Temperature-gradient-aware bionic optimization method for heat source distribution in heat conduction

Heat source distribution optimization is a fundamental problem for electronic cooling, which influences the performance and the life of the electronics. In this paper, the temperature-gradient-aware bionic optimization (TGBO) is developed for heat source distribution problem in heat conduction. Mathematical analysis is conducted to figure out the influence of the added heat source on the temperature gradient field. Based on the analyzed result, the placement strategy of the heat sources is proposed for bionic optimization, which suggests the added heat source be located at the position with the maximum magnitude of temperature gradient. Three typical numerical cases are used to validate the effectiveness of the proposed optimization method and to compare the solutions with the ones of the temperature-aware bionic optimization (TBO) in previous study. The optimized results indicate that TBO is likely to arrange the heat sources with some certain distances among each other and is more suitable for the problems with relatively symmetric boundary conditions. While for TGBO, the final solutions are likely to locate the heat sources adjacent to each other. TGBO performs better for the problems with asymmetric boundary conditions. Moreover, the total calculation of TGBO is less than the one of TBO in general, especially for the cases with symmetric boundary conditions. The influence of the initial position of the first heat source on the performance of bionic optimization method is also investigated. The results show that the first heat source is better to be placed at the position with the best heat transfer condition, which can help to improve the performance of the final solution and reduce the number of the steps during the adjusting stage. It is suggested that temperature-gradient-aware bionic optimizations is an effective method to optimize the heat source distribution for heat conduction enhancement. (C) 2016 Elsevier Ltd. All rights reserved.