Test Methods Of Clothing Dynamic Moisture Transfer Based On Color Parameters And Application In Comfort Prediction

Purposes on this study are to explore in depth the color parameters based on fabric dynamic moisture transfer test methods and establish predictive models of subjective comfort for stretch knit pants. The method "Dynamic surface moisture transfer" test water transfer of fabric surface by the principle that cobalt chloride changes its color adding water. It research the vapor change in the fabric surface and the instantaneous response in this method.Water weight of the fabric surface may be visually seen from color change of the fabric after soaking cobalt chloride.It was found that white fabric soaked with cobalt chloride showed in blue after drying, with the fabric absorbing vapor, the color gradually from blue to purple and pink as the fabric soaked with cobalt chloride was tested on the simulation sweating system device. It offers a simple and effective method for dynamic moisture transfer to distinguish the distribution of fabric moisture content by color parameter. The work of this thesis is as follows:(1)Selecting 47 fabrics such as cotton, hemp, viscose, polyester, acrylic, Modal and wool to conduct the dynamic surface moisture transfer experiments with their own simulation sweating device, establish the positive and negative photo database of the dynamic wet transfer discoloration for 47 kinds of fabrics, the database of Munsell Color HV/C value, database of spectral photometer reflectivity, the database of L* A* B* value, moisture content database, the Munsell color photo database of positive and negative, the HV/C value database of spectral photometer.(2) Test Munsell color HV/C value to the simulated skin of simulated sweating devices and conduct analysis of variance, discover different test result between the different simulated skins by analysis of variance. Therefore, conduct water retention test to the seven common alternative simulated skin, select synthetically cotton fabric as the surface of simulated skin through the water retain rate and fabric structure, the faux suede as Inner layer of simulated skin.(3)It is discovered that there is a characteristic time appearing difference at fabric moisture absorption properties as reaching 15 minutes and 25 minutes in experimental though analyzing variation Monsell HVC of fabric sample. However, in the initial time period, such as 0 minutes, the Monsell color index is 2 in 37 minutes for 47 kinds of fabrics, which indicates that the index can not distinguish water features. The overall view from the line chart, the number of duplicate samples of three Monsell color HVC indicators are less than 5, accounting for about 10% of experimental samples. Therefore, only the color index measuring Monsell dynamic moisture transfer fabrics can not distinguish enough the water vapor permeability of fabrics, which is necessary to consider comprehensively the three indicators Monsell HVC color information. It is discovered that the proportion of number of varieties of maximum correlation coefficient value between the value of hue and lightness and moisture content value is close to by analyzing the grey correlation between Monsell colors HV/C value and water content in 52 fabrics. The evaluation methods of only Monsell color H values conversing the dynamic moisture transfer properties of fabrics miss the related information of the Monsell color brightness V and Chroma C, so it is necessary to consider the complete label information of Monsell color HV/C value and establish the composite indicator of a dynamic evaluation of the surface moisture transfer.(4) Using the principle of fuzzy comprehensive evaluation to construct the the fuzzy synthesis algorithm of the Monsell HV/C value of fabrics, and applying the algorithm to calculate HV/C fuzzy comprehensive index of fabrics between seven time point in 0-30 minutes. The results showed that:the fuzzy composite index of Monsell color HV/C value can be more intuitive distinction color changes of fabrics after absorbing water than that of Monsell color H value, and it provides a new scientific evaluation approach for the dynamic water vapor permeability of fabrics.(5)The linear regression model between the color and the fuzzy composite index for the Monsell color HV/C value and moisture content of fabrics is established. The linear regression model determination coefficient Rz2 of the monsell color index and moisture content of fabrics is 0.8798, less than the determination coefficient RM20.9002 for the color HV/C value and moisture content, which further proves that the fuzzy composite index of the Monsell color HV/C value is more accurate to response the water vapor permeability of fabrics than the Monsell color index.(6) By analyzing of all samples of the reflectance spectrophotometer and wavelength curve, the results show that the sample reflectance discrimination is the largest in the wavelength of 600-700 nm. So the reflectivity at the wavelength of 650 nm can be used as the Eigen value to determine the dynamic surface moisture transfer values of the fabrics. By analyzing the linear correlation determination coefficient R2of L*, a*, b* values and moisture content, extract L* the highest correlation coefficients of water content and the most value as another characteristic value measuring the fabric dynamic surface moisture transfer. By establishing linear mode between moisture content and reflectance, moisture content and L* value, it is revealed that the predicted moisture content of 650nm reflectivity is higher accuracy than L* values.(7) By cluster analysis methods, combining with moisture line graph from Figure 4-15 it is found that experimental time analysis to 15 minutes, the discrimination for the moisture content of each sample is the most clear, so the fifteen minutes is chosen as a representative of the moisture content value. The BP neural network model under the spectral photometer L*(X1)、a* (X2)-b* (X3). C* (X4)、ho* (X5) and moisture content in 15 minutes is established. Prediction results show that:the neural network model to predict the moisture content of the fabric has good accuracy (relative error<5%), not only has a high degree of learning sample fit, but also has very close effect between the prediction values and the experimental test values for the two samples.(8) It is discovered through the partial correlation analysis between physical indexes and Color Parameters of the Clothing Dynamic Moisture Transfer:The partial correlation is the highest between the reflectivity and the thermal resistance, followed by the thickness, between the water vapor transmission rates, return to tide rate, air permeability rate and the reflectivity there is a certain relationship. Between absorbing water height and reflectivity the partial correlation is non-existent. Between thermal resistance and reflectivity there are positive bias correlation, Between the water vapor transmission rate, return to tide rate, air permeability rate and the reflectivity there are negative bias correlation.(9) Select seven kinds knitted fabrics to make stretch pants and test subjective comfort, and design the subjective measure questionnaire through Likert Scale. Likert Scale selected the initial scale evaluation of 18 kinds of subjective feeling, after the initial investigation of 38 subjects, removed the low differentiate questions, retained 9 high differentiate subjective feeling evaluation statement such as humidity sensitive, hot feel, sense of restraint, prickle, sticky feel, a sense of weight, smooth feel, static integrated comfort feel and comprehensive comfort evaluation after 5 minutes movement.The formal subjective test questionnaire were designed by the nire subjective feel evaluation statement. Conduct the factor analysis to the subjective comfort of fitted pants is, obtain the 0-minute linear combination of the two principal components, and rename the first factor Z01 for the static touch comfort factor, the second factor Z02 for wet sticky binding comfort factor; 15-minute linear combination of the two principal components, rename the first factor Zi51 for the integrated comfort factor, the second factor Z152 for the wet and heat comfort factor.30-minute linear combination of the two principal components, rename the first factor Z301 for the hot weight feeling comfortable factor, the second factor Z302 for the thorn sticky comfort factor, the third factor Z303 for wet binding comfort factor.(10)Through the analysis of 22 prediction model for subjective comfort sporting by 0 minute,15 minutes,30 minutes, it shows that:the use of the color characteristic values of the fabric dynamic moisture transfer may predict perfectly subjective comfort of knit stretch pants, and the prediction is high accuracy, the fitting error between the measured and predicted values is less than 5%. It is found by analysis that fuzzy index (X5) on subjective comfort is the most significant. It appears 8 times as main feature values in the 22 models. Followed by the L value (X3) which appears 3 times as main feature values in the 22 models. Moisture content (X1), G median (X9) appears 2 times as main feature values in the 22 models. Reflectance (X2), subjective index (X4), objective index (X6), G average (X7), G bias (X8) appearsl times as main feature values in the 22 models. RGB mean (X10), RGB standard deviation (X11), RGB intermediate value (X12), LAB average (X13), LAB standard deviation (X14) LAB intermediate values (X15) do not appear as the main features value.(11) Considering the factor analysis and stepwise regression analysis it shows that:Exercise 0 minute, in the static, the feeling wearing the stretch pants are mainly static comprehensive sense of comfort and tactile threshold region. At this point the main characteristics values are fuzzy composite index value (X5),reflectance (X2),G average (X7).Exercise 15 minutes, the feeling are the weight of the subjective sense and stick in a sensory threshold area. At this point the main characteristics values are fuzzy composite index value (X5), moisture content (X1), G median (X9), objective index (X6).Exercise 30 minutes, the feeling are subjective thermal feel, a sense of weight and sticky feeling in sensory threshold region. At this point the main characteristics values are L value (X3), moisture content (X1), Fuzzy Composite Index (X5), the subjective index(X4).