Subsurface Imaging Using Accumulated Wavefield Energy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing subsurface imaging techniques face challenges in accurately estimating seismic propagation velocity using waveform inversion, particularly due to the inefficiency in utilizing low-frequency components and the lack of a unique solution in iterative least-squares methods, which often require extreme assumptions and prior geological information.
Innovation Solution
A subsurface imaging apparatus and method that acquires an observed wavefield, calculates accumulated energies of both observed and modeled wavefields, and updates characteristic parameters to minimize the difference between these energies, using a gradient-based approach with a Jacobian calculation to iteratively refine the seismic velocity model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative least-squares method is used for waveform inversion, then subsurface structure can be estimated, but unique solution does not exist and extreme assumptions are required
Solution Approach 1:
The patent transforms the waveform inversion problem by changing the parameter space from traditional seismic wavefield parameters to accumulated energy parameters. This transformation creates a new formulation where the objective function is based on accumulated energy differences between observed and modeled wavefields, enabling unique solution identification without requiring extreme assumptions about subsurface structure.
Solution Approach 2:
The patent replaces the traditional mechanical iterative least-squares optimization system with an energy-based accumulation approach. Instead of minimizing squared residuals through complex iterative adjustments, the method accumulates energy over time and minimizes the difference in accumulated energies, simplifying the optimization process while maintaining measurement precision.
2Measurement precision
If low frequency components are used in waveform inversion, then subsurface imaging can be performed, but insufficient low frequency components reduce imaging accuracy
Solution Approach 1:
The patent applies preliminary action by accumulating energy over time before performing the inversion analysis. This time accumulation process enhances the low frequency components in the observed wavefield, effectively preparing the data to contain sufficient low frequency information even when the original seismic data is insufficient, thereby improving velocity model accuracy.
Solution Approach 2:
The patent implements continuity of useful action through the time accumulation of energy. By continuously integrating energy over the entire time series rather than using instantaneous wavefield values, the method continuously enhances the signal-to-noise ratio and preserves low frequency components throughout the inversion process, maximizing the utilization of available low frequency information.
3Productivity
If traditional waveform inversion is used, then seismic data can be processed, but reliance on prior geological information is required
Solution Approach 1:
The patent enables self-service by formulating the inversion problem in terms of accumulated energy differences, which naturally incorporates the seismic data's own characteristics without requiring external prior geological information. The method uses the data's inherent energy accumulation properties to drive the inversion process, allowing the system to serve itself without relying on pre-assumed geological models.
Data Source
AI summary
Provided is an apparatus of imaging a subsurface of a target area. The apparatus includes: an observed wavefield acquiring unit configured to acquire an observed wavefield for a target area based on seismic data; a parameter storage configured to store a characteristic parameter for the target area; a modeled wavefield creator configured to create a modeled wavefield corresponding to the observed wavefield using the characteristic parameter stored in the parameter storage; an energy calculator configured to calculate accumulated energies of the observed wavefield and the modeled wavefield, wherein the accumulated energies are defined as energies accumulated over time with respect to the observed wavefield and the modeled wavefield; and a parameter updating unit configured to update the characteristic parameter stored in the parameter storage such that a difference between the accumulated energy of the observed wavefield and the accumulated energy of the modeled wavefield is minimized.