Crisp and Fuzzy Classifiers in the Two-Phase Gas-Liquid Flow Diagnostics

2012 ◽  
Vol 17 (4) ◽  
pp. 385-394
Author(s):  
Paweł Fiderek ◽  
Tomasz Jaworski ◽  
Robert Banasiak ◽  
Jacek Kucharski
Keyword(s):  
Pattern Recognition ◽  
Flow Pattern ◽  
Liquid Flow ◽  
Confusion Matrix ◽  
Clustering Algorithms ◽  
Prediction Performance ◽  
Two Phase ◽  
Fuzzy Classifiers ◽  
Gas Liquid Flow ◽  
Liquid Flows

Abstract The following paper presents results of common clustering algorithms use, both crisp and fuzzy, for flow pattern recognition of two-phase gas-liquid flows observed in horizontal pipeline. Obtained results of HCM, FCM, and kNN clustering algorithms were presented in a form of confusion matrix and compared via its prediction performance.

GAS-LIQUID FLOW PATTERN RECOGNITION USING DIMENSIONLESS NUMBERS

2019 ◽  
Author(s):  
Caio Araujo ◽  
Tiago Ferreira Souza ◽  
Maurício Figueiredo ◽  
valdir estevam ◽  
Ana Maria Frattini Fileti
Keyword(s):  
Pattern Recognition ◽  
Flow Pattern ◽  
Liquid Flow ◽  
Dimensionless Numbers ◽  
Gas Liquid Flow

Experimental Study on Flow Patterns of Decaying Swirling Gas-Liquid Flow in a Horizontal Pipe

2020 ◽  
Author(s):  
Shuai Liu ◽  
Li Liu ◽  
Jiarong Zhang ◽  
Hanyang Gu
Keyword(s):  
Flow Pattern ◽  
Liquid Flow ◽  
Separation Efficiency ◽  
Swirling Flow ◽  
Flow Patterns ◽  
Two Phase ◽  
Horizontal Pipe ◽  
Gas Removal ◽  
Swirl Vane ◽  
Gas Liquid Flow

Abstract Swirling flow is one of the well-recognized techniques to control the working process. This special flow is widely adopted in swirl vane separators in nuclear steam generator (SG) for water droplet separation and the fission gas removal system in Thorium Molten Salt Reactor (TMSR) for gas bubble separation. Since the parameters such as separation efficiency, pressure drop and mass and heat transfer rate are strongly dependent on the flow pattern, the accurate prediction of flow patterns and their transitions is extremely important for the proper design, operation and optimization of swirling two-phase flow systems. In this paper, using air and water as working fluids, a visualization experiment is carried out to study the gas-liquid flow in a horizontal pipe containing a swirler with four helical vanes. The test pipe is 5 m in length and 30 mm in diameter. Firstly, five typical flow patterns of swirling gas-liquid flow at the outlet of the swirler are classified and defined, these being spiral chain, swirling gas column, swirling intermittent, swirling annular and swirling ribbon flow. Being affected by the different gas and liquid flow rate of non-swirling flow, it is found that the same non-swirling flow can change into different swirling flow patterns. After that, the evolution of various swirling flow patterns along the streamwise direction is analyzed considering the influence of swirl attenuation. The results indicate that the same swirling flow pattern can transform into a variety of swirling flow patterns and subsequent non-swirling flow patterns. Finally, the flow pattern maps at different positions downstream of the swirler are presented.

scholarly journals Gas–liquid Flow Pattern Recognition Based on Wavelet Packet Energy Entropy of Vortex-induced Pressure Fluctuation

2013 ◽  
Vol 13 (2) ◽  
pp. 83-88 ◽  
Author(s):  
Zhiqiang Sun ◽  
Shuai Shao ◽  
Hui Gong
Keyword(s):  
Pattern Recognition ◽  
Flow Pattern ◽  
Liquid Flow ◽  
Pressure Fluctuation ◽  
Bluff Body ◽  
Flow Direction ◽  
Wavelet Packet ◽  
Flow Patterns ◽  
Energy Entropy ◽  
Gas Liquid Flow

Here we report a novel flow-pattern map to distinguish the gas-liquid flow patterns in horizontal pipes at ambient temperature and atmospheric pressure. The map is constructed using the coordinate system of wavelet packet energy entropy versus total mass flow rate. The wavelet packet energy entropy is obtained from the coefficients of vortex-induced pressure fluctuation decomposed by the wavelet packet transform. A triangular bluff body perpendicular to the flow direction is employed to generate the pressure fluctuation. Experimental tests confirm the suitability of the wavelet packet energy entropy as an ideal indicator of the gas-liquid flow patterns. The overall identification rate of the map is 92.86%, which can satisfy most engineering applications. This method provides a simple, practical, and robust solution to the problem of gas-liquid flow pattern recognition.

Flow Pattern Transition in Pipes Using Data-Driven and Physics-Informed Machine Learning

2020 ◽  
Author(s):  
André M. Quintino ◽  
Davi L. L. N. da Rocha ◽  
Roberto Fonseca Jr. ◽  
Oscar M. H. Rodriguez
Keyword(s):  
Machine Learning ◽  
Experimental Data ◽  
Flow Pattern ◽  
Liquid Flow ◽  
Two Phase Flow ◽  
Data Driven ◽  
Phase Flow ◽  
Two Phase ◽  
Flow Pattern Transition ◽  
Gas Liquid Flow

Abstract Flow pattern is an important engineering design factor in two-phase flow in the chemical, nuclear and energy industries, given its effects on pressure drop, holdup, and heat and mass transfer. The prediction of two-phase flow patterns through phenomenological models is widely used in both industry and academy. In contrast, as more experimental data become available for gas-liquid flow in pipes, the use of data-driven models to predict flow-pattern transition, such as machine learning, has become more reliable. This type of heuristic modeling has a high demand for experimental data, which may not be available in some industrial applications. As a consequence, it may fail to deliver a sufficiently generalized transition prediction. Incorporation of physics in machine learning is being proposed as an alternative to improve prediction and also to reduce the demand for experimental data. This paper evaluates the use of hybrid-physics-data machine learning to predict gas-liquid flow-pattern transition in pipes. Random forest and artificial neural network are the chosen tools. A database of experiments available in the open literature was collected and is shared in this work. The performance of the proposed hybrid model is compared with phenomenological and data-driven machine learning models through confusion matrices and graphics. The results show improvement in prediction performance even with a low amount of data for training. The study also suggests that graphical comparison of flow-pttern transition boundaries provides better understanding of the performance of the models than the traditional metric

scholarly journals Experimental and Numerical Study on Gas-Liquid Flow in Hilly-Terrain Pipeline-Riser Systems

2021 ◽  
Vol 2021 ◽  
pp. 1-15
Author(s):  
Zhaoyang He ◽  
Limin He ◽  
Haixiao Liu ◽  
Dan Wang ◽  
Xiaowei Li ◽  
...  
Keyword(s):  
Flow Pattern ◽  
Liquid Flow ◽  
Slug Flow ◽  
Gas Transport ◽  
Numerical Study ◽  
Pipeline System ◽  
Hilly Terrain ◽  
Two Phase ◽  
Gas Liquid Flow ◽  
Stable Flow

In offshore oil and gas transport, gas-liquid mixed transport is a basic flow phenomenon. In general, pipeline undulations are caused by seabed topography; therefore, it is of great significance to study the mechanisms underlying gas and liquid flows in hilly-terrain pipeline-riser systems. This study established a hilly-terrain pipeline-riser experimental system in an indoor laboratory. The flow pattern and its flow mechanism were studied via experimental observation and pressure detection. Experimental results showed that the gas-liquid flow pattern in the hilly-terrain pipeline-riser system can be divided into four types: severe slugging, dual-peak slug, oscillation flow, and stable flow, where dual-peak slug flow is a special flow pattern in this pipeline system. Hilly-terrain units obstruct the downstream gas transport, weaken the gas-liquid eruption in the riser, and increase the cycle of severe slugging. In this paper, gas is regarded as power in the flow of gas and liquid, and the accumulation of liquid in low-lying areas is regarded as an obstacle. Then, the moment of gas-liquid blowout is studied as main research object, and the mechanism of flow pattern transformation is described in detail. This study investigated the accuracy of the OLGA 7.0 simulation results for the gas-liquid two-phase flow in the hilly-terrain pipeline-riser. The results show that OLGA 7.0 achieves a more accurate calculation of severe slugging and stable flow and can predict both the pressure trend and change characteristics. However, the simulation accuracies for dual-peak slug flow and oscillation flow are poor, and the sensitivity to gas changes is insufficient.

Sign in / Sign up

Export Citation Format

Share Document