Modelling the Relationships between the Processing Parameters and Properties of Rotor Spun Wool/Acrylic Blended Yarns
Rotor spinning is the most widely used new spinning technology nowadays. However, applications of the rotor spinning technology in wool spinning industry are still very scanty. There are great quantities of noils in China. If the rotor spinning technology is adopted to process these noils, considerable economic benefits will be achieved. Wool/acrylic blended yarns are spun on a rotor spinning machine and the yarn properties are tested. The systematic methodology for modelling the yarn properties from the processing parameters is explored in this paper. The response surface methodology and Adaptive Network-based Fuzzy Inference System (ANFIS) are employed to predict the effects of rotor speed, opening roller speed and twist multiplier on the yarn properties. Performance of these two models is evaluated by comparing the root mean square error and surface diagrams. The results revealed that ANFIS had better predictive accuracy, but the response surface methodology is superior in handling data distributions in terms of depicting the relationship between inputs and outputs.
rotor spinning yarn properties response surface methodology ANFIS
AKOLO Barrison Paul CHEN Ting WU Lili
College of Textile and Clothing Engineering, National Engineering Laboratory for Modern Silk, Soochow University, Suzhou, 215021, P.R.China
国际会议
The 12th International Wool Research Conference(第十二届国际羊毛会议12th IWRC)
上海
英文
138-142
2010-10-19（万方平台首次上网日期，不代表论文的发表时间）