This training course is available upon request. When requesting a course, please know there is a minimum number of participants needed for the course to be held. Request more information in the form to notify our team of your interest.
This course presents an introduction to the practical use of deep neural networks (DNNs) to predict elastic and rock properties as well as an overview of relevant machine learning theory. In these supervised learning workflows, the relationship mapping the seismic to the properties of interest are learned from the data itself. Key to deriving robust operators is big data. This course shows how to generate and incorporate synthetic data (WellGen) in the machine learning workflow (Emerge) to get more reliable estimates.
Geophysicists, geologists, engineers and technical staff who want to understand the theory and learn how to apply these increasingly critical techniques. It would be helpful for the student to have experience with HampsonRussell AVO, Strata and Emerge as thissoftware is used in the exercises. Having said this, the exercises are self-contained and do not require prior knowledge of this software.
Introduction to deep neural networks (DNNs) and the need for big datasets.
Creating big data by generating synthetic well and seismic data.
Examples showing how the inclusion of synthetic data improves the property estimates.
Lithofacies Classification, Rock Physics Modeling, Statistics, Variogram modelling, Well Simulations.
Generating synthetic seismic data using AVO modelling
Seismic data preparation including correlation, generation of angle gathers and scaling.
Supervised learning, multi-linear regression and deep neural networks.
Generation and selection of seismic attributes for use with supervised learning.
Quality controlling the results including cross-plotting.
Explains both the theory and practice of deep neural networks using WellGen and Emerge.
The workshop demonstrates a complete workflow showing how to estimate elastic (e.g. P-wave impedance, Vp/Vs and density) and rock properties (e.g. Porosity, Saturation) using both real and synthetic data. These data science-based (DNN) estimates are compared with theory-based (inversion) estimates.
Teaches the user how to apply WellGen and Emerge using a real Gulf of Mexico gas sand example.
Duration: 1 day Software used: HampsonRussell WellGen Course Format: Instructor-led, workflow-based, in-person classroom or virtual Prerequisites: It would be helpful to have experience with HampsonRussell AVO, Strata and Emerge as these software modules are used in the exercises. Having said this, the exercises are self-contained and do not require prior knowledge of the software.