Application of artificial neural network for mixed convection in a square lid-driven cavity with double vertical or horizontal oriented rectangular blocks

Abstract The practicability of using Artificial Neural Network (ANN) to predict the thermal behaviour due to mixed convection is established. Numerical simulations are conducted first for a laminar mixed convection problem in a lid-driven square cavity with two internal rectangular blocks, oriented vertically or horizontally. CFD results are used for training and testing the ANN to predict new cases; thus, saving effort and computation time and validate the obtained numerical results of Nusselt number. A wide range of Reynolds (100 ≤ Re ≤ 1500), Grashof numbers (1.5 × 104 ≤ Gr ≤ 105), Richardson number (0.00667 ≤ Ri ≤ 10) and the distance between the two blocks (0.2 ≤ W/L ≤ 0.8) are considered. Results indicated that varying the distance W/L has an important influence on the Nusselt number. It was observed that increasing Re and Gr numbers magnitudes leads to an increased Nusselt number and that Nusselt number obtained for the case of vertical blocks is higher than the case of horizontal blocks. Furthermore, the maximum Nu number obtained for the vertical blocks was at W/L = 0.5 and Ri = 0.044 and for the horizontal blocks case was at W/L = 0.2 and Ri = 0.044. Finally, new correlations of Nu number versus Re, Gr and the spacing ratio between the two blocks W/L are derived for possible utilization in engineering design.