Multiphase Flow Rate Measurement Using Sensor Data and Reference Meter Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring flow rates of multiphase and multicomponent fluids from oil and gas wells lack accuracy and are costly due to reliance on low-precision flow meters and equipment not originally designed for metrology, limiting real-time monitoring and increasing operational expenses.
Innovation Solution
A system and method utilizing a set of sensors to collect primary measurements of pressure, temperature, and additional parameters, combined with a reference multiphase flow meter for training, establishing a relationship between these parameters and flow rates, allowing for accurate, continuous measurement of flow rates without continuous use of a high-precision flow meter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If multiple low-precision flow meters are used to monitor flow rates from multiple wells, then the cost effectiveness of the monitoring system is improved, but the measurement precision deteriorates
Solution Approach 1:
A machine learning model is introduced as an intermediary between the low-precision flow meters and the reference measurements. The model learns to translate readings from inexpensive flow meters into accurate flow rate estimates by training on data collected when a reference flow meter is temporarily installed, thereby resolving the contradiction between cost-effectiveness and measurement precision.
Solution Approach 2:
The system creates a virtual copy of the reference flow meter's measurement capability through machine learning. Instead of physically installing expensive reference flow meters at all wells, the system learns to replicate their measurement accuracy using data from cheaper flow meters, achieving both cost savings and high precision.
2Measurement precision
If a single high-precision flow meter is used and switched between wells, then the measurement precision is improved, but the productivity of continuous monitoring deteriorates
Solution Approach 1:
The system enables low-precision flow meters to serve themselves by learning from reference measurements. The machine learning model allows the inexpensive flow meters to automatically achieve high-precision measurement capabilities without requiring physical intervention or switching with reference meters, maintaining continuous monitoring at all wells simultaneously.
Solution Approach 2:
The system performs preliminary training by collecting data from both low-precision flow meters and reference flow meters during an initial phase. This preliminary action enables the machine learning model to learn the relationship between different measurement devices, allowing continuous high-precision monitoring to begin afterward without requiring reference meters to be physically present.
3Device complexity
If equipment not originally designed for metrology is used to determine flow rates, then the device complexity is reduced, but the measurement precision deteriorates
Solution Approach 1:
The system changes the parameters of existing flow meters by applying machine learning transformations to their readings. Instead of requiring specialized metrology equipment, the system transforms the output parameters of standard flow meters through learned relationships, achieving high precision while maintaining low device complexity.
Data Source
AI summary
A method and system for determining a flow rate of at least a phase or a component of a fluid produced from an oil and gas well are presented hereinafter. The fluid is one of a multiphase and of a multicomponent fluid. The method comprises, in a training phase, collecting primary measurements of pressure, temperature, and additional flow parameter of the produced fluid. The primary measurements are carried out at the wellhead by a set of sensors installed in a flow line for the produced fluid. In the training phase, the method also comprises collecting a flow rate of at least one of the phases or components of the produced fluid simultaneously measured by a reference multiphase flow meter installed in the flow line. It also includes establishing a relationship between the pressure, temperature, and the additional flow parameter and the flow rate of the at least one of the phases or components of the produced fluid. The method also comprises, in a subsequent production phase, determining the flow rate of the at least one of the phases or components of the produced fluid based on the primary measurements of the pressure, temperature, and the at least one additional flow parameter and on the established relationship.