Modeling of Thermal Conductivity of Stretch Knitted Fabrics Using an Optimal Neural Networks System

Elastic knitted fabrics are gaining growing popularity for clothing use due to its enhanced comfort properties. In this study, the modeling of thermal conductivity of knitted fabrics made from pure yarn cotton (cellulose) and viscose (regenerated cellulose) fibers and plated knitted with elastane (Lycra) fibers using an Artificial Neural Network (ANN) was investigated. Knitted fabric structure type, yarn count, yarn composition, gauge, elastane fiber proportion (%), elastane yarn linear density, fabric thickness, loop length and fabric areal density, were used as inputs to the ANN model. Two types of model were built by utilizing multilayer feedforward neural networks which took into account the generality and the specificity of the stretch knitted fabric families. A virtual leave one out approach dealing with over fitting phenomenon and allowing the selection of the optimal neural network architecture was used. The proposed ANN technique was compared to the linear regression analysis. The generalization ability of the selected ANN model was calculated. It has revealed an excellent robustness in prediction with good accuracy, superior than that of the linear model. The developed model was able to predict accurately the thermal conductivity of stretch knitted fabrics by selecting the optimum operating parameters and characteristics of yarn and fabric for a particular end-use.