棉织物洗涤柔顺性评价模型构建OA

Fabric softness,a paramount attribute influencing consumer preference and perceived quality in textiles,is profoundly affected by household laundering practices and the use of chemical softeners.Despite its significance,the objective and quantitative evaluation of fabric softness remains a considerable challenge within both industry and academia.Traditional reliance on subjective hand-feel assessments,while direct,is inherently susceptible to individual variability and lacks reproducibility.Although instrumental methods using devices like the PhabrOmeter or KES-F system provide objective data,they often fall short of fully capturing the complex,multi-dimensional nature of human tactile perception,which integrates sensations of bending,compression,and surface friction.Previous research has explored the relationship between fabric properties and handle,yet a comprehensive predictive model specifically tailored for quantifying the softness of cotton fabrics under diverse,real-world home washing conditions—encompassing variations in softener chemistry,dosage,temperature,and cumulative treatment cycles—is notably absent.This gap impedes the systematic development and efficacy verification of softener products.The primary objective of this study was,therefore,to construct a reliable and quantifiable regression model that effectively bridges the gap between subjective human perception and objective instrumental measurements,so as to establish a scientific methodology for predicting and quantifying the softness of cotton fabrics after household washing. Pure cotton plain weave fabrics were selected as the substrate.A total of 108 distinct fabric samples were prepared through a controlled process involving standard preconditioning,pre-washing with a standardized detergent to eliminate manufacturing finishes,and subsequent softening treatments.Four prevalent types of commercial softener actives—esterquat,imidazolinium quat,di-octadecyl dimethyl ammonium chloride(D1821),and amide quat—were employed.These were applied at three concentrations(1g∕L,3g∕L and 5 g∕L),across three temperature levels(25℃,35℃and 45℃),and for one,three,or five treatment cycles to simulate cumulative washing and softening effects.For the subjective evaluation,a panel of eight trained assessors evaluated all samples under standard atmospheric conditions according to established AATCC guidelines.Each assessor scored four key hand-feel attributes:bending stiffness handle(BSH)to assess resistance to folding,friction roughness handle(FRH)to evaluate surface smoothness,compression fullness handle(CFH)to gauge bulk and compressibility,and the overall total handle value(THV)to represent integrated comfort.The consistency and reliability of the subjective ratings were statistically verified using Kendall's coefficient of concordance(W).In parallel,a suite of objective physical measurements was conducted.The subsequent data analysis involved several steps.First,the weight coefficients of the three primary hand-feel attributes(BSH,FRH and CFH)in constituting the THV were determined through regression analysis.Second,the normality of the data distribution for the primary handles was confirmed using the Kolmogorov-Smirnov test.Finally,stepwise regression analysis was employed as the core statistical technique to build the predictive models,which aimed to relate the six objective physical indicators to each of the three primary subjective hand-feel scores(BSH FRH and CFH). The subjective evaluations demonstrated exceptional consistency among the eight panelists,with all Kendall's W values for BSH FRH CFH,and THV being significant at the 0.01 level,thereby firmly establishing the validity of the subjective data.The contribution of the primary hand-feel attributes to the overall THV was successfully quantified,with weight coefficients determined as 0.475 for BSH,0.222 for CFH,and 0.302 for FRH,indicating that bending stiffness and surface friction are the most critical perceptual factors for softness in cotton fabrics.The stepwise regression analysis produced highly accurate predictive models.The optimal regression equation for BSH achieved an R2 of 0.907,incorporating softness,drape coefficient,and smoothness as key predictors.The model for CFH attained an even higher R2 of 0.913,relying on fabric thickness and areal density.The FRH model,based on smoothness and stiffness,also showed strong performance with an R2 of 0.833.These high R2 values confirmed that the selected objective parameters effectively captured the nuances of subjective hand feel.Most importantly,a subsequent independent validation experiment confirmed the model's practical utility and stability,demonstrating a strong and significant correlation(Pearson coefficient=0.899,and R2=0.807)between the model-predicted THV and the averaged subjective panel ratings.