基于NSGA-II优化的流形微通道散热器在SiC功率模块中实现更优散热与热均匀性

Silicon carbide (SiC) power modules are a favorable option for renewable energy and electric vehicles, thanks to the superior performance of SiC bare dies. However, thermal management issues, such as high heat flux and poor thermal uniformity, have been identified as major constraints on the performance improvement of SiC power modules in practical applications. To address these challenges, this article proposes an automated optimization methodology based on the nondominated sorting genetic algorithm with an elite strategy and finite-element analysis for the manifold microchannel (MMC) heat sink. The optimized MMC heat sink is fabricated on a three-phase SiC power module for a thermal performance evaluation through a dedicated thermal test platform. Experimental results show that the optimized MMC heat sink improves the thermal uniformity by 55.6%, reduces the maximum junction-to-fluid thermal resistance of the SiC power module by 9.2% in comparison to the traditional pin-fin heat sink, and simultaneously decreases the total weight of the SiC power module with a cold plate by 8.7%.