Machine Learning Test Location Recommendations for Borehole Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for selecting test locations in subsurface reservoirs are inefficient, leading to a high rate of invalid tests and increased operational time, which results in tool wear and resource wastage.
Innovation Solution
A system utilizing machine learning models to analyze petrophysical data and generate recommendations for optimal test locations along a borehole, integrating data from various sources to improve the accuracy of test validity predictions and mobility index calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional methods are used for selecting test locations in subsurface reservoirs, then operational simplicity is maintained, but the rate of invalid tests increases and operational time is wasted
Solution Approach 1:
The system performs preliminary analysis of petrophysical data before conducting formation tests to identify optimal test locations. The machine learning model predicts test validity in advance by analyzing multiple petrophysical parameters, allowing operators to plan test sequences that maximize validity while minimizing wasted operational time at invalid locations.
2Reliability
If traditional methods are used for selecting test locations, then operational complexity remains low, but tool wear increases due to repeated invalid tests
Solution Approach 1:
The system introduces an intermediary machine learning model that acts as a decision support layer between the operator and the formation testing process. This intermediary analyzes petrophysical data and provides predictions about test validity, reducing the need for operators to manually evaluate complex geological parameters while improving test selection accuracy and reducing tool wear.
3Productivity
If machine learning models are used to analyze petrophysical data and generate test location recommendations, then test validity is enhanced and operational efficiency is optimized, but computational complexity and data processing requirements increase
Solution Approach 1:
The system implements self-service capabilities where the machine learning model automatically processes petrophysical data and generates test location recommendations without requiring extensive manual intervention. The model is trained on historical data to independently identify patterns and make predictions, reducing the computational burden on operational systems while maintaining high productivity and accuracy in test location selection.
Data Source
AI summary
A method can include receiving petrophysics data acquired along a borehole in a subsurface region; generating test location recommendations along the borehole using the petrophysics data as input to a machine learning model; and outputting, based on the test location recommendations, selected locations for performing tests using a downhole tool disposed in the borehole