摘要详情

ID / 提交时间

83 / 2024-08-17 12:00:03

标题

HIGH ACCURAY CHF PREDICTION USING NEWEST AI STRATEGIES

关键字

Critical heat flux; Heat transfer; Deep learning; Neural network

主题及专题

TRACK 5: NUCLEAR ENGINEERING

状态

全文录用

作者

KAI WANG / THE SUN YA-SEN UNIVERSITY

Da Wang / Iwhalecloud Co., Ltd. Building B, Yihua Industrial Park, Yuhuatai District, Nanjing City

Xiaoxing Liu / 1 Sino-French Institute of Nuclear Engineering and Technology, Sun Yat-Sen University, Zhuhai 519082, China,

Songbai Cheng / Harbin Engineering University

摘要

One of the most important and challenging job in the two-phase thermohydraulic filed is the prediction of Critical Heat Flux (CHF) as it serves a critical role in preventing the heater surface form burnout, thus ensuring the safety of the system. As the phenomena itself is so complex, a universal theory to explain all the CHF related topic is still not feasible. Thus, different methods such as empirical or semi-empirical correlations, look-up tables (LUT), CFD methods are proposed to predict CHF. Though people have tried to improve the accuracy of the CHF prediction, traditional methods always face the problem of low accuracy and universalness. AI strategies, which are developed in the recent years, have shown the great potential in accurate prediction of CHF. Compared to traditional methods like LUT, AI can greatly enhance the accuracy, showing the tremendous potential of the method. In this study, we used three different AI strategies to predict CHF, it was found that transformers outweigh among the three methods, achieving the best performance.