Machine learning multi-attribute prediction for reservoir properties with Emerge
This course covers the theory and practical use of Emerge. Emerge is an interactive program that is fully linked within HampsonRussell Software and performs multi-attribute seismic analysis for seismic reservoir characterization using multivariate statistics and neural networks
Learning Objectives
- A comprehensive overview of the generation of seismic attributes
- Provides a mechanism for the user to derive complex relationships between seismic attributes and petrophysical parameters
- Understanding of how to recognize reliable attributes when estimating reservoir parameters.
- Basic theory of neural network technologies
- Application of neural network technology in well log prediction, petrophysical volume generation, and seismic lithology classification
- Structured to teach theory alongside practical exercises, equipping the user in the operation of the Emerge software
Audience
Geophysicists, geologists, engineers, and technical staff who want to understand the theory and learn how to apply these increasingly critical techniques.
Content
Topics covered in this course include:
- Theory of seismic attributes, linear, non-linear, and neural network methodologies for attribute selection, cross-validation, and attribute ranking
- Application of attributes to convert seismic data volumes into geological or petrophysical volumes.
- Application of attributes to predict missing log data
- Attributes exercises using seismic data and well logs
Duration: 1 day Software used: HampsonRussell Emerge
Course Format: Instructor-led, workflow-based, in-person classroom or virtual Prerequisites:Experience with HampsonRussell Software is not a prerequisite for this workshop, although students will become familiar with the functionality through attendance.