GAN-Based Carbon Storage Box Identification in Fault Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to accurately identify and analyze CO2 storage conditions in fault zone areas due to insufficient analysis precision and disturbance of fault zones on seismic waveforms, making it difficult to assess carbon storage capabilities.
Innovation Solution
A method utilizing a GAN network to process pre-stack single-shot seismic data and well logging data, performing pre-stack time migration, building an isochronous stratigraphic framework, and using a background waveform data filling model to generate fine stable sedimentary background seismic waveform data, which helps in identifying carbon storage boxes by calculating wave impedance and delineating their geometric structure and internal characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional seismic data processing methods are used, then the analysis process is simple, but the analysis precision of fault zone areas is insufficient
Solution Approach 1:
The patent introduces a GAN-based background waveform data filling model as an intermediary tool between seismic data processing and fault zone analysis. This model generates synthetic background waveforms that represent normal sedimentary conditions, which are then subtracted from actual seismic data to highlight fault zone characteristics. The intermediary model enables precise identification of fault zones without requiring complex direct analysis methods.
Solution Approach 2:
The patent replaces traditional mechanical seismic data processing techniques with a neural network-based computational approach. Instead of relying on conventional signal processing algorithms that struggle with fault zone complexity, the system uses deep learning models (GANs) to automatically learn and subtract background patterns, thereby achieving superior analysis precision in fault zone areas.
2Adaptability or versatility
If fault zone areas are included in seismic data analysis, then the coverage is comprehensive, but the disturbance of fault zones makes it difficult to analyze CO2 storage conditions
Solution Approach 1:
The patent extracts and removes the harmful influence of fault zones from the seismic data analysis by using the GAN model to generate and subtract background waveforms. The background waveform data filling model learns from fault-free areas and generates synthetic waveforms that represent normal conditions. By subtracting these synthetic waveforms from actual data, the method effectively removes fault zone disturbances while maintaining comprehensive coverage.
Solution Approach 2:
The patent converts the harmful disturbance caused by fault zones into a beneficial analysis opportunity. Instead of trying to directly analyze the complex fault zone signals, the system uses the GAN model to create synthetic background waveforms from fault-free areas. The difference between actual and synthetic waveforms then highlights fault zone characteristics, transforming the problem of fault zone disturbance into a method for identifying and characterizing fault zones for CO2 storage assessment.
3Quantity of substance
If the background waveform data filling model is trained on limited stable sedimentary area data, then the training data requirement is reduced, but the model accuracy may be compromised
Solution Approach 1:
The patent applies dimensionality change by transitioning from 2D seismic profiles to 3D volumetric data processing. The GAN model processes seismic data in three dimensions, allowing it to learn from stable sedimentary areas and generate background waveforms throughout the entire study volume. This dimensional approach enables the model to extrapolate accurately from limited training data to large untrained regions, maintaining high model accuracy while reducing the quantity of required training data.
Data Source
AI summary
The present disclosure belongs to the field of capture, utilization, and storage of carbon dioxide, particularly relates to a method and system for intelligently identifying a carbon storage box based on a GAN network, and aims at solving the problem that the analysis accuracy of a fault zone area in the prior art is insufficient. The method comprises the steps: delineating seismic waveform data of a stable sedimentary area through a GAN network, and removing seismic waveform data points in the fault zone area; obtaining a stable sedimentary background seismic waveform data invertomer; obtaining a three-dimensional wave impedance prediction data volume; making a difference to obtain an abnormal wave impedance data volume; retaining abnormal wave impedance data of fault-karst in the three-dimensional variance attribute volume to obtain a fault-karst wave impedance data volume; and then obtaining a carbon storage box interpretation model.