Abstract: In order to meet the intelligent development needs of the ventilation system in the heading face,aiming at overcoming the problems that the traditional local ventilation system cannot monitor and intelli-gently control the airflow state of the air outlet in real time,which leads to the unreasonable distribution of the wind speed field,the serious gas accumulation in the dead corner area and the safety hazards such as dust pollution,an airflow intelligent control system based on the digital twin technology at the air outlet of the heading face is proposed to optimize the airflow field distribution.The overall framework,operation process and key technologies of the system implementation is established,using the Zigbee self-organizing network function to monitor and collect real-time data such as wind speed,gas and dust concentration in the roadway.The ARIMA time series prediction model is used to intelligently predict and analyze the gas and dust concentration at the next moment,moreover the niche four-segment coding genetic algorithm is introduced to extract the corresponding intelligent control rules of airflow,and combined with GPRS wireless transmission technology to realize the intelligent control of airflow status of air outlet.On this basis,Unity3D is used to construct the virtual model of the system and realize the mapping interaction between the physical entity and the virtual twin.Through the design and construction of the digital twin system experimental test platform,the technical feasibility of system’s key function such as real-time monitoring,decision-making evaluation,intelligent control,and virtual-real integration are verified,besides combined with specific cases to compare and analyze the intelligent control effect of system.The results shows that after adjustment,the wind speed at the driver’s position increases significantly.The gas concentration at the top corner is reduced from 0.623% to 0.306%,and the reduction rate reaches 50.8%.The dust concentration at the driver’s position is reduced from 1 180 mg/m3 to 695 mg/m3,and the single-point dust reduction rate reached 41.1%,the dust concentration along the pedestrian height at the return air side is reduced from 430 mg/m3 to 150 mg/m3,and the dust re-duction rate reaches 65%,which further optimize the distribution of wind speed,gas and dust field in the heading face.