Abstract: To address the heat dissipation bottleneck of high power density industrial controllers, this paper proposes a multi-level heat dissipation optimization design based on the Icepak simulation platform. Both component-level and system-level optimizations are conducted with experimental validation. At the component level, an innovative extended heat sink structure is designed to build parallel thermal conduction paths. The junction temperature rise of the CPU and FPGA chips is reduced by 26.6% and 16.2%, respectively. At the system level, based on the flow field analysis to diagnose the air stagnation problem inside the cabinet, a ducted fan solution is designed. The junction temperature rise of the CPU and FPGA chips is reduced by 36.9% and 28.8%, respectively. The study shows that collaborative heat dissipation optimization can effectively improve the thermal management performance of industrial controllers and has strong engineering application value.