Subsurface Velocity and Density Modeling with Wavefield Gradients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating accurate velocity and density models of subsurface reservoir structures are inadequate, leading to imperfect subsurface imaging and reduced efficiency in hydrocarbon exploration and production.
Innovation Solution
A method involving wavefield propagation and gradient-based updates is employed to generate velocity and density models using seismic data, utilizing wavefield separation techniques and iterative frequency band processing to refine these models iteratively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods are used for generating velocity and density models, then the process is simpler, but the imaging accuracy and reliability of subsurface models are insufficient
Solution Approach 1:
The patent segments the velocity model and density model generation into distinct computational steps: wavefield propagation, gradient calculation for velocity, reflectivity gradient calculation for density, and iterative updates. This segmentation allows each component to be optimized independently while maintaining overall accuracy, resolving the contradiction between imaging precision and method complexity.
Solution Approach 2:
The patent employs parameter changes by iteratively updating velocity and density models based on calculated gradients from wavefield data. The method transforms the inversion problem into a gradient-based optimization process, changing model parameters systematically to improve imaging accuracy while providing a structured approach to manage complexity.
2Reliability
If gradient-based iterative updates are employed to refine models, then the reliability of subsurface models improves, but the computational time and processing complexity increase
Solution Approach 1:
The patent performs preliminary wavefield propagation and gradient calculations before model updates, preparing all necessary computational components in advance. This preliminary action structures the iterative process to minimize redundant calculations, thereby improving model reliability while reducing overall computational time.
Solution Approach 2:
The patent implements feedback mechanisms where updated velocity and density models are used to generate new wavefields, which then feed into the next iteration of gradient calculations. This feedback loop systematically improves model reliability through iterative refinement while the structured feedback process prevents wasteful computations by using previous iteration results.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method provides highly accurate subsurface structure imaging, enhancing hydrocarbon exploration success and reducing geohazards by improving the resolution and reliability of subsurface models.
Implementation Method 1
a velocity model maps out the layers of rock in a geographic region and the expected speed at which seismic waves (for example, P-waves or S-waves) will travel in each layer
Implementation Method 2
The wavefield separation may be performed using a Stokes' theorem-based approach
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method (400) including: receiving (402), by a data processing apparatus, a set of seismic data of the hydrocarbon reservoir; setting (404), by the data processing apparatus, an initial velocity model and an initial density model; generating (406), by the data processing apparatus, wavefields of the hydrocarbon reservoir based on the set of seismic data; selecting (408), by the data processing apparatus, a spatial direction; generating (410), by the data processing apparatus, a velocity gradient and a reflectivity gradient of the selected spatial direction based on the wavefields; and updating (412), by the data processing apparatus, the velocity model and the density model using the velocity gradient and the reflectivity gradient of the selected spatial direction.