Scribd
Upload
31 views

ML Introduction

Uploaded by

thebhavikbaruah
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
31 views

ML Introduction

Uploaded by

thebhavikbaruah
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Introduction To Digital Age Data Modeling Using

Machine Learning
Nashat J. Omar
Petroleum Production Engineer
Agenda
• The Art of Prediction
• Historical Context
• Industrial Revolution and the Current Iteration
• What is Machine Learning
• Machine Learning Applied to Oil and Gas
• ML workflow and Python
The Art of Predicting the Unpredictable

Pforecasting
rediction is the process of
or estimating future
events, trends, or behaviors
based on current and historical
data, patterns, or other
information. The goal of
prediction is to anticipate
outcomes in various contexts
such as science, economics,
weather, health, and daily life. Predicting Things is Not a
New Thing
Historical Context
Clay Tablets
The Babylonians
Babylonians have predicted
the movement of the sun and
the moon by using a
systematic thinking an
scientific modeling approach
Saros Cycle
Historical Context
The Egyptians
Egyptians have observed the
sun/moon, seasonal changes,
Nile Flood and advanced in
mathematics to predict the
climate change
Historical Context
19th Century
London Cholera
outbreak
In 19th Century, an early
implementation of new data
science approach was utilized.
Dr.John Snow

Soho, London
Industrial Revolution
We are here : A pivotal Time
Introduction to ML
What is Machine Learning?
Machine learning is a branch
of artificial intelligence (AI)
that allows computers to
learn from data and make
decisions or predictions
without being explicitly
programmed for each
specific task.
Why The Fuss?
Why Machine Learning for Oil and
Gas?
• Understand Oil and Gas Data in a better way.
• Reveal Insights From Data.
• Create Predictive Models.
• Become a leader in the new revolution.
• Combine Domain and ML knowledge to
create cutting edge technology.
• Identify Well and Reservoir Problems early.
• Generate Well Logs For Undrilled Field Areas
Why you should Care?
• Machine Learning is Advancing
very fast.
• Large Operators started to
implement(Shell, BP, Aramco,
ADNOC)
• Service Companies are offering ML
packages(Haliburton,
Schlulmberger, etc)
• ML is away forward to shape the
Digital Petroleum engineer.
• Non Orthodox Predictive Models
very fast to model and deploy
• Improved Decision
• Reduce Maintenance Cost
Machine Learning Use Cases From the Industry:
Oil and Gas Industry Has Imposed Interesting Use
Cases for machine learning that helped to improve the operational efficiency.
1-Stuck Pipe prediction and prevention | https://doi.org/10.2118/208778-MS , 2022
2-Production Allocation Algorithm | https://doi.org/10.2118/219392-MS, 2024
3-ESP Failure Prediction and Prod Opt. | https://doi.org/10.2118/216990-MS, 2023
4-Missing Data Generation and Synthesizing | https://doi.org/10.2118/219384-MS, 2024
5-EOR and Waterflood Optimization and management | https://doi.org/10.2118/218258-MS, 2024
Type of Machine Learning Task
Machine Learning Can be used for various
task, here are major uses of machine
learning as task based classification:
1. Classification Tasks
2. Regression Task
3. Clustering Tasks
4. Time Series Analysis and Forecasting
Classification
Classification refers to the process of
organizing or arranging Oil and Gas
Data into categories or groups based on
shared characteristics or particular
domain criteria. This concept can be
applicable to number of domains in Oil
and Gas applications
1. Well Type Classification
2. Production Problem Labeling
3. Drilling Operation Monitoring
4. Problem imposing Reservoir Sections
5. Well Log facies classficiation and zone identification
Regression
Regression is a statistical method
used for modeling the relationship
between a dependent variable and
one or more independent variables.
It aims to predict or explain the
dependent variable based on the
values of the independent
variables.
1. Well Log Value vs. Depth Prediction
2. Production Forecasting
3. Flow assurance onset
4. Optimal Operation
5. Stuck Pipe Prediction
Time Series Analysis
Analysis and Forecasting
Time series analysis is a statistical
technique that deals with time-
ordered data points. This type of
analysis is used to understand the
underlying structure and function
of the data and to make forecasts.
1. Production Forecast
2. Reservoir performance
3. Reservoir Pressure Prediction
4. Data Filling and Gap Imputation
5. Stuck Pipe Prediction
6. Lost circulation Prediction
Clustering
Clustering is a machine learning technique used to group a set of objects in such a
way that objects in the same group (or cluster) are more similar to each other than to
those in other groups. It is an unsupervised learning method
Machine Learning Project Flow
A Series of predefined agile steps can be followed to create a
machine learning solution:

You might also like