Advances in the modeling of fire development in passive houses

One of the greatest challenges the world is facing is climate change. According to the Energy Performance of Buildings Directive (EPBD 2010/31/EU), the construction of passive houses, thanks to their characteristics of airtightness and thermal insulation, is a way to reduce pollution and to increase energy efficiency. However, the airtightness induces new fire risks since it prevents the hot fumes of a fire inside the building from flowing out, leading to a pressure increase inside the house. It must be remembered that 130 Pa is the threshold value of pressure at which the occupants might have difficulties in opening an inward-opening door, remaining trapped inside the house and exposed to the toxic and hot fumes, as well as to the flames. Researches aimed at evaluating the behaviour of passive houses in case of an internal fire are carried out at the University of Mons; the present thesis represents a contribution to these researches. A total of twenty-one full-scale experiments was conducted in a 70 m3 two-room building and two fire experiments were also performed outdoors. In all tests, poolfires of n-heptane were considered. Firstly, it was studied how the internal pressure is influenced by: different configurations of the doors, different areas of the poolfire, and safety measures, i.g. the opening of pressure relief vents or shutdown of the ventilation fan. The trend over time of the HRR, of several internal temperatures, of the pressure in the rooms, of the flow rate in the ventilation system, and of the gas concentrations was determined, verifying also the repeatability of the tests. Secondly, a mathematical model aimed at reproducing the tests was built inside FDS, allowing to compare the results of the simulations with the experimental ones. The comparison was largely satisfactory and, where differences were evident in the trend of the parameters, it was possible to find out their causes and to suggest ways to improve the accuracy of the modeled results.