Automated Salt Body Interpretation via Smooth Surface Wrapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional salt interpretation in seismic data processing is slow, complex, and subjective due to manual methods, which are tedious and time-consuming, especially when dealing with poorly imaged sections caused by imperfect velocity models.
Innovation Solution
The implementation of automated smooth surface wrapping methods that classify seismic data into feature, uncertain, and non-feature categories using volume attributes, allowing for the automatic determination of salt and sediment regions, and smoothing uncertain areas to connect clear boundaries, thereby reducing the need for manual interpretation and improving turnaround time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual interpretation methods are used for salt interpretation, then interpretation accuracy can be maintained through operator expertise, but the process becomes tedious, subjective, and time-consuming
Solution Approach 1:
The system performs automated salt interpretation by itself without requiring manual operator intervention. The automated workflow classifies seismic data, identifies salt bodies, and generates velocity models automatically, making the system self-sufficient while maintaining interpretation quality through algorithmic consistency
Solution Approach 2:
The patent replaces the manual mechanical interpretation process with an automated computational system. The DNA-inspired search algorithm and level set methods substitute human operator actions with automated computational workflows, eliminating subjectivity and reducing time while preserving interpretation accuracy through rigorous mathematical frameworks
2Productivity
If automated feature extraction is used, then interpretation time is reduced, but the methods require complex algorithms and more computational resources
Solution Approach 1:
The patent segments the interpretation process into distinct automated stages: DNA-inspired search for initial feature detection, level set evolution for boundary refinement, and smooth surface wrapping for final boundary definition. This segmentation allows each stage to be optimized independently, managing algorithmic complexity while maintaining high productivity
Solution Approach 2:
The DNA-inspired search algorithm performs preliminary action by quickly identifying potential salt body locations and control points before the more computationally intensive level set and surface wrapping operations. This preliminary classification reduces the search space for subsequent algorithms, managing overall computational complexity while maintaining speed
3Extent of automation
If conventional automated tracking is used, then some automation is achieved, but poorly imaged sections due to imperfect velocity models remain difficult to interpret
Solution Approach 1:
The patent changes the approach from direct seismic event tracking to attribute-based classification. By transforming seismic data into classified attributes (salt, non-salt, uncertain) and using these as inputs for level set evolution, the system becomes more reliable in poorly imaged sections where traditional event tracking fails
Solution Approach 2:
The patent introduces classified volume attributes as an intermediary between the raw seismic data and the final salt body interpretation. This intermediary layer provides a more robust representation that handles poor image quality better, allowing the automated workflow to reliably interpret salt bodies even when velocity models are imperfect
4Manufacturing precision
If manual control point determination is performed, then detailed boundary definition is possible, but the process requires extensive operator time and subjective judgment
Solution Approach 1:
The system automatically determines control points and boundary definitions without operator intervention. The DNA-inspired search algorithm automatically identifies control points, and the level set evolution automatically refines boundaries, eliminating the need for manual control point placement while maintaining high boundary definition precision
Solution Approach 2:
The patent replaces manual control point determination with automated algorithms. The DNA-inspired search and level set methods substitute human operator actions with computational processes that objectively identify control points and define boundaries, reducing operator effort while maintaining or improving precision
Data Source
AI summary
The present disclosure provides an automated interpretation workflow for smooth surface wrapping of an imaged volume. The methods use a volume attribute to classify data into regions corresponding to feature/uncertain/non-feature parts and use smoothing through the uncertain parts to connect the clear boundaries of the feature. Any volume attribute can be used as long as feature/uncertain/non-feature categories can be identified in terms of continuous or discontinuous threshold values or ranges. The workflow can be combined with interpretation of well-defined boundary parts to resolve uncertainty and can be used with both explicit single-z and implicit multi-z (level set) boundary representations.