Numerical Simulation of the Impact of Rainfall on Tunnel Fire

An improved understanding of tunnel fire dynamics is crucial for fire and life safety. This work underscores the significance of Computational Fluid Dynamics (CFD) techniques in addressing the interaction between tunnel fires and rainfall. Discrete phase model based on the Lagrangian approach is applied to simulate raindrops, and turbulence models of RANS (Reynolds Averaged Navier-Stokes) and LES (Large Eddy Simulation) are applied to simulate the fire field. This numerical work reveals the detailed fluid flow, temperature fields, and turbulence characteristics of tunnel fires under the impact of rainfall in portal. Results showed that the RANS model can predict the flow field better near the wall, while the LES model captures the small eddies in fire behaviors. Comparison between simulation and model-scale tunnel experimental data suggests that the LES model has a better prediction of temperature distribution for tunnel fires under no rainfall condition, while the RANS model performs better in predicting temperature distribution for tunnel fires under the effect of rainfall. Finally, this work simulates the full-scale tunnel fire under the impact of rainfall and demonstrates the CFD simulation of tunnel fire under extreme weather. Correlations between the local increased pressure and maximum airflow velocity and rainfall parameters are developed.

深入理解隧道火灾动力学特性，对于火灾防控与生命安全保障至关重要。本研究强调了计算流体动力学（Computational Fluid Dynamics，CFD）技术在解析隧道火灾与降雨相互作用问题中的重要价值。本研究采用基于拉格朗日方法的离散相模型模拟雨滴，并分别采用雷诺平均纳维-斯托克斯（Reynolds Averaged Navier-Stokes，RANS）与大涡模拟（Large Eddy Simulation，LES）两类湍流模型对火灾流场进行仿真。本次数值仿真揭示了降雨影响下隧道洞口区域隧道火灾的详细流场、温度场及湍流特性。仿真结果表明，RANS模型对壁面附近流场的预测精度更优，而LES模型则能够捕捉火灾行为中的小尺度涡旋结构。通过将仿真结果与模型尺度隧道试验数据进行对比，研究发现：无降雨条件下的隧道火灾温度分布预测以LES模型效果更佳，而降雨影响下的隧道火灾温度分布预测则以RANS模型表现更优。最后，本研究针对降雨影响下的全尺寸隧道火灾开展了仿真，验证了极端天气条件下隧道火灾CFD仿真方法的可行性。本研究还建立了局部升压、最大气流速度与降雨参数之间的定量关联式。