Geosteering Recommendations Using Real-Time Geological Model Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current geosteering methods rely on static geological models that require manual updating and are insufficient for predicting complex geology, leading to inefficiencies and increased drilling risks due to human intervention and the inability to account for unforeseen geological features.
Innovation Solution
An ensemble geosteering model architecture utilizing two machine learning models: one to update the geological model in real-time based on drilling data, and another to generate geosteering recommendations incorporating updated geological models, drilling data from offset wells, and operational requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual updating and manipulation of static geological models is used, then human expertise can be applied to interpret data, but the workflow becomes slow and human intervention becomes a bottleneck
Solution Approach 1:
The patent replaces the manual mechanical process of geologist intervention with an automated machine learning system. The ML model automatically updates geological models and generates geosteering recommendations in real-time, eliminating the bottleneck of human intervention while maintaining or improving accuracy through data-driven predictions.
Solution Approach 2:
The system enables self-service by allowing the geological model to automatically update itself using real-time drilling data without requiring manual manipulation. The ML model continuously refines the geological model and generates geosteering recommendations autonomously, making the system self-sufficient and eliminating dependency on continuous human intervention.
2Device complexity
If static geological models are used, then model simplicity is maintained, but the models fail to predict challenging and complex geology
Solution Approach 1:
The patent transforms static geological models into dynamic models that automatically update in real-time as drilling progresses. The ML model continuously incorporates new drilling data to refine geological predictions, allowing the model to adapt to complex and changing subsurface conditions while maintaining computational efficiency through automated updates.
3Adaptability or versatility
If existing geological models are used, then current data can be processed, but unforeseen geology and operational limitations cannot be accounted for
Solution Approach 1:
The system implements continuous feedback by automatically incorporating real-time drilling data into the geological model updates. The ML model uses this feedback loop to detect unforeseen geological conditions and adjust predictions accordingly, improving reliability by accounting for actual subsurface conditions as they are encountered during drilling operations.
Data Source
AI summary
Certain aspects of the disclosure provide for systems and methods for geosteering a wellbore using an ensemble of machine learning models. The method may include processing one or more inputs with a first machine learning model trained to infer an updated geological model associated with the wellbore. The method may further include processing with a second machine learning model trained to generate a geosteering recommendation for the wellbore, one or more of: the updated geological model, drilling data from one or more offset wells, drilling requirements associated with the wellbore, or completion requirements associated with the wellbore to the second machine learning model. The method may further include outputting the geosteering recommendation for the wellbore.