Neural Network Flow Rate Prediction for Drilling Influx Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting fluid influxes and losses during drilling from a floating vessel is complicated by the changing volume of the riser string due to wave motion and tides, making it difficult to determine whether fluid is entering or leaving the wellbore.
Innovation Solution
A system using an accelerometer and an adaptive neural network filter to predict the flow rate from the well, compensating for changes caused by vessel movement, allowing for accurate detection of kicks and losses without requiring connection to the vessel's motion compensation system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional flow rate monitoring is used on a floating vessel, then fluid influxes and losses can be detected, but the changing riser string volume due to wave motion and tides causes false readings and reduces detection accuracy
Solution Approach 1:
The system uses accelerometers to continuously monitor vessel motion and feeds this data back to the neural network filter, which dynamically adjusts flow rate predictions based on detected motion patterns. This feedback loop enables real-time compensation for wave-induced riser volume changes, maintaining detection accuracy despite environmental disturbances.
Solution Approach 2:
An adaptive neural network filter acts as an intermediary between raw flow rate measurements and final detection results. The filter processes accelerometer data and flow rate data separately, then combines them to produce corrected flow rate predictions that account for vessel motion effects, effectively mediating between disturbed measurements and accurate detection.
2Measurement precision
If complex motion compensation systems are connected to the floating vessel, then detection accuracy improves, but system complexity and cost increase
Solution Approach 1:
The system performs self-service by using its own accelerometer measurements and neural network processing to automatically compensate for motion effects without requiring external motion compensation systems. The neural network adapts to vessel-specific motion patterns during operation, enabling the system to correct its own measurements independently.
Solution Approach 2:
The patent replaces complex mechanical motion compensation systems with a computational approach using accelerometers and neural networks. Instead of using mechanical devices to physically compensate for vessel motion, the system uses software-based filtering and prediction algorithms to correct flow rate measurements, significantly reducing mechanical complexity.
3Reliability
If the riser string volume changes due to vessel movement, then the volume of fluid in the riser changes, but this makes it difficult to distinguish between normal volume changes and actual fluid influxes or losses
Solution Approach 1:
The neural network performs preliminary calculations to predict what the flow rate should be under normal conditions, accounting for expected riser volume changes due to vessel motion. By establishing this baseline prediction before actual detection, the system can then compare predicted versus actual flow rates to reliably identify true influxes or losses.
Solution Approach 2:
The system continuously compares predicted flow rates with actual measurements and uses accelerometer feedback to dynamically adjust predictions. This feedback mechanism enables the system to distinguish between flow rate changes caused by vessel motion and those caused by actual fluid influxes or losses, preserving critical flow information.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables reliable and inexpensive detection of fluid influxes and losses during closed wellbore pressure controlled drilling operations, improving safety by quickly identifying potential issues before they become major events.
Implementation Method 1
an accelerometer and an adaptive neural network filter to predict the flow rate from the well
Data Source
AI summary
A system for detecting fluid influxes and losses can include a sensor which detects floating vessel movement, and a neural network which receives a sensor output, and which outputs a predicted flow rate from a wellbore. A method can include isolating the wellbore from atmosphere with an annular sealing device which seals against a drill string, inputting to a neural network an output of a sensor which detects vessel movement, the neural network outputting a predicted flow rate from the wellbore, and determining whether the fluid influx or loss has occurred by comparing the predicted flow rate to an actual flow rate from the wellbore. Another method can include inputting to a neural network actual flow rates into and out of the wellbore, and an output of a sensor which detects vessel movement, and training the neural network to output a predicted flow rate from the wellbore.