A computational fluid dynamics approach for predicting the transportation of subway tunnel particles to platform in train movement-induced unsteady fluid flow

Air quality at subway platforms has raised increasing concerns nowadays. Tunnel particulate matter (PM) can be carried into platforms by piston wind and worsen the air quality at platforms. The national standards in China specify the maximum acceptable PM10 (with a diameter of less than 10 μm) concentration at the platform. In this paper, a computational fluid dynamics model with a discrete phase model was developed and validated through experiments. The model was used to study the dynamic distribution of airflow and PM10 during the train passage through a station. The influence of five key factors on the PM10 concentration at the platform was evaluated through gray correlation analysis. The results showed that train deceleration had the largest impact on PM10 concentration at the platform with a maximum gray correlation degree of 0.85, followed by the blockage ratio, train top speed, tunnel length, and train length. The maximum PM10 concentration increased by 254.74% when the train deceleration increased from 0.5 to 2.5 m/s2 after excluding the background concentration. A case study was conducted to observe PM10 concentrations during the entire operation hours. The two peaks of PM10 concentration were 188 and 180 μg/m3, which occurred during the morning and evening rush hours, respectively. PM10 concentrations during the morning and evening rush hours exceeded the specified concentration limit for 2.1 and 2.6 h, respectively. Based on the above results, four relevant suggestions for ventilation and dust removal were proposed for improving the air quality at platforms.