Parameters as Gas-Sand Predictors
7.1 Introduction
7.2 Petrophysical Parameters for Gas-Sand Identification
7.3 Lithologic and Fluid Content Dissimilarities of Values of Petrophysical Parameters
7.4 Parameter Ranking and Efficiency of Identification of Gas-Sands
7.5 ROC Curve Analysis with Cross Validation
7.6 Ranking Parameters According to AUC Values
7.7 Classification with Multidimensional Parameters as Gas Predictors
7.8 Conclusions
Definitions and Notations
References
15 8 Use of Type Curve for Analyzing Non-Newtonian Fluid Flow Tests Distorted by Wellbore Storage Effects 8.1 Introduction 8.2 Objective 8.3 Problem Analysis 8.4 Use of Finite Element 8.5 Analysis Methodology 8.6 Test Data Examples 8.7 Conclusion Nomenclature References Appendix A: Non-Linear Boundary Condition and Laplace Transform Appendix B: Type Curve Charts for Various Power Law Indices
16 Part 3: Reservoir Permeability Detection
17 9 Permeability Prediction Using Machine Learning, Exponential, Multiplicative, and Hybrid Models 9.1 Introduction 9.2 Additive, Multiplicative, Exponential, and Hybrid Permeability Models 9.3 Combination of Basis Function Expansion and Exhaustive Search for Optimum Subset of Predictors 9.4 Outliers in the Forecasts Produced with Four Permeability Models 9.5 Additive, Multiplicative, and Exponential Committee Machines 9.6 Permeability Forecast with First Level Committee Machines. Sandstone Dataset 9.7 Permeability Prediction with First Level Committee Machines. Carbonate Reservoirs 9.8 Analysis of Accuracy of Outlier Replacement by The First and Second Level Committee Machines. Sandstone Dataset 9.9 Conclusion Notations and Definitions References
18 10 Geological and Geophysical Criteria for Identifying Zones of High Gas Permeability of Coals (Using the Example of Kuzbass CBM Deposits) 10.1 Introduction 10.2 Physical Properties and External Load Conditions on a Coal Reservoir 10.3 Basis for Evaluating Physical and Mechanical Coalbed Properties in the Borehole Environment 10.4 Conclusions Acknowledgement References
19 11 Rock Permeability Forecasts Using Machine Learning and Monte Carlo Committee Machines 11.1 Introduction 11.2 Monte Carlo Cross Validation and Monte Carlo Committee Machines 11.3 Performance of Extended MC Cross Validation and Construction MC Committee Machines 11.4 Parameters of Distribution of the Number of Individual Forecasts in Monte Carlo Cross Validation 11.5 Linear Regression Permeability Forecast with Empirical Permeability Models 11.6 Accuracy of the Forecasts with Machine Learning Methods 11.7 Analysis of Instability of the Forecast 11.8 Enhancement of Stability of the MC Committee Machines Forecast Via Increase of the Number of Individual Forecasts 11.9 Conclusions Nomenclature Appendix 1- Description of Permeability Models from Different Fields Appendix 2- A Brief Overview of Modular Networks or Committee Machines References
20 Part 4: Reserves Evaluation/Decision Making
21 12 The Gulf of Mexico Petroleum System – Foundation for Science-Based Decision Making Introduction Basin Development and Geologic Overview Petroleum System Reservoir Geology Hydrocarbons Salt
