A multi-task learning model with feature separation constraint for parameter prediction of gas–liquid two-phase flow
- 7 January 2025
- journal article
- research article
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (2) , 025301
Abstract
Precise measurement of gas void fraction is crucial for the control and optimization of various industrial processes. However, due to the intricate dynamics inherent in two-phase flow, measuring gas void fraction remains a challenging task. In this work, we propose a novel deep learning model called Feature Separation Constraint Network (FSCN). First, we conduct gas–liquid two-phase flow experiments in vertical upward pipelines, where temporal and spatial signals about the fluid are collected using a four-sector distributed conductivity sensor. Second, a spatio-temporal convolutional neural network is employed to extract features, and flow pattern classification serves as an auxiliary task to provide global features for gas void fraction prediction. Subsequently, an expert network structure with a feature separation constraint is designed to enhance prediction performance. Finally, comparative analyses with competitive baselines demonstrate FSCN’s superior capability in gas void fraction estimation. A series of ablation experiments confirm the efficacy of each component.Keywords
Funding Information
- TaiShan Industrial Experts Program
- National Natural Science Foundation of Tianjin, China (21JCJQJC00130)
- National Natural Science Foundation of China (62203079)
- China Postdoctoral Science Foundation (2023M732610)
This publication has 34 references indexed in Scilit:
- Experimental study of gas-liquid flow patterns and void fraction in prototype 5 × 5 rod bundle channel using wire-mesh sensorAnnals of Nuclear Energy, 2022
- Exact Riemann solutions for the drift-flux equations of two-phase flow under gravityJournal of Differential Equations, 2022
- Multi-task learning for digital rock segmentation and characteristic parameters computationJournal of Petroleum Science and Engineering, 2022
- Experimental investigation and new void-fraction calculation method for gas–liquid two-phase flows in vertical downward pipeExperimental Thermal and Fluid Science, 2021
- Two-phase flow and mass transfer in microchannels: A review from local mechanism to global modelsChemical Engineering Science, 2021
- Studies of gas-liquid two-phase flows in horizontal mini tubes using 3D reconstruction and numerical methodsInternational Journal of Multiphase Flow, 2020
- A novel complex network-based deep learning method for characterizing gas–liquid two-phase flowPetroleum Science, 2020
- A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memoryScience of The Total Environment, 2019
- Void Fraction Measurement of Gas-Liquid Two-Phase Flow Based on Empirical Mode Decomposition and Artificial Neural NetworksHeat Transfer Engineering, 2018
- A new contactless impedance sensor for void fraction measurement of gas–liquid two-phase flowMeasurement Science and Technology, 2016