(9) Multiphase Flows

[ICJWSF2022-I06] Identification of Flow Regime in Transition Region of High-Pressure Gas-Liquid Two-Phase Flow Using Image Reconstruction and Machine Learning

*Naoto Shibata^1,3, Shuichiro Miwa², Kazuhiro Sawa³, Masahiro Takahashi¹, Tetsuro Murayama¹, Norio Tenma⁴ (1. Mitsui E&S Machinery Co., Ltd. (Japan), 2. The University of Tokyo (Japan), 3. Hokkaido University (Japan), 4. National Institute of Advanced Industrial Science and Technology (Japan))

Keywords:gas-liquid two-phase flow, time-strip methods, machine learning, flow regime, convolutional neural network, Gaussian Mixture Model