심층퍼지신경망을 이용한 임계유동 예측

Abstract

In order to understand and quantify reactor coolant leakage in nuclear power plants, it is important to identify and simulate the leak phenomena of the fluid. When reactor coolant leaks from a high pressure and temperature reactor coolant system, two-phase mixture or vapor is ejected by the flashing phenomenon and has critical flow features. Therefore, critical flow analysis is needed to identify and quantify the leakage phenomena. In this study, deep fuzzy neural networks (DFNNs) based on the artificial intelligence methods are developed to predict the critical flow. The total acquired data used for predicting critical flow consist of 5116 input and output data pairs. The DFNNs refer to a methodology that includes the previously developed cascaded fuzzy neural networks (CFNNs) and the simplified cascade fuzzy neural networks (SCFNNs). The proposed DFNN model uses the data obtained from the Henry–Fauske model to understand the fluid properties and predict the critical flow. This model performed very well when compared to existing the machine learning methods. It is expected that effective information will be provided to quantify the reactor coolant leak from the DFNN model. Also, the developed DFNN model can be used as a universal standalone program to predict critical flow, which runs faster without iterative calculations, and without a need for steam tables.