Numerical simulations of the separated flow around a freight train passing through a tunnel using the sliding mesh technique

Research output: Contribution to journalArticlepeer-review


The main aim of this investigation is to analyse the flow around a freight train as it passes through a tunnel. The separated flow around the train nose is related to energy losses, lateral vibration, noise and streamline deviation, and it also influences the velocity magnitudes around the train. Such effects are expected to become more important with the prospect of increasing freight train speeds. The numerical simulations performed in this study use a Class 66 locomotive connected to eight container wagons, scaled to 1/25th, moving at a train speed of 33.5 m/s through a tunnel with a blockage ratio of 0.202. The k–ω model combined with a high advection scheme solves the governing equations on a structured hexahedral mesh using the sliding mesh technique. The pressure histories at the tunnel walls and train surface as well as the velocity field around the train were validated with experimental data obtained using a moving model. The longest separation bubble is found at the middle-height and middle-width of the locomotive due to extended corners at these regions. When the train enters the tunnel, the separation length is reduced by 32% at the roof and 31% at the sides, compared to open air. The maximum separation length is found at the sides of the train where it reattaches at 19% of the locomotive length, influencing the velocity peak at a short distance from the train surface. The larger the separation length, the higher the length/duration of this peak. When the train head is halfway through the tunnel, the nose velocity peak reduces by 30% compared to open air. The position of the nose inside the tunnel affects not only the slipstream velocity but also the velocity field at the tunnel portal and exit. These novel findings can be used as a benchmark for designing new freight train and tunnel shapes.


Original languageEnglish
Number of pages17
JournalProceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit
Early online date8 Jun 2019
Publication statusE-pub ahead of print - 8 Jun 2019


  • Computational fluid dynamics (CFD), Sliding-mesh, RANS, Flow Separation, Compression wave, Train aerodynamics, k-omega SST, k–ω shear stress transport, Computational fluid dynamics, train aerodynamics, compression wave, Reynolds-averaged Navier–Stokes, sliding-mesh, flow separation

ASJC Scopus subject areas