Thermal-hydraulic analysis of a circular-finned pin-fin heat sink and performance prediction using Artificial neural networks

As the power density of electronic devices increases, the demand for efficient thermal management is also increasing. This study proposed to evaluate whether a new structure, in which circular pins are attached to a conventional pin-fin heat sink, can contribute to improved thermal performance. The thermal-hydraulic characteristics of the proposed shape are numerically investigated using the finite volume method with the standard k-ε turbulence model at various Reynolds number(8500, 13000, 17500, and 22000), height of fin(7.5 mm, 12.5 mm, and 17.5 mm), and fin angles(0°, 10°, 20°, 30°, and 40°). The circular fins disturb the flow and generate vortices, guiding flow closer to the heat source as the fin angle( θ ) increases, thereby enhancing heat transfer. The results show that under optimal conditions, the performance evaluation criteria( PEC ) is improved by up to 32% compared to the existing pin-fin heat sink, and the total entropy generation is reduced by up to 35%. In particular, the optimal design resulted in a 2% improvement in PEC and a 10% reduction in total entropy generation, compared to the configuration with a fin angle of θ = 0°. To enable fast and accurate prediction, an Artificial Neural Network(ANN) prediction model was developed. The ANN model demonstrated high accuracy, achieving an average relative error of less than 2% compared to the numerical simulation results. This study investigates the underlying physical mechanisms of the circular-finned pin-fin heat sink and offers guidelines for the optimization of electronic cooling system design through rapid and accurate performance prediction.