CFD-based evaluation of air induction nozzle performance: Integration with AI for predictive modeling

Air induction nozzles are commonly utilized in agricultural settings to reduce chemical drift, with internal fluid dynamics playing a critical role in their efficiency. However, experimental investigation of internal flows is often hindered by dimensional constraints. This study aims to identify optimal design parameters for air-induction nozzles to achieve a specified air-to-liquid ratio (ALR), employing a combined approach of computational fluid dynamics (CFD) and artificial intelligence methods. The impacts of several geometric factors, including the venturi air inlet diameter, length of air passage, mixing chamber length, V-cut angle, and nozzle inlet diameter, were assessed through computational fluid dynamics (CFD) analyses and AI modeling. The findings demonstrated that subtle variations in the design parameters distinctly influenced the internal two-phase flow and the resulting air-liquid ratio within the nozzle. An optimum venturi inlet diameter of 7 mm was found to produce the target air-liquid ratio of 0.00055 when combined with a throat diameter of 1.4 mm, a mixing chamber length of 7 mm, air passage length of 1.5 mm, an inlet nozzle diameter of 3 mm, and a V-cut angle of 32° The importance ranking highlighted that the venturi inlet air diameter was the most influential factor (35 %), with mixing chamber length and nozzle inlet diameter contributing equally (20 % each). These results offer practical design recommendations for refining air induction nozzle performance and demonstrate the value of integrating CFD with AI to enhance nozzle development and agricultural spraying methods.