The 10th Asian Symposium on Computational Heat Transfer and Fluid Flow ( ASCHT2025)

This study proposes a Genetic Algorithm-based Topology Optimization (GATO) framework for enhancing the thermal management of High Voltage Direct Current (HVDC) equipment. The methodology integrates genetic algorithms with topology optimization to automatically generate complex internal cooling channel structures within a fixed heat sink geometry. Using binary matrix representation, the framework iteratively evolves efficient flow paths under multi-source heat loads. The optimized designs are evaluated via 3D CFD simulations demonstrating superior performance in reducing peak temperatures and improving thermal uniformity. Dual termination criteria—objective function convergence and structural stabilization—ensure both accuracy and robustness. Results highlight the GATO method's scalability and effectiveness, offering a systematic and intelligent strategy for advanced thermal design in high-power electronic systems.

Genetic algorithm,Topology Optimization (TO),Computational fluid dynamics,Multi-source heat dissipation,High-voltage direct current