Artificial Intelligence, Machine Learning and Geoscience Data Analytics for Hydrocarbon Prospect Generation

Workshop Values/Significance

1. Unlocking remaining hydrocarbon resources of your concession/field by drilling robust prospects

2. Boosting hydrocarbon production of the brown fields by targeting the untapped resources that can be delineated by advanced machine learning techniques

3. Building robust reservoir models using machine learning-based reservoir characterization for better development planning

4. Predicting the missing well-logs that did not recorded in old wells

Course Content

Part 1: Introduction

• 1 Chapter-1: Introduction
• Introduction to the Course
• 4th Industrial Revolution Elements
• Descriptive Vs. Quantitative Geosciences
• Exploiting Big Data
• Making Business Decision with Uncertainty Analysis
• Be Modern Geoscientist

Part 2: Geoscience Data Analytics for Prospect Generation and Reservoir Characterization

2 Chapter-2: Probabilistic Analytics for Geoscience Data

• Introduction to Data Analytics
• Probabilistic Analytics for Geoscience Data
• Introduction to Probability Theory
• Basic Concepts
• Probability Density Functions
•  Monty Hall Problem
•  Bayesian Inference for Data Analytics and Integration
• Bayesian Classifications and Statistics

Applications and Workflows of Probabilistic Analytics to Prospect Generation and Reservoir characterization

• Probabilistic Resources Assessment of Hydrocarbon Prospects and Leads
• Probabilistic Rock Physics Analysis and Interpretation for Reservoir Characterization
• Bayesian Seismic AVO Inversion for Reservoir Facies Characterization
• Reservoir Rock Characterization using integrated Seismic Inversion and Probabilistic Rock Physics
• Probabilistic Assessment of the Reservoir Connectivity for Water Flooding Optimization

Practical Sessions using Software with Real Datasets

Chapter-3: Statistical Analytics for Geoscience Data

• Introduction to Geostatistical Analytics
• Descriptive Statistics and Their Uses in Subsurface Data Analytics
• Correlation Data Analytics
• Correlation and Covariance
• Geological Correlation Vs. Statistical Correlation
• Correlation and Covariance Matrices
• Regression-Based Predictive Analytics
• Bivariate Regression (Linear and Nonlinear)
• Multivariate Linear Regression
• Principal Component Regression
• Geostatistical Variography Analysis and Modeling
• Variogram and Spatial Correlation
• Variogram Modeling Workflow
• Geological Interpretations of the Variograms
• Geostatistical Estimation Methods
• Kriging Method
• Co-kriging Method
•  Factorial Kriging

Applications and Workflows of Geostatistical Analytics for Prospect Generation and Reservoir Characterization

• Geostatistical Seismic Inversion Methods
• Frequency Domain of Geostatistical Seismic Inversion
• Trace-by-Trace Geostatistical Seismic Inversion
• Global Geostatistical Acoustic Inversion
• Global Geostatistical Elastic Inversion
• Geostatistical AVO Seismic Inversion
• Geostatistical Seismic Petrophysical Inversion
• Geostatistical Seismic AVO Inversion to Reservoir Facies
• Integrating Rock Physics Analysis with Geostatistical Seismic Inversion for Reservoir Properties Mapping
• Geostatistical Modeling of the Reservoir Facies
• Facies Modeling Methods
• Practical Consideration in Facies Modeling
• Automatic Fault and Fractures Delineations

Practical Sessions using Software with Real Datasets

• Chapter-4: Geoscience Data Analytics with Principal Component Analysis
• Introduction to Principal Component Analysis (PCA)
• Aims of PCA
• Procedures of PCA
• Geological Interpretations of PCA
• PCA for Classifications

Applications and Workflows of Geoscience Data Analytics with PCA

• Well Logs Predictions
• Faults and Fractures Identification
• Seismic Multi-Attribute Analysis

Practical Sessions using Software with Real Datasets

Machine Learning Applications to Prospect Generation and Reservoir Characterization

• Machine Learning Applications to Prospect Generation and Reservoir Characterization
• Introduction
• Artificial Intelligence-Based Predictions and Classification Methods
• Geoscience Data Analytics by Machine Learning
• Linear Regression
• Multi-linear Regression
• Support Vector Regression
• Support Vector Regression
• Random Forest Regression
• Geoscience Data Classification and Clustering by Machine Learning
• K-Nearest Neighbors
• Support Vector Machine
• Decision Tree Classification
• Random Forest Classification
• Deep Learning for Geoscience Data Analytics
• Artificial Neural Network
• Convolutional Neural Network
• Challenges in Machine Learning and Artificial Intelligence

Applications of Machine Learning to Prospect Generation and Reservoir Characterization

• Missing Logs Predictions
• Reservoir Facies Classifications
• Seismic Facies Classification
• Supervised Seismic Geobodies Delineation
• Seismic Image Segmentation
• 3D Seismic Reservoir Properties Predictions
• 3D Seismic Velocity Calibration and Depth Conversion

Practical Sessions using Software with Real Datasets