Correlation functions between online and offline detected yarn

Abstract

Electronic yarn clearer online detected data can be used to justify whether the evenness performance of yarn meets the standard for further processing. The reliability of offline detection is limited by the sample numbers. Based on the analysis of 100 sets of yarn performances (both online and offline) and assuming that external factors are fixed, linear fitting was performed to obtain the correlation functions between online and offline data of yarn performance. Finally, the correlation functions between the online detected data, coefficient of variation value and hairiness H value and offline detected data were established. The correlations between these parameters are significant. In particular, the function-deduced confidence and prediction intervals are very useful to forecast abnormal conditions during real yarn production. The production efficiency could be significantly improved under the guiding of these correlation functions.