Machine Learning for Petroleum Engineers using Python
WHEN
18 - 20 OCTOBER 2021
WHERE
MS TEAMS
TIME
9.00AM - 5.00PM
RM 1,350 FOR PROFESSIONALS
10% Discount for Early Bird (until 18 September 2021) / Group / Students
CONTENT SUMMARY
INTRODUCTION
Digitization and machine learning are buzz words in the oil and gas industry. However, many petroleum engineers find it difficult to catchup with the implementation programming languages and machine learning concepts, even though they are very fluent in the physics-based understanding of the subject. This course aims to provide those engineers with a comprehensive introduction to Python programing and builds a hands-on experience on using Python as a tool for petroleum engineering applications. The course is designed to provide participant, the confidence to implement supervised machine learning algorithms in their day-to-day working life. It also paves the way to dig dipper and catch up the many other machine learning algorithms on their own. The course will begin by introducing Python and many useful libraries and proceeds to exploratory data analysis. Then it covers Regression and Classification methods as applied to problems encountered in petroleum engineering. Timeseries forecasting of production data, well log analysis and predictions followed by application of Artificial Neural Network (ANN) and Multiple regression will follow using a hands-on exercise.
No | Topic |
1 |
- Exploratory data analysis with example 1. Missing data 2. Outliers - 3-sigma rule |
2 |
- Regression 1. Curve fitting [Scipy - optimize (curve-fit)] 2. Linear regression [Scipy - stats.linregress] 3. Polynomial regression [Numpy – polyfit] 4. Multiple regression [Sklearn linear- model (Linear regression)] - Classification 1. Logistic classification 2. Random forest |
3 |
- Time series forecasting - Well Log analysis. - Prediction of missing log - Pore pressure prediction using ANN (compare with polynomial and Multiple regression) |
OBJECTIVES
Upon completion of this course, participants will be able to:
WHO SHOULD ATTEND?
1. Dr Berihun Mamo Negash (UTP)
Dr Berihun Mamo Negash is a senior lecturer at Universiti Teknologi PETRONAS. He has been researching and working on machine learning and its application in the petroleum engineering field for over five years. In addition to experiment-based works, he has also published few articles in areas of applied machine learning. Proxy modeling of a reservoir using system identification concept and missing production and pressure data reconstruction are his main area of specializations
2. Mr Cheang Hoi Him (PETRONAS)
Mr Cheang Hoi Him is a data scientist in PETRONAS Digital Sdn Bhd. He graduated from Universiti Teknologi PETRONAS as Petroleum Engineering graduate and has 3 years working experience in PETRONAS solving industry pain points with digital approach and data science methodology. His area of expertise is Python programming, building machine learning model in Python, cloud computing and model deployment.
*fee quoted does not include SST, GST/VAT or withholding tax (if applicable)
*fee quoted does not include SST, GST/VAT or withholding tax (if applicable)
Centre for Advanced & Professional Education (CAPE)
Level 16, Menara 2,
Menara Kembar Bank Rakyat,
50470, Jalan Travers,
Kuala Lumpur.
+605 - 368 7558 /
+605 - 368 8485
cape@utp.edu.my