Acoustic detection of boiling states by deep convolutional neural networks; Visual explanation of identification basis

Ueki, Yoshitaka*; Hirako, Itsuki*; Tezuka, Kosuke*; Aizawa, Kosuke ; Ara, Kuniaki*

With a final goal of early detection and understanding of the transition of coolant boiling events in the core of sodium-cooled fast reactors, our present aim is to obtain and maintain the basic knowledge necessary for developing anomaly detection technology associated with local anomalies in the core and to demonstrate basic feasibility. We constructed a deep learning method and evaluated its performance to detect the occurrence and understand the transition of subcooled boiling using acoustic identification. In this research, we aim to acquire acoustic data during subcooled boiling of ultrapure water and learn feature quantities of the boiling in time-frequency expression. A deep learning model of a convolutional neural network for label classification was constructed. In addition to being able to identify the occurrence of boiling with high accuracy, the visualization of the identification basis using the gradient-weighted class activation mapping (Grad-CAM) method revealed the acoustic frequency bands that the deep learning model determined to be of high importance. We also constructed a regression analysis-type deep learning model and demonstrated that boiling heat flux values can be predicted with high accuracy.