A Study of Reproducibility of a Full-Scale Multi-Room Compartment Fire Experiment

摘要

A study of 45 pool fire tests is presented in this paper. All tests have been conducted in the same experimental setup but with four different ventilation scenarios. The experimental setup consisted of a three-room compartment connected to a stairway. The tests have been conducted during a 6-year period under slightly different ambient conditions. The purpose of this paper is to study and quantify the reproducibility of these tests with the aim to illustrate the degree of variability that can be expected in this kind of experiment and how the variability is affected by different ventilation scenarios. The 95% confidence interval of the measured temperature rise in the tests covered +/- 7% to 35% around the mean, depending on place of measurement and studied scenario. The variation in the results is due to both the variation of different weather conditions and other unknown parameters. It has not been possible to statistically distinguish between these two sources of variation. The data set allows to estimate the effect of random variations in experiments of fire behavior and smoke movement across compartments. The fuel of the pool fires is well defined and the mass loss is known, which makes the data of use in validation of multi-room compartment fire models.