4D Seismic Inversion for Saturation and Pressure Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 4D seismic inversion methods face challenges in accurately separating and quantifying saturation and pressure changes in hydrocarbon reservoirs due to limitations in using time shift or amplitude information alone, particularly in complex scenarios like reservoirs under gas clouds and stacking reservoirs, where velocity and impedance changes do not align properly.
Innovation Solution
A method involving iterative elastic AVO inversion that combines time shift and amplitude information to generate updated physical property models, using time strain as low-frequency information and rock physics models to convert impedance changes into saturation and pressure changes, allowing for simultaneous inversion of P- and S-impedance changes and improved interpretation of 4D seismic data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If time shift or amplitude information alone is used for 4D seismic inversion, then the inversion process is simpler, but the accuracy of saturation and pressure change separation deteriorates
Solution Approach 1:
The patent combines time shift information and amplitude information into a unified 4D seismic inversion framework. The objective function integrates both time shift misfit and amplitude misfit terms, allowing simultaneous utilization of both data types to improve saturation and pressure change separation accuracy while maintaining a coherent inversion process.
Solution Approach 2:
The patent extends the inversion from traditional single-dimension amplitude analysis to multi-dimensional analysis by incorporating time shift as an additional dimension. This is achieved by formulating the objective function to include both time shift residuals and amplitude residuals, effectively adding a temporal alignment dimension to the inversion process.
2Measurement precision
If iterative elastic AVO inversion combining time shift and amplitude information is used, then the accuracy of saturation and pressure change interpretation is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent employs an iterative inversion process that continuously refines the saturation and pressure change estimates by repeatedly minimizing the objective function. Each iteration uses the results from the previous iteration to update the model, ensuring continuous improvement of the solution until convergence criteria are met, thereby maximizing interpretation accuracy.
Solution Approach 2:
The inversion framework incorporates feedback mechanisms where the misfit between observed and predicted time shift and amplitude data is calculated, and this misfit information is used to update the physical property models in subsequent iterations. The feedback loop ensures that the inversion process systematically reduces errors and converges toward the optimal solution.
3Manufacturing precision
If time strain is used as low-frequency information in the inversion, then the vertical distribution of changes is improved, but the method becomes more complex to implement
Solution Approach 1:
The patent performs preliminary calculation of time strain from the time shift data before incorporating it into the main inversion process. This preliminary action prepares the low-frequency information in advance, which is then used as a constraint or prior information in the iterative inversion, improving the vertical distribution of changes without requiring complex real-time calculations during the inversion.
Solution Approach 2:
Time strain serves as an intermediary variable that bridges the time shift data and the physical property changes. By introducing time strain as a mediator, the patent facilitates the incorporation of temporal alignment information into the inversion framework in a mathematically tractable way, improving vertical resolution while maintaining implementation feasibility.
Data Source
AI summary
A method for inversion of 4D seismic data, including: determining time shift between baseline and monitor geophysical datasets; determining time strain from the time shift; iteratively repeating until a stopping criteria is satisfied, performing an iterative elastic AVO inversion with a 4D difference providing an update, from an initial model including the time strain, to generate an updated time strain and an updated physical property model, wherein the stopping criteria is a misfit between synthetic data generated from the updated physical property model and the 4D difference being within a predetermined noise level generating final values for the physical property model; and converting, with a rock physics model or a reservoir simulation model, the final values to saturation and/or pressure changes for a subsurface region.