Visual Fire Power: An Algorithm for Measuring Heat Release Rate of Visible Flames in Camera Footage, with Applications in Facade Fire Experiments

• Published in: Fire technology Vol. 59; no. 1; pp. 191 - 215
• Main Authors: Bonner, Matthew, Węgrzyński, Wojciech, Rein, Guillermo
• Format: Journal Article
• Language: English
• Published: New York Springer US 2023; Springer Nature B.V
• Subjects: Algorithms; Calorimetry; Cameras; Characterization and Evaluation of Materials; Civil Engineering; Classical Mechanics; Engineering; Experiments; Facades; Heat; Heat release rate; Heat transfer; Mathematical analysis; Outdoor air quality; Physics; Turbulent flames; Computer vision; Heat release rate; Flame volume; Cameras; Facades; Fire testing
• ISSN: 0015-2684; 1572-8099
• DOI: 10.1007/s10694-022-01341-z

This paper presents a novel algorithm, called Visual Fire Power, for measuring the heat release rate of a turbulent flame using video footage taken from two cameras, located at an approximate right angle to each other and at a known distance from the fire. By measuring the time-averaged volume of the fire, Visual Fire Power can measure heat release rate in situations where traditional calorimetry may be impractical (such as experiments outdoors), as well as uniquely providing a method for comparing the heat release rates of different flames in the same experiment, e.g. externally venting flames from different windows of the same compartment. The algorithm was benchmarked against synthetic data and calculated the volume of common solids with approximately 30% uncertainty. The relationship between volume and heat release rate was then calibrated from videos of burners at known heat release rates. These experiments were used to calculate the Orloff-DeRis constant γ , which linearly relates flame volume and heat release rate. The value for γ was found to be 1505 ± 183 kW/m 3 . The algorithm was demonstrated on recordings of a standard Polish facade fire test. Improving the range of data measured in both fire testing and fire experiments could help to increase our knowledge of fire dynamics and provide better data for researchers and engineers in the future.