Machine-Learning ICV Flow Control Using Acoustic Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hydrocarbon well operations face challenges in maintaining target flow rates due to changes in reservoir properties, making it difficult to detect and attribute flow rate deviations to control valves, leading to variations in hydrocarbon production rates.
Innovation Solution
A machine-learning-based system using acoustic sensing and distributed acoustic sensing (DAS) to record acoustic signals from intelligent completion valves (ICVs), correlating them with fluid flow rates, and training a machine-learning model to predict and adjust ICV positions or pressures for precise flow control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physics-based models are used to estimate control valve settings, then the initial flow rate control is achieved, but the flow rate deviation occurs when reservoir properties change
Solution Approach 1:
The system continuously monitors actual flow rates and compares them with target flow rates, using the deviations to adjust control valve settings in real-time. This closed-loop feedback mechanism ensures that flow rate consistency is maintained even when reservoir properties change, resolving the contradiction between initial control accuracy and long-term consistency.
Solution Approach 2:
The system dynamically adjusts control parameters (valve positions, opening degrees) based on changing reservoir conditions and observed flow rate deviations. By continuously updating operational parameters rather than relying on static physics-based models, the system maintains accurate flow rate control despite property changes in the reservoir.
2Loss of time
If traditional monitoring methods are used, then basic flow rate measurement is possible, but timely detection of flow rate deviation is difficult
Solution Approach 1:
The system implements continuous real-time monitoring of flow rates with automated comparison against target values. This persistent feedback loop enables immediate detection of deviations as they occur, eliminating the time delays associated with traditional periodic monitoring methods.
Solution Approach 2:
The system replaces manual or periodic mechanical measurement methods with automated electronic sensing and computational analysis. This substitution enables continuous, real-time detection of flow rate deviations through electronic sensors and digital processing, significantly reducing detection time and improving measurement reliability.
3Manufacturing precision
If manual control methods are used, then operational simplicity is maintained, but precise flow control is difficult to achieve
Solution Approach 1:
The system uses automated feedback control where measured flow rates are continuously compared with target values and used to automatically adjust valve positions. This eliminates the need for manual intervention while achieving precise flow control, as the system self-corrects deviations without operator input.
Solution Approach 2:
The control system is designed to automatically monitor, detect deviations, and adjust valve settings without requiring manual operation. The system serves itself by using its own measurements to drive its control actions, achieving precise flow control while reducing operational complexity from manual to automated control.
4Reliability
If reservoir properties are assumed constant, then initial flow rate targets are met, but actual flow rate deviates from target
Solution Approach 1:
The system continuously monitors actual flow rates and detects deviations caused by changing reservoir properties. This feedback information is used to dynamically adjust control valve settings, enabling the system to adapt to reservoir changes and maintain consistent flow rates despite varying geological conditions.
Solution Approach 2:
The control system transitions from static control based on fixed reservoir assumptions to dynamic control that continuously adapts to changing conditions. By making real-time adjustments based on monitored performance, the system maintains reliability even as reservoir properties evolve over time.
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
Enhances the efficiency of wellbore operations by accurately maintaining target flow rates, preventing issues like water breakthrough and loss of caprock integrity, and improving hydrocarbon production and sweep efficiency.
Implementation Method 1
The flow of pressurized injection fluid through the ICVs produces resulting acoustic signals. An acoustic sensing system may be used to record the acoustic signals produced by the ICVs.
Data Source
AI summary
A system for controlling intelligent completion valves (ICVs) used in a hydrocarbon well operation is disclosed. The ICVs can be located downhole in a wellbore, and an outflow of pressurized injection fluid from the wellbore may be used to drive hydrocarbons toward one or more offset producing wells. A trained machine-learning model can be generated by training a machine-learning model on training data comprising acoustic signal data generated by a pressurized injection fluid flowing through the ICVs at various flow rates and ICV positions. When applied to new acoustic sensing system sensor data associated with an ICV of multiple ICVs in the well, the trained machine-learning model can generate a result indicating a predicted flow rate of the pressurized injection fluid through the ICV. The result may be output and used to control the flow rate of pressurized injection fluid through the ICVs.