Repeatability Assessment in Tunnel Fire Experiments

Abstract

A medium-scale experimental setup located in Revinge, Sweden, is used to generate new repeated experimental datasets focussed on measuring a set of key variables, including time-temperature curves and temperature profiles. A total of 6 repeated fire tests were conducted and are here used to exemplify typical issues associated with repeatability, uncertainty and variability in tunnel fire experiments. A key aspect considered in this work is the smoke stratification among repeated experiments and how fire behaviour can be affected by boundary conditions. This work is important to facilitate future efforts to collect high-quality tunnel fire data including a quantitative assessment of the experimental uncertainty, data that will be useful for validation of tunnel fire models.