A high heat transfer performance of inclined rib mini-channel heat sink designed by machine learning and laser powder bed fusion

Advances in additive manufacturing technology expanded the possibilities for heat sink design. This work introduces the design, optimization, and performance of a novel inclined rib mini-channel heat sink (IR-MCHS) fabricated by additive manufacturing, which is developed based on traditional rectangular straight mini-channel heat sinks (RS-MCHS) and straight rib mini-channel heat sinks (SR-MCHS). The proposed rib structure has a non-upright configuration in the height direction. Comprehensive optimization of the mini-channel was performed using a genetic algorithm. A neural network model was trained to predict the performance of all possible structures within the parameter range and verify the accuracy of the optimization. Furthermore, the performance of the optimized IR-MCHS was evaluated under various inlet flow rates. For the considered flow rate range, the maximum performance evaluation criteria of the proposed optimized IR-MCHS are 1.26 and 1.14, respectively, compared to the baseline RS-MCHS and SR-MCHS. Additionally, the optimized IR-MCHS was fabricated using laser powder bed fusion, and its performance was validated. The experimental results were consistent with the simulation results, which verified the effectiveness and reliability of the optimized design. This paper provides a new method for designing the heat sink.