Automated Well Marker Prediction Using Neural Network Facies Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The traditional manual process of determining well markers in reservoir modeling is time-consuming and costly, leading to inaccuracies and wasteful expenditures due to the reliance on human interpretation of well log data, which is labor-intensive and prone to errors.
Innovation Solution
An automated method using artificial neural networks (NNs) to predict well markers by separating NN-predicted facies output into training and target wells, calculating sameness and coverage scores, and iteratively determining the top and depth positions of dominant facies zones, thereby reducing manual participation and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual interpretation of well log data by geoscientists is used to determine well markers, then expertise and knowledge can be applied to identify zones, but the process becomes time-consuming and costly
Solution Approach 1:
The patent replaces the manual mechanical process of well picking with an automated computer-based system that uses neural networks and machine learning algorithms. The system automatically identifies lithostratigraphic zones and determines well markers by processing well log data, substituting human geoscientist analysis with computational methods that maintain accuracy while dramatically reducing time requirements.
Solution Approach 2:
The system enables self-service automation where the computer-based platform independently performs well marker determination without requiring continuous human intervention. The neural network model automatically analyzes well log data, identifies zones, and generates well markers, allowing the process to serve itself rather than relying on manual geoscientist labor for each determination.
2Productivity
If manual well picking processes are used, then geoscientists can apply their knowledge to interpret data, but inaccuracies and errors can still occur due to human factors
Solution Approach 1:
The patent replaces manual interpretation with an automated neural network-based system that consistently applies the same analytical criteria to all well data. This substitution eliminates human errors such as fatigue, bias, and inconsistency, while maintaining the expertise needed for accurate zone identification through trained machine learning models that have learned from extensive geological data.
3Reliability
If more manual review and verification of well markers is performed to improve accuracy, then better 3D lithofacies models can be created, but the process becomes even more time-consuming and costly
Solution Approach 1:
The patent replaces complex manual review processes with automated computational verification. The neural network system not only determines well markers but also automatically validates results through consistent application of geological criteria, reducing the need for multiple rounds of manual review while maintaining or improving model accuracy.
Solution Approach 2:
The system incorporates feedback mechanisms where the neural network continuously learns from and refines its analysis of well log data. By processing results through automated validation loops and using calculated metrics such as sameness scores and coverage scores, the system provides feedback that improves accuracy without requiring increased manual intervention or process complexity.
Data Source
AI summary
This disclosure generally describes methods and systems, including computer-implemented methods, computer-program products, and computer systems, for predicting well markers. One computer-implemented method includes separating neural-network (NN)-predicted facies output associated with a plurality of wells into two sets, a first set of NN-predicted facies output of training wells and a second set of NN-predicted facies output of target wells, calculating, for each training well of the plurality of wells, a sameness score between zones of NN-predicted facies output and human-identified lithostratigraphic units (finer zones), calculating a mean sameness score for the finer zones for all training wells, identifying finer zones with a mean sameness score greater than a threshold value as dominant facies zones, and iterating over each target well to calculate a top and depth position of each dominant facies zone determined based upon the NN-predicted facies output of the target well.