Laplace Domain Seismic Imaging Gradient Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Full waveform inversion methods face limitations due to local minima in their objective functions and insufficient low-frequency information, particularly in Laplace-domain seismic imaging, which can produce artifacts near the seafloor and require longer streamers for deep-water structure recovery.
Innovation Solution
A new gradient scaling method for Laplace-domain waveform inversion is proposed, calculating a scaled gradient direction using the sum of squares of conventional gradient values accumulated with depth to enhance recovery of high-velocity deep-water structures while suppressing seafloor artifacts, utilizing a seismic imaging system with a scaled gradient calculating unit and modeling parameter updating unit for iterative processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Laplace-domain waveform inversion is used to recover subsurface information, then low-frequency information is improved and local minima are reduced, but artifacts are produced near the seafloor and longer streamers are required
Solution Approach 1:
The patent applies local quality by using depth-dependent gradient scaling where the scaling factor varies with depth. The accumulated gradient is calculated differently at different depths, allowing the inversion to adapt to local variations in the subsurface structure. This enables better resolution of deep-water structures while suppressing seafloor artifacts through localized optimization of the gradient direction at each depth level.
Solution Approach 2:
The patent changes the parameter of gradient scaling by introducing a depth-dependent scaling factor that modifies the gradient direction at each iteration. The scaling factor is calculated using the accumulated gradient from previous iterations, dynamically adjusting the inversion process to prioritize deep structures while reducing sensitivity to seafloor irregularities.
2Device complexity
If conventional gradient inversion is used, then computation is simpler, but recovery of deep-water structures is insufficient and seafloor artifacts are not suppressed
Solution Approach 1:
The patent applies preliminary action by calculating the accumulated gradient before using it to scale the current gradient. The accumulated gradient is computed from previous iterations and stored for use in subsequent iterations. This preliminary computation of the scaling factor enables the inversion to progressively enhance deep-water structure recovery while suppressing seafloor artifacts without requiring complex real-time calculations during the inversion process.
Data Source
AI summary
The description relates to a seismic imaging technology technique for modeling a subsurface structure through waveform inversion in the Laplace domain. The seismic imaging system comprises a scaled gradient calculating unit calculating a scaled gradient, a modeling parameter updating unit updating the model parameters using the scaled gradient direction, and an iteration control unit controlling the scaled gradient calculating unit and the modeling parameter updating unit to repeat processing iteratively until a stopping criteria is met.