Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
植木 祥高*; 平子 樹*; 手塚 晃輔*; 相澤 康介; 荒 邦章*
AI Thermal Fluids (Internet), 4, p.100021_1 - 100021_12, 2025/12
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.
植木 祥高*; 平子 樹*; 手塚 晃輔*; 相澤 康介; 荒 邦章*
Proceedings of 12th International Conference on Multiphase flow (ICMF2025) (Internet), 2 Pages, 2025/05
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 of convolutional neural network and evaluated its performance to detect the occurrence and understand the transition of subcooled boiling using acoustic identification. In addition to being able to identify the occurrence of boiling with high accuracy, the visualization of the identification basis based on Grad-CAM 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 it is possible to predict boiling heat flux values with high accuracy.
手塚 晃輔*; 植木 祥高*; 相澤 康介; 荒 邦章*
no journal, ,
In Japan, the government has announced a carbon neutral declaration and formulated a green growth strategy. Nuclear power generation is being pursued as one of the options to realize decarbonization in power sector, and one of the goals is to utilize sodium-cooled fast reactors (SFR), which are positioned as next-generation reactors. The recent design of demonstration SFR requires a higher level of accident tolerance in response to the demand for enhanced safety, so an early detection of anomalies during in-service operation is important. In this study, we assume the case where the coolant flow rate is reduced due to local blockage and coolant boiling occurs in the fuel assembly due to overheating. To understand the physical mechanism of the boiling acoustics, we investigate the correlation between the sound generated during the subcooled pool boiling of water and the solid vibration based on frequency analysis. As a result, we confirm depending on the wire diameter, that is a geometric parameter of the heat transfer surface, boiling-acoustic frequency peaks shift and overlap with the band of resonance frequencies of the secondary mode. It suggests that some features of the boiling acoustics are highly correlated with the resonance phenomena of the heating solid.
