Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 27, 2026Scientific ReportsOpen Access

A non-intrusive framework using acoustic signals and deep learning for boiling diagnostics in visual-limited environments

View Full Paper
Ask AI
Bookmark
Share

Authors

PHPei-Hsun HuangJSJee Hyun SeongJCJonathan Mario Castro-Aguilar

Discussion

Loading...

Member takes

Overview

Demonstrates a novel approach for monitoring boiling characteristics in inaccessible environments, suggesting improved thermal safety assessment techniques.

Key Points

  • The aim is to develop a non-intrusive framework for monitoring boiling in situations where visual access is restricted or impossible.
  • Utilized hydroacoustic sensing to capture boiling-induced acoustic emissions.
  • Transformed signals into Short-Time Fourier Transform (STFT) spectrograms.
  • Employed a convolutional neural network (CNN) to infer key boiling parameters.
  • Integrated results with an ANSYS CFX wall-boiling model for validation.
  • The CNN achieved high predictive accuracy for heat flux and wall superheat.
  • Demonstrated robustness under varying acoustic noise conditions down to 0 dB SNR.
  • Inferred boiling parameters matched benchmarks from image-based methods.
  • Confirmed generalizability across varied operational conditions including temperature and flow rate.

Cite This Study

Huang et al. (2026) studied this question.

synapsesocial.com/papers/69c61fa915a0a509bde18163https://doi.org/10.1038/s41598-026-41757-z
View Full Paper
Ask AI
Bookmark
Share