Effects of Bronze Element Surface on Bubble Departure Size in Pool Boiling and its Prediction through ANN and GA Approaches

The present study investigates the surface roughness of the Bronze element on heat transfer in pool boiling process. The experiments were carried out with a solution of deionezed water (50%) and isopropanol (50%) in a specified container containing a hollow cylinder of bronze metal. The results indicated that the roughness index had a significant effect on the bubble dynamic. Increasing the surface roughness led to promote the bubble generation points and bubble departure diameter subsequently heat transfer enhancement through fluctuations in the solution. In addition, artificial neural network (ANN) and genetic algorithm (GA) were applied for developing the bubble departure diameter that the roughness index was one of the independent variables. Although the ANN was more capable than GA in data prediction, the GA could be employed as a more efficient and easy available approach. Therefore, the both models were powerful with acceptable errors (R2ANN=0.9982, R2GA =0.9929). Finally, the processed models were compared to Cole, Van Stralen, Lee and Stephane models. The results depicted that the ANN and GA methods had superior agreement with the experimental data than other models.