What question did this study set out to answer?

The aim is to evaluate the effectiveness of deep learning models in identifying boiling states using acoustic signals.

Deep-learning-based acoustic state sensing of microbubble emission boiling by image recognition using the spectrogram and the cepstrogram

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial applied deep learning to boiling sound analysis in cooling tubes, indicating effective anomaly detection.

The aim is to evaluate the effectiveness of deep learning models in identifying boiling states using acoustic signals.
Utilized three deep convolutional neural networks: AlexNet, VGG-16, and ResNet-50.
Extracted features from acoustic signals using spectrograms and cepstrograms as input data.
Conducted comparisons under subcooling conditions to assess performance in boiling state detection.
Cepstrograms achieved higher prediction accuracy than spectrograms.
ResNet-50 obtained an F1-score of 0.9971 for classification tasks and an R² of 0.9967 for regression tasks.
The proposed deep-learning approach effectively identifies and predicts boiling states in cooling systems.

Haruyama et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark