Statistical and ANN analysis of thermal and evaporative resistances of multilayered fabric ensembles

Abstract

In this paper, thermal and evaporative resistance of multilayered fabrics meant for cold weather conditions have been reported. Polyester hollow fibers of 6 denier and 15 denier were used to produce needle-punched fabrics. Full factorial experimental design was used to produce 30 different needle-punched nonwoven fabrics from two different linear densities of fiber by varying mass per unit area and punch density at three and five different levels, respectively. The needle-punched nonwovens were used as insulative middle layer. Single jersey knitted fabric and polytetrafluoroethylene-coated fabric was used as inner and outer layer, respectively. The multilayered fabric ensembles were evaluated for thermal and evaporative resistance using sweating guarded hot plate (SGHP) system. Regression equations were developed to draw the contour plots and to analyze the effect of different parameters on thermal and evaporative resistance of fabrics. Two independent one-way analysis of variance (ANOVA) were conducted to find the significance of linear density of fiber and effect of inner and outer layers on thermal and evaporative resistance of fabrics. The two-way ANOVA was conducted to analyze the effect of mass per unit area and punch density on thermal properties of fabrics and ‘F’ values were calculated. Mean square values of pure error and lack of fit were studied to analyze the fitness of the developed model for thermal properties of fabrics. An artificial neural network (ANN) model was developed to predict the thermal and evaporative resistance of multilayered fabrics and compared with the experimental values. It has been observed that the ANN model predicts the thermal and evaporative resistance of multilayered fabrics with high degree of accuracy.