Time-Variant Wavefield Propagation for Accurate Subsurface Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seismic data processing methods, such as Full Waveform Inversion (FWI), struggle to accurately calculate modeled data, leading to inaccurate images of subsurface formations, particularly in complex 3D models, and fail to represent amplitude variations with reflection angle accurately.
Innovation Solution
Propagate wavefields through a parameter volume that varies as a function of travel-time, using elastic, acoustic, or elasto-acoustic propagators, allowing for iterative updates of the earth model to minimize misfit between recorded and modeled data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Full Waveform Inversion (FWI) methods are used to generate modeled data, then the processing can be completed with standard static subsurface models, but the accuracy of the generated images and amplitude variations are insufficient
Solution Approach 1:
The patent applies the dynamics principle by transitioning from static subsurface models to dynamic models where earth parameters (velocity, density) vary as a function of travel-time. This allows the model to adapt and change during wavefield propagation, capturing the time-variant nature of seismic wave interactions with subsurface formations, thereby improving image accuracy and amplitude representation.
Solution Approach 2:
The patent implements parameter changes by making earth parameters functions of travel-time rather than static values. The velocity model v(x,t) and density model ρ(x,t) are updated iteratively based on travel-time information from wavefield propagations, allowing the model parameters to evolve and improve accuracy throughout the inversion process.
2Reliability
If static subsurface models are used for wavefield propagation, then the computational process is simpler, but the amplitude variations with reflection angle are not accurately represented
Solution Approach 1:
The patent makes the propagation model dynamic by using time-variant earth parameters that change during wavefield propagation. This dynamic approach accurately captures amplitude variations with reflection angle by allowing the velocity and density models to evolve based on travel-time, improving reliability of amplitude representation.
Solution Approach 2:
The patent implements feedback through iterative updates of the earth model using misfit functions that compare recorded data with modeled data. The travel-time information from each wavefield propagation feeds back into updating the velocity and density models, creating a closed-loop system that progressively improves amplitude variation accuracy.
3Measurement precision
If conventional FWI with static models is applied, then the computational cost is lower, but the misfit between recorded and modeled data remains high
Solution Approach 1:
The patent applies preliminary action by using travel-time information from initial wavefield propagations to guide subsequent model updates. The travel-time data is extracted and used to inform the next iteration of the inversion process, allowing the model to progressively converge to a better solution that reduces data misfit.
Solution Approach 2:
The patent maintains continuity of useful action through iterative wavefield propagations and model updates. Each iteration builds upon the previous one, with the earth model continuously evolving based on accumulated travel-time information, ensuring progressive reduction of misfit between recorded and modeled data.
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 generates more accurate images of subsurface formations by improving the representation of amplitude variations with reflection angle, enhancing the quality of seismic data interpretation.
Implementation Method 1
waves 112 emitted by the source 110, at a known location, penetrate an explored formation 121-127 and are reflected, refracted or diffracted at interfaces 120, 122, 124, and 126
Implementation Method 2
waves 112 emitted by the source 110, at a known location, penetrate an explored formation 121-127 and are reflected, refracted or diffracted at interfaces 120, 122, 124, and 126
Implementation Method 3
waves 112 emitted by the source 110, at a known location, penetrate an explored formation 121-127 and are reflected, refracted or diffracted at interfaces 120, 122, 124, and 126
Data Source
AI summary
A method for generating an image (IF) of a subsurface includes receiving an initial earth model of the subsurface, receiving a recorded dataset d associated with the subsurface, generating a modeled dataset p based on the initial earth model and recording positions corresponding to the recorded dataset d, wherein the modeled dataset p is calculated based on a parameter volume PV that varies in time, updating the initial earth model, to generate an updated earth model, based on a misfit function that depends on the recorded dataset d and the modeled dataset p, and generating the image IF of the subsurface based on the updated earth model.