The recent development of high speed trains over the last decade led to a growing interest in their aerodynamics. Beside the aerodynamic loading on the train itself, some issues related to the train-infrastructure interaction have to be carefully taken into account. Among them, there are the pressure waves generated by a train passing inside a tunnel and the ballast lifting phenomenon. A train at full operational speed generates a strong induced airflow that may interact with the infrastructures near the trackside and may damage them, and they may also endanger the workers near the rails. In order to ensure the safety of both train passengers and trackside workers specific requirements imposed by the European regulations have to be fulfilled. Nowadays the tests required for the train homologation are based on full scale measurements of the airspeed in specified positions along the railway line. These ones are very expensive to be performed, they suffer from repeatability errors and they are time consuming. The purpose of this research is to focus on the slipstream issue and to analyse the problem both using the experimental approach required by the European standards and numerical CFD simulations. An important part of the work was initially dedicated to the post processing of an extensive experimental campaign performed on the Italian high speed line, considering more than 200 recordings of high speed train passages, in order to obtain a reliable results to be used as for comparison with CFD results. Since the flow is highly turbulent and there are several parameters affecting the results, a statistical approach has to be adopted. CFD simulations have been performed on the full scale train geometry using both RANS and URANS approaches with a wall treatment based on the wall functions. Different setups have been compared in the CFD analysis, checking both mesh and turbulence models dependencies. These simulations highlighted the great complexity of the problem with an experimental measurement performed close to the shear layer of the boundary layer of the flow around the train. Encouraging results are obtained from the numerical analysis indicating the ability of the URANS with the SST turbulence model in the prediction of both the train induced flow and the turbulent structures around it.
Il recente sviluppo dei treni ad alta velocità ha portato ad un aumento dell'interesse nello studio della loro aerodinamica. Oltre al carico aerodinamico del treno stesso, alcuni problemi di interazione con le infrastrutture devono essere attentamente presi in considerazione. Tra loro, vi sono onde di pressione generate da un treno passante in galleria e il fenomeno del ballast lifting. Un treno a piena velocità genera un forte flusso d'aria indotta che può interagire con le infrastrutture vicino ai binari e può danneggiarle, e possono anche mettere a rischio i lavoratori vicino le rotaie. Per far in modo di assicurare la sicurezza di entrambi i passeggeri e dei lavoratori sui binari dei requisiti specifici imposti dai regolamenti europei devono essere soddisfatti. Al giorno d'oggi i test richiesti per l'omologazione del treno sono basati su misurazioni in scala reale della velocità dell'aria in punti specifici lungo i binari. Questi test sono molto costosi, soffrono di errori di ripetibilità e richiedono molto tempo. Lo scopo di questa ricerca è di concentrarsi sul problema dello slipstream e di analizzarlo usando sia approcci sperimentali richiesti dagli standard europei che simulazioni CFD. Una parte considerevole del lavoro è stata inizialmente dedicata al post processing di un'ampia campagna sperimentale condotta sulla linea ad alta velocità italiana, considerando più di 200 passaggi del treno ad alta velocità, in modo da ottenere un risultato affidabile da utilizzare come confronto con i risultati CFD. Dato che il flusso è altamente turbolento e dato che ci sono diversi parametri che influenzano i risultati, deve essere adottato un approccio statistico. Delle simulazioni CFD sono state effettuate sulla geometria in scala reale usando entrambi gli approcci RANS e URANS con condizioni al contorno basate sulle wall functions. Diversi setup sono stati comparati nell'analisi CFD, controllando le dipendenze di entrambe le mesh e dei modelli di turbolenza. Queste simulazioni sottolineano la grande complessità del problema con una misurazione sperimentale effettuata vicino allo shear layer dello strato limite del flusso intorno al treno. Sono stati ottenuti risultati incoraggianti dalle analisi numeriche che indicano l’abilità dell'URANS con il modello di turbolenza SST nella previsione del flusso indotto dal treno e delle strutture turbolente che lo circondano.
Numerical-experimental analysis of the slipstream produced by a high speed train
ZAMPIERI, ALESSANDRO
2016/2017
Abstract
The recent development of high speed trains over the last decade led to a growing interest in their aerodynamics. Beside the aerodynamic loading on the train itself, some issues related to the train-infrastructure interaction have to be carefully taken into account. Among them, there are the pressure waves generated by a train passing inside a tunnel and the ballast lifting phenomenon. A train at full operational speed generates a strong induced airflow that may interact with the infrastructures near the trackside and may damage them, and they may also endanger the workers near the rails. In order to ensure the safety of both train passengers and trackside workers specific requirements imposed by the European regulations have to be fulfilled. Nowadays the tests required for the train homologation are based on full scale measurements of the airspeed in specified positions along the railway line. These ones are very expensive to be performed, they suffer from repeatability errors and they are time consuming. The purpose of this research is to focus on the slipstream issue and to analyse the problem both using the experimental approach required by the European standards and numerical CFD simulations. An important part of the work was initially dedicated to the post processing of an extensive experimental campaign performed on the Italian high speed line, considering more than 200 recordings of high speed train passages, in order to obtain a reliable results to be used as for comparison with CFD results. Since the flow is highly turbulent and there are several parameters affecting the results, a statistical approach has to be adopted. CFD simulations have been performed on the full scale train geometry using both RANS and URANS approaches with a wall treatment based on the wall functions. Different setups have been compared in the CFD analysis, checking both mesh and turbulence models dependencies. These simulations highlighted the great complexity of the problem with an experimental measurement performed close to the shear layer of the boundary layer of the flow around the train. Encouraging results are obtained from the numerical analysis indicating the ability of the URANS with the SST turbulence model in the prediction of both the train induced flow and the turbulent structures around it.| File | Dimensione | Formato | |
|---|---|---|---|
|
2018_04_Zampieri.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
47.41 MB
Formato
Adobe PDF
|
47.41 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/140423