Dr. Tom Smith, the President and CEO of Geophysical Insights, presents to the University of Houston Wavelet Society on the impact of machine learning on seismic interpretation for the Oil and Gas Industry. In this presentation, Dr. Smith explains the concept of 3D seismic survey for interpretation, the advantages of seismic surveys by using multi-attribute, how to use machine learning to analyze seismic surveys, and how seismic interpretation with machine learning of multi-attribute seismic surveys have been conducted successfully around the world.
- Machine Learning in Seismic Interpretation
- Deep Learning for Seismic Facies Classification
- Understand how Patch-Based Classification network provides a high-quality seismic facies delineation quality
- Understand how Encoder-Decoder Segmentation Network provides the best seismic facies delineation quality
- Deep Learning for Fault Identification
- Difference Between Blenders and Classifiers Thinking
- How machine learning is used to help geologists and geophysicists to get through the grunt work
- Machine Learning Terminologies
- Supervised Machine Learning
- Unsupervised Machine Learning
- Self-Organizing Map (SOM)
- Winning Neuron
- Multi-attribute Seismic Sample
- Multi-attribute 3D Surveys
- Attribute Selection List (ASL)
- Multi-attribute Seismic Interpretation
- What is it and what are its demonstrated areas of application?
- What is attribute space?
- What are attributes?
- How machine learning classifier identifies natural clusters in attribute space.
- Example of how unsupervised neural network operates
- 100 Synthetic Seismograms
- 4 basic interpretation wavelets.
- Sample density in attribute space.
- Identifying the winning neurons.
- The stacking process in attribute space.
- Self-Organizing Map (SOM) Analysis of a Wedge Model in Two Attributes
- Achieving finer reservoir detail and yielding higher accuracy with machine learning on multiple attributes simultaneously.
- The Classic ThoughtFlow of The New Seismic Interpretation
- What is a Geobody is and Why is it Important?
- A Case Study: The Eagle Ford
- The Wheeler Diagram on the Eagle Ford
- Comparing Self-Organizing Maps (SOM) with Instantaneous and Inversion Attribute Selection Lists (ASLs).
Paradise is an evolving platform of geoscience tools for seismic interpretation. The Paradise Workbench includes components to develop new machine learning technology with Python and/or the new Paradise Scripting Language (PSL, PSE, PSLi).
Machine Learning Essentials for Seismic Interpretation, a 16-hours e-Learning course with certification, is available HERE.
Most Popular Papers
Systematic Workflow for Reservoir Characterization in Northwestern Colombia using Multi-attribute Classification
Net Reservoir Discrimination through Multi-Attribute Analysis at Single Sample Scale
Seismic Facies Classification Using Deep Convolutional Neural Networks