Computational Civil Engineering

Visibility Prediction Framework – A next-generation model for visibility in smoke in build environment

Smoke is the greatest threat in building fires. The first effect that smoke has on humans is the reduction of visibility – the smoke conceals evacuation signs and exits, and makes the orientation in space a very difficult task. The loss of visibility happens at a very low concentration of smoke. In fact, this concentration is so low, that if the visibility through smoke is not significantly reduced, it also means the smoke is not a direct threat to life in a short exposure. Assessment of visibility is fundamental to building safety. Yet, we do not have adequate tools to do this. The existing model of T. Jin was created in the 1970s and has not been significantly modified since then. The tools at our disposal have certainly evolved for the last 50 years, but in the field of visibility in smoke, we can still only estimate visibility, rather than model it. The NextVIS project is aiming to close that methodological gap. Replacing the existing relationship between the concentration of smoke and visibility, we intend to implement a physical model of photon and smoke particle interaction. This will enable us to model the appearance of a target, e.g. sign, through a layer of smoke, as in the eye of the observer. Additionally, our tool will allow to evaluate the impact of the characteristics of an evacuation sign (size, colour, light intensity) or the architectural context of the building. This will enable safety solutions tailor-fit for the building. The existing model of Jin does not allow for that. Our next-generation model for visibility in smoke will have a considerable impact on fire science. The scientists base their conclusions related to safety on parameters, especially visibility, which can be measured. Many recent research projects in fire safety aim to improve visibility in smoke in the most efficient way. Our model will be built into a software package – the Visibility Prediction Framework (VPF). This package will be open- source and available to everyone. This model will be useful to the researchers in the most innovative research fields in fire science, such as the use of virtual reality to investigate human behaviour in fires. The project is a joint Polish-German collaboration planned for three years. In Poland, the Building Research Institute (ITB) team will perform novel experiments on the visibility in smoke conditions. Observers will participate in these experiments, to help us determine at what distortion conditions the evacuation sign stops being visible. This data will be used by the German team of the University of Wuppertal (BUW). The German team will be building the heart of the project – the VPF framework. After two years in development, we will use the VPF in the built environment. We will conduct real-fire experiments to validate the model. The visibility in smoke model is a key tool for fire safety engineering. We envision that our new model will become a part of this toolbox and will help to increase safety in buildings.

Funded by NCN (Polish partner) and DFG (German partner)
Projectduration: 2022 - 2024
DFG project number: 465392452

More information about #UniWuppertal: