Genetic Algorithm Based Optimization of PCM Based Heat Sinks and Effect of Heat Sink Parameters on Operational Time

Using an approach that couples genetic algorithm (GA) with conventional numerical simulations, optimization of the geometric configuration of a phase-change material based heat sink (PBHS) is performed in this paper. The optimization is done to maximize the sink operational time (SOT), which is the time for the top surface temperature of the PBHS to reach the critical electronics temperature (CET). An optimal solution for this complex multiparameter problem is sought using GA, with the standard numerical simulation seeking the SOT forming a crucial step in the algorithm. For constant heat dissipation from the electronics (constant heat flux) and for three typical PBHS depths $(A)$⁠, predictive empirical relations are deduced from the GA based simulation results. These correlations relate the SOT to the amount of phase change material to be used in the PBHS $(φ)$⁠, the PBHS depth $(A)$⁠, and the heat-spreader thickness $(s)$⁠, a hitherto unconsidered variable in such designs, to the best of the authors’ knowledge. The results show that for all of the typical PBHS depths considered, the optimal heat-spreader thickness is 2.5% of the PBHS depth. The developed correlations predict the simulated results within 4.6% for SOT and 0.32% for $ϕ$ and empowers one to design a PBHS configuration with maximum SOT for a given space restriction or the most compact PBHS design for a given SOT.