The characterization of microbubbles for venturi tube is important for the associated industrial applications, but still challenging due to the coupling effects of numerous operating factors. Here, we report a machine learning (ML)-aided approach for predicting the characteristics of microbubbles generated by venturi tube. Full factorial design of experiments (DOE) was first carried out, followed by the image post-processing to obtain multi-dimensional dataset. After data cleaning, MLP (Multi-Layer Perception), random forest (RF) and Catboost models were trained to correlate the Sauter mean diameter (d_s) to five operating features, namely, throat-to-outlet ratio β, divergent angle θ, gas-to-liquid ratio α, gas Reynolds number Re_g and liquid Reynolds number Re_l. All three ML models provide excellent predictability on d_s, while the Catboost model displays the best extrapolation performance in three investigated scenarios. Internal importance analysis shows that the throat size and Re_g play the greatest and least influence on d_s, respectively. We also explored the mathematical fitting approach based on obtained experimental dataset. The results show that ML models deliver improved predictive performance over mathematical model, but the latter provides better mechanistic interpretability. This work demonstrates the great potential of ML in the gas–liquid multiphase flow.

文丘里管微气泡的表征对于相关的工业应用很重要，但由于众多操作因素的耦合效应仍然具有挑战性。在这里，我们报告了一种机器学习 (ML) 辅助方法，用于预测文丘里管产生的微气泡的特征。首先进行全因子实验设计（DOE），然后进行图像后处理以获得多维数据集。数据清理后，训练 MLP（多层感知）、随机森林 (RF) 和 Catboost 模型，将 Sauter 平均直径 ( d _s ) 与五个操作特征相关联，即喉道与出口比β、发散角θ , 气液比 α, 气体雷诺数Re_g和液体雷诺数Re _l。所有三个 ML 模型都对d _s提供了出色的可预测性，而 Catboost 模型在三个调查场景中显示了最佳外推性能。内部重要性分析表明，喉管尺寸和Reg _g分别对d _s影响最大和最小我们还探索了基于获得的实验数据集的数学拟合方法。结果表明，ML 模型提供了比数学模型更好的预测性能，但后者提供了更好的机械解释性。这项工作证明了 ML 在气液多相流中的巨大潜力。