Modeling to predict thermal aging for flame-retardant fabrics considering thermal stability under fire exposure

The performance of firefighters’ clothing will deteriorate due to various exposures. Predicting its service life before decommissioning is essential to guide the use and maintenance of the uniform. The aim of this study is to introduce a model to predict the tensile strength of flame-retardant fabrics under fire exposure. The thermal degradation and microstructure of Kevlar/polybenzimidazole and polyimide/Kevlar fabrics were investigated. The decrease of tensile strength was attributed to the chemical changes and the development of microstructure cracks and charring of the fibers. Multiple linear regression (MLR) and artificial neural network (ANN) models were established to predict the tensile strength after thermal aging. The ANN model presented a better prediction result ( R 2 = 0.88, root mean square error (RMSE) = 96.91) than the MLR method ( R 2 = 0.76, RMSE = 138.61). The addition of fabric backside temperature ( T), glass transition temperature ( T g ), and degradation temperature ( T d ) further increased the R 2 (4%) and decreased the RMSE (14.99) of the ANN model, which was recommended as a prediction approach with better accuracy. The findings of this study will contribute to estimating the continuous performance of firefighting clothing.