Probabilistische Bewertung von Entrauchungsanlagen

• Probabilistic evaluation of smoke extraction systems

Wesseling, Mark Thomas; Müller, Dirk (Thesis advisor); Kriegel, Martin (Thesis advisor)

1. Auflage. - Aachen : E.ON Energy Research Center, RWTH Aachen University (2021)
Book, Dissertation / PhD Thesis

In: EBC, Energy efficient buildings and indoor climate 87
Page(s)/Article-Nr.: 1 Online-Ressource : Illustrationen, Diagramme

Dissertation, RWTH Aachen University, 2020

Abstract

There are many deaths worldwide from fires in buildings, often caused by the resulting smoke. Smoke is not only a respiratory poison; it also reduces visibility and thus prevents people from finding their way out of the building. Smoke and heat exhaust systems can be used to remove smoke from the building in the event of a fire. These include the natural smoke extraction systems (NSE), which transport the smoke through roof openings using hydrostatic pressure, and the mechanical smoke extraction systems (MSE), which are operated by fans. The performance of the systems depends on many boundary conditions. These are hardly predictable and subject to large fluctuations, such as the weather. According to the current state of the art, despite the importance of the boundary conditions, their unstable nature is not sufficiently considered in the building planning process. To close this gap in the process, this thesis presents a new simulative approach that considers the boundary conditions and their realistic variations in the calculation of performance criteria. For this purpose, a flow simulation model for the CFD code Fire Dynamics Simulator (FDS) is created and coupled with the Monte Carlo method. The performed sensitivity analysis indicates that different parameters for the description of the weather and the source of the fire have a particularly large influence on the performance criteria. The probabilistic simulation model therefore includes the wind speed and direction as well as the ambient temperature, the maximum heat release rate, the fire development factor, the soot yield and the position of the fire source with corresponding distribution functions. To consider the wind at acceptable calculation times, this model applies a decoupling of the environmental simulation from the interior flow. This thesis performs a Monte Carlo simulation with 1 000 parameter combinations for a simple building geometry with a square base area. Two simulations, testing the smoke control performance of NSE and MSE respectively, are carried out for each parameter combination. The simulations are not only an initial application for the calculation tool but also a comparison between NSE and MSE to determine the personal safety provided by each. In the first six minutes after the start of the fire, the personal safety in the interior is greater with MSE as compared to NSE. Two minutes after the start of a fire, there is still 100% probability of safe conditions for people with MSE; for those with NSE, it is just 80%. After the seventh minute, the selected parameter limits confirm equal conditions for NSE and MSE, in which a safe condition can be assumed in 50% of the cases.

Institutions

• E.ON Energy Research Center [080052]
• Chair of Energy Efficient Buildings and Indoor Climate [419510]