Acoustic Array Coherence Filtering for Signal Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for processing signal data from arrays, such as in acoustic logging, struggle to effectively filter out unwanted signals, especially when the signal-to-noise ratio is poor, leading to difficulties in extracting formation properties through poorly bonded or detached casings.
Innovation Solution
A coherence-filtering technique is applied in the frequency-wavenumber domain, where a coherence function is calculated and convolved with the signal data to suppress non-coherent signals, enhancing the coherence of the remaining data and allowing for the extraction of formation properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If routine semblance method is applied directly to data, then processing is simple, but formation signals are difficult to distinguish from noises due to low coherence
Solution Approach 1:
The patent applies coherence filtering as a preliminary processing step before semblance analysis. The coherence function is calculated from the raw data and used to weight the data prior to applying the semblance method, thereby pre-enhancing the visibility of coherent formation signals while suppressing incoherent casing signals and noise.
2Measurement precision
If maximum likelihood method is used to enhance resolution, then formation signal resolution improves, but low-coherence formation signal with poor signal-to-noise ratio remains difficult to resolve
Solution Approach 1:
The patent introduces a coherence function as an intermediary element that mediates between the raw data and the semblance processing. This coherence function acts as a weighting factor that enhances the contribution of coherent signals (formation waves) while suppressing incoherent signals (casing waves and noise), thereby improving the effective signal-to-noise ratio before final analysis.
3Measurement precision
If waveform subtraction method is applied to suppress casing signals, then formation signal coherence is enhanced, but the method does not work well when casing and formation signals overlap in time
Solution Approach 1:
The patent transforms the data from the time domain to the frequency-wavenumber (f-k) domain, where signals are represented by different frequency and wavenumber characteristics rather than just time. In this transformed domain, the coherence function can effectively distinguish between casing and formation signals based on their different propagation characteristics, even when they overlap in time.
Data Source
AI summary
A method waveform processing technique utilizing signal coherence of the array data for processing signals having poor signal-to-noise ratio. Raw waveform data is first transformed into f-k (frequency-wavenumber) domain. A coherence function is then calculated and convolved with the data in the f-k domain, which effectively suppresses non-coherent signals in the data. For the remaining coherent data, the unwanted part is muted and the wanted part is retained and inverse-transformed to yield the coherence-filtered array waveform data. After this processing, small signals that are hidden in the original data are extracted with much enhanced coherence. Subsequent processing of the data yields reliable information about formation acoustic property.