Machine Learning for Reservoir and Production Engineering Application

Module 1: Petroleum Engineering Data & Python Overview

• Integrated Production System & Data Integration

• Subsurface Engineering Data

• Data Mining & Analytics

• Python for Oil & Gas Industry

• Data Pre-processing

• Python Data Types & Basic Python Functions

• Variables, Expressions, and Statements

• Values and Data Types

• Statements, Expressions & Operators

• Lists, Indexing & Slicing

• Data Containers (Lists and Dictionaries)

• Creating Functions

• Complex Data Structures & Control Flow

• Dictionaries

• Data Loops: For & While Loops

• Conditional (IF) Statements

Module 2: Python Data Analysis & Visualizations

• Python Tools & Package Options

• Anaconda Navigator

• Python IDEs (VS Code, Spyder, Jupyter Notebook, etc.)

• Python Libraries

• Data Analysis, Manipulation & Visualizations

• NumPy

• Pandas

• Matplotlib

• SciPy

• Seaborn

• Descriptive Statistics

• Histograms

• Bar Plots

• Scatter Plots

• Pie Charts

• Box Plots

• Bubble Chart

• Violin Plot

• Heat Map

• Tree Map

• Python Machine Learning Libraries

• Scikit‐learn

• Tensorflow

• Keras

• Pytorch

• Scikit-learn Machine Learning Algorithms

• Workflow of the Machine Learning Model

• Deploy Machine Learning Models as Interactive APIs

Module 3: Python & Machine Learning

• Python Libraries for Oil & Gas

• NeqSim: PVT Modeling

• GemPy: 3D Geological Modeling

• Pyscal: Rel-Perm & Cap-Pressure

• Lasio: Log Reading & Visualizations

• XTGeo: Subsurface Reservoir Modelling

• DeepField: Reservoir Simulation

• Psapy: Production System Analysis

• PySand: Sand Management

• Overview of Machine Learning

• Supervised and Unsupervised Learning

• Classification and Regression Algorithms

• Unsupervised Learning

• K-Mean Clustering using Scikit-learn

• Clustering Oil Wells based on Petrophysical Properties

• Lithofacies Clustering

• Electrofacies Grouping

• K-Nearest Neighbors

• Random Initialization

• Choosing the Number of Clusters

• Hierarchal Clustering and Dendrogram

• Hierarchal Cluster for Water Cut

• Classification Techniques

• Logistic Regression

• Support Vector Machines (SVM)

• Decision Tree

• Random Forest

• Supervised Learning

Module 4: Artificial Neural Networks (ANN)

• Overview of Artificial Neural Networks (ANN)

• Model Evaluation and Validation

• Neural Network Architecture & Components

• Activation Functions in Neural Networks

• Back-propagation & Optimization Algorithms

• ANN Model Validation & Testing

• Under-fitting vs Over-fitting

• Bias-Variance Trade-off

• Analysis & Prediction of Reservoir Fluid Properties

• Prediction of PVT Data using Machine Learning Techniques

• Permeability Measurement Techniques

• Permeability Prediction Techniques

• Applications of Artificial Neural Networks for Core-Calibrated Permeability Predictions using the Petrophysical Data.

Module 5: Machine Learning & Python Dashboarding

• Python Web-Based Dashboarding Libraries:

• Streamlit

• Dash (Plotly)

• Panel (Anaconda)

• Production Analysis Dashboards

• Core & PVT Data Dashboards

• Case Studies and Examples

• Streamlit Library

• Designing Dynamic Python Applications with Streamlit

• Interactive Web Applications & Dashborads

• Streamlit Layout Features

• State Management and Dynamic Interactions with Streamlit

• Useful Tools for Efficient Coding

• Setting Up Your Development Environment: Anaconda Distribution

• Introduction to Python Programming Basics

• Hands-on Practice: Python Basics using Jupyter Notebook

• Visualizing and Presenting Data Insights

• Artificial Neural Networks: Definitions, Architectures, Types, Training, and Validation.

• Python Project 1: Create Dashboard for Nodal Analysis & Vertical Lift Performance (VLP) Calculations

• Machine Learning Project 2: Drilling Data Optimization

Module 6: Applications of Machine Learning in the Petroleum Industry

• Machine Learning Applications in the Petroleum Industry

• Machine Learning Projects in Various Petroleum Engineering Aspects

• Machine Learning Workflows

• Deep Learning Python Libraries Tensorflow & PyTorch

• Case Studies and Examples

Module 7: Python Programming Tools

• Machine Learning Project 3: Pump Intake Pressure Estimation using Regression Techniques

• Hands-on Project: Data Analysis with Real Petroleum Industry Data

• Handling Data from Various Sources: Files and Formats

• Data Cleaning and Preprocessing Techniques

• Estimating Missing Data and Data Transformation Methods

Module 8: Data & Time Series Analysis

• Time Series Analysis

• Time Series Data and Its Characteristics

• Long Short-Term Memory (LSTM) Networks for Time Series

• Evaluating Time Series Models

• Optimization of Surface Network Models using Machine Learning

• Machine Learning Project 4: Flow Rate Estimation from Pump Intake Pressure

• Simple Model Dashboard: Create Simple Data Input Dashboard uses the Trained Model and Generates the Outputs.

Module 9: Machine Learning Projects (Regression)

• Introduction to Regression Machine Learning Methods.

• Machine Learning Project 5: Learning for Oil Properties Estimation: Oil FVF or Viscosity

• Classification Machine Learning Methods.

• Machine Learning Project 6: Decline Curve Analysis (DCA) & prediction.

• Machine Learning Project 7: Well Logging Analysis Dashboard & Core-calibrated Permeability Predictions.

Module 10: Machine Learning Projects (Classification) & Time Series Analysis

• Machine Learning Project 8: Rock Typing & Hydraulic Flow Units (HFU) Prediction.

• Machine Learning Project 9: Flow Assurance & Scale Formation Prediction using Classification Techniques

• Machine Learning Project 10: Predicting Production Performance using Time series.