Convolutional Neural Network for Hydrocarbon Well Log Differencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hydrocarbon well logs from different databases often have varying levels of completeness and are challenging to process and analyze due to their splintered nature, requiring time-consuming manual comparison and merging processes.
Innovation Solution
A machine learning-based method using a convolutional neural network (CNN) to differentiate between hydrocarbon well logs stored in different databases, determining actionable differences and performing actions such as splicing or replacing logs to create a merged dataset for generating a subsurface model, thereby facilitating efficient data management and analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual comparison and merging processes are used for hydrocarbon well logs from different databases, then data completeness and accuracy can be ensured, but processing time and operational complexity increase significantly
Solution Approach 1:
The patent replaces manual mechanical comparison processes with an automated machine learning system using convolutional neural networks. The system automatically compares well logs from different databases, identifies actionable differences, and performs merging operations without human intervention, thereby reducing processing time while maintaining accuracy through algorithmic analysis of log data patterns
Solution Approach 2:
The machine learning system performs self-service by autonomously identifying differences between well logs, determining which differences require action, and executing merging operations. The system trains on historical data to improve its own performance over time, automatically adapting to different log formats and databases without requiring manual reconfiguration
2Productivity
If automated machine learning methods are used to process and merge well logs, then processing speed and productivity improve, but system complexity and difficulty of implementation increase
Solution Approach 1:
The patent introduces an intermediary machine learning layer that sits between the raw well log data and the final merged output. This intermediary system handles the complexity of comparison and decision-making algorithms, translating complex log data into actionable differences and merged results, thereby shielding users from the underlying system complexity while maintaining high processing speed
Solution Approach 2:
The system performs preliminary action by pre-training the machine learning models on historical well log data before actual merging operations. This preliminary training phase establishes the system's capability to quickly process new data, and the pre-processed training data serves as a foundation that reduces the complexity of subsequent real-time merging operations
3Loss of information
If comprehensive analysis of all well log differences is performed, then data completeness improves, but computational resources and processing time increase
Solution Approach 1:
The patent applies partial action by focusing the machine learning system's analysis only on actionable differences between well logs rather than examining every single data point. The system identifies and processes only those differences that require merging or correction, leaving non-actionable variations unprocessed, thereby reducing computational resource consumption while maintaining data completeness for critical elements
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods for machine learning-based differencing for hydrocarbon well logs are disclosed. A computer system generates an input layer of a machine learning module. The input layer includes a graphical representation of datasets obtained from one or more hydrocarbon wells. The computer system generates an encoding layer of the machine learning module from the graphical representation. The encoding layer includes a two-dimensional array based on the multiple datasets and a merged dataset of the one or more hydrocarbon wells. The computer system determines that a difference between a first hydrocarbon well log of the multiple datasets and a second hydrocarbon well log of the merged dataset warrants an action. The determining is performed using a convolutional layer. Responsive to determining that the difference warrants an action, the computer system performs the action on the first hydrocarbon well log and the second hydrocarbon well log to modify the merged dataset.