Velocity Model Uncertainty Estimation via Seismic Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining the uncertainty of a velocity model in seismic imaging are limited, as they rely on well data that can only be applied at discrete depths, resulting in a best-guess seismic velocity estimate without uncertainty, which is not comprehensive.
Innovation Solution
A method that involves constructing a velocity model from seismic data, performing seismic migration, generating a semblance panel, and estimating uncertainty by taking the derivative of the velocity model output to define lower and upper velocity bounds, which vary across different depths, allowing for a more accurate representation of seismic velocity uncertainty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods using well data are used to determine velocity model uncertainty, then the method is simple to implement, but the measurement precision is insufficient because uncertainty cannot be determined at continuous depths
Solution Approach 1:
The patent replaces the mechanical/well-based measurement system with a seismic signal processing system. Instead of relying on physical well data at discrete depths, the invention uses seismic migration data and semblance analysis to continuously determine velocity model uncertainty across all depths, substituting a computational approach for the limited physical measurement approach
Solution Approach 2:
The patent transitions from one-dimensional discrete depth measurements (well data at specific depths) to two-dimensional continuous depth coverage by incorporating the time dimension through seismic migration. The semblance analysis across migrated seismic data provides velocity uncertainty information at every depth level, not just at discrete well locations
2Loss of information
If well data is used to determine velocity uncertainty, then the data acquisition is straightforward, but the information completeness is insufficient because only discrete depth points are covered
Solution Approach 1:
The patent makes the seismic migration process multi-functional by extracting not only the velocity model itself but also the velocity model uncertainty from the same migrated seismic data. The semblance analysis applied to the migration output simultaneously provides both velocity information and uncertainty information across all depths, eliminating the need for separate data acquisition campaigns
3Reliability
If a best-guess velocity estimate is provided without uncertainty bounds, then the result is simple to obtain, but the reliability is insufficient for accurate hydrocarbon exploration
Solution Approach 1:
The patent implements feedback by using the migrated seismic data to evaluate and quantify the reliability of the velocity model through semblance analysis. The process feeds back uncertainty information to the velocity model, allowing users to assess confidence levels and make more reliable exploration decisions based on quantified uncertainty rather than simple point estimates
Data Source
AI summary
A method for estimating uncertainty of an output of a velocity model of a subsurface region includes receiving seismic data associated with a subsurface region and captured by one or more seismic receivers, constructing a velocity model of the subsurface region based on the received seismic data, performing a seismic migration of the received seismic data based on the constructed velocity model to obtain migrated seismic data, generating a semblance panel from the migrated seismic data, and estimating an uncertainty of the output of the velocity model based on the generated semblance panel.