Novel design to enhance the thermal performance of plate-fin heat sinks based on CFD and artificial neural networks

Abstract

• A novel design to enhance thermal performance of plate-fin heat sinks. • Artificial neural network. • Numerical analysis based on Computational Fluid Dynamic. • Plate-fin heat sink with symmetrical half-round hollow pins in vertical arrangement. Over the last decades, intensive attentions have been spent for thermal performance enhancement of the heat sinks as a result of the heat dissipation problems in an extremely competitive industry of electronics. In the present work, an efficient thermal design of a plate-fin heat sink with symmetrical half-round hollow pins vertically arranged and subjected to parallel flow is implemented. In particular, a computational fluid dynamic (CFD) analysis was performed for evaluating the thermal performance of the different possible designs, i.e. different values for inside and outside diameters of the attached hollow pins. These limited data points are then fed into a feed-forward back-propagation neural network to predict the base temperature and pressure drop that can be based on to judge the most effective geometry of the hollow pins. This is next followed by a CFD analysis of the pin’s pitch effect on the thermal performance of the studied heat sinks to find the most efficient design. The study has shown that the proposed design shows higher thermal performance over the other configurations in the literature while maintaining the bare minimum change in fabrication and implementation. Therefore, this design has a promising potential to enhance the thermal efficiency of the electronic devices.