Physics-Based Hybrid Machine Learning for Critical Heat Flux Prediction with Uncertainty Quantification
This work addresses the problem of accurate and reliable critical heat flux prediction for nuclear reactor safety and performance, representing an incremental improvement by integrating existing methods.
The study tackled predicting critical heat flux in nuclear reactors by developing a hybrid model combining physics-based correlations with machine learning for uncertainty quantification, achieving a mean absolute relative error of 1.846% with the best-performing model and showing that hybrid models outperform pure machine learning, especially in data-scarce scenarios.
Critical heat flux is a key quantity in boiling system modeling due to its impact on heat transfer and component temperature and performance. This study investigates the development and validation of an uncertainty-aware hybrid modeling approach that combines machine learning with physics-based models in the prediction of critical heat flux in nuclear reactors for cases of dryout. Two empirical correlations, Biasi and Bowring, were employed with three machine learning uncertainty quantification techniques: deep neural network ensembles, Bayesian neural networks, and deep Gaussian processes. A pure machine learning model without a base model served as a baseline for comparison. This study examines the performance and uncertainty of the models under both plentiful and limited training data scenarios using parity plots, uncertainty distributions, and calibration curves. The results indicate that the Biasi hybrid deep neural network ensemble achieved the most favorable performance (with a mean absolute relative error of 1.846% and stable uncertainty estimates), particularly in the plentiful data scenario. The Bayesian neural network models showed slightly higher error and uncertainty but superior calibration. By contrast, deep Gaussian process models underperformed by most metrics. All hybrid models outperformed pure machine learning configurations, demonstrating resistance against data scarcity.