This thesis illustrates and describes the results of a modelling work investigating the effect of fire dynamics, queue vehicle formation, tunnel geometry, ventilation system and human behaviour on the evacuation time in road tunnels. The modelling approach is based on CFD simulations of the fire development and interaction with tunnel geometry and vehicles using FDS+Evac [1,2]. The evacuation module of FDS+Evac (an agent-based egress model) is also used to run several simulations aimed at performing and evaluating the safety of several evacuation strategies. Finally, the vehicle queue formation dynamics is simulated using a specifically conceived model which is coupled with the FDS+Evac to assess the impact of queue formation on the evacuation process. Four key aspects have been analysed through these modelling tools: 1) the impact of modelling assumptions on the emergency exit choice and the evacuation time; 2) the effect of the queue formation model and distribution of the vehicles in the tunnel on the evacuation time; 3) the effect of emergency ventilation conditions on agent’s intoxication; 4) the impact of different type of vehicles (i.e. cars, buses) on the evacuation process. The modelling approach was first compared and validated using the evacuation experiment performed in the a Göta road tunnel (Sweden) [1]. The complete geometry of this emergency scenario was reproduced, including smoke generation, distribution of vehicles along the tunnel and location of the emergency exits. A satisfactory agreement was obtained, and a sensitivity analysis on model assumptions was performed to tune the evacuation sub-models. The calibrated model is finally used to simulate new tunnels and different fire scenarios.
Questa tesi illustra e descrive gli effetti della modellizzazione delle geometrie, del sistema di ventilazione e del comportamento umano sui tempi di evacuazione in caso di incendi in Gallerie autostradali. L'approccio alla modellazione si basa sull’ utilizzo del software FDS+Evac [1,2]. Il modulo di evacuazione di FDS+Evac (un modello di uscita basato su agenti) viene anche utilizzato per eseguire diverse simulazioni volte a stabilire e valutare la sicurezza di diverse strategie di evacuazione. Infine, la dinamica della formazione delle code dei veicoli viene simulata utilizzando un modello appositamente concepito che è accoppiato con FDS+Evac per valutare l'impatto del processo di formazione della coda sul processo di evacuazione. Quattro aspetti chiave sono stati analizzati attraverso questi strumenti di modellazione: 1) l'impatto delle ipotesi sulla modellazione della scelta dell'uscita di emergenza e il tempo di evacuazione; 2) l'effetto del modello di formazione della coda e la distribuzione dei veicoli nel tunnel sul tempo di evacuazione; 3) l'effetto delle condizioni di emergenza sull'intossicazione degli agenti; 4) l'impatto di diversi tipi di veicoli (ad esempio automobili, autobus) sul processo di evacuazione. L'approccio di modellizzazione è stato prima comparato e convalidato utilizzando l'esperimento di evacuazione eseguito nel tunnel stradale di Göta (Svezia) [1]. La geometria completa di questo scenario di emergenza è stata riprodotta, compresa la generazione di fumo, la distribuzione di veicoli lungo il tunnel e la posizione delle uscite di emergenza. È stato ottenuto un accordo soddisfacente ed è stata eseguita un'analisi di sensitività sulle ipotesi di modello per verificare la dipendenza del modello dai parametri implementati. Il modello calibrato viene infine utilizzato per simulare nuove gallerie e diversi scenari di incendio.
Queue formation and evacuation modelling in road tunnels during fires
BOSCO, DAVIDE
2016/2017
Abstract
This thesis illustrates and describes the results of a modelling work investigating the effect of fire dynamics, queue vehicle formation, tunnel geometry, ventilation system and human behaviour on the evacuation time in road tunnels. The modelling approach is based on CFD simulations of the fire development and interaction with tunnel geometry and vehicles using FDS+Evac [1,2]. The evacuation module of FDS+Evac (an agent-based egress model) is also used to run several simulations aimed at performing and evaluating the safety of several evacuation strategies. Finally, the vehicle queue formation dynamics is simulated using a specifically conceived model which is coupled with the FDS+Evac to assess the impact of queue formation on the evacuation process. Four key aspects have been analysed through these modelling tools: 1) the impact of modelling assumptions on the emergency exit choice and the evacuation time; 2) the effect of the queue formation model and distribution of the vehicles in the tunnel on the evacuation time; 3) the effect of emergency ventilation conditions on agent’s intoxication; 4) the impact of different type of vehicles (i.e. cars, buses) on the evacuation process. The modelling approach was first compared and validated using the evacuation experiment performed in the a Göta road tunnel (Sweden) [1]. The complete geometry of this emergency scenario was reproduced, including smoke generation, distribution of vehicles along the tunnel and location of the emergency exits. A satisfactory agreement was obtained, and a sensitivity analysis on model assumptions was performed to tune the evacuation sub-models. The calibrated model is finally used to simulate new tunnels and different fire scenarios.File | Dimensione | Formato | |
---|---|---|---|
Tesi.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Testo della tesi
Dimensione
5.45 MB
Formato
Adobe PDF
|
5.45 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/10589/139755