LI Q, CAO Y X, TANG Y Q,et al.Prediction model for gas volume fraction in high-altitude tunnels based on multi-factor coupling[J].Journal of Henan Polytechnic University(Natural Science) ,2026,45(4):86-94.

doi:10.16186/j.cnki.1673-9787.2025120037

Received:2025/12/11

Revised:2026/03/31

Published:2026/06/17

Prediction model for gas volume fraction in high-altitude tunnels based on multi-factor coupling

Li Qi1, Cao Yuxiang1, Tang Yuqiu2, Lei Peng1, Li Min1

1.School of Architecture and Urban-Rural Planning， Sichuan Agricultural University， Chengdu  611830，Sichuan，China;2.Yibin Tianyuan Group Co.， Ltd.， Yibin  644000， Sichuan，China

Abstract: Objectives To investigate the effects of altitude， gas emission rate，  distance between the air supply outlet and the tunnel face， and ventilation velocity on the diffusion and migration of gas in high-altitude gas tunnels during construction， a multi-factor coupled prediction model for gas concentration at the tunnel face was established. Methods A three-dimensional numerical model was established using computational fluid dynamics （CFD） software. Nineteen working conditions were designed using a single-factor control variable method to analyze the influence of each key factor on gas migration in the tunnel. Based on a four-factor， four-level orthogonal experimental design and multivariate nonlinear regression analysis， a multi-factor coupled prediction model for gas concentration at the tunnel face was constructed. Field measurements and numerical simulation results were used to validate the model accuracy.  Results The results show that the gas concentration at the tunnel face increases exponentially with altitude. When altitude increases from 0 km to 5 km， the average gas concentration in the stable flow region rises from 0.072% to 0.128%. A strong positive linear correlation exists between gas concentration and gas emission rate. When the gas emission rate increases from 1 m³/min to 7 m³/min， the average gas concentration at different sections in the stable flow region stabilized within the range of 0.03%-0.23%. The minimum gas concentration at the tunnel face occurs when the distance between the air duct outlet and the face is 10 m. Gas concentration shows a clear inverse power-law relationship with ventilation velocity， and a reduction of up to 44.4% is observed when velocity exceeds 16 m/s. The proposed model achieves a coefficient of determination of 0.903， and the relative prediction error is within 2.79%-4.54%.  Conclusions The proposed model can effectively predict the gas concentration at the tunnel face under varying altitudes， gas emission rates，  ventilation velocities， and air duct positions， providing a scientific basis for ventilation design and gas hazard prevention in high-altitude gas tunnels.

Key words:high altitude;construction ventilation;gas migration;prediction model;computational fluid dynamics