KAGAKU KOGAKU RONBUNSHU, Vol.27, No.4, 461-465, 2001
Prediction of dynamic heat transfer of a 2-D bubble column using ANNs
The instantaneous local heat transfer rates in a 2-D bubble column were measured by using a hot -wire probe. An artificial neural network (ANN) was applied. to correlate the fluctuation of heat transfer rates with dynamic motions of bubble and liquid. The optimal structure of the artificial neural network was determined for the prediction of heat transfer rates. The trained ANN model gives a good prediction of dynamic heat transfer with errors of less than 5%.