11th SFPE Conference on Performance-Based Codes and Fire Safety Design Methods, Warsaw

Back from Warsaw after quite a hectic week, starting on Sunday with the ECCG meeting (European Chapters Coordination Group) invited to attend the Board of Directors meeting.

It was a great conference, with new ideas to be implemented in our day-to-day work and other to be developed for future applications. 

I am very proud of the JVVA involvement in the 11th SFPE PBD Conference held in Warsaw this week. My colleague Olaf Perez, Fire Engineer at JVVA, presented the paper "Sensitivity Analysis for Modelling Parameters Used for Advanced Evacuation Simulations – How Important Are the Modelling Parameters When Conducting Evacuation Modelling?", Jimmy Jönsson presented the Spanish case study on Industrial Buildings and I presented the paper "Influence of Variability of Soot Yield Parameter in assessing the safe conditions in Advanced Modelling Analysis. Results of Physical and Numerical Modelling Comparison” together with Wojciech Wegrzynski, Head of Smoke Control, Signalling and Fire Automation Division at the Building Research Institute (ITB) of Poland.

I have had also the pleasure of moderating the Smoke session and being part of the panel for the Shopping mall case study. 

I want to spend a few lines describing what was my paper was about. It is a research exercise and thus a work in progress one, but I think it could be already quite useful.

During this year we performed several experimental tests in the ITB Research Centre of Warsaw, Poland, using different fuels with a wide range of soot yield (from 0.001 g/g fort Methanol to 0.178 g/g for Toluene). The results have been compared to numerical models processed with both Ansys Fluent and Fire Dynamics Simulator. The results showed good agreement for Heptane and thus this fuel was used as a base for a numerical study looking at 20 different cases, each one with a soot yield difference of 0.01 g/g (from 0.01 g/g to 0.20 g/g).

The numerical analysis, both for Ansys Fluent and FDS, show an hyperbolic trend of the Visibility when increasing the soot yield with a cut-off point around 0.12 g/g where a variation of the soot yield can produce a relevant (below 0.12 g/g) or a negligible (above 0.12 g/g) change in the Visibility.

Based on the numerical analysis performed, for the same fire scenario, the following value of visibility has been calculated from the models: 

Soot Yield (g/g)	0.01	0.02	0.03	0.04	0.05	0.06	0.07	0.08	0.09	0.10
Visibility (m) – FDS*	14.87	10.90	7.34	5.39	4.29	3.55	2.98	2.60	2.31	2.06
Visibility (m) – Fluent*	15.15	7.64	4.91	4.04	2.96	2.44	2.12	1.90	1.62	1.48
Soot Yield (g/g)	0.11	0.12	0.13	0.14	0.15	0.16	0.17	0.18	0.19	0.20
Visibility (m) – FDS*	1.85	1.71	1.56	1.43	1.31	1.21	1.14	1.07	1.00	0.94
Visibility (m) – Fluent*	1.34	1.30	1.23	1.09	1.02	N/A	0.91	0.849	0.82	0.74

*Value for reflecting signs

As can be seen in the table using a chemical reaction with a soot yield value of 0.02 g/g the Visibility (for reflecting signs) would be about 10m, whilst with a soot yield value of 0.04 g/g, the Visibility (for reflecting signs) would be about 5m and with a soot yield of 0.10 g/g, the Visibility (for reflecting signs) would be about 2m. This means that for the same fire scenario the implications of choosing a reaction instead of another can lead to a huge over prediction of the Visibility and therefore a huge under estimation of the Fire Safety.

One of the main conclusions of this study is that unless the materials (and thus the chemical reactions) in the building to be assessed by a Fire Engineering analysis are well known, value of soot yield below about 0.10 g/g should be used with extremely caution when performing an ASET/RSET exercise.

Although the overall conclusions still be valid, more accurate experiments will be performed in order to better understand the deviation from the numerical results.

Thanks to Charles Fleischmann, Professor at the University of Canterbury, New Zealand, for the interesting comments and the new ideas provided. 

I hope you enjoy the presentation.