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

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