Evaluation of optimization schemes and determination of solid fuel properties for CFD fire models using bench-scale pyrolysis tests

CFD fire modeling tools are continuously developed and improved to increase their predictive capability of phenomena observed in practical applications. Such models require that effective material properties be provided so that the pyrolysis codes used in the models can properly estimate the thermal degradation of solid fuels involved in a fire situation. This paper presents analyses aimed at evaluating the plausibility of obtaining material properties numerically from pyrolysis data collected in a Fire Propagation Apparatus (FPA). A theoretical pyrolysis model is used to simulate the experimental data and the input parameters (i.e. the material properties) are adjusted to provide the best possible agreement between simulations and experiments. This is done via the application of evolutionary optimization methodologies. First, available optimization techniques are evaluated using synthetic data and it is shown that the Shuffled Complex Evolution approach ([17]) can recover the input parameters with high accuracy, efficiency, and robustness. Second, the algorithm is applied to experimental FPA pyrolysis data of practical materials: polymethyl methacrylate (PMMA), single-wall corrugated board, and chlorinated polyvinyl chloride (CPVC). (C) 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.