摘要
To solve the multi-variable and multi-objective optimization problem in the thermal design process of the dual-input aeronautic static inverter,an optimization method based on the combination of the multi-objective evolutionary algorithm based on decomposition(MOEA/D)and the fuzzy set theory is proposed.The heat transfer path of the power device is analyzed and an equivalent heat circuit is conducted.We take junction temperature of the power device,mass,and cost of the heat sink as optimization goals,and take the heat sink structure parameters as design variables to conduct thermal optimization based on MOEA/D.This paper carries out a comparative study,and the results show that the proposed improved algorithm can meet the different requirements for multi-objective weights,and have good rapidity and robustness.
提出基于MOEA/D和模糊集理论相结合的优化方法,解决双输入航空静止变流器热设计过程中多变量多目标的寻优问题。分析功率器件的传热路径并构建等效热路,以功率器件结温、散热器质量、散热器成本为优化目标,以散热器的结构参数为设计变量,基于MOEA/D开展热优化。本文进行了比较研究,结果表明所提出的算法不仅能够满足多目标权重的不同要求,且快速性和鲁棒性好。
基金
supported by the National Natural Science Foundation of China(Nos.U1933115,U2133203)