Supercritical water heat transfer coefficient prediction analysis based on BP neural network

Abstract

14 groups of experimental data have been collected for determining the heat transfer coefficient within the scope of supercritical water pressure. On the basis of experimental data, the BP neural network predictive model has been built for the determination of supercritical water heat transfer coefficient. Using the BP neural network prediction model, a study has been conducted to determine effect of changing certain parameters such as heat flux, mass flux, pipe diameter and pressure on the heat transfer coefficient of supercritical water. The prediction results show that the mean error, standard deviation and the root mean square error are 0.179032187%, 4.128897482% and 0.179032187% respectively, the regression coefficient R is 0.97165 and the maximum error is 16.06332%. The trained BP neural network prediction model can applied for better prediction and understanding of the heat transfer coefficient of supercritical water. The prediction range is as follow: specific enthalpy is 451.30–3135.87kJ/kg, mass flux is 400–3000kg/m2s, heat flux is 200–2960kW/m2, pressure is 22.6–31MPa and diameter is 0.7–38.1mm.