Multi-Stage Seismic Interpolation for Weak Energy Mode Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Acquired seismic data often loses weaker energy modes due to the dominance of stronger modes, leading to biased interpolation and loss of coherent energy, which affects the quality and resolution of post-processing steps.
Innovation Solution
A multi-stage process using Matching Pursuit Fourier Interpolation (MPFI) techniques combined with surface-wave analysis, modeling, and inversion (SWAMI) to progressively reconstruct and remove specific seismic energy modes, ensuring other modes are preserved and accurately interpolated.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If single-stage interpolation is performed on seismic data, then processing speed is improved, but weaker energy modes are lost due to dominance of stronger modes
Solution Approach 1:
The patent divides the interpolation process into multiple stages, where each stage handles a specific subset of energy modes. The seismic data is processed in stages, with each stage focusing on interpolating weaker modes after stronger modes have been handled, thereby preventing loss of weaker energy modes while maintaining processing efficiency.
Solution Approach 2:
The patent performs preliminary processing to identify and separate different energy modes before interpolation. By pre-characterizing the seismic data to distinguish between strong and weak energy modes, the system can apply appropriate interpolation strategies to each mode, ensuring weaker modes are preserved.
2Device complexity
If interpolation is performed in the presence of strong noise, then processing simplicity is maintained, but interpolation quality deteriorates due to biased sparse priors
Solution Approach 1:
The patent extracts and removes strong noise components from the seismic data before performing interpolation. By separating the noise from the signal, the system can apply interpolation to the cleaned data, resulting in higher quality results without the biasing effects of strong noise on sparse priors.
Solution Approach 2:
The patent performs preliminary noise removal and data cleaning operations before the main interpolation process. This preliminary action prepares the data by eliminating strong noise components, ensuring that subsequent interpolation operates on cleaned data and produces accurate results.
3Loss of time
If all energy modes are interpolated simultaneously, then processing time is reduced, but resolution of post-processing steps deteriorates
Solution Approach 1:
The patent segments the interpolation process into multiple stages, where each stage processes specific energy modes. This segmentation allows the system to handle different modes with appropriate detail, improving the resolution of post-processing steps while managing processing time through structured execution.
Solution Approach 2:
The patent implements a dynamic interpolation strategy where the processing approach adapts to the characteristics of different energy modes. By adjusting the interpolation parameters and methods based on the specific mode being processed, the system optimizes both time and resolution requirements.
Data Source
AI summary
A method for generating resolved data is disclosed. The method receives captured data in a first signal space from sensors at a resource site and determines a signal characteristic associated with a first signal component, a second signal component, or a noise component within the captured data. The method transforms the captured data from the first signal space to a second signal space using a first transform operator. The method further extracts a first signal component from the transformed captured data in the second signal space. The extracted first signal component may be transformed back to the first signal space to generate a first extracted data which may be subtracted from the captured data. The method reconstructs the first extracted data to generate a first reconstructed data included in the resolved data. The resolved data includes a minimal amount of a noise component associated with the captured data.