Application of an Adaptive Neuro-fuzzy System for Predicting Rotor Spinning Yarn Quality Properties

摘要

One of the problems in rotor spinning is to know the speed of the opening roller and rotor to be used in setting machine in order to produce the desired yarn quality. The opening roller and navel have been selected by using multi-objective fuzzy pattern recognition model. The purpose of this study was to model and optimize a manufacturing process. The rotor spinning machine was run at random sets of rotor speed and opening roller speed and the data obtained trained by using an Adaptive Neuro-Fuzzy Inference System (ANFIS). According to overall satisfaction degree and utility value, the gradation of the samples shows that the best samples are produced from combination of low rotor speed and high rotor speed.