Multi-objective design majorization of a plate-fin natural convective radiator for the LED application using particle swarm majorization algorithm

A plate-fin natural convective radiator is a general cooling system that is used for LED lights. The geometric parameters majorization of the radiator is important to thermal management and materials cost. Thermal modeling and optimal design of the plate-fin natural convective radiator is present in this paper. Multi-objective particle swarm majorization is used to solve the minimum entropy generation rate and minimal material cost. The result of optimal design is called Pareto optimal solution, which is a set of multiple optimal solutions. The geometric variables like fin number, fin height and fin thickness effect on the entropy generation rate and materials cost are studied. The three-dimension heat transfer model is used to evaluate the thermal performance of the original radiator and optimized radiator. The simulation results show that the junction temperature of the optimized radiator is lower than that of the original radiator. Compared with other algorithms, the results of the MOPSO algorithm are better or competitive.