Sonic Wavefield Separation via Cross-Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for wavefield separation in sonic data analysis, particularly in the oil and gas industry, face challenges in accurately estimating direct phases to extract event signals, which are often overwhelmed by direct phases in waveforms.
Innovation Solution
The method employs cross-correlation of waveform traces at adjacent sensor locations to estimate and remove direct phases, using polynomial fitting and time shift corrections to improve accuracy, and applies a k-f filter to select up-going or down-going waves, enabling effective wavefield separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods (f-k filter, median filter, parametric estimation) are used to remove direct phases, then processing can be performed with standard algorithms, but the accuracy of direct phase estimation deteriorates, leading to poor event signal extraction
Solution Approach 1:
The patent introduces cross-correlation as an intermediary technique between conventional filtering methods and direct phase estimation. By using cross-correlation of waveform traces at adjacent sensor locations, the method obtains a reference direct phase signal that mediates the estimation process, improving accuracy without requiring complex adaptive algorithms
Solution Approach 2:
The method creates a copy of the direct phase signal through cross-correlation operations on adjacent traces. This copied reference signal is then used to estimate and remove direct phases from the original waveforms, enabling accurate event signal extraction without directly processing the complex overlapping signals
2Loss of information
If direct phases are not removed from waveforms, then the complete waveform information is preserved, but event signals cannot be effectively extracted because they are overwhelmed by direct phases
Solution Approach 1:
The patent applies the extraction principle by separating and removing the direct phase component from the total waveform signal. Through cross-correlation-based estimation, the direct phases are extracted and subtracted, leaving the event signals (reflected and transmitted waves) isolated for accurate analysis
Solution Approach 2:
The waveform signal is segmented into distinct components: direct phases and event signals. The cross-correlation method enables identification and separation of these segments based on their different temporal and spatial characteristics, allowing selective processing of each component
Data Source
AI summary
A method for wavefield separation of sonic data is provided. The method comprises estimating direct phases of waveforms of sonic data observed with two or more sensors by using cross-correlation of waveform traces at adjacent sensor locations, removing the direct phases from the observed waveforms, and extracting event signals from the waveforms after removing the direct phases.