DAS Zigzag Noise Mitigation via Autocorrelation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In fiber optic distributed acoustic sensing (DAS) for vertical seismic profile (VSP) data acquisition, inadequate acoustic contact between the fiber optic cable and the borehole wall leads to contamination of seismic signals with reverberatory noise, known as zigzag noise, which hinders accurate formation evaluation.
Innovation Solution
The implementation of a method to identify and remove zigzag noise by analyzing autocorrelation and crosswise lag summation functions to determine the noise's periodicity, followed by techniques such as predictive deconvolution, adaptive subtraction, time-reversed deconvolution, and the use of a time-limited comb operator to subtract the noise from the data, thereby preserving the seismic event's arrival time and improving data quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the fiber optic cable is not in adequate acoustic contact with the borehole wall, then the cable can be easily deployed and retrieved, but the seismic signal is contaminated with reverberatory noise (zigzag noise)
Solution Approach 1:
The patent extracts and removes the zigzag noise component from the DAS traces through spectral analysis and filtering techniques, separating the harmful noise from the useful seismic signal while preserving the ability to retrieve clean formation data
Solution Approach 2:
The patent converts the harmful zigzag noise into a detectable pattern through autocorrelation analysis, where the periodic nature of the noise becomes a identifiable feature that can be targeted for removal, transforming the problem into a solvable pattern recognition task
2Measurement precision
If the fiber optic cable is in good acoustic contact with the borehole wall, then the seismic signal quality is high, but the cable deployment and retrieval becomes more difficult
Solution Approach 1:
The patent segments the DAS data into different depth intervals and processes each segment separately, allowing identification of specific zones with zigzag noise problems while preserving data quality in good contact zones, enabling selective noise mitigation without affecting overall cable performance
3Object-affected harmful factors
If standard noise filtering techniques are applied to remove zigzag noise, then the noise is reduced, but the seismic event arrival time information may be distorted
Solution Approach 1:
The patent uses feedback mechanisms where the filtered DAS traces are compared with the original traces, and the filtering process is iteratively adjusted to minimize distortion of the seismic event arrival times while maximizing noise removal, ensuring preservation of critical timing information
Solution Approach 2:
The patent changes the parameter of noise periodicity detection by using autocorrelation analysis to identify the specific frequency characteristics of zigzag noise, allowing targeted filtering that preserves the lower frequency seismic signal components containing arrival time information
Data Source
AI summary
To mitigate zigzag noise and increase the quality of data provided from DAS VSP in wells with significant vertical sections, zigzag noise characteristics are identified and quantified. The zigzag noise properties can be extracted from an analysis of an autocorrelation of DAS VSP traces. The zigzag noise has a characteristic time period or repeat time delay that is the time period for the noise to propagate along the wireline through a zone of the wellbore with poor acoustic coupling between the fiber optic cable and formation. This period can be identified from analysis of the autocorrelation referred to herein as a crosswise lag summation function. The crosswise lag summation function identifies groups of DAS data traces containing zigzag noise and outputs zigzag noise periodicity for each group of traces. Once it has been identified, the zigzag noise can be removed from the VSP data and improve formation evaluation.