engleski

Developing ANN Model for the Prediction of Comfort-related Indicator Based on Yarn Parameters

The presented research focuses on knitted fabrics produced of cotton and polyester yarns in different compositions. The IR thermographic method is used to capture the duration of the wetness period after simulated intensive sweating. The dataset is further used for the ANN modelling. It was shown that all the ANN predicted values are very close to the corresponding actual value. Moreover, the absolute error was less than 4% in all the samples and the correlation coefficient between the actual and predicted values was 0.99. Therefore, the ANN is able to capture the relationship between the observed input and output parameters.

yarn, polyester, cotton, modelling, ANN, sweating

nije evidentirano