Automated Well-Log Correlation via Dynamic Warping Descriptors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual correlation of well logs is time-consuming, subjective, and prone to errors, as it requires geologists to align and identify strata across multiple well logs, which can be costly and inefficient.
Innovation Solution
Automated well-log correlation using dynamic warping with descriptors, such as shape, magnitude, and frequency information, to align well logs, which includes generating combinations of descriptors, parameters, and weights to achieve accurate alignments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual correlation of well logs is performed, then geologists can identify and align strata across multiple well logs, but the process is time-consuming, subjective, and prone to errors
Solution Approach 1:
The patent replaces the manual mechanical process of visual inspection and alignment with an automated computational system. The system uses dynamic time warping algorithms and shape descriptors to automatically correlate well logs, eliminating the need for manual geometric alignment while improving both speed and objective accuracy of strata correlation across multiple well logs.
Solution Approach 2:
The patent transforms the correlation problem from a geometric alignment task to a parameter-based comparison task. By extracting shape descriptors (parameters) from well log data and using dynamic time warping to compare these parameters, the system achieves accurate correlation without manual intervention, resolving the contradiction between automation and accuracy.
2Reliability
If manual review of well logs is performed, then geologists can identify structures or features of interest, but the process is costly and inefficient
Solution Approach 1:
The patent enables the well log correlation system to perform its own alignment and identification tasks autonomously. The automated system extracts features, computes shape descriptors, performs dynamic time warping, and identifies correlated strata without requiring human geologist intervention, thereby achieving both high reliability through consistent algorithmic application and high productivity through rapid automated processing.
3Measurement precision
If dynamic warping is performed using only node magnitudes, then well logs can be aligned, but the alignment accuracy is insufficient compared to using descriptors
Solution Approach 1:
The patent segments the well log data into discrete features and computes shape descriptors for each feature. By dividing the continuous well log into identifiable geological features and characterizing each with multiple descriptors (magnitude, slope, curvature), the system achieves more accurate alignment than using raw magnitudes alone, while the automated computation manages the increased complexity.
Solution Approach 2:
The patent transitions from one-dimensional magnitude comparison to multi-dimensional descriptor comparison. By incorporating additional dimensions such as slope, curvature, and other shape characteristics alongside magnitude, the system achieves superior alignment accuracy. The dynamic time warping algorithm efficiently handles this increased dimensionality through automated computation.
Data Source
AI summary
Well logs can be automatically correlated using dynamic warping with descriptors. For example, a computing device can receive user input indicating a manual correlation between a first node in a first well-log and a second node in a second well-log. The computing device can use the manual correlation to automatically determine a combination of descriptors usable to correlate the first node to the second node. The computing device can then perform dynamic warping on other well logs using the combination of descriptors determined by using the first well log and the second well log. This may provide more accurate correlations between well logs.