QSI - Seismic Facies Classification
In this course, students learn techniques to incorporate prestack and poststack seismic data and well data to create seismic facies probability models. Students are taught how to work with wireline and facies logs in the Crossplot utility to analyze lithoseismic relationships, and apply those relationships to inverted attributes to create probabilistic lithology models. In addition, use of the workflow engine to apply different machine learning techniques to generate facies models is taught. Students gain a solid understanding of the different parameter settings in the workflows, and the QC options available to analyze and optimize the results.
Familiarity with Integrated Canvas.
Who Should Attend:
Geoscientists and geophysicists interested in facies visualization and probability analysis for interpretation and reservoir characterization workflows.
Crossplot Indexing and Lithoseismic Classification in Integrated Canvas
: This one-to-two-hour course includes two exercises that provide a hands-on introduction to manual and statistical crossplotting tools available in Emerson E&P Software.
- The first exercise covers how to work with multiplots to create indexed volumes, logs and maps.
- The second exercise covers how to use the Lithoseismic Classification utility to create lithologic probabilities based on a crossplot of log attributes and user-defined parameters. The probabilities are saved as a crossplot relationship, which can then be applied to volume attributes using the Seismic Attributes calculator to generate Most Probable Facies volumes.
The course also includes tips for visualizing the results in Integrated Canvas.
Waveform Classification in Integrated Canvas
: The Waveform Classification workflow in Integrated Canvas uses a neural network approach to create facies volumes from poststack seismic attributes. This two-hour course focuses on the theory behind the workflow and how to define and QC the workflow parameters to ensure the best results. In addition to learning how to run the workflow, students learn different visualization tools for examining the results. Students learn how to use the Waveform Classification workflow to perform:
- Unsupervised classification
- Supervised classification using:
- Seismic traces or well logs
- 2D Models (wedge models)
Rock Type Classification in Integrated Canvas: The Rock Type Classification workflow uses a machine learning approach to create a probabilistic facies model based on lithology logs and prestack and/or poststack seismic attributes. This one-hour course focuses on the theory behind the workflow and understanding how to define and QC the workflow parameters to ensure the best results. In addition to learning how to run the workflow, students learn different visualization tools for examining the results.
Attribute Clustering in Integrated Canvas: The Attribute Clustering workflow uses a machine learning approach (Self Growing Neural Network) to create a facies model based on poststack and/or prestack seismic attributes. This one-hour course focuses on the theory behind the machine learning method and on understanding how to define and QC the workflow parameters to ensure the best results. In addition to learning how to run the workflow, students learn different visualization tools for examining the results.
||Products / Solutions
Are you looking for a course that cannot be found? Request a Course