Abstract: To address the issue of uneven airflow distribution in the thermal management of containerized energy storage systems, this paper proposes an optimization design method for air duct structure based on response surface methodology (RSM). Design variables such as inlet height and width are used to establish response surface models of airflow non-uniformity and air resistance through coupled CFD simulations, enabling multi-objective optimization to determine optimal structural parameters. Results show that the optimized air duct significantly improves outlet airflow uniformity and effectively reduces the system temperature gradient. Simulations indicate that the maximum temperature difference in a dual-cluster model is controlled within 4.5 °C, with temperature rise not exceeding 8.5 °C. Field test data under dynamic variable operating conditions show a maximum temperature difference of approximately 4 °C and a maximum temperature rise of 7 °C, with uniform temperature distribution consistent with simulation trends. The proposed asymmetric air duct design—featuring top-side supply and bottom return airflows—combined with terminal outlet regulation strategy, achieves efficient airflow organization within a compact space, validating the effectiveness and engineering feasibility of the thermal management solution under complex operating conditions.