Geographic Fire Mapping Guide

The short answer is that you can used points or volumes. The default is to use points (as this is still the most common way of performing fire mapping in the industry) but you can easily choose to use volumes when setting up the fire zone. This capability has been in Detect3D for about two years now, and is commonly used on projects.

 

The coverage of the volumes is determined by the % area of the faces that are visible to the detector, and the distance from the detector. When we resolve whether the volume is visible or not, it’s just an application of the inverse square law.

 

We did some experimental validation of this method with JGC (note we call it the “box method”), which was presented at FABIG last year – see attached. I believe that this is the only validation for a fire mapping software that exists anywhere.

 

The size of the volume is normally determined by the performance standard, i.e. it is normally specified. You could of course do something more advanced.

Long answer…it’s important to remember that we are not really attempting to model fires when performing geographic fire mapping. Even if you could run a CFD simulation, the combustion, radiation and soot models are not good enough to give you an accurate power spectrum that would be visible to the flame detector. Not only that, but the detectors also work on the “flicker” of the flame, which again, is virtually impossible to model. And even if you could, you cannot model the inner workings of the detector because it’s proprietary. There are too many unknowns to realistically have a go at modeling the physics.

 

Geographic mapping accepts that the physics cannot be accurately resolved, and instead creates a proxy method where coverage can be sensibly evaluated and easily calculated. With a point method, nobody is saying that points are infinitesimally small fires! We are saying that the volume of a zone that is visible to a detector is a fairly good proxy for calculating its actual coverage – and we use points to calculate that volume. The “box method” again does not actually model fires using the volumes – it addresses some of the shortcomings of the point method to more accurately calculate the volume that is visible to the detector, using the dimensions of a flame.