Geologic Models for Seismic Interpretation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current geophysical inversion methods in hydrocarbon exploration face challenges due to non-uniqueness and incomplete information, leading to inaccurate subsurface modeling and fluid prediction, particularly in complex features like faults and salt structures, and rely on simplistic priors lacking geometric information.
Innovation Solution
A computer-implemented method that performs data preparation and machine learning to generate initial geologic models incorporating geometric information and AVO behaviors, allowing for dimensionality reduction and improved inversion processes, including the use of neural networks for batch prediction and dynamic simulation in hydrocarbon management and carbon sequestration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional geophysical inversion methods are used, then the process is simpler and faster to implement, but the accuracy of subsurface modeling and fluid prediction deteriorates due to non-uniqueness and incomplete information
Solution Approach 1:
The patent applies preliminary action by generating initial geologic models with geometric information and AVO behaviors before performing the main inversion process. This pre-processing step prepares the data in advance, reducing the complexity and improving the accuracy of the subsequent inversion by providing better starting conditions and constraints.
Solution Approach 2:
The patent introduces neural networks as an intermediary between the seismic data and the inversion process. These neural networks capture complex AVO and geometric relationships, acting as a mediator that transforms raw seismic data into more informative representations that improve inversion accuracy while managing computational complexity.
2Measurement precision
If simplistic priors are used in inversion, then the process is easier to implement, but the prediction accuracy deteriorates due to lack of geometric information
Solution Approach 1:
The patent applies dimensionality change by incorporating geometric information into the prior models. Instead of using only simple statistical priors, the method adds spatial and geometric dimensions to the priors, allowing them to capture complex subsurface structures and improve prediction accuracy while maintaining manageable complexity through systematic approaches.
3Measurement precision
If detailed analysis of seismic volumes is performed, then the accuracy of hydrocarbon prediction improves, but the cycle time for seismic interpretation increases
Solution Approach 1:
The patent applies preliminary action by performing batch predictions using trained neural networks before detailed analysis. This pre-computation step generates initial results that guide subsequent detailed analysis, reducing the overall cycle time by avoiding exhaustive detailed analysis of all seismic volumes and focusing computational effort only where needed.
Data Source
AI summary
A method and a system for generating one or more initial geologic models for hydrocarbon management or carbon capture sequestration is disclosed. The initial geologic models may be generated that describe distribution of AVO (Amplitude-Variation-with-Offset) behaviors of one or more subsurface features, such as sands or rock types, into one or more probability volumes, resulting in a dimensionality reduction of information. The initial geologic models may be used in a variety of contexts, such as for geologic interpretation or as prior information for input to a seismic inversion process.