Numerical assessment of train slipstream in tunnels: Stochastic analysis from CFD data

The train slipstream, referring to the dynamic airflow induced by moving trains, presents significant safety risks in confined environments like tunnels. While much research has focused on slipstream effects in open air, studies in tunnels are limited due to the challenges of simulating these complex aerodynamic conditions. This study aims to address these challenges by validating a CFD model based on URANS for train slipstream analysis in tunnels, comparing it against experimental data. A novel numerical statistical approach is introduced, enabling the robust characterization of slipstream phenomena using extended tunnel configurations, allowing the collection of multiple independent velocity profiles from a single simulation. The results highlight the differences in slipstream behavior between short and extended tunnels, emphasizing the impact of tunnel length on piston wind and wake development. By focusing on statistical comparisons, including ensemble mean, standard deviation, and peak distribution, the study demonstrates that the multiple-probe approach offers a robust and detailed representation of slipstream behavior. This methodology provides a general and replicable framework for characterizing slipstream flow statistics, proving especially valuable during early train and tunnel design stages where experimental data are lacking, and showing promising potential in the context of train homologation processes.