Aiming to address the airflow regulation challenges during visual detection and foreign fiber (e. g., film) removal in seed cotton, finite element software was used to simulate and calculate relevant parameters, analyze their impact on flow field characteristics, in order to optimize the flow field structure and improve impurity removal efficiency and conveying stability. A three-dimensional multiphysics geometric model was developed, incorporating key components such as the seed cotton inlet, impurity removal outlet, cotton outlet, nozzles, and auxiliary air inlets. Based on the SST k-ω turbulence model, the systematic simulation analysis was conducted on the angle and velocity of the auxiliary air inlets. The simulation results revealed that when the seed cotton inlet velocity was set as 9 m/s, the impurity removal air inlet velocity as 1 m/s, the cotton outlet auxiliary air inlet velocity as 6 m/s, and the auxiliary airflow angle as 15°, the airflow velocity distribution at both the impurity removal and cotton outlets was optimized, effectively meeting the operational requirements for foreign fiber separation and cotton discharge, while minimizing flow blockage and recirculation within the conveying pipeline. Prototype testing under the optimized simulation parameters confirmed that the integrated system, combined with machine vision-based recognition, achieved a foreign fiber removal rate of 90% and a processing capacity of 4 tons of seed cotton per hour.