahab D. Mohaghegh Shale Analytics Data-Driven Analytics in Unconventional Resources Shale Analytics Shahab D. Mohaghegh Shale Analytics Data-Driven Analytics in Unconventional Resources 123 Shahab D.Mohaghegh Petroleum andNatural GasEngineering West Virginia University Morgantown,WV USA and Intelligent Solutions, Inc. Morgantown,WV USA ISBN978-3-319-48751-9 ISBN978-3-319-48753-3 (eBook) DOI 10.1007/978-3-319-48753-3 LibraryofCongressControlNumber:2016955428 ©SpringerInternationalPublishingAG2017 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpart of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission orinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilar methodologynowknownorhereafterdeveloped. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publicationdoesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfrom therelevantprotectivelawsandregulationsandthereforefreeforgeneraluse. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authorsortheeditorsgiveawarranty,expressorimplied,withrespecttothematerialcontainedhereinor foranyerrorsoromissionsthatmayhavebeenmade. Printedonacid-freepaper ThisSpringerimprintispublishedbySpringerNature TheregisteredcompanyisSpringerInternationalPublishingAG Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland To Dorna that gives my life meaning Foreword It is an honor and a pleasure to be asked to write the Foreword to this much anticipated book on the soft-computing, data-driven methodologies applied across unconventional reservoirs so as to harness the power of raw data and generate actionableknowledge.Wearetakenalongawell-documentedbutstillbumpyroad that starts with an introduction to the shale revolution and draws salient compar- isons between the traditional modeling of these unconventional resources and the non-deterministic and stochastic workflows prevalent in all industries that strive to analyze vast quantities of raw data to address and solve business problems. We are enlightened as to an array of analytical methodologies that have suc- cessfully proven to be not only pertinent in the oil and gas industry but also computer resource friendly. Methodologies drawn from artificial intelligence and data mining schools of thought, such as artificial neural networks, fuzzy logic, fuzzy cluster analysisand evolutionary computing, the last ofwhich isinspired by the Darwinian Theory of Evolution through Natural Selection. The book inspires geoscientists entrenched in first principles and engineering concepts to think outside the box shaped by determinism, and to marry their experienceandinterpretations ofthedatawiththeresults generatedbydata-driven advanced analytical workflows. The latter approach enables ALL the data to be exploitedandopensthedoortohiddenrelationshipsandtrendsoftenmissedbythe formerapproachexecutedinisolation.Weareencouragedasengineerstonotonly to break out of the silos and put ALL our data into the context of experience and interpretation but also to allow the data to do the talking and ask questions of our traditional workflows. The convincing conversational tone is enhanced by some in-depth case studies that have proven successful to oil and gas operators from a business value proposition perspective. The Shale Production Optimization Technology (SPOT) chapteraloneisvalidationenoughtoprovidecredibleendorsementofthestrategic and tactical business insight into hydraulic fracturing practices essential to exploit the unconventional resources. The book provides plausible and cogent arguments for a top-down modeling methodology. We are introduced to a very convincing vii viii Foreword presentation of material that details a “formalized, comprehensive, multivariate, full-field, and empirical reservoir model”. Wearefortunatetohavethisbookpublishedatsuchanopportunetimeintheoil and gas industry as it climbs slowly out of a big trough left by the erratic price fluctuationsofthepasttwoyearssince2014.Operatorsandservicecompaniesalike will see incredible value within these pages. The authors have diligently provided an index to very important soft-computing workflows to ensure ALL the hard and soft data detailed in the early chapters are mined to surface powerful knowledge. This original knowledge can help design best practices for drilling and completing the wells in unconventional reservoirs. It also sheds light on different modeling practices for field re-engineering and well forecasting in reservoirs that necessitate innovative workflows as traditional interpretive approaches honed in the conven- tional reservoirs prove inadequate. Thestrengthofthemessageslogicallylaidoutinthisbookintroducethereader to a collection of soft-computing techniques that address some critical business issues operators in the unconventional reservoirs are facing on a daily basis: quantifying uncertainty in the shale, forecasting production, estimating ultimate recovery, building robust models and formulating best practices to exploit the hydrocarbons. We are indebted to Shahab Mohaghegh for his original thought, passion and innovation across the many years as he evangelizes the application of the ever-increasing popularity of data-driven methodologies. As one of the earliest pioneersinthisfield,wearegratefulthathehasputpentopaperandprovidedthe industry with a very valuable book. Cary, USA Keith R. Holdaway FGS Advisory Industry Consultant SAS Global O&G Domain Acknowledgements First and foremost, I like to acknowledge and thank the contribution of my col- leaguesatIntelligentSolutions,Inc.RaziGaskariandMohammadMaysami.Their hardworkandcommitmenthasplayedamajorroleinshapingtheconceptsthatis presented in this book as “Shale Analytics”. IalsowouldliketothankProfessorSamAmeri,ChairofPetroleumandNatural Gas Engineering Department at West Virginia University for his help and coop- eration during the time that I was busy working on this book. Some Chapters of this book were prepared in collaboration with the following co‐authors: Maher J. Alabboodi, West Virginia University Dr. Soodabeh Esmaili, Devon Energy Faegheh Javadi, Mountaineer Keystone Dr. Amir Masoud Kalantari, University of Kansas Dr. Mohammad Omidvar Eshkalak, University of Texas ix Contents 1 Introduction... .... .... ..... .... .... .... .... .... ..... .... 1 1.1 The Shale Revolution.... .... .... .... .... .... ..... .... 2 1.2 Traditional Modeling .... .... .... .... .... .... ..... .... 4 1.3 A Paradigm Shift .. ..... .... .... .... .... .... ..... .... 4 2 Modeling Production from Shale... .... .... .... .... ..... .... 7 2.1 Reservoir Modeling of Shale .. .... .... .... .... ..... .... 9 2.2 System of Natural Fracture Networks.... .... .... ..... .... 10 2.3 System of Natural Fracture Networks in Shale. .... ..... .... 13 2.4 A New Hypothesis on Natural Fractures in Shale... ..... .... 14 2.5 Consequences of Shale SNFN . .... .... .... .... ..... .... 16 2.6 “Hard Data” Versus “Soft Data”.... .... .... .... ..... .... 18 2.7 Current State of Reservoir Simulation and Modeling of Shale . .... .... ..... .... .... .... .... .... ..... .... 19 2.7.1 Decline Curve Analysis.... .... .... .... ..... .... 20 2.7.2 Rate Transient Analysis.... .... .... .... ..... .... 21 2.8 Explicit Hydraulic Fracture Modeling.... .... .... ..... .... 22 2.9 Stimulated Reservoir Volume.. .... .... .... .... ..... .... 24 2.10 Microseismic . .... ..... .... .... .... .... .... ..... .... 27 3 Shale Analytics .... .... ..... .... .... .... .... .... ..... .... 29 3.1 Artificial Intelligence .... .... .... .... .... .... ..... .... 33 3.2 Data Mining.. .... ..... .... .... .... .... .... ..... .... 33 3.2.1 Steps Involved in Data Mining .. .... .... ..... .... 34 3.3 Artificial Neural Networks .... .... .... .... .... ..... .... 35 3.3.1 Structure of a Neural Network... .... .... ..... .... 36 3.3.2 Mechanics of Neural Networks Operation.. ..... .... 38 3.3.3 Practical Considerations During the Training of a Neural Network.. .... .... .... .... ..... .... 41 xi xii Contents 3.4 Fuzzy Logic.. .... ..... .... .... .... .... .... ..... .... 55 3.4.1 Fuzzy Set Theory .... .... .... .... .... ..... .... 57 3.4.2 Approximate Reasoning.... .... .... .... ..... .... 59 3.4.3 Fuzzy Inference.. .... .... .... .... .... ..... .... 60 3.5 Evolutionary Optimization .... .... .... .... .... ..... .... 62 3.5.1 Genetic Algorithms... .... .... .... .... ..... .... 63 3.5.2 Mechanism of a Genetic Algorithm... .... ..... .... 64 3.6 Cluster Analysis... ..... .... .... .... .... .... ..... .... 66 3.7 Fuzzy Cluster Analysis... .... .... .... .... .... ..... .... 68 3.8 Supervised Fuzzy Cluster Analysis.. .... .... .... ..... .... 70 3.8.1 Well Quality Analysis (WQA)... .... .... ..... .... 71 3.8.2 Fuzzy Pattern Recognition.. .... .... .... ..... .... 74 4 Practical Considerations. ..... .... .... .... .... .... ..... .... 83 4.1 Role of Physics and Geology.. .... .... .... .... ..... .... 84 4.2 Correlation is not the Same as Causation. .... .... ..... .... 84 4.3 Quality Control and Quality Assurance of the Data . ..... .... 86 5 Which Parameters Control Production from Shale .... ..... .... 91 5.1 Conventional Wisdom ... .... .... .... .... .... ..... .... 92 5.2 Shale Formation Quality.. .... .... .... .... .... ..... .... 93 5.3 Granularity... .... ..... .... .... .... .... .... ..... .... 98 5.4 Impact of Completion and Formation Parameters... ..... .... 98 5.4.1 Results of Pattern Recognition Analysis ... ..... .... 99 5.4.2 Influence of Completion Parameters .. .... ..... .... 102 5.4.3 Important Notes on the Results and Discussion... .... 106 5.5 Chapter Conclusion and Closing Remarks .... .... ..... .... 106 6 Synthetic Geomechanical Logs. .... .... .... .... .... ..... .... 109 6.1 Geomechanical Properties of Rocks . .... .... .... ..... .... 109 6.1.1 Minimum Horizontal Stress. .... .... .... ..... .... 110 6.1.2 Shear Modulus .. .... .... .... .... .... ..... .... 110 6.1.3 Bulk Modulus... .... .... .... .... .... ..... .... 110 6.1.4 Young’s Modulus .... .... .... .... .... ..... .... 111 6.1.5 Poisson’s Ratio .. .... .... .... .... .... ..... .... 112 6.2 Geomechanical Well Logs .... .... .... .... .... ..... .... 112 6.3 Synthetic Model Development . .... .... .... .... ..... .... 113 6.3.1 Synthetic Log Development Strategy.. .... ..... .... 115 6.3.2 Results of the Synthetic Logs ... .... .... ..... .... 116 6.4 Post-Modeling Analysis .. .... .... .... .... .... ..... .... 124 7 Extending the Utility of Decline Curve Analysis... .... ..... .... 127 7.1 Decline Curve Analysis and Its Use in Shale.. .... ..... .... 127 7.1.1 Power Law Exponential Decline. .... .... ..... .... 129 7.1.2 Stretched Exponential Decline... .... .... ..... .... 130