Study Of Thermal Protective Performance And Comfort For Firefighter-clothing Fabrics

Fire-fighter protective clothing is of great importance to firemen, as they are routinely exposed to heat and sometimes contacted with flames. So protective clothing should be comfortable and provide protection against fire and heat. Therefore, the objective of this paper is to study the thermal protective performance and comfort of firefighter-clothing fabrics and the factors which affect the thermal protective performance and comfort of fabrics. It is also to investigate optimal assembly of the overall materials in terms of their heat protection and moisture transmission.In this paper, twelve outer fabrics such as NomexIIIA, PBI/Kevlar, Kermel, polysulfonamide, flame-resistant cotton, etc, are selected and two moisture barriers with PTFE, and three thermal liners are studied. The physical properties of each fabric are tested.To simulate three-layer firefighter protective clothing system, the appropriate fabrics for firefighter protective clothing are selected, and the orthogonal design method is used. Thermal protective performance and moisture transmission of firefighter clothing are studied through different material combinations. Then experiment data are discussed by range analysis and variance analysis to study the effects of three layers for heat protection and moisture comfort. Results show that the priority order of three factors for TPP rating is outer fabric> moisture barrier> thermal liner, and the priority order of three factors for WVTR value is moisture barrier> thermal liner>outer fabric. Outer fabric and moisture barrier have significant effects on thermal protective performance and moisture transmission at 90 percent confidence level respectively.Based on the above experiments, factors which influence thermal protective performance on outer fabrics are studied. Results show that fiber content, mass weight, and thickness of fabrics are dominant factors determining thermal protective performance in high intensity. Besides, exposure conditions such as intensity of the heat and property of the heat also affect thermal protective performance. With the greater heat exposure, time of the 2^nd degree burn is shorter. With 84Kw/m² mixtures of convective/radiant heat exposure, the TPP rating of 70/30 condition is better than that of 50/50 condition.The effect of moisture on thermal protective performance is investigated in different radiant and convective heat exposures. Based on NI virtual instrument and Labview graphical programming language, a virtual instrument determining temperature has been developed. And four fabric preparations are evaluated. At the 16.8kw/m² radiant exposure, three-layer assemblies sprayed with little water (33%) on outer fabrics provide the weakest thermal protection. Time of the 2^nd degree burn for assemblies which are sprayed with much water (66%) on outer fabrics is similar to the assemblies without water. And three-layer assemblies with outer fabric soaked with water (100%) provide the greatest thermal protection. However, moisture enhances the thermal protection of three-layer assemblies at the 84kw/m2 convective and radiant exposure.At last, a nonlinear correlation between fabric physical properties and thermal protective performance is developed in this paper. Based on Matlab neural network toolbox, BP neural networks using to predict the thermal protective performance of fabrics are developed. For outer heat resistant fabric (single-layer), a BP neural network with a single hidden layer is constructed including nine input nodes, eleven hidden nodes, and one output node. The input variables are mass weight, thickness, weave, thread density of warp and weft direction, yarn count of warp and weft direction, limited oxygen index, conductive coefficient, and damage length. And TPP rating is used as output variable. In the training process, the connection weights are modified with gradient-descent algorithm and adaptive learning rate to solve the two defects of the BP network. After training, the predicted ability of the proposed BP neural network is tested. The results indicate that correlation coefficient between the predicted and experimental value is 0.952, and average error is 3.93 percent.For three-layer assemblies, a BP neural network with a single hidden layer is constructed including twelve input nodes, six hidden nodes, and one output node. The input variables are mass weight, thickness, limited oxygen index of each layer fabric. And TPP rating of three-layer assembly is used as output variable. The results show that correlation coefficient between the predicted and experimental value is 0.962, and average error is 4.11 percent. Therefore, evaluation of the fabric's thermal protective performance can be economical and accurate through the proposed BP prediction model.