A multiobjective optimization of area‐to‐point heat conduction to minimize both mean temperature and temperature variance is conducted based on a decomposition‐based multiobjective binary quantum‐behaved particle swarm optimization (PSO) method (MOMBQPSO/D). The MOMBQPSO/D adopts the framework of the multiobjective evolutionary algorithm based on decomposition and modifies the binary quantum‐behaved PSO. In the first step of the MOMBQPSO/D, the multiobjective area‐to‐point problem is divided into a series of subproblems using Tchebycheff decomposition method. Next, all the subproblems are solved simultaneously using the modified binary quantum‐behaved PSO. Finally, a series of Pareto optimal solutions representing the conducting path structures are stepwise selected from the solutions to the subproblems. The features of the Pareto optimality‐based conducting paths and cooling performance are described. In addition, the effects of the conductive material quantity, optimization objective, heat sink location, and heat source distribution on the conducting path structure and cooling performance are discussed.

中文翻译：

---

基于二元量子行为的粒子群算法和切比雪夫分解方法的多点优化点对点导热结构

基于分解的多目标二进制量子行为粒子群优化（PSO）方法（MOMBQPSO / D），进行了一个区域到点导热的多目标优化，以最小化平均温度和温度变化。MOMBQPSO / D采用基于分解的多目标进化算法框架，并对二进制的量子行为PSO进行了修改。在MOMBQPSO / D的第一步中，使用Tchebycheff分解方法将多目标点对点问题划分为一系列子问题。接下来，使用改进的二进制量子行为PSO同时解决所有子问题。最后，从子问题的解中逐步选择一系列代表传导路径结构的帕累托最优解。描述了基于帕累托最优性的传导路径和冷却性能的特征。此外，还讨论了导电材料的数量，优化目标，散热片位置和热源分布对导电路径结构和冷却性能的影响。