Wavelet Transform Coherent Noise Reduction in Distributed Acoustic Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed Acoustic Sensing (DAS) data collected in hydrocarbon wellbores often suffers from noise artifacts, particularly coherent noise caused by cable slapping and zig-zag patterns, which reduce the accuracy of seismic data and hinder hydrocarbon recovery operations.
Innovation Solution
The implementation of a continuous wavelet transform (CWT) method to identify and mitigate coherent noise in DAS shot gathers by suppressing noise wavelet coefficients, using reference coefficients from less noisy traces, and applying inverse CWT to obtain noise-reduced seismic data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If DAS data is collected using fiber optic cable in subterranean operations, then seismic data can be detected along the wellbore, but coherent noise artifacts (cable slapping and zig-zag patterns) are introduced that reduce data accuracy
Solution Approach 1:
The patent extracts and removes coherent noise components from DAS seismic data through signal processing techniques. Specifically, it identifies and separates the harmful coherent noise (cable slapping and zig-zag patterns) from the useful seismic signals, then eliminates the noise components while preserving the actual formation responses, thereby improving measurement precision without sacrificing data collection capability
Solution Approach 2:
The patent introduces intermediate processing steps between data collection and final analysis. It uses reference traces and correlation techniques as mediators to identify and mitigate coherent noise. The reference traces serve as intermediaries to detect noise patterns, and the correlation process acts as an intermediary mechanism to separate noise from actual seismic signals, allowing accurate subsurface evaluation despite the presence of cable-induced artifacts
2Object-generated harmful factors
If conventional noise filtering methods are applied to DAS data, then some noise is reduced, but significant seismic signals may also be attenuated along with the noise
Solution Approach 1:
The patent applies local quality by treating different portions of the DAS data differently based on their characteristics. It identifies specific time windows and depth intervals where coherent noise is predominant versus where actual seismic signals are strong. The noise mitigation is applied locally in the time-depth domain, adjusting processing intensity based on the local signal-to-noise ratio, thereby reducing noise in affected regions while preserving seismic signals in clean regions
Solution Approach 2:
The patent uses partial action by applying noise mitigation only to specific coherent noise components rather than uniformly filtering all frequencies. It selectively targets the coherent noise patterns (cable slapping and zig-zag) identified through reference trace correlation, applying attenuation only where and when these specific noise patterns are detected, rather than applying broad-spectrum filtering that would attenuate valuable seismic signals
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively reduces coherent noise in DAS data, enhancing the precision of seismic profiles and facilitating more accurate hydrocarbon recovery operations by improving the quality of seismic data used for subsurface formation evaluation.
Implementation Method 1
converting, from a time domain to a wavelet domain, traces of a set of channels of a plurality of channels within the first region
Implementation Method 2
applying an inverse wavelet transform to the modified wavelet domain traces of the set of channels to generate time domain denoised traces
Implementation Method 3
DAS may be used to acquire the seismic data necessary to form the VSP. Acoustic sensing based on DAS may use the Rayleigh backscatter property of a fiber's optical core
Data Source
AI summary
A distributed acoustic sensing (DAS) system is coupled to an optical fiber along a plurality of channels. The system generates a DAS seismic profile of the subsurface formation based on detected seismic data, identifies at least one region having coherent noise, and identifies which of the plurality of channels are within the identified at least one region. For each trace of data associated with the plurality of noisy channels, the system converts, from a time to a wavelet domain, the trace of data and a reference trace having less coherent noise, and suppresses the wavelet coefficients of the trace of data based on the wavelet coefficients of the reference trace. After the system mitigates the noise in the wavelet domain, an inverse wavelet transform is applied to the trace of data to convert back to the time domain and create a reduced noise DAS seismic profile.