Modern building fire sensors are capable of supplying substantially more information to the fire service than just the simple detection of a possible fire. In order to use this information, a decision support system needs to be developed that interprets a range of sensor signals and provides information about the building environment to the fire panel or building information server in real time. Typical fire models useful for predicting the impact of fire in a building utilize a prescribed heat release rate (HRR) for the fire and can predict sensor response and smoke spread. For the inverse problem, a sensor-driven fire model uses sensor signals to estimate the HRR of the fire, identify areas where hazardous conditions are developing, and predict the spread of smoke and the development of the fire. This type of model may then be used as part of a decision support system for emergency responders that provide real time predictions for the development of fire in a building. A sensor-driven fire model (SDFM) is being developed at NIST for the NIST Virtual Cybernetic Building Test-bed to investigate the feasibility of such a model in buildings with HVAC systems. Version 1.2 of SDFM uses ceiling jet algorithms for temperature and smoke or gas concentrations to convert the analog or digital data from heat, smoke, and carbon monoxide alarms to a HRR. The Zone Fire Model (ZFM) is then used to obtain layer temperatures, layer heights, and gas concentrations for the room of fire origin as well as surrounding rooms. With this information, the growth and spread of the fire and the location of hazardous conditions can be estimated.
A Sensor-Driven Fire Model, Version 1.2, Special Publication (NIST SP), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.sp.1110
(Accessed May 28, 2023)