Prediction of Free Convection from Vertical and Inclined Rows of Horizontal Isothermal Cylinders Using ANFIS

The current study is conducted to predict the laminar free convection from horizontal isothermal cylinder sets in vertical and inclined rows, using an adaptive neuro-fuzzy inference system (ANFIS). The effects of the Rayleigh number, vertical and horizontal separation distances on average heat transfer from the rows are considered via this prediction. The training data for optimizing the ANFIS structure is based on available experimental data. A hybrid learning algorithm consisting of the gradient descends method and least-squares method is used for ANFIS training. In the view of the physics of the problem, it was observed that in the vertical row, by increasing the vertical separation spacing heat transfer increases. Moreover, the heat transfer increases by increasing the horizontal spacing ratio. In addition, according to good consistency obtained between the experimental and predicted results, it can be concluded that ANFIS can be used to predict the experimental results accurately.