Novel models for predicting the shape and motion of an ascending bubble in Newtonian liquids using machine learning

Abstract

As a conventional and persistent topic, a single bubble freely ascending in Newtonian liquids is investigated based on its shape and motion predictions using the strategy of machine learning. The dataset for training, validating, and testing neural networks is composed of the current experimental results and the extensively collected data from previous research works, which covers a broad range of dimensionless parameters that are [Formula: see text], and [Formula: see text]. The novel models of the aspect ratio E and drag coefficient [Formula: see text] are proposed based on a backpropagation neural network. The comparisons of the conventional correlations indicate that the new E model presents a significant superiority. This E model also has a good capability to predict the minimum E as about 0.26 that is consistent with the theoretical value [Formula: see text]. Moreover, the [Formula: see text] models are divided into E-independent and E-dependent types. The performances of these two type models are quite similar and both agree well with the experimental results. The errors of the [Formula: see text] predictions for Re > 1 are mostly in the range of [Formula: see text].

Deep Learning Bubble Segmentation on a Shoestring

“…But how do we take these AI-based techniques into our own lab? − “Bubble” is one of the thousand categories used in ImageNet; however, unfortunately, it refers to soap bubbles instead of gas bubbles dispersed in a liquid, the subject of this paper. In ImageNet there are roughly one million images divided over 1000 categories.…”