Abstract

Image-based deep learning (DL) models are employed to enable the detection of critical heat flux (CHF) based on pool boiling experimental images. Most machine learning approaches for pool boiling to date focus on a single dataset under a certain heater surface, working fluid, and operating conditions. For new datasets collected under different conditions, a significant effort in re-training the model or developing a new model is required under the assumption that the new dataset has a sufficient amount of labeled data. 

This research explores supervised, semi-supervised, and unsupervised machine learning strategies that are formulated to adapt to different scenarios. The project consists mainly of two phases:

Phase I: studies the scenario of when the new dataset has limited labeled data available. This scenario was addressed by employing Convolutional Neural Networks (CNNs) and Transfer learning (TL) to tackle such situations.

Phase II: studies the scenario of when the new dataset has no labeled data available at all. In such cases, this research presents a generalized framework where any Unsupervised Image-to-Image (UI2I) translation model could be deployed to generalize a pre-existing classification model by adapting the new dataset to look like it came from a domain similar to the one that the classification model is familiar with.

The approaches presented are the first of their kind to utilize TL and UI2I to solve the boiling heat transfer problem within the heat transfer community and are a step forward towards obtaining a one-for-all general model.