Subsurface Velocity Model Spectral Parameterization for FWI Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seismic exploration methods face challenges in efficiently obtaining accurate initial velocity models for subsurface structures, which are crucial for reducing computational burden and ensuring convergence in Full Waveform Inversion (FWI) processes.
Innovation Solution
An apparatus and method utilizing a Gauss Newton method through spectral parameterization of the velocity model, including a nonvolatile memory, processor, display, and user input unit, to optimize the velocity model by repeatedly updating it until residual minimization is achieved.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a good initial velocity model is used in FWI, then the number of iterations is reduced and convergence is guaranteed, but obtaining an accurate initial velocity model requires complex preprocessing and increases device complexity
Solution Approach 1:
The patent applies preliminary action by performing spectral parameterization and Gauss-Newton optimization to generate an accurate initial velocity model before the main FWI process. This preprocessing step creates a better starting point for inversion, ensuring convergence while reducing the number of iterations needed in the subsequent FWI process.
Solution Approach 2:
The patent introduces spectral parameterization as an intermediary method between raw seismic data and the final velocity model. By parameterizing the velocity model in the spectral domain and optimizing it using Gauss-Newton method, it creates an intermediate representation that facilitates convergence in the main FWI process.
2Measurement precision
If spectral parameterization with Gauss Newton method is applied to obtain accurate initial velocity model, then inversion accuracy is improved, but computational burden increases
Solution Approach 1:
The patent applies parameter changes by transforming the velocity model into the spectral domain through spectral parameterization. This changes the representation of the velocity model from spatial to frequency domain, allowing the Gauss-Newton method to more efficiently optimize the initial model and improve accuracy while managing computational requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus and method for optimizing a subsurface velocity model are provided to be able to optimize a velocity model by applying the Gauss Newton method through parameterization of the velocity model in exploration seismology.