Application of an adaptive neuro-fuzzy system for the prediction of cotton yarn strength from HVI fibre properties

摘要

This paper presents the application of a hybrid neuro-fuzzy system for the prediction of cotton yarn strength from HVI fibre properties. The proposed system possesses the advantages of both artificial neural networks and fuzzy logic, and is thus more intelligent. HVI fibre test results are used to train the neuro-fuzzy inference system and its prediction performance is compared with those of artificial neural network and regression models. The prediction accuracy of the proposed neuro-fuzzy system is superior to that of a conventional multiple regression model and comparable with an artificial neural network model.