Seismic Imaging with Synthetic Data for Multiples Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Previous seismic survey techniques fail to compensate for missing recorded data when processing free-surface multiples as signal, leading to noise and limited seismic image quality, particularly in resolution and areal illumination.
Innovation Solution
A method that generates synthetic data to compensate for incomplete seismic recordings by using a one-way or two-way wave propagator based on a current model of the subsurface structure, modifying it to reduce differences between synthetic and observed data while maintaining the velocity component unchanged, allowing for improved processing of both primaries and free-surface multiples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If free-surface multiples are used as signal to improve seismic image quality, then resolution and areal illumination are enhanced, but noise and cross-talk are introduced due to missing recorded data
Solution Approach 1:
A synthetic data generator acts as an intermediary component that creates complete seismic data by combining recorded data with synthetically generated data. This mediator fills in the missing data gaps and enables proper processing of free-surface multiples without introducing noise, as the synthetic data is generated from a subsurface model rather than from incomplete recorded data.
Solution Approach 2:
The patent creates a synthetic copy of the complete seismic data by generating data from a subsurface model. This synthetic data copy is then used to fill in missing recorded data and to properly process free-surface multiples, allowing the system to work with complete data representations without the noise issues associated with incomplete recorded data.
2Loss of information
If more seismic sources and receivers are deployed to improve data coverage, then missing data is reduced, but survey complexity and cost increase
Solution Approach 1:
The synthetic data generator is a self-service system that automatically generates complete seismic data from a subsurface model without requiring additional physical sources or receivers. The system self-completes the data by synthesizing missing portions based on the model, eliminating the need for more complex survey deployments.
Solution Approach 2:
The patent replaces the mechanical deployment of additional seismic sources and receivers with a computational system that generates synthetic data. Instead of physically adding more survey equipment to improve data coverage, the system uses computer-generated synthetic data to fill in missing information, thereby reducing survey complexity.
3Ease of manufacture
If recorded free-surface multiples are mapped into the subsurface without data compensation, then processing is simplified, but image quality is degraded due to incorrect mapping
Solution Approach 1:
The synthetic data generator serves as an intermediary that properly maps free-surface multiples into the subsurface by using a subsurface model as a guide. This mediator ensures correct mapping by reference to the model, preventing the incorrect mapping that occurs when processing incomplete recorded data directly.
Solution Approach 2:
The system uses feedback from the subsurface model to guide the mapping of free-surface multiples. The synthetic data generator continuously refines the mapping process by comparing recorded data with model predictions, ensuring that multiples are correctly positioned and mapped into the subsurface structure, thereby improving image quality.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances seismic image quality by treating free-surface multiples as signal, increasing resolution and areal illumination, and enabling the use of fewer or more widely spaced seismic sources and receivers without compromising image quality.
Implementation Method 1
generates synthetic survey data using a one-way or two-way wave propagator based on a current model of a target structure
Implementation Method 2
seismic sources are activated to generate seismic waves directed into a subsurface structure. The seismic waves generated by the seismic sources travel into the subsurface structure of the Earth, with a portion of the seismic waves being reflected back by the subsurface rock structures to the surface
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Synthetic survey data is generated using a two-way or one-way wave propagator based on a current model of a target structure. The current model is modified to reduce a difference between the synthetic survey data and observed survey data, while maintaining unchanged a velocity component of the current model, where the modifying of the current model produces a modified model. The modified model is used to reduce an adverse effect of multiples in the target structure, or to promote a favorable effect of multiples in the target structure.