Stick-Slip Prediction via Surface and Downhole Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems in the oilfield industry are unable to predict stick-slip events before they occur, leading to potential damage to drill strings and downhole tools, as well as non-productive time due to uncontrolled rotational speeds and accelerations.
Innovation Solution
A method and system that measure surface and downhole properties using sensors and a computing system to identify patterns that precede previously detected stick-slip events, training a model to determine the likelihood of a stick-slip occurrence based on matched distributions of these properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If stick-slip is detected after occurrence, then damage can be mitigated, but the damage has already occurred and non-productive time is lost
Solution Approach 1:
The system performs preliminary action by detecting precursor patterns in rotational speed and acceleration data before stick-slip events occur. The machine learning model analyzes historical data to identify predictive features and triggers alerts or automated responses in advance, allowing preventive measures to be taken before damage occurs and non-productive time is lost.
2Productivity
If rotational speed is increased to maintain drilling productivity, then productivity improves, but stick-slip events occur more frequently causing damage
Solution Approach 1:
The system implements feedback by continuously monitoring rotational speed and acceleration data, comparing real-time measurements against predicted stick-slip conditions using machine learning models. When precursor patterns are detected, the system provides feedback signals that can trigger automated responses to adjust drilling parameters, creating a closed-loop control system that maintains productivity while preventing damage.
3Stability of the object's composition
If friction between drill string and wellbore is increased to control slip, then stick-slip is reduced, but drilling efficiency decreases
Solution Approach 1:
The system performs preliminary action by detecting precursor patterns in rotational speed and acceleration data before stick-slip events occur. The machine learning model analyzes historical data to identify predictive features and triggers alerts or automated responses in advance, allowing preventive measures to be taken before damage occurs and non-productive time is lost.
Data Source
AI summary
A method for predicting a stick-slip event includes measuring one or more surface properties using a sensor at the surface. The method also includes measuring one or more downhole properties using a downhole tool in a wellbore. The method also includes determining that the one or more surface properties and the one or more downhole properties match a distribution. The distribution occurs before two or more previously-detected stick-slip events. The method also includes determining a likelihood that a stick-slip event will occur based at least partially upon the distribution that the one or more surface properties and the one or more downhole properties match.