A NEURAL NETWORK MODEL TO PREDICT THERMAL CONDUCTIVITY OF STRETCH KNITTED FABRICS

Elastic knitted fabrics are gaining popularity for apparel use due to its improved comfort functional properties. So a lot of researchers are interesting about these structures. In this paper, we presents an artificial neural network (ANN) modeling thermal conductivity of knitted fabrics made from pure yarn viscose (regenerated cellulose) and cotton (cellulose) �bers and plated knitted with elasthane (Lycra) fibers. Yarn count, fabric thickness, knitted fabric structure type, elasthane fiber proportion (%), elasthane yarn linear density, yarn composition, fabric areal density, and gauge, were used as inputs to the ANN model. A virtual leave one out technique allowing the selection of the optimal ANN architecture was used. The generalization ability of the chosen ANN model was calculated. It has revealed a good robustness in prediction with good accuracy. The developed model was able to accurately predict the thermal conductivity of stretch knitted fabrics by selecting the optimum operating parameters and intrinsic features of structure of fabric and yarn.