<ul><li>Critical heat flux (CHF) prediction is crucial for efficiency and safety in nuclear reactors to prevent equipment damage.</li><li>Traditional machine learning approaches have shown potential for CHF prediction but lack interpretability and reliability in diverse operational conditions.</li><li>Hybrid models, combining data-driven ML with physics-based models, demonstrate lower error metrics and improved accuracy compared to empirical correlations.</li><li>Integration of ML-based CHF models into subchannel codes shows promise in enhancing performance over conventional methods.</li></ul>

Deployment of Traditional and Hybrid Machine Learning for Critical Heat Flux Prediction in the CTF Thermal Hydraulics Code

Discover more