Ultrasonic Flow Regime Identification via Face Recognition Bayesian Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imaging techniques struggle to accurately identify three-phase flow regimes in conduits, particularly when high levels of free gas are present, due to multiple reflections and clutter, which impede flow regime identification in hydrocarbon production systems.
Innovation Solution
An ultrasonic transceiver array mounted around the conduit transmits and receives energy, with a data processing system that organizes fluid transit data into matrices for Bayesian classification and face recognition, enabling accurate identification of flow regimes by comparing actual conditions to a database of test conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional tomographic reconstruction methods are used for three-phase flow imaging, then the imaging capability for two-phase flows is adequate, but the identification accuracy deteriorates when high levels of free gas are present due to multiple reflections and clutter
Solution Approach 1:
The patent segments the flow identification process into distinct phases: data acquisition from ultrasonic transceivers, organization into data matrices, comparison with database templates, and Bayesian classification. This segmentation allows each phase to be optimized independently, improving overall accuracy despite the presence of free gas reflections
Solution Approach 2:
The patent introduces an intermediary database of pre-acquired flow regime data matrices that serve as reference templates. The actual flow data is compared against these intermediaries to identify flow regimes, effectively mediating the identification process and reducing the impact of harmful reflections
2Adaptability or versatility
If conventional imaging techniques are applied to three-phase flow, then the system can handle two-phase flows adequately, but it fails to provide satisfactory identification for oil-water-gas mixtures simultaneously
Solution Approach 1:
The patent creates a universal identification system that handles all three-phase combinations (oil-water, oil-gas, water-gas) and two-phase flows through a single Bayesian classification framework. The system uses the same data matrix comparison approach regardless of which phases are present, providing versatile and accurate identification across all flow conditions
Solution Approach 2:
The patent changes the identification parameters from traditional tomographic reconstruction parameters to ultrasonic transit time and attenuation parameters organized in data matrices. This parameter transformation enables the system to distinguish between different phase combinations by comparing characteristic patterns in the matrix data against the database
3Ease of operation
If standard tomographic reconstruction is used, then the imaging process is straightforward for transmission-based methods, but the presence of gas bubbles results in clutter that impedes flow regime identification
Solution Approach 1:
The patent extracts the essential flow regime identification information from the ultrasonic data by organizing it into structured data matrices that capture characteristic patterns. This extraction process separates the useful flow regime information from the harmful clutter caused by gas bubbles, enabling accurate identification despite the presence of free gas
Solution Approach 2:
The patent performs preliminary organization of ultrasonic transit time and attenuation data into data matrices before comparison with flow regime templates. This preliminary structuring action prepares the data in an optimal format for Bayesian classification, ensuring that flow regime information is preserved and enhanced before the final identification step
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
This approach effectively differentiates between various flow regimes, such as bubbly and continuous water flows, even in challenging conditions like wet gas or saturated bubbly flows, providing clearer images and improved accuracy over traditional tomographic reconstruction methods.
Implementation Method 1
an array of a plurality of ultrasonic transceivers mounted about the periphery of the conduit transmitting and receiving energy for travel through the fluid in the conduit
Data Source
AI summary
Multiphase flow regimes in downhole or surface flow lines are made identifiable by forming images of the flow based on data obtained by flow regime metering of the flow. The flow regime metering data are gathered and processed in a format in which it can be subjected to face recognition processing of the type used for recognition of the faces of persons, and also to Bayesian classification techniques. Identification capabilities are particularly enhanced in flow regimes where gas bubbles or large amounts of free gas cause clutter and multiple reflections in the metering data.