Concurrent Velocity Model and Depth Image Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional seismic imaging methods are limited in concurrently generating both a velocity model and a depth image, which are crucial for accurate subsurface exploration and geologic interpretation, often resulting in unsatisfactory depth images due to data or model-based issues such as noise and inaccurate velocity models.
Innovation Solution
The proposed method involves an iterative process that concurrently produces velocity models and depth images by acquiring seismic data, generating shallow and stacking velocity models, and iteratively refining these models through prestack depth migration and congruency tests to improve signal-to-noise and achieve convergence within interpretational uncertainties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional two-step velocity model building and depth migration are used, then velocity models can be estimated, but depth images are limited in quality due to data or model-based issues
Solution Approach 1:
The patent merges velocity model building and depth migration into a single simultaneous inversion process. The migration operator is integrated directly into the inversion algorithm, allowing both velocity model estimation and depth image generation to occur concurrently rather than sequentially, thereby improving both velocity model accuracy and depth image quality
Solution Approach 2:
The patent implements an iterative inversion process where depth images are used to guide velocity model updates, and velocity models are used to improve depth image migration. This feedback loop continues until convergence is achieved, allowing each component to refine the other and overcome limitations of conventional one-way approaches
2Measurement precision
If iterative concurrent generation of velocity models and depth images is implemented, then imaging quality improves, but processing time increases
Solution Approach 1:
The patent performs preliminary actions by estimating initial velocity models and generating preliminary depth images in early iterations, which then guide subsequent inversion steps. This allows the algorithm to converge more efficiently by using preliminary results to constrain later iterations, reducing overall processing time while maintaining high imaging quality
Data Source
AI summary
In various embodiments, the present disclosure describes methods for processing seismic data to concurrently produce a velocity model and a depth image. Various embodiments of the methods include: a) acquiring seismic data; b) generating a shallow velocity model from the seismic data; c) generating a stacking velocity model using the shallow velocity model as a guide; d) generating an initial interval velocity model from the stacking velocity model; and e) generating an initial depth image using the initial interval velocity model. The methods also include iterative improvement of the initial depth image and the initial interval velocity model to produce improved depth images and improved interval velocity models. Improvement of the depth images and the interval velocity models is evaluated by using a congruency test.