Manufacturing method of luminous blanket with noctilucence
By establishing a prediction model based on the LSSVR model, the problem that traditional luminous luminescent blanket manufacturing process is difficult to achieve efficient dyeing on cotton fabric carpets is solved, and more efficient and accurate dyeing process optimization is achieved.
Patent Information
- Application Number
- CN202510012167.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-06
- Publication Date
- 2025-05-06
AI Technical Summary
The traditional luminous luminous blanket manufacturing process is difficult to achieve efficient dyeing on cotton-planted fabric carpets, and requires a lot of experience-dependent experimental optimization, so it is impossible to systematically observe the interaction effect of dye and luminous powder.
The least squares support vector regression (LSSVR) model was established using a nonlinear prediction method to predict the depletion rate, dilution rate, total dilution rate and color intensity of dyed cotton fabrics, and the dyeing parameters were optimized through previous experiments and machine learning.
It improves the commercial dyeing efficiency of luminous luminescent blankets, reduces the dependence of manual experiments, can more accurately predict and adjust dyeing parameters, and improves the repeatability and efficiency of the dyeing process.
Smart Images
Figure CN119943215A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of luminous blanket manufacturing, and in particular to a manufacturing method of a self-luminous luminous blanket. Background Art
[0002] Luminous blankets can provide partial lighting in the dark, which is convenient for users to move around. The traditional manufacturing of luminous blankets is a process of mixing luminous powder with printing dyes for dye bath or screen printing. The luminous principle depends on the luminous elements embedded in the inner layer of the blanket fabric. Before printing the luminous pattern, the luminous powder needs to be mixed with the printing dye to ensure that the color intensity meets the standard. Generally, this process is mainly used on non-woven fabrics or other chemical fiber fabrics, but the luminous dyeing of cotton fabric carpets is still a technical difficulty at present, because most cotton fabric dyeing parameters are determined based on past experience and require experienced workers. The dye value after mixing with luminous powder changes, and it is often necessary to retest. Then, according to the adopted dye parameters and the given dye formula, various performance indicators related to the dyeing process are obtained during the dyeing process. The usual method of traditional optimization process is to modify each parameter one by one while keeping other parameters at their current levels. This allows the influence of a single parameter to be studied while keeping the overall optimization intact, but it requires a lot of experiments. The results provided by this method are also not very accurate, and the interaction effect of dye and luminous powder cannot be observed, which makes it more difficult to determine the performance of process parameters. Therefore, systematic process optimization of dyeing of cotton-planted fabrics using reactive dyes is a key to improving the efficiency of commercial dyeing of luminous blankets. Summary of the invention
[0003] In view of the above technical problems, the present invention provides a method for manufacturing a self-luminous luminous blanket, comprising establishing a prediction model and adjusting parameters according to the prediction model to perform process dyeing; The prediction model adopts a nonlinear prediction method to predict output parameters, and the output parameters include exhaustion rate (E%), dilution rate (F%), total dilution rate (T%) and color strength (K / S) of the dyed cotton fabric. The prediction model is a least squares support vector regression (LSSVR) model based on input parameters, and the input parameters include dye dosage, fixation temperature, fixation time, dye bath pH value, material-liquid ratio and salt concentration. The prediction accuracy of the prediction model is calculated using root mean square error (RMSE), mean absolute error (MAE) and determination coefficient ( ) for evaluation; (1); (2); (3); (4); (5); Among them, the Indicates the actual output, the Represents the predicted output, Indicates the total number of samples collected. represents the prediction error, represents the approximation error, and Represents the target value and the actual value in the training and test data sets. and represent , and represent ; Establishing the prediction model includes preliminary experiments and machine learning. The preliminary experiments adopt the Taguchi method to design experiments. The preliminary experiments include parameter optimization and robust design.
[0004] This method of combining traditional dyeing textiles with machine learning predictive modeling can greatly improve the upgrading of the textile industry and the efficiency of existing commercial dyeing, gradually shifting from traditional reliance on manual labor and experience to the use of artificial intelligence. BRIEF DESCRIPTION OF THE DRAWINGS
[0005] The present invention will be further described below in conjunction with the accompanying drawings.
[0006] Figure 1 The predictive modeling process in the present invention; Figure 2 It is the correlation between the actual value and the predicted value of the parameter of the model in the present invention. DETAILED DESCRIPTION
[0007] Various aspects of the present invention are described in further detail below.
[0008] Unless otherwise defined or indicated, all professional and scientific terms used herein have the same meaning as those familiar to those skilled in the art. In addition, any method and material similar or equivalent to the described content can be applied to the method of the present invention.
[0009] The present invention provides a method for manufacturing a self-luminous luminous blanket, comprising establishing a prediction model and adjusting parameters according to the prediction model to perform process dyeing; like Figure 1As shown, the prediction model adopts a nonlinear prediction method to predict output parameters, and the output parameters include exhaustion rate (E%), dilution rate (F%), total dilution rate (T%) and color strength (K / S) of dyed cotton fabric. The prediction model is a least squares support vector regression (LSSVR) model based on input parameters, and the input parameters include dye dosage, fixation temperature, fixation time, dye bath pH value, material-liquid ratio and salt concentration. The prediction accuracy of the prediction model is calculated using root mean square error (RMSE), mean absolute error (MAE) and determination coefficient ( ) for evaluation; (1); (2); (3); (4); (5); Among them, the Indicates the actual output, the Represents the predicted output, Indicates the total number of samples collected. represents the prediction error, represents the approximation error, and Represents the target value and the actual value in the training and test data sets. and represent , and represent ; The root mean square error (RMSE), a scale-dependent error metric, largely determines the effectiveness of the prediction model. Because it allows a single variable to be compared across multiple different configurations, this metric is used to determine whether the data split ratio is appropriate. In other words, the root mean square error (RMSE) is a statistic that evaluates the degree to which the model deviates from the correct answer, and the lower the value, the higher the prediction accuracy. The RMSE is calculated by taking the square root of the sum of the squares of the differences between the actual output and the expected output. In addition, it can also be viewed as an indicator of the difference between the expected value and the actual observed value, so a lower RMSE indicates a higher accuracy and ability to predict the results.
[0010] The mean absolute error (MAE) is a statistic that does not consider either the directionality or severity of the error when considering a set of predictions. This value is the weighted mean of the absolute errors between the expected and actual values of all observations in the test dataset.
[0011] Coefficient of determination It is a statistic of the goodness of fit between the model test values and the predicted values, ranging from zero to one. If it is close to one, the selected inputs produce the desired outputs; if it is far from one, some adjustments are needed. The calculation of the coefficient of determination R² is based on a comparison of the sum of squared residuals and the sum of squared deviations from the mean of the variable in question.
[0012] Establishing the prediction model includes preliminary experiments and machine learning. The preliminary experiments adopt the Taguchi method to design experiments. The preliminary experiments include parameter optimization and robust design.
[0013] The preliminary experiment included 27 separate tests at 6 different input parameters and 3 different factor levels, using the rotation method to dye the cotton fabric, adding sodium chloride to heat the mixed sodium carbonate to make a dye bath at 20°C, and using soap solution to clean the unfixed dye. The soap solution was a non-ionic detergent with a concentration of , the material-liquid ratio was 1:10, the temperature was 95°C, and the duration was 15 min. After the soap solution treatment, the dyed cotton fabric was dried in a drying oven at 80°C for 30 min.
[0014] The individual tests formed a data set. After uploading the data set to MATLAB, it was split 80 / 20 between the training set and the test set, and the root mean square error (RMSE), mean absolute error (MAE) and determination coefficient ( ).
[0015] Using a spectrophotometer, the light absorbance of the dye bath solution before and after dyeing and the residual soap solution was measured and recorded at the maximum absorption wavelength of 560nm. The exhaustion rate (E%), dilution rate (F%) and total dilution rate (T%) were determined using the following equations: (6); (7); (8); The light absorbances of the soap solution before and after dyeing are expressed as , and express.
[0016] Tables 1, 2, 3, and 4 below show the results of the machine learning model, E%, F%, T%, and K / S, respectively. Figure 2 The fitting of each parameter of the model in the present invention is shown, and R² is between 0.6606 and 0.9819, which shows that the prediction accuracy of the model is excellent.
[0017] Table 1 shows the prediction modeling results of E%:
[0018] Table 2 shows the prediction modeling results of F%:
[0019] Table 3 shows the prediction modeling results of T%:
[0020] Table 4 shows the prediction modeling results of K / S:
[0021] A reflectance spectrometer was used to evaluate the color intensity, i.e., the K / S ratio, of cotton fabric samples randomly extracted from 20 different positions. The color intensity value was determined by measuring at the wavelength where the dye absorbs the lightest light. The average color intensity was taken, and the surface morphology of the undyed and dyed samples was analyzed using a scanning electron microscope. The samples were trimmed to a size of no more than 1 square centimeter each and then fixed individually on a standard sample holder with an accelerating voltage of 10 kV and a working distance of 15-17 mm. An X-ray diffractometer was used to measure the X-ray diffraction patterns of the undyed and dyed samples, which facilitated microscopic observation of the state of the fabric after printing and dyeing.
[0022] The parameter optimization includes conducting experiments according to the Taguchi design method in Minitab software to ensure the repeatability of the dyeing process, and finding the parameter optimization of each response, namely E%, F%, T% and K / S, according to the signal-to-noise ratio analysis.
[0023] The LSSVR model also includes tuning parameters, which are represented by the symbol , and Indicates that the values are set to 32, 0.0625, and .
[0024] The input parameters are adjusted in the model to obtain the output parameters, wherein the input parameters are the values after the luminous powder and the printing dye are mixed.
[0025] The final input parameters and output parameters are determined according to the LSSVR model to guide the dyeing process. After dyeing, a luminous fabric layer is made. The luminous elements are embedded in the luminous fabric layer to make a blanket with both luminous and electronic components.
[0026] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for manufacturing a self-luminous luminous blanket, characterized in that: It includes establishing a prediction model and adjusting parameters for process dyeing according to the prediction model; The prediction model adopts a nonlinear prediction method to predict output parameters, and the output parameters include exhaustion rate (E%), dilution rate (F%), total dilution rate (T%) and color strength (K / S) of the dyed cotton fabric. The prediction model is a least squares support vector regression (LSSVR) model based on input parameters, and the input parameters include dye dosage, fixation temperature, fixation time, dye bath pH value, material-liquid ratio and salt concentration. The prediction accuracy of the prediction model is calculated using root mean square error (RMSE), mean absolute error (MAE) and determination coefficient ( ) for assessment; （1）； （2）； （3）； （4）； （5）； Among them, the Indicates the actual output, the Represents the predicted output, Indicates the total number of samples collected. represents the prediction error, represents the approximation error, and Represents the target value and the actual value in the training data and the test data. and represent , and represent ; Establishing the prediction model includes preliminary experiments and machine learning. The preliminary experiments adopt the Taguchi method to design experiments. The preliminary experiments include parameter optimization and robust design.
2. The method for manufacturing a luminous blanket according to claim 1, characterized in that: The preliminary experiments included separate tests at different input parameters and different factor levels, using the rotation method to dye cotton fabrics, adding sodium chloride to heat the mixed sodium carbonate at 20°C to make a dye bath, and using soap solution to clean the unfixed dye, the soap solution was a non-ionic detergent with a concentration of , the material-liquid ratio was 1:10, the temperature was 95°C, and the duration was 15 min. After the soap solution treatment, the dyed cotton fabric was dried in a drying oven at 80°C for 30 min.
3. The method for manufacturing a self-luminous luminous blanket according to claim 2, characterized in that: The individual tests formed a data set, and after uploading the data set to MATLAB, it was split between the training set and the test set in a ratio of 80 / 20.
4. The method for manufacturing a luminous blanket according to claim 1, characterized in that: Using a spectrophotometer, the light absorbance of the dye bath solution before and after dyeing and the remaining soap solution was measured and recorded at the maximum absorption wavelength of 560nm. The exhaustion rate (E%), dilution rate (F%) and total dilution rate (T%) were determined using the following equations: （6）； （7）； （8）； The light absorbances of the soap solution before and after dyeing are expressed as , and express.
5. The method for manufacturing a self-luminous luminous blanket according to claim 4, characterized in that: A reflectance spectrometer was used to evaluate the color intensity, i.e., the K / S ratio, of cotton fabric samples randomly extracted from 20 different positions. The color intensity value was determined by measuring at the wavelength where the dye absorbs the lightest light. The average color intensity was taken. A scanning electron microscope was used to analyze the surface morphology of the undyed and dyed samples. The samples were trimmed to a size of no more than 1 square centimeter each and then fixed individually on a standard sample holder. An accelerating voltage of 10 kV was used and the working distance was between 15 and 17 mm. An X-ray diffractometer was used to measure the X-ray diffraction patterns of the undyed and dyed samples.
6. The method for manufacturing a luminous blanket according to claim 5, characterized in that: The parameter optimization includes conducting experiments according to the Taguchi design method in Minitab software to ensure the repeatability of the dyeing process, and finding the parameter optimization of each response, namely E%, F%, T% and K / S, according to the signal-to-noise ratio analysis.
7. The method for manufacturing a luminous blanket according to claim 6, characterized in that: The LSSVR model also includes tuning parameters, which are represented by the symbol , and Indicates that the values are set to 32, 0.0625, and .
8. The method for manufacturing a luminous blanket according to claim 7, characterized in that: The input parameters are adjusted in the model to obtain the output parameters, wherein the input parameters are the values after the luminous powder and the printing dye are mixed.
9. The method for manufacturing a self-luminous luminous blanket according to claim 8, characterized in that: The final input parameters and output parameters are determined according to the LSSVR model to guide the dyeing process. After dyeing, a luminous fabric layer is made. The luminous fabric layer is embedded with light-emitting elements to make a luminous blanket.