Thermal–hydraulic performance optimization of Tesla-type minichannel structures for AUV battery thermal management using genetic algorithms

A Tesla-type minichannel liquid cooling strategy is proposed to address thermal management challenges of lithium-ion batteries in compact autonomous underwater vehicles (AUVs) operating under high-rate discharge. A coupled thermal–fluid simulation model is established, incorporating internal heat generation and variations in state of charge. Response surface methodology (RSM) is employed to quantify the influence of key geometric parameters on thermal resistance and pumping power. A multi-objective optimization is performed using the NSGA-II algorithm, and the optimal configuration is identified through the TOPSIS method with entropy weighting. At a discharge rate of 3C, the optimized reverse Tesla valve (RTV)-type channel achieves a 48 % (0.009 K/W) reduction in thermal resistance and a 21 % (0.023 W) decrease in pumping power compared to baseline designs. At a Reynolds number of 2400, the RTV configuration reduces the average battery temperature by up to 14.8 % versus the I-type channel, and by 41.3 % compared to a non-cooled system. Experimental validation is conducted using a custom test platform with the cold plate placed between lithium-ion batteries. At a discharge rate of 1C and a coolant flow rate of 1080 mL/min, the RTV plate maintains the battery surface temperature below 25.9 °C, with a cold plate surface temperature difference of only 1.3 °C, lower than that of the forward Tesla value (FTV) configuration. Although a higher pressure drop is observed, the RTV channel provides a favorable balance between heat dissipation and energy efficiency. These results confirm the feasibility and effectiveness of the proposed AUV battery thermal management design and are expected to promote the broader application of Tesla-valve structures in the thermal control of underwater energy systems.