Engineering simulations in the design and analysis of tunnels for fire safety and smoke ventilation

Abstract

In the design of tunnel ventilation systems, ranging from transport to utilities tunnels, engineering simulations are used by engineers globally. The simulations include 1D model and 3D computational fluid dynamics (CFD). CFD is popular in fire safety engineering, and the Fire Dynamics Simulator (FDS) is widely used. This thesis sets out to answer one question: Are engineering simulations the right tools for tunnels fire modelling? The short answer is yes, with a longer answer in this thesis. Carbon monoxide, or CO is the principal culprit in fire deaths, and is an important design parameter. The research shows when considering CO production and simulation, a range of CO yield should be considered. This is due to the inherent variability with CO production in experiments. Secondly, this thesis demonstrates because engineering simulations are a simplified reality, model validation is critical. With observations of oscillation when modelling large fires in a longitudinally ventilated tunnel, this thesis shows there is a significant oscillatory behaviour for fire sizes over 30 MW. This resulted in identifying numerical issues in FDS, and contributed to FDS developers fixing this. Using engineering simulations, this thesis shows a 1D model remains important in design by using a simplified 1D model developed in this thesis, TE1D, to describe fire throttling in tunnels. The effects of fire throttling are assessed, that is the additional pressure drops in the tunnel due to a fire, occurring at the fire’s location, and downstream of the fire. This thesis resulted in a novel quantifiable model that permits this effect to be calculated in design. This was not the case with previous models. 1D and 3D engineering simulations remain critical in design. This thesis concludes engineers need to understand the right time to use the right tool, and the limitations of these tools.