Well Log Marker Propagation Using FastDTW and Minimum Spanning Tree
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for interpreting subsurface geologic structures are inefficient in accurately propagating markers across multiple well logs, leading to incomplete characterization of subsurface regions, which hampers resource extraction operations.
Innovation Solution
A system and method that iteratively propagates markers from a well log to other well logs in a geographic region using a FastDTW approach, constructing a minimum spanning tree and adjusting search ranges to enhance marker propagation and confidence calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual marker identification and propagation methods are used across multiple well logs, then human expertise can ensure accuracy in marker placement, but the process requires extensive human intervention and is time-consuming
Solution Approach 1:
The system performs preliminary actions by pre-processing well log data to identify potential marker locations and prepare propagation paths before actual marker placement. The FastDTW algorithm pre-calculates similarity metrics between well logs, and the minimum spanning tree pre-establishes the propagation hierarchy, enabling rapid automated propagation without extensive manual intervention during the main interpretation process.
Solution Approach 2:
The system enables self-service by implementing automated marker propagation that operates independently after initial setup. The algorithm automatically identifies corresponding markers across well logs by comparing log characteristics, calculates confidence scores, and propagates markers through the minimum spanning tree structure without requiring continuous human guidance, significantly reducing manual intervention time.
2Productivity
If automated marker propagation algorithms are implemented across all well logs, then the speed of marker propagation increases, but the accuracy and reliability of marker identification may decrease
Solution Approach 1:
The system applies local quality by adjusting propagation parameters and search ranges based on local well log characteristics and geological conditions. Different regions of the well log data receive tailored processing parameters, and the minimum spanning tree allows different propagation strategies for different well pairs based on their similarity and spatial relationships, maintaining high accuracy while achieving automated speed.
Solution Approach 2:
The system incorporates feedback mechanisms by calculating confidence scores for each propagated marker and using these scores to guide further propagation decisions. The algorithm monitors propagation success rates and adjusts parameters accordingly, with the ability to flag low-confidence markers for manual review, thereby maintaining reliability while operating at automated speeds.
3Loss of information
If a comprehensive search range is used for marker propagation across all wells, then all potential markers can be identified, but the computational complexity and processing time increase significantly
Solution Approach 1:
The system applies segmentation by dividing the set of all well logs into hierarchical groups based on the minimum spanning tree structure. Instead of comparing every well log against every other well log simultaneously, the algorithm segments the propagation task into manageable stages following the tree hierarchy, reducing computational complexity from O(n²) to approximately O(n log n) while maintaining comprehensive marker identification through iterative propagation.
Solution Approach 2:
The system implements dynamics by making the search range adaptive rather than fixed. The search range dynamically adjusts based on the current propagation stage, the specific well pair being compared, and the confidence scores obtained during propagation. This allows the algorithm to expand search ranges when needed to ensure completeness while contracting them to reduce computational complexity in routine propagation steps.
Data Source
AI summary
A system can include a processor; memory operatively coupled to the processor; and processor-executable instructions stored in the memory to instruct the system to: receive a marker on a well log for a well in a geographic region; and iteratively propagate the marker automatically to a plurality of well logs for other wells in the geographic region.