Back to Search Start Over

A temperature and mass dependent thermal model for fire response prediction of marine composites

Authors :
Y. Wu
Christopher T. Key
James Lua
Brian Y. Lattimer
Jeff O'Brien
Source :
Composites Part A: Applied Science and Manufacturing. 37:1024-1039
Publication Year :
2006
Publisher :
Elsevier BV, 2006.

Abstract

An accurate assessment of accumulative damage of a composite ship structure subjected to fire is strongly reliant on the accurate characterization of the time dependent temperature distribution within the composite system. Current state-of-the-art analysis tools assume that thermal–mechanical properties of composite structures are independent of temperature change and mass loss. In this paper, a temperature and mass dependent heat diffusion model is developed to characterize the temperature and mass dependent heat conduction, energy consumption resulting from the decomposition, and the energy transfer associated with vaporous migration. The temperature dependent thermal conductivity and specific heat capacity are determined for the composite at a given resin decomposition stage using a recently developed small-scale test apparatus. Given the temperature dependent thermal properties of the fiber and resin materials, the resulting temperature dependent thermal properties of a woven fabric composite are computed from the newly developed thermal–mechanical analysis tool (TMAT). To assess the effect of using single heating rate based mass loss on the composite fire response prediction, the kinetic parameters are obtained from the post-processing of thermogravimetric analysis (TGA) test data conducted at various heating rates. The effects of temperature dependent thermal conductivity, specific heat capacity, and kinetic parameters determined at different heating rates are explored through the application of the temperature and mass dependent fire model to a composite plate subjected to a hydrocarbon fire. The accuracy of the temperature and mass dependent thermal model is demonstrated by comparing its response prediction with available experimental data.

Details

ISSN :
1359835X
Volume :
37
Database :
OpenAIRE
Journal :
Composites Part A: Applied Science and Manufacturing
Accession number :
edsair.doi...........19b6173ce95ec8cdbde4e7c01f5f6f1a
Full Text :
https://doi.org/10.1016/j.compositesa.2005.01.034