Risks in using CFD-codes for analytical fire-based design in buildings with a focus on FDS:s handling of under-ventilated fires

Abstract

The use of CFD-models as an engineering tool for fire based analytical design of buildings has increased over the last few years. The reason for this is partly that the legislation allows for analytical dimensioning but also because the legislation demands a high degree of verification for the analytical dimensioning. The biggest reason is, however, that the computer power has increased and reached a point where it is applicable for engineering problems in terms of both time and money. The most common CFD-program in Sweden is FDS (Fire Dynamics Simulator) which is developed by NIST (National Institute of Standards and Technology), today’s version is the fifth large release. To ensure the correctness of the program it needs to be validated against experimental data. In this report, FDS is validated for under-ventilated fires. It is, however, not only how the program simulates the reality that is associated with risks but also how the program is handled by its users and their reviewers. If incorrect results are used and reviewed in an incorrect way that means that fire safety design in buildings may be incorrectly dimensioned which can have a negative impact on people’s health during evacuations in case of fire.

The process to decide if incorrectness in how FDS simulates under-ventilated fires is made in four steps. First, under-ventilated fires are described, how and when they arise and what consequences they may have for people’s health during fires, which is made through a literature study. The next step is to validate how FDS works for under-ventilated fires, which is made by comparing FDS output data with experimental tests performed by the SP Technical and Research Institute of Sweden. The test is a part of a larger research project (BRANDFORSK) financed by the Swedish Fire Research Board. The next step is to decide how the users handle under-ventilated in FDS. This is done through a series of telephone interviews with fire and safety design consultants in Sweden. The fourth step is to investigate how the reviewer (rescue services) handles their role as a reviewer.

The results show that the empirical expression concerning when a fire is allowed to burn or not together with the mixture fraction combustion makes the heat release rate and thereby the temperatures very sensitive for changes of oxygen level. Visibility and toxicity (carbon monoxide level), which is based on the soot and carbon monoxide yields, are much harder to apply and the FDS output for these parameters should be used very cautiously. The results also show that the fire and safety design consultants generally have little understanding about how FDS treats under-ventilated fires but that the basic use is handled in a good way. It is clear that although the rescue services in many cases can do a good review of an analytical dimensioning, they lack the knowledge and the resources for doing a good review of an analytical dimensioning with an under-ventilated fire involved. In total this means that incorrectness in how FDS simulates an under-ventilated fires do constitute a risk when it is used in analytical dimensioning of a buildings fire and safety design. The consequence can be that people in the building may be exposed to conditions dangerous (in a higher degree) to their health.

Measures to reduce the risk are for example that the CFD course given at the Department of Fire and Safety Engineering and Systems Safety at Lund University contains more elements about how FDS handles under-ventilated fire and how the output is affected. This is however no guarantee for that the students/users actually learn more and there should also exist certified CFD/FDS able users. A suitable party for this is a trade organisation like BIV. It is also appropriate to improve the conditions for the rescue service in their reviewing role

• author: Björklund, Anders
• supervisor: Patrick Van Hees
• organization: Risk Management and Safety Engineering (M.Sc.Eng.); Division of Fire Safety Engineering; Division of Risk Management and Societal Safety
• course: VBR920 20082
• year: 2009
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: validation, usage, reviewer, simulation, uncertainty, FDS, fire, under-ventilated, combustion, CFD, risks.
• publication/series: LUTVDG/TVBB--5281--SE
• report number: 5281
• ISSN: 1402-3504
• language: English
• additional info: Rapport 5281
• id: 1384434
• date added to LUP: 2009-04-22 12:07:34
• date last changed: 2020-12-03 14:26:03