Mitigating Cycle-Skipping in Full Waveform Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current full waveform inversion (FWI) methods face challenges in avoiding cycle-skipping, which leads to convergence to local minima due to the oscillatory nature of seismic data, resulting in inaccurate subsurface models.
Innovation Solution
A probabilistic approach is introduced to identify seismic traces prone to cycle-skipping by calculating a probabilistic measure based on the relationship between recorded and estimated seismic data, using statistical functions to determine travel-time differences and phase differences, and applying these measures to select and weight data for FWI, ensuring accurate subsurface modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional full waveform inversion is applied to seismic data, then subsurface imaging is performed, but cycle-skipping occurs leading to convergence at local minima
Solution Approach 1:
The patent applies preliminary quality control measures before performing full waveform inversion. It calculates travel-time differences between observed and modeled seismic data, identifies traces exceeding threshold values (indicating cycle-skipping), and excludes these problematic traces from the inversion process. This preliminary filtering action prevents cycle-skipping from causing convergence to local minima, thereby improving both imaging accuracy and convergence reliability.
2Quantity of substance
If all recorded seismic data is used in FWI, then more data is available for inversion, but traces with large travel-time differences cause cycle-skipping
Solution Approach 1:
The patent applies local quality control by evaluating each seismic trace individually for cycle-skipping conditions. It calculates travel-time differences for each trace against a threshold value and applies different treatment: traces within the threshold are included in FWI with full weight, while traces exceeding the threshold are excluded or down-weighted. This local differentiation ensures that high-quality traces contribute maximally to imaging precision while problematic traces do not degrade the overall result.
3Device complexity
If a fixed frequency cut-off is used in FWI, then computational complexity is reduced, but cycle-skipping cannot be effectively mitigated
Solution Approach 1:
The patent introduces dynamic adaptivity by making the frequency cut-off parameter variable rather than fixed. It calculates travel-time differences for each trace and dynamically adjusts the frequency cut-off based on these differences: traces with small travel-time differences use higher frequencies for better resolution, while traces with large differences apply lower frequency cut-offs or exclusion to prevent cycle-skipping. This dynamic approach maintains computational efficiency while effectively mitigating cycle-skipping across diverse trace conditions.
Data Source
AI summary
Computing device, computer instructions and method for identifying seismic traces prone to cycle-skipping in a full waveform inversion method. The method includes receiving recorded seismic data recorded with seismic sensors over a subsurface of interest; selecting a model that describes the subsurface; calculating, based on the model and the recorded seismic data, estimated seismic data; and choosing a probabilistic measure that characterizes a relationship between the recorded seismic data and the estimated seismic data. The probabilistic measure includes at least one statistical function.