Analytic Filter Aligns Seismic Traces for 4D Subsurface Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 4D seismic data processing techniques face challenges in accurately estimating time shifts due to ill-posed inversion problems and non-linearities, particularly in subsidence scenarios with large time shifts, leading to multiple solutions and interpretation difficulties.
Innovation Solution
The method employs an analytic filter based on plane wave destruction filters parameterized by time shifts, which are a time domain analogue of frequency domain phase shift operators, to align seismic traces from base and monitor surveys, allowing for more efficient and accurate estimation of time shifts by reducing noise and improving computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional inversion techniques are used to estimate time shifts, then the method can be applied to 4D seismic data, but the estimation becomes ill-posed with multiple solutions and interpretation difficulties
Solution Approach 1:
The patent introduces an intermediary all-pass filter as a mediator between the base and monitor surveys. This filter serves as a transformation tool that converts the ill-posed inversion problem into a well-posed filtering problem, eliminating multiple solutions while preserving time shift estimation accuracy
Solution Approach 2:
The patent transforms the problem by changing parameters from direct time shift estimation to filter coefficient determination. By parameterizing the all-pass filter with coefficients that can be uniquely estimated from data, the method converts an ill-posed problem with multiple solutions into a well-posed problem with a unique solution
2Measurement precision
If cross-correlation methods are used to align seismic traces, then time shifts can be estimated, but computational complexity increases and noise sensitivity rises
Solution Approach 1:
The patent replaces the mechanical cross-correlation process with an all-pass filter-based approach. Instead of computationally intensive cross-correlation operations, the method uses filter coefficient estimation followed by simple filter application, significantly reducing computational complexity while maintaining accuracy
Solution Approach 2:
The patent extracts the essential time shift information through filter coefficient estimation, separating the core alignment function from complex computational procedures. This extraction allows time shift estimation without requiring full cross-correlation computations
3Reliability
If velocity changes are used to characterize reservoir evolution, then subsurface changes can be detected, but large time shifts in subsidence scenarios cause non-linearities and estimation errors
Solution Approach 1:
The patent applies a dynamic all-pass filter that can adapt to varying time shifts across different time samples and spatial locations. This dynamic filtering approach handles large time shifts in subsidence scenarios without suffering from non-linearities, maintaining estimation accuracy where conventional methods fail
Data Source
Figure 1a
Figure 1b
Figure 2a~2b
AI summary
Disclosed is a method for characterising the evolution of a subsurface volume. The method comprises providing a first and second surveys of the reservoir with a first and second sets of seismic traces, constructing an analytic filter operable to shift one or more seismic traces in dependence of a model parameter; and performing an inversion to obtain estimates of the model parameter such that the analytic filter aligns said first survey and said second survey. The analytic filter may be a plane wave destruction filter, and be such that the model parameter is a function of the analytic filter and is not a function of one or more of said second set of seismic traces.