Management of heat transfer and hydraulic characteristics of a micro-channel heat sink with various arrangements of rectangular vortex generators utilizing artificial neural network and response surface methodology

It is common to use micro-channel heat sinks (MCHSs) in equipment such as; ICs, transistors, LEDs, and high-power lasers, which generate heat due to the passage of electric current. This heat is often a menace to harm these devices and their internal parts. For this reason, heat rejection in the MCHSs is an endless challenge for researchers. Placing vortex generators (VGs) within the MCHS improves the cooling capacity but incurs a considerable pressure drop. Meanwhile, the shape, geometric dimensions, and arrangement of the VGs significantly affect this heat transfer. In the current study, the placement angle (theta), the longitudinal distance (dl), and the transverse distance (dt) of the VGs were chosen to be altered. The Artificial Neural Networks (ANN) and Response Surface Methodology (RSM) were exerted to study their variation's effect on the Nusselt number (Nu) and pressure drop (Delta P) of an MCHS. The presented data illustrated that the results of the ANN model were closer to the data provided by the numerical simulation. With the coefficient of determination of 0.995 and 0.992 in forecasting the Nu and Delta P, the ANN exhibited better performance than the RSM model. Besides, the ANN model recommended that to acquire the highest relative efficiency index, the optimum values of placement angle, the longitudinal and transverse distances of the VGs should be 60, 0.151 mm, and 0.166 mm, respectively.

Automated summary

Using micro-channel heat sinks (MCHSs) in electronic devices is common but the generated heat poses challenges in heat rejection. Placing vortex generators (VGs) improves cooling but increases pressure drop. In this study, ANN and RSM models were used to analyze the effect of varying VGs' placement angle, distances on the Nusselt number and pressure drop of MCHSs.