Optimization of multi-orifice synthetic jet configuration using flow and heat transfer characteristics coupled with machine learning

Abstract

This study investigates the heat transfer and flow characteristics of multi-orifice synthetic jet (SJ) to identify the most efficient configuration for enhanced cooling, particularly in regions where conventional single-orifice jets are limited by recirculation effects. The heat transfer characteristics are evaluated using infrared thermography, while flow behavior is analyzed through hot-wire anemometry and smoke-wire flow visualization. A total of 23 orifice configurations, including single, two, three, four, and eight-satellite arrangements, are examined while maintaining a constant total orifice area. Results indicate that the two, three, and four-satellite configurations exhibit enhancements of 19.8 %, 27.25 %, and 26.63 %, respectively, over the single-orifice configuration at Z/D = 2, while the eight-satellite configuration shows no significant improvement. The peak heat transfer rate is observed for the two-satellite configuration at Z/D = 4, which is 13.75 % higher than that of the single-orifice configuration. At larger jet-to-surface spacings, the single-orifice configuration demonstrates superior performance due to the formation of more coherent and stable vortex structures. Hot-wire anemometry measurements reveal that differences in the velocity ratio between the central and satellite jets strongly affect flow interaction and vortex formation, thereby influencing the heat transfer rate. Furthermore, flow visualization confirms that smaller central orifices reduce flow recirculation and promote enhanced near-wall mixing, supporting the observed heat transfer trends. An artificial neural network (ANN) model developed to predict the heat transfer behavior of SJ impingement achieves a maximum prediction error of 6.14 % and an R² value of 0.99. Overall, the findings provide valuable insights for optimizing multi-orifice synthetic jet configurations to achieve efficient cooling in compact thermal management systems.