Automated Wellbore Landing Zone Prediction System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manual and tedious process of determining the wellbore landing zone in multiple subterranean layers within a hydrocarbon reservoir is prone to human errors, lacking efficiency and accuracy.
Innovation Solution
A computer-implemented system that creates independent subsurface geological models from directional and vertical well surveys to predict the landing zone of wellbores, correlating the predictions and updating the models with new data for accurate assignment to specific formations, thereby reducing manual intervention and enhancing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual methods using subsea depth maps and vertical well logs are used to determine wellbore landing zones, then geologists can identify targeted formations, but the process is very manual and tedious and can lead to human errors
Solution Approach 1:
The patent replaces manual geological analysis methods with an automated computer-implemented system that uses machine learning models and algorithms to determine wellbore landing zones. The system automatically processes well survey data, seismic data, and geological models to assign wellbores to formations, eliminating the need for manual interpretation of depth maps and well logs by geologists.
Solution Approach 2:
The system enables self-service by allowing the automated landing zone determination process to operate independently without requiring continuous manual intervention. The machine learning models automatically train on historical data and continuously improve their accuracy, and the system can process new well data as it becomes available without requiring geologist involvement for each individual wellbore assignment.
2Reliability
If manual methods are used to determine wellbore landing zones, then geologists can interpret formation targets, but the process is prone to human errors
Solution Approach 1:
The patent segments the complex task of wellbore landing zone determination into multiple independent components: data acquisition modules, machine learning model training, prediction generation, and result validation. The system uses separate models for different geological scenarios and can process different data types independently, making the overall complex system manageable and reliable.
Solution Approach 2:
The system implements feedback mechanisms where prediction results are validated against known geological data and historical well outcomes. The machine learning models continuously learn from new well data and update their parameters to improve accuracy. Geologists can provide feedback on prediction accuracy, which is used to refine the models further, creating a self-improving system that reduces errors over time.
Data Source
AI summary
Techniques for predicting a landing zone of a wellbore include identifying a first subsurface geological model of a first subterranean layer located under a terranean surface that includes an upper boundary depth of the first subterranean layer and a lower boundary depth of the first subterranean layer; identifying a second subsurface geological model of a second subterranean layer deeper than the first subterranean layer that is independent of the first subsurface geological model and includes an upper boundary depth of the second subterranean layer; correlating a predicted landing zone for a plurality of wellbores using the first and second subsurface geological models that is based on a location of a horizontal portion of each wellbore; and generating data that comprises a representation of the correlated plurality of wellbores for presentation on a graphical user interface (GUI).