Abstract: In order to design high performance heat sink under engineering constraints, the optimization objective function of heat sink is established based on the thermal resistance network model of the heat sink. The best geometry parameters of the heat sink under engineering constraints are obtained through genetic algorithm objective function optimization. The results are in agreement with the simulation results by commercial software, which verifies the effectiveness and accuracy of the model. The research of this paper provides a good reference for improving the design capability of high performance heat sink and the development efficiency of new products.