The Fire Navigator: smoke and flame sensors in smart buildings

The Fire Protection Engineering magazine has recently published our article reporting exciting research on the theme of smart buildings and fire protection. In this work, sponsored by Chief Donald J. Burns Memorial Research Grant, my student Nahom and I developed an algorithm that uses data arriving from building sensors to detect and map an ongoing fire. The algorithm, called the Fire Navigator, then provides forecasts of future smoke and flame spread within the building, allowing to see where and how the fire would propagate if not stopped before hand.

We envision that the forecasting of fire dynamics in buildings will lead to a paradigm shift in the response to fire emergencies, providing the fire service with essential information about smoke propagation and flame spread ahead of time (i.e. minutes before it happens). Disposing of information on fire events before they actually happen would have a positive effect on the fire service efficiency and safety, therefore saving human lives and mitigating property losses and environmental damage. Smart buildings anticipate the occupants’ needs with the help of various sensors. Control of heating and air conditioning, energy consumption and lighting are now common examples of how sensors allow control over key aspects of the built environment. We want to extend this to fire safety engineering and enhanced fire fighting. Already existing smoke and heat sensors, as well as sprinklers, generate data that has yet to be harnessed and used in smart buildings. This is what our article proposes and shows how to do it.

Our work is based on combining new and old ideas. The new ideas are the use of a very quick fire model based on cellular automata theory, and the integration of  the whole system into building information models (BIM). You can read the full article at the SFPE website:

The Fire Navigator: Forecasting the Spread of Building Fires on the Basis of Sensor Data 


NOTE: This research was sponsored by SFPE and Bentley Systems via the Chief Donald J. Burns Memorial Research Grant. We thank Arup, specially Judith Schulz, for sharing their expertise in BIM and fire protection systems, and thank KPF for permitting the use of their architectural BIM models.

Breakthrough in the understanding of flaming wildfires

I wrote a commentary article in the Proceedings of the US National Academy of Sciences (PNAS) about a recent stellar contribution to our understanding of how wildfires spread. In doing so, I have written in short the scientific context of wildland fires and also I put forward the possible impacts of the work on the field..
It can be read here ((10.1073/pnas.1512432112), and an except follows.

Breakthrough in the understanding of flaming wildfires

The rise of humanity was intimately bounded to fire. Humans first observed flames when fleeing wildland fires, the natural version of the phenomenon that would then become the most important technological achievement of the human race: the mastery of fire for cooking, lighting, settlement, hunting, and warfare (Bird 1995).
Wildfires are important to the natural sciences. Since deep time, the top surface of the Earth’s crust has been the interface where abundant plant organic matter meets an atmosphere rich in oxygen. This interface is flammable, especially in dry, windy and hot conditions, and leads to wildfire after an ignition event. Not only has fire contributed to shaping most ecosystems on Earth, but it plays essential roles supporting life through the regulation of atmospheric oxygen, the carbon cycle, and the climate (Bowmand et al. 2009, Watson et al. 1978).
As part of the current anthropogenic age, humans have also modified the fire regimes of many ecosystems, and have contributed for example to its cessation in certain regions (e.g., in the USA National Parks until 1960), or to increasing its frequency and severity through drainage (e.g., peatlands) and possibly through climate change (e.g., arctic fires). Of note, multiple US$ billions are spent annually across the world to fight wildfires for the protection of communities and valuable ecosystems.
Despite its central importance to the planet and to humanity, our understanding of fire remains very limited. For example, we currently cannot accurately forecast the location of a fire in 30 min time. To quote Hottel (1984): “A case can be made for fire being, next to the life processes, the most complex of phenomena to understand”. It comes as no surprise, then, that the discipline of fire science is less mature than other Earth science topics. For example, a quick look at the literature shows that there are three times more scientific studies published per year on volcanoes than on wildfires. Fire science requires more decades of fruitful research to mature and gain full understanding of this natural phenomenon.

Rate of Spread

The fate of a flaming wildfire starts with its genesis at ignition, by natural means like a lightning strike, or by anthropogenic means like slash-and-burn. Once ignited, part of the heat released by the flames will drive the spread over connected fuel beds of grass, shrubs, and trees. Another mechanism of propagation is by lofting burning embers that land farther away, but flame spread is more important. The dynamics of spread are such that wildfires accelerate with tail winds, dry weather, or up-slopes; and decelerate with head winds, rain or down-slopes.
The most lasting contribution to the science of wildland fires is the pioneering work of Rothermel in 1972 (Rothermel, 1972). He formulated an empirical model for predicting the spread rate of a wildfire. This formulation is ubiquitous and can be found at the core of most wildfire behaviour simulations. These simulations are currently in use by forestry agencies and firefighting command centres across the world. For example, Rothermel’s model is part of the US Wildland Fire Decision Support System, used in planning of every large and long duration federal wildland fire incident. However, Rothermel’s formulation is empirical: Whilst it can provide rough predictions of the rate of spread by calibration to previous laboratory data, it does not explain how fire spreads. Its empirical nature hinders scientific progress and does not allow for improvements to simulations. Until very recently, there was no valid scientific theory of wildfire spread that could complete Rothermel’s model.

Sketch of flame spread of a fire with tail wind over a fuel bed of fine particles. The paths for heat transfer by
radiation, convection, and flame contact are noted. According to Finney et al. (2015), the vortices are created by buoyant
instabilities and lead to ignition of the fuel by flame contact. Modified from Rothermel, 1972.

Finney et al. 2015

In this context, we see that the recent work of Finney et al. (2015) is a scientific breakthrough. Finney et al. have discovered the long-missing piece of the puzzle to understand wildfire dynamics. Their seminal work puts forward for the first time a fundamental, comprehensive and verifiable theory of flaming wildfire spread. Finney’s theory relates the rate of spread to basic fluid mechanics and heat transfer, and it is strongly supported by laboratory data and field observations across a wide range of scales from 10 cm to 15 m.

Let me put this in the framework of a simple theory. Fire dynamics dictate that spread can be seen as the succession of ignition events (Emmons 1963). This way, the rate of spread s of a fire is given in Eq. (1) by two terms, the length of fuel bed heated by the flames (expressed as δ) and the time that a fuel particle takes to ignite (expressed as tig) (Drysdale 2011).

 s=δ/tig   (Equation 1)

We know that mostly depends on flame inclination and the slope of the terrain, whereas depends mostly on fuel properties like particle size, moisture and plant composition. The scientific contributions of Finney et al. are cast around the novel identification of the two terms in Eq. 1 that govern wildfires.
First, by careful inspection of visual images of fire across scales, they show that vortex flows and peaks-and-troughs generated by the buoyancy of the flames are responsible for heating the fuel bed length δ. Then, temperature measurements then show that the intermittency of the peaks-and-troughs causes the flames to instantaneously touch the thin fuel particles, which in turn produces the contact ignition governing  tig. Figure 1 shows a sketch including these mechanisms.

Convection vs. Radiation

Their work feeds into a long-standing debate in the field on whether it is radiation or convection that controls the heat transfer to the fuel bed ahead (see Fig.1). The specific heat transfer mechanism affects the interpretation of experimental observations, and is critical in correctly formulating physically based models (Morvan 2011). Finney et al. settle the debate by identifying with strong evidence that heat transfer is controlled by flame contact, the phenomenon where both radiation and convection heat transfer are combined, but with the distinctiveness that the timing of flame contact is driven by convective flows.

Profound impact in fire science

Finney’s theory can have a profound impact in the field. The impact is four-fold regarding i) previous scientific studies, ii) wildfire simulations, iii) new technologies, and iv) multi-disciplinarity. These are explained in the following.
Previous scientific studies on wildfire spread should be revisited to help put Finney’s theory into a broader context. experimental and computational studies might need to be reinterpreted in the light of
the roles of flame intermittency and flame contact. The state of the art should naturally revisit and replace Rothermel’s model to give way to a new physically based Rothermel–Finney’s model.

Rothermel-Finney’s model would improve simulations of fire behaviour and help them gain in both accuracy and consistency. This in turn would allow the simulations to provide a more reliable layer of information during fire incidents.
The increased accuracy of simulations should eventually allow for high-fidelity forecasting technologies. A technology able to rapidly forecast the movement of a wildfire would lead to a paradigm shift in the response to emergencies, providing the Fire Service with essential information about the ongoing fire (Rios et al 2014).
The topic of wildfires is currently fragmented among the fields of biology, ecology, meteorology, chemistry, and combustion. These fields have a lot to offer one another, but better communication and cooperation are essential to move it forward. It is hoped that by strengthening the importance of fundamental knowledge and by settling long-standing debates, Finney et al. will serve as the basis for developing new multidisciplinary collaborations in the study of wildfires.

Finally, I foresee that after reading their work, many readers might start seeing the peaks-and-troughs reported by Finney et al. in every wildfire, as I already do now. As the English poet John Milton once said, “so easy it seem'd, once found, which yet unfound most would have thought impossible”.

References

• MA Finney, JD Cohen, JM Forthofer, SS McAllister, MJ Gollner, DJ Gorham, K Saito, NK Akafuah, BA Adam, JD English (2015) The role of buoyant flame dynamics in wildfire spread. Proc. Natl. Acad. Sci. USA, 10.1073/pnas.1504498112.
• MI Bird, Fire, prehistoric humanity, and the environment, Interdisciplinary Science Reviews 20(2), 141-154, 1995. DOI:10.1179/isr.1995.20.2.141A.
• DMJS Bowman, JK Balch, P Artaxo, WJ Bond, JM Carlson, MA Cochrane, CM D’Antonio, RS DeFries, JC Doyle, SP Harrison, FH Johnston, JE Keeley, MA Krawchuk, CA Kull, JB Marston, MA Moritz, IC Prentice, CI Roos, AC Scott, TW Swetnam, GR van der Werf, SJ Pyne, Science 324 (5926), 481-484, 2009. DOI:10.1126/science.1163886. 
• JE Watson, Lovelock, L Margulis, Methanogenesis, fires and the regulation of atmospheric oxygen, Biosystems 10 (4),pp 293-298,1978. 
• HC Hottel, Stimulation of fire research in the United States after 1940, Combustion Science and Technology 39:1–10, 1984. doi:10.1080/00102208408923781.
• RC Rothermel, A mathematical model for predicting fire spread in wildland fuels, USDA Forest Service, Intermountain Forest and Range Experiment Station, Ogden, Utah, Research Paper INT-115, 1972. 
• HW Emmons, Fire in the forest, Fire Research Abstracts and Reviews 5, 163, 1963. 
• D Drysdale, An introduction to fire dynamics, 3rd edition. John Wiley and Sons Ltd, Chichester, 2012. 
• D Morvan, Physical Phenomena and Length Scales Governing the Behaviour of Wildfires: A Case for Physical Modelling, Fire Technology 47 (2), pp 437-460, 2011. doi:10.1007/s10694-010-0160-2. 
• O Rios, W Jahn, G Rein, Forecasting wind-driven wildfires using an inverse modelling approach, Natural Hazards and Earth System Sciences 14, pp. 1491-1503, 2014. doi:10.5194/nhess-14-1491-2014

Forecasting fire dynamics: tomorrow's infrastructure protection

Imagine a technology able to forecast fires. It would lead to a paradigm shift in the response to emergencies and provide the Fire Services with essential information about the ongoing blaze with some lead time (i.e. seconds or minutes ahead of the event). It would also allow for the future of infrastructure protection to be implemented in smart buildings.

316 s after ignition. could this be forecasted ahead of time? [Rein 2012]

But despite advances in the understanding of fire dynamics over the past decades and despite the advances in computational capacity, our ability to predict the behaviour of fires in general and building fires in particular remains very limited. The state-of-the-art of computational fire dynamics is not fast or accurate enough to provide valid forecasts on time...

Paleofuture: forecast made in 1900 of the fire-fighting in the year 2000.
By Villemard, 1910, National Library of France

But we found a way to solve this problem. In a recently finished PhD thesis and set of published  papers, we show the technology is possible. We  propose to use sensor measurements of the ongoing fire to steer and accelerate computer simulations. This takes advantage of the concept of data assimilation (similar to what meteorologists do to forecast the weather).

Our method consists on combining a simplified spread mechanism with a fire model, and use sensor data to find the fire parameters that dominate the spread. This way, the model automatically recovers information lost by approximations in the physics, chemistry and the maths.

Concept of data assimilation and the sensor steering of model predictions [Cowlard et al 2010]

 A series of compartment fire cases haven been studied this way, and we investigated two different fire models. First a simple two-zone model, and then a state-of-the-art computational fluid dynamics (CFD) model.
 
For the simple two-zone forecast model, the firepower and the growth rate were estimated correctly up to 30 s ahead of the event: the model was faster than the fire. This was the very first time a fire forecast technology was demonstrated and the first time positive lead times were reached. The results show that the simple model is able to deliver fast and useful information about the ongoing fire thanks to the sensor data. This initial work demonstrated that the new methodology is effective, and allowed us to move to the next level of complexity. 

Computational domain of the fire
compartment [Jahn et al 2012]

For the CFD forecast model, we use a coarse grid that provides short computation times. Spatially resolved forecasts were obtained in reasonable time. It is even possible to estimate the growth rates of several different spreading fires simultaneously. Although actual positive lead times were not reached here with CFD, it is shown that the use of relatively coarse grid size in the forward model significantly accelerates the assimilation (up to 100 times faster) without loss of forecast accuracy. Actual positive lead times with CFD are possible by reducing the computational time by at least another order of magnitude in the near future using high performance computing techniques.

Dalmarnock Fire Test One conducted on July 25th.

Our latest bit on the topic was a test case using the measurement data  from a real fire. We forecasted in near real time the Dalmarnock Test One, conducted in 2006 inside the 3.5 x 4.7 x 2.4 m living room in a high rise building in the city of Glasgow. It was possible to find a good fit between the observations and the forecast using CFD.


The results are a fundamental step towards the development of forecast technologies able to lead the fire emergency response. The work opens the door to forecasting fire dynamics, but it is an on-going research topic.

We are happy that the work has been featured in the media and  people is being exposed to this novel idea:

• Interview for Scottish TV News (go to minute 19 here). Aired on 29 Nov 2010.
• Interview for BBC Radio Scotland (or go minute 42.20 here).  Aired on 29 Nov 2010.
• Articles in CORDIS-EU and Xinhuanet (in Chinese).

Our research resources on the topic (in reverse chronological order):

1) Jahn, Rein and Torero (2012), Forecasting fire dynamics using inverse
Computational Fluid Dynamics and Tangent Linearisation, Advances in
Engineering Software 47 (1), pp. 114-126. doi:10.1016/j.advengsoft.2011.12.005

2) Rein (2012), Plenary Keynote: Numerical forecasting of fire dynamics: tomorrow's infrastructure protection - Young Investigators Conference of the European Community on Computational Methods in Applied Sciences (ECCOMAS), Aveiro. See below.

3) Jahn, Rein and Torero (2011), Forecasting Fire Growth using an Inverse CFD Modelling Approach in a Real-Scale Fire Test, Fire Safety Science 10, pp 1349-1358, doi:10.3801/IAFSS.FSS.10-1349 

4) Jahn, Rein and Torero (2011), Forecasting Fire Growth using an Inverse Zone Modelling Approach, Fire Safety Journal 46, pp. 81–88. doi:10.1016/j.firesaf.2010.10.001. Paper shortlisted for 2010 Lloyd's Science of Risk Prize.

5) Jahn (2010), Inverse Modelling to Forecast Enclosure Fire Dynamics, PhD Thesis, School of Engineering, University of Edinburgh.

6) Cowlard, Jahn, Empis, Rein and Torero (2010), Sensor Assisted Fire Fighting, Fire Technology 46 (3), doi:10.1007/s10694-008-0069-1


Numerical forecasting of fire dynamics (Plenary YIC ECCOMAS)