Wood scientistForest Products and Harvesting Research staff, USDA Forest Service, Forest Products LaboratoryUSDA Forest Service, MadisonWashington, DC, WI
Pages: 16 Published: Jan 1994
This work demonstrates the complexity of modeling wall and corner fires in a compartment. The model chosen for this purpose is the Ohio State University (OSU) room fire model. This model was designed to simulate fire growth on walls in a compartment and therefore lends itself to direct comparison with standard room test results. The model input were bench-scale data obtained from the ASTM Test Method for Heat and Visible Smoke Release Rates for Materials and Products (E 906). Six wood materials were tested in the bench-scale test and also in an ASTM room fire test (proposed method). The simulations from the OSU model were compared with the database of 26 room tests representing a range of conditions. We treated the model as a black box and only varied the input data. The criteria used for comparison were heat release rate, radiative heat flux to the floor, and upper layer temperature. The agreement between model predictions and experiments varied. We conclude that the OSU model in its present state of development is not able to track fire growth in scenarios of the burner source against the back wall. The accuracy of the bench-scale data and methods to reduce the input data play a very important role in the simulation results. This work also raises many important issues such as the need for clear documentation of the modeling process and the definition of criteria for determining good agreement between the model simulation and experimental results.
fire models, validation, predictive capability
Paper ID: STP12955S