This paper is to report a prediction model for thermal protective performance of multilayer fabrics based on Matlab neural network toolbox. Then a back propagation (BP) neural network model is developed to predict the...This paper is to report a prediction model for thermal protective performance of multilayer fabrics based on Matlab neural network toolbox. Then a back propagation (BP) neural network model is developed to predict thermal protective performance of multilayer fabrics for firefighters. The network consists of twelve input nodes, six hidden nodes, and one output node. The inputs are weight, thickness, density of warp and weft, limited oxygen index (LOI), and heat conductivity of each-layer fabric. Thermal protective performance (TPP) rating of multilayer fabrics is the output. In this paper, the data from the experiments are used as learning information for the neural network to develop a reliable prediction model. Finnally the model performance is verified, and the proposed model can be applied to predict the thermal protective performance of multilayer fabrics for firefighters.展开更多
文摘This paper is to report a prediction model for thermal protective performance of multilayer fabrics based on Matlab neural network toolbox. Then a back propagation (BP) neural network model is developed to predict thermal protective performance of multilayer fabrics for firefighters. The network consists of twelve input nodes, six hidden nodes, and one output node. The inputs are weight, thickness, density of warp and weft, limited oxygen index (LOI), and heat conductivity of each-layer fabric. Thermal protective performance (TPP) rating of multilayer fabrics is the output. In this paper, the data from the experiments are used as learning information for the neural network to develop a reliable prediction model. Finnally the model performance is verified, and the proposed model can be applied to predict the thermal protective performance of multilayer fabrics for firefighters.
