A Variable-Fidelity Multi-Objective Evolutionary Method for Polygonal Pin Fin Heat Sink Design

For the multi-objective design of heat sinks, several evolutionary algorithms usually require many iterations to converge, which is computationally expensive. Variable-fidelity (VFO) methods were suggested improve efficiency algorithms. However, problems are seldom optimized using VFO. Therefore, a variable-fidelity method (VFMEM) was suggested. Similar other algorithms, VFMEM solves high-fidelity model low-fidelity model. Compared with distinctive characteristic its application in optimization. First, uses locate region where global optimal solution might be found. Sequentially, both high- and models can integrated find real solution. Circulation distance elimination (CDE) uniformly obtain PF. To evaluate feasibility VFMEM, two classical benchmark functions tested. widely used particle swarm optimization (MOPSO), significantly improved Pareto frontier (PFs) could also obtained. algorithm’s feasibility, polygonal pin fin sink (PFHS) carried out VFMEM. initial design, results showed that mass, base temperature, temperature difference designed optimum decreased 5.5%, 18.5%, 62.0%, respectively. More importantly, if completed directly by MOPSO, computational cost entire procedure would increased.