Fire safety in buildings usually relies on the withdrawal of smoke or the preservation of stratification to allow proper evacuation of occupants. In order to assess safety levels, much work has been done in the past, leading to regular prescribed solutions. However, in the case of large enclosures such as theatres or sports arenas, the degree of precision of these classical methods is not well known, leading to high calculated safety coefficients and thus high construction and operating costs. This paper is devoted to a specific project where smoke motion simulations were run to help conceive a more cost-effective design. After completion of the building, a large-scale fire test was performed inside to validate the conclusions of the numerical study. The paper focuses on this fire test and subsequent simulations performed on the fire test scenario and aims at comparing the numerical and large-scale experimental results. Several simplifying modelling hypotheses (geometry, fire modelling, boundary and initial conditions) are examined and compared with the test. Provided enough care is taken in the simulations, these results show that good agreement can be reached. Recommendations are drawn on this basis.
A neuro-human thermal model was optimized to increase the prediction accuracy of the physiological variables of a group of 15 healthy male students exposed to transient environmental conditions. The effect of both the passive and active systems parameters was studied using a sensitivity analysis, and the parameters that had the most influence on the neuro-human thermal model outputs were established. A genetic algorithm was then used to optimize the model in order to determine the parameters that corresponded to the studied population. The results showed that the optimization increased the precision of the neuro-human thermal model. The mean absolute error and the maximum error between the experimental data and the numerical results for mean skin temperature were 0.13°C and 0.56°C, respectively, and we obtained 0.03°C and 0.11°C, respectively, for rectal temperature. These results show that the neuro-human thermal model can be accurately adjusted for the rectal, mean and local skin temperatures of a targeted population by using a genetic algorithm to determine the values of the parameters that correspond to this population.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.