Shrinkage and Porosity of Cotton and Hemp Fabrics Based on ELM Algorithm

Abstract

Cotton and linen fabric is a kind of mixed textile of half linen and half cotton, which has the characteristics of both linen and cotton, so it is widely used in summer clothes. Shrinkage and bulk properties are important indicators of the quality of cotton and linen fabrics, so it is necessary to accurately predict their shrinkage and bulk properties. In this study, the ELM is introduced to predict its fluctuation range, and the PSO algorithm is adopted to optimize the ELM. The FIG technology is introduced to realize the dynamic extraction of its characteristics, obtain the FIG-PSO-ELM algorithm, and verify its validity. The experimental results show that within 1500 seconds, the shrinkage rate and elastic viscosity of cotton and linen fabrics change greatly, and the values are in the range of 49.2 to 51.2, and the maximum value is 51.15. In addition, the prediction results of the FIG-PSO-ELM method in Low, R, and Up sequences are basically consistent with the real values, showing good prediction accuracy. Compared with the prediction results of the other four algorithms, the FIG-PSO-ELM method has the best prediction effect, and the errors on the three sequences are the smallest, among which the minimum is 0.0019 and the minimum is 0.0368. Overall, the FIG-PSO-ELM method has a good prediction effect, accurately predicts the polycondensation area and porosity of cotton and linen fabrics, and has a good effect on the actual weaving of cotton and linen fabrics.