Critical Heat Flux During Boiling in Channels and Rod Assemblies: Problems of Data Description and Generalization

Abstract

Some problems arising in the description and generalization of experimental data on the critical heat fluxes appearing upon boiling in channels are due to the lack of a strict definition of the critical heat flux and strict criteria for recording the conditions under which it occurs. It is shown that the desire to reduce the number of variables in the description and generalization of the data often leads to an incorrect interpretation of the influence of individual parameters. The problem of generalizing the experimental data is discussed in connection with the presence of a strong coupling between variables. This connection is what explains the large number and the diversity of empirical relations for the critical heat flux. It is suggested that the problem of multicollinearity should be solved at the stage where the parameters of the regression model are chosen, preference being given to independent variables. Attention is called to the drawbacks of look-up tables, derived by averaging and smoothing the primary experimental data, in describing the critical heat flux.