Acoustic Log Peak Tracking with Rock-Physics Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Acoustic logging in the oil and gas industry faces challenges in accurately tracking signal peaks due to noise and sudden changes in formation properties, leading to gaps or incorrect data in logs, particularly in identifying compressional wave velocities.
Innovation Solution
The method improves peak tracking by supplementing semblance methods with consistency checks based on rock-physics constraints and historical/future data, using dipole shear-wave slowness peaks and empirical relations to predict compressional wave velocities, thereby reducing false positives and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If semblance methods are used to detect signal peaks in acoustic logs, then automated peak identification is achieved, but false positives and gaps occur due to noise and sudden formation changes
Solution Approach 1:
The patent implements feedback mechanisms by using previously identified peaks to guide subsequent peak searches. The system uses a tracking algorithm that incorporates historical peak locations and formation property changes to predict and validate peak positions in real-time, reducing false positives while maintaining automation. This feedback loop allows the system to adapt to sudden formation changes and correct for noise interference.
Solution Approach 2:
The patent applies preliminary action by pre-establishing constraints based on rock-physics relationships and expected velocity ranges before peak detection occurs. These pre-defined physical constraints serve as validation criteria that filter out false peaks detected by the semblance method, ensuring that only physically plausible peaks are accepted. This preliminary framework is set up in advance to guide the automated detection process.
2Productivity
If automated tracking methods are used to identify compressional wave peaks, then processing efficiency is improved, but tracking accuracy deteriorates during sudden velocity changes
Solution Approach 1:
The patent implements dynamics by making the peak tracking algorithm adaptive to changing formation conditions. The system dynamically adjusts its search parameters and validation criteria based on detected changes in formation properties. When sudden velocity changes are detected, the algorithm automatically modifies its behavior to accommodate the new conditions while maintaining continuous tracking, thus preserving both efficiency and accuracy through dynamic adaptation.
Solution Approach 2:
The system uses feedback from velocity change detection to adjust tracking parameters in real-time. When the algorithm detects a sudden change in compressional wave velocity, it uses this feedback to recalibrate its peak identification criteria, ensuring accurate tracking continues despite the abrupt formation changes. This feedback mechanism allows the automated system to maintain precision during dynamic geological transitions.
3Measurement precision
If the compressional-wave signal is too weak to be detectable, then signal-to-noise ratio deteriorates, but the peak with highest velocity may be mistakenly identified as compressional peak
Solution Approach 1:
The patent introduces an intermediary validation step that uses rock-physics constraints and shear-wave velocity relationships to verify peak identifications. When the compressional wave signal is weak or undetectable, this intermediary system uses the known physical relationships between different wave types and formation properties to infer the correct peak location, preventing misidentification of other peaks as compressional peaks. This intermediary validation layer ensures reliability even when direct signal detection fails.
Solution Approach 2:
The system establishes preliminary expectations of compressional wave velocity ranges and characteristics before signal detection. These pre-defined physical constraints serve as a reference framework that helps identify and reject false peaks even when the actual compressional signal is weak or absent. The preliminary physical model provides a benchmark against which detected peaks are validated, ensuring correct identification despite poor signal-to-noise conditions.
Data Source
AI summary
Disclosed herein is an approach to improving the accuracy of signal-peak tracking in acoustic logs by supplementing a semblance method used to detect signal peaks in individual coherence-, correlation- or amplitude-based semblance maps with consistency checks based on rock-physics constraints and/or based on history and/or future data for the signal of interest.