Nuclear reactors generally operate at conditions where critical heat flux (CHF) does not occur. However, it is most likely to occur during the transient conditions following an accident such as LOCA, LOFA, or power excursion. Under such conditions, the fuel rod surface temperature becomes too high to maintain a continuous liquid contact that causes an abrupt rise of the cladding temperature, which can lead to a rupture of the cladding material. Therefore, the system's structural integrity is affected, resulting in severe hazards and significant economic losses.Accurate prediction of CHF is of significant importance for the safe operation of nuclear power reactors. It is a helpful tool in defining the optimal operating conditions and maintaining the system integrity within safety limits. CHF prediction is usually based on empirical correlations since theoretical models are rather complex, and the associated physical mechanisms are not yet fully understood. Empirical correlations are derived from experimental data performed in very limited conditions in terms of mass velocity, pressure, and flow geometry.Three Hybrid artificial neural networks (ANN) models namely radial basis function (RBF), generalized regression neural networks (GRNN), and multi-layer perceptron (MLP) combined with empirical mode decomposition (EMD) are developed for CHF predictive modeling using CHF experimental databases.First, the original experimental inputs data series are decomposed into several intrinsic mode functions (IMFs) and one residual by EMD, whose components are divided into high, medium, and low components.The performance parameters indicate that the mean absolute percentage error (MAPE) of the EMD-RBF, EMD-GRNN, and EMD-MLP are 20.6403, 12.6121, and 5.0624; the root mean square error (RMSE) are 0.8862, 0.6303, and 0.3993; the mean absolute error (MAE) are 0.6678, 0.4470, and 0.1880. The values of the R square of the developed prediction approach utilizing EMD-RBF, EMD-GRNN, and EMD-MLP models are 0.8415, 0.9275, and 0.9793, and the index of the agreement are 0.9443, 0.9709, and 0.9898., The value of the R-square and the index of agreement of the proposed models are much higher than those of the simple models.Pearson's test results show that the association strength between the actual value and the predicted values of the proposed model EMD-MLP is the strongest. These results show the following: (a) compared with other related, recent studies, the prediction accuracy of the hybrid model EMD-MLP proposed in this research is the best hybrid model; (b) the proposed hybrid model (EMD-MLP) attains superior performance compared with single models.

Hybrid Improved Empirical Mode Decomposition and Artificial Neural Network Model for the Prediction of Critical Heat Flux (Chf)

Paper Proceedings: https://asmedigitalcollection.asme.org/ICONE/ICONE28/volume/85246

2021 28th International Conference on Nuclear Engineering (ICONE28)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.