Adaptive Computational Grid for Marine Seismic Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current marine seismic survey techniques face challenges in acquiring high-quality, high-resolution data due to the complexity and expense of determining physical properties of subterranean formations, often resulting in false positives or negatives when identifying anomalies like oil reservoirs, leading to inefficient resource allocation and potential waste.
Innovation Solution
An adaptive computational grid approach is implemented, which dynamically adjusts its resolution based on measured physical field data to optimize grid resolution, reducing false positives and negatives by determining the minimal spatial resolution capable of differentiating anomalies, thereby improving the reliability of physical property determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution computational grid is used to improve anomaly detection accuracy, then measurement precision improves, but device complexity and computational cost increase
Solution Approach 1:
The patent implements an adaptive computational grid that dynamically adjusts its resolution based on the detected features in the physical field data. The grid transitions from a coarse initial state to a fine detailed state in regions where anomalies are detected, while maintaining coarse resolution in homogeneous regions. This dynamic adaptation allows the system to achieve high measurement precision for anomaly detection without the computational burden of uniformly high-resolution grids throughout the entire domain.
Solution Approach 2:
The patent applies different grid resolutions to different spatial regions based on their characteristics. In regions containing anomalies or features of interest, the grid is refined to high resolution to ensure accurate detection and characterization. In regions with homogeneous properties, the grid remains coarse to reduce computational complexity. This local differentiation of quality allows the system to optimize the trade-off between measurement precision and device complexity by concentrating computational resources only where necessary.
2Measurement precision
If uniform fine grid resolution is applied throughout the domain, then measurement precision improves, but productivity and computational efficiency deteriorate
Solution Approach 1:
The computational domain is segmented into multiple regions with different grid resolutions based on their importance and complexity. The domain is divided into coarse regions and fine regions, where fine regions are further subdivided into sub-regions of varying resolution. This segmentation allows the system to apply high computational effort only to specific segments that require detailed analysis, while other segments are processed with lower computational resources, thereby maintaining productivity while improving measurement precision where needed.
Solution Approach 2:
Instead of applying fine grid resolution uniformly across the entire domain (excessive action), the patent applies fine resolution only to specific regions where anomalies are detected or where high precision is required for physical property determination (partial action). This selective application of computational resources achieves the necessary measurement precision without the excessive computational cost of全域 fine grids, thus maintaining productivity.
3Reliability
If adaptive grid refinement is performed iteratively, then reliability of anomaly detection improves, but loss of time increases
Solution Approach 1:
The patent performs preliminary grid refinement by detecting anomalies at a coarse grid level first, then using this preliminary information to guide subsequent refinement steps. The initial coarse grid provides a preliminary view of the physical field that identifies regions requiring further investigation. This preliminary action allows the system to focus computational time on refining only the necessary regions, rather than uniformly refining the entire domain, thus improving reliability while minimizing time loss.
Solution Approach 2:
The adaptive grid refinement process incorporates feedback mechanisms where the results from each refinement iteration are used to guide the next iteration. The system monitors the detected features and physical property variations, and uses this feedback to determine whether further refinement is necessary in specific regions. This feedback-driven approach ensures that refinement continues only as long as and where it improves reliability, preventing unnecessary iterative refinement that would increase time loss without providing additional value.
Data Source
AI summary
Techniques are described for determining an optimal resolution for a grid to be used in solving inverse problems. Reference physical fields may be computed based on model data for the computational grid at a starting resolution. Cells in the computational grid may be split in a plurality of iterations to provide finer resolution. The model data may be perturbed by introducing different physical property values to the cells. The physical fields may be calculated based on the perturbed model data. A comparison may be made between the reference physical fields and the calculated physical fields based on the perturbed model data for the purpose of determining whether cell splitting should continue.