Analysis and Prediction of the Wearing Comfort Performance of an Assembly of Fabric by Optimization ANN

This article is to report a study based on fabric physical properties measured on the KES system. Grey incidence (GI) analysis, as a mathematic method that ranks the sequence of importance of lots of variables in complicated factors has been applied, In order to select the efficient input variables of ANN (artificial neural network) during the prediction of wearing comfort performance. A series of experiments and analyses were performed to study the heat-moisture comfort property of fabric during exercise in a standard environmental chamber conditions. The optimization ANN models with the parameters selected by using the GI analysis are investigated, which construct on the convergence speed and the prediction accuracy The result indicates that the optimization model of BP neural network is an efficiency technique and has a wide prospect in the application to prediction of wearing comfort performance.