Analysis of the factors affecting heat transfer performance and prediction of heat transfer coefficient for MVR evaporator: A case study

Abstract

The heat transfer coefficient of evaporation is one of the key factors affecting the design and management of evaporators. This paper presented an industrial case to investigate the influencing factors on the evaporator's heat transfer coefficient, as well as the interaction between the influencing factors. Besides, a comparison of accuracies was made between the heat transfer coefficients predicted by physical models and the support vector regression (SVR) model. The study results showed that due to the interaction between the influencing factors, changes in the heat transfer coefficient varied greatly from the laboratory results. The impact of other variables must be controlled for in industrial research. There were big errors between the results predicted by the early experimental or numerical models and the observed results. The root mean square error (RMSE) values and mean absolute percentage error (MAPE) values of the physical models were all greater than 150 and 5%, respectively, and their R2 values were all lower than 0.50. The data mining-based SVR predictions were more accurate than the predictions by physical models, with a maximum relative error of 5.38% and a minimum relative error of −5.60%.