|A Swedish Best Practice Guideline for Proper Use of CFD-Models|
A Swedish Best Practice Guideline for Proper Use of CFD-Models With ASET Analysis
By Daniel Rosberg and Johan Norén
IntroductionIn Sweden, there are performance-based regulations when constructing a building. In this performance-based code, compliance with the fire safety regulations can be demonstrated in two ways: either by constructing the building in accordance with pre-accepted solutions (as defined by the Swedish National Board of Housing, Building and Planning), or by means of fire safety engineering methods showing that the fire safety is satisfactory according to the societal level of safety. Fire safety engineering methods are used when pre-accepted solutions are not met due to building-specific conditions (i.e. if the building is over 16 floors) or if there are specific stakeholder requests (i.e. large fire compartsments due to occupancy).
Performance based regulations were first implemented in Sweden in 1994. In 2012, new fire safety regulations were implemented . These were the most comprehensive revisions to the Swedish Building Code since the transition to performance-based regulations. In connection to the implementation of these new regulations, the Swedish National Board of Housing, Building and Planning also published general guidelines on the use of fire safety engineering methods .
The national guidelines for ASET-analysis
In the guidelines from the Swedish National Board of Housing, Building and Planning , scenario analysis may be used to analyze means of egress from a building. Other methods, such as qualitative comparative analysis and quantitative risk analysis are also permitted. The scenario analysis approach is based on comparing available safe egress time (ASET) with required safe egress time (RSET). The design process for the ASET analysis is based on pre-defined fire scenarios where parameters such as type of occupancy and available systems (i.e. sprinkler systems) are considered. The prescribed scenarios are chosen to represent a probable worst case scenario and a number of robust scenarios.
Fire scenarios to be analyzedThe national guidelines  specify three required fire scenarios that are generally applicable to the majority of ASET analyses. These scenarios are selected to represent a reasonable stress on the building’s fire protection.
Fire Scenario 1 is characterized by a severe fire with rapid development, high maximum heat release rate, and a high production of byproducts - a probable worst case. The installed fire protection systems are assumed to function as intended and the impact of these may be included in the design fire. Also, fire propagation shall be selected as “conservative” (see Tables 1 and 2). The impact of active systems is also specified in the new regulations.
If the building is not equipped with a full automatic fire and evacuation alarm, the analysis should include Fire Scenario 2. This scenario includes a fire in an area that is normally unoccupied, but which is adjacent to an area where there is a large number of people.
Fire Scenario 3 is characterized by a fire progression which is expected to have a smaller stress effect on the building’s fire protection. On the other hand, in this scenario individual systems (such as sprinkler or smoke control systems) are not functioning as intended. The systems in the analyzed building should all be made inaccessible separately. In this scenario, the fire progression shall be selected “non-conservative” (see Tables 1 and 2).
Design values to be usedDesign values for the required fire scenarios according to growth rate, maximum heat release and heat of combustion should not be less than what is defined in in the national guidelines. The heat release rate should be calculated according to the well-known t-squared fires with defined α-values. In the guidelines there is also design values defined for by-products in the early stage of the fire. The soot yield, CO- and CO2- production is dependent on what fire scenario analyzed.
Table 1 presents the design values according to the different occupancies. Suggested design values for byproducts, are presented in Table 2.
In the national guidelines there are also defined levels for tenable conditions to determine when untenable conditions occur.
The Swedish best practice for ASEt-analysis with CFD-modelsRegarding the difficulty to fully understand the defined design process for ASET analysis in the national guidelines  and the Swedish National Board of Housing, Building and Planning doesn’t give any guidance concerning how to do the analysis (it is assumed that the fire safety engineer knows how to do it), there was a need for more in-depth guidance. As a result, the Swedish SFPE Chapter (BIV, föreningen för brandteknisk ingenjörsvetenskap) initiated a project in 2012 to develop a Swedish best practice to ensure better use of CFD-modelling. The starting point for the project was to pick-up where the national guidelines ended concerning ASET-analysis and to give more user-friendly recommendations.
The work in the project was carried out similar to the development of open-source codes, standardizing committees and how the SFPE-organizations work with the development of best practice guidelines. The project was completely non-profit and the project group consisted of 8 members with representatives from the consulting industry, academic institutions and research institutes. The project was completed in one year and took about 600 person-hours to complete. In order to increase the quality, raise awareness of the project and get a wider distribution and legitimacy for the work, the best practice, in a preliminary form, was sent out for referral. All members of the BIV and other relevant organizations within the fire safety community in Sweden were invited to give consultation responses. The comments received were considered before the final version of the best practice was published.
The overall purpose with the best practice was to be a supporting guide for practising fire safety engineers, reviewers and clients to achieve a sufficient quality level and to increase the understanding for the process when analysing ASET. From this point of view, the best practice included both technical guidance concerning how to work with CFD-models, as to describe a well-functioning working process and to provide suggestions for quality assurance.
Most fire safety engineers in Sweden working in the design of buildings use the CFD-model Fire Dynamics Simulator (FDS)  when performing advanced ASET-analysis. Based on this, the best practise was developed for FDS version 5.5.3 (SVN 7031), which was the latest official version when the project was initiated. The best practice was limited to only describing aspects concerning the early stages of fire during well ventilated conditions and to fulfil the requirements given by the Swedish National Board of Housing, Building and Planning. However, some parts of the best practice guidance can still be used for other CFD-programs or versions of FDS and for different kinds of analysis which don’t follow the Swedish way of performing ASET-analysis. Especially, parts concerning the working process and the quality assurance are areas that in some extent are universal for all kinds of CFD-modelling.
The content of the best practice consists of a suggested working process, different technical aspects concerning fire characteristics (based on the national guidelines), and example on parameters to control quality assurance. But, the best practice also gives guidance to important aspects concerning verifying and validation of FDS, smoke management, how to handle input and output data and what sensitivity analysis to perform to ensure reliable results. However, the best practice doesn´t give hands-on tips to programming a FDS input file or how to do specific functions in pre-process programs such as PyroSim. But there are other guidelines developed in Sweden  and Denmark  that give more hands-on tips for these kinds of issues.
It is also important to note that the guidelines are not written from the perspective on how to properly model fire. They are written with a designer´s perspective, about how the defined requirements in the building regulations should be met.
Daniel Rosberg is with WSP Sweden and Johan Norén is with Briab Brand & Riskingenjörerna AB