Textile production quality prediction method and system based on data identification
By pre-processing and performance detection of textile data, combined with temperature, humidity and light sequences, the problem of low accuracy in textile production quality prediction is solved, product consistency and stability are improved, and quantitative evaluation of environmental factors is achieved.
Patent Information
- Application Number
- CN202411556541.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-04
- Publication Date
- 2024-12-03
AI Technical Summary
The accuracy of textile production quality prediction in the prior art leads to problems with product consistency and stability, especially the impact of environmental factors such as temperature and humidity is difficult to accurately evaluate.
Textile data is collected through data acquisition equipment, pre-processing and performance testing, including temperature, humidity and light detection, combined with the detection results to predict textile production quality, and quantitative analysis is performed using temperature, humidity and light sequences to ensure the accuracy and consistency of the data. sex.
The accuracy of textile production quality prediction is improved, quantitative analysis of environmental factors and key production parameters is realized, and the accuracy of quality inspection and data analysis efficiency is enhanced.
Smart Images

Figure CN119067266A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of textile production quality prediction, and in particular to a textile production quality prediction method and system based on data recognition. Background Art
[0002] With the development of economic and technological textiles, labor costs are gradually increasing, and the transformation of the textile industry to automation and intelligence has become an inevitable trend. The quality of textiles is the lifeline of textile enterprises, which directly affects the brand image and market competitiveness of enterprises. Textile enterprises attach great importance to quality control in the production process and need to take various measures to ensure that product quality meets the standard requirements. Thereby improving the stability and reliability of product quality. The textile production quality prediction method and system based on data recognition can timely discover potential quality problems and take measures to improve them by collecting and analyzing data in the production process.
[0003] In the prior art, the production environment, equipment and product quality parameters are monitored in real time through detection equipment, characteristic points in the parameters are extracted, the main characteristics affecting product quality are found, and a quality prediction model is established. The model is used for quality prediction and real-time monitoring to improve the efficiency of textile production.
[0004] For example, the invention patent with announcement number CN115409642B announces a textile printing intelligent manufacturing production management system, which includes: by setting a textile placement status monitoring module, a printing dye usage status monitoring module, a finished printed textile quality monitoring module, a textile data storage library and a production management early warning terminal, the rationality of the placement status of the textile, the stability coefficient of the usage status of the printing dye in each working time period and the comprehensive printing quality coefficient of the finished textile are analyzed, and production management early warning is carried out for the manufacture of printed textiles based on this.
[0005] For example, the invention patent with announcement number: CN117115147B announces a textile detection method and system based on machine vision, including: based on the textile image, using a convolutional neural network to extract image features and generate a preliminary recognition image feature library; through convolutional neural networks, recurrent neural networks, multispectral image acquisition technology, computer vision technology and blockchain technology, a textile detection system is formed. First, the combination of convolutional neural networks and recurrent neural networks provides a high-accuracy basis for textile defect detection. Multispectral image acquisition technology enables hidden defects to be detected more accurately, and real-time quality control instructions are generated and applied, as well as intelligent defect analysis.
[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems: In the prior art, the environment has a significant impact on the production quality of textiles. Excessively high temperatures may cause certain fibers to soften or deform. In an environment with excessively high humidity, the fibers will absorb moisture and swell, affecting the strength and elasticity of the fibers. When the humidity is too low, the fibers will become brittle and easy to break. Changes in temperature and humidity can lead to quality differences between different batches of textiles, affecting the consistency and stability of the products. There is a problem of low accuracy in predicting the quality of textile production. Summary of the invention
[0007] The embodiments of the present application solve the problem of low accuracy of textile production quality prediction in the prior art by providing a textile production quality prediction method and system based on data recognition, thereby improving the accuracy of textile production quality prediction.
[0008] An embodiment of the present application provides a textile production quality prediction method based on data identification, comprising the following steps: S1, collecting initial textile data through a data acquisition device and preprocessing to obtain textile data, wherein the textile data includes textile material type data, production environment data and sensor data; S2, performing performance testing on the textile according to the textile data to obtain a quality test result, wherein the performance testing includes temperature testing, humidity testing and light testing, wherein the temperature testing is used to analyze the influence of temperature on the textile quality during the production process, the humidity testing is used to analyze the influence of humidity on the textile quality during the production process, and the light testing is used to analyze the influence of light on the textile quality during the production process; S3, performing textile production quality prediction according to the textile data and the quality test results to obtain a textile prediction result, wherein the textile production quality prediction is used to predict the quality performance of the textile.
[0009] Furthermore, the specific steps of the preprocessing are as follows: A1, performing data cleaning on the initial textile data to obtain first textile data, and the data cleaning includes missing value filling, outlier processing, duplicate value removal and noise data smoothing; A2, performing data transformation on the first textile data to obtain textile data, and the data transformation includes data format conversion and unit standardization; A3, integrating the textile data with the corresponding timestamp to obtain a textile time series.
[0010] Furthermore, the specific process of the performance test is as follows: performing temperature testing on the textile to obtain a temperature influence value, and obtaining a temperature sequence according to the obtained temperature influence value, wherein the temperature influence value includes a first tensile strength, a first fiber diameter, and a first pilling number; performing humidity testing on the textile to obtain a humidity influence value, and obtaining a humidity sequence according to the obtained humidity influence value, wherein the humidity influence value includes a second tensile strength, a second fiber diameter, and a second pilling number; performing light testing on the textile to obtain a light influence value, and obtaining a light sequence according to the obtained light influence value, wherein the light influence value includes a third tensile strength, a third fiber diameter, and a third pilling number.
[0011] Furthermore, the specific method for obtaining the temperature sequence is as follows: measuring the temperature influence value of the textile sample in the temperature simulation environment at a preset time point within a preset time period through a data detection device, integrating the temperature influence value with the corresponding timestamp to obtain a temperature sequence, the temperature simulation environment is used to simulate the temperature detection of textile samples in different temperature ranges under preset humidity and preset light intensity, and the temperature sequence includes a first tensile strength sequence, a first fiber diameter sequence and a first pilling number sequence.
[0012] Furthermore, a specific method for obtaining the humidity sequence is as follows: measuring the humidity impact value of a textile sample in a humidity simulation environment through a data detection device at a preset time point within a preset time period, and integrating the humidity impact value with the corresponding timestamp to obtain a humidity sequence, wherein the humidity simulation environment is used to simulate the humidity detection of textile samples within different humidity ranges under a preset temperature and a preset light intensity, and the humidity sequence includes a second tensile strength sequence, a second fiber diameter sequence, and a second pilling number sequence of the textile.
[0013] Furthermore, the specific method for obtaining the illumination sequence is as follows: measuring the illumination impact value of the textile sample in the illumination simulation environment at a preset time point within a preset time period through a data detection device, integrating the illumination impact value with the corresponding timestamp to obtain an illumination sequence, the illumination simulation environment is used to simulate the illumination detection of textile samples within different light intensity ranges under a preset temperature and a preset humidity, and the illumination sequence includes a third tensile strength sequence, a third fiber diameter sequence and a third pilling number sequence of the textile.
[0014] Furthermore, the specific method for obtaining the quality inspection result is as follows: the tensile strength coefficient, the fiber diameter coefficient and the pilling coefficient are obtained according to the temperature sequence, the humidity sequence and the light sequence, and the quality inspection result is obtained by combining the textile data and the detection weight obtained from the detection weight sequence in the preset database, wherein the detection weight sequence includes a tensile strength weight sequence, a diameter weight sequence and a pilling weight sequence, wherein the tensile strength weight sequence is an ordered set of tensile strength weights of different types of textile materials, the diameter weight sequence is an ordered set of diameter weights of different types of textile materials, and the pilling weight sequence is an ordered set of pilling weights of different types of textile materials; the quality inspection result is calculated by the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the quality test result of the i-th performance test of the p-th textile sample, represents the tensile strength weight of the pth textile specimen, represents the diameter weight of the pth textile sample, represents the pilling weight of the pth textile sample, represents the tensile strength coefficient of the i-th property test of the p-th textile sample, represents the fiber diameter coefficient of the i-th property test of the p-th textile sample, It represents the pilling coefficient of the i-th performance test of the p-th textile sample.
[0015] Furthermore, the specific steps for setting the detection weight are as follows: B1, arranging the influence of tensile strength on the types of textile materials in ascending order, and obtaining the tensile strength weight of the textile material through the serial number of the textile material type and the total weight of tensile strength; B2, arranging the influence of fiber diameter on textile materials in ascending order, and obtaining the fiber diameter weight of the textile material through the serial number of the textile material type and the total weight of fiber diameter; B3, arranging the influence of pilling quantity on textile materials in ascending order, and obtaining the pilling weight of the textile material through the serial number of the textile material type and the total weight of pilling.
[0016] Furthermore, the specific method of obtaining the textile prediction result is as follows: obtaining a reference weight from a preset database, and obtaining the textile prediction result in combination with the quality inspection result, wherein the reference weight includes a temperature weight, a humidity weight, and a light weight; and the textile prediction result is calculated by the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the weight of the i-th performance test, represents the quality test result of the i-th performance test of the p-th textile sample, represents the textile prediction result of the p-th textile sample.
[0017] An embodiment of the present application provides a textile production quality prediction system based on data identification, comprising: a data acquisition module, a performance detection module and a quality prediction module; wherein the data acquisition module is used to collect initial textile data through a data acquisition device and perform preprocessing to obtain textile data, wherein the textile data includes textile material type data, production environment data and sensor data; the performance detection module is used to perform performance detection on the textile according to the textile data to obtain a quality detection result, wherein the performance detection includes temperature detection, humidity detection and light detection, wherein the temperature detection is used to analyze the influence of temperature on the quality of the textile during the production process, the humidity detection is used to analyze the influence of humidity on the quality of the textile during the production process, and the light detection is used to analyze the influence of light on the quality of the textile during the production process; the quality prediction module is used to perform textile production quality prediction according to the textile data and the quality detection result to obtain a textile prediction result, wherein the textile production quality prediction is used to predict the quality performance of the textile.
[0018] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The performance of textiles is tested through textile data to obtain quality test results, and the textile production quality is predicted based on the textile data and quality test results to obtain textile prediction results, thereby realizing the quantitative analysis of environmental factors and key production parameters, and then realizing the improvement of the accuracy of textile production quality prediction, and effectively solving the problem of low accuracy of textile production quality prediction in the prior art.
[0019] 2. The tensile strength, fiber diameter and pilling quantity of textiles in temperature simulation environment, humidity simulation environment and light simulation environment are measured by data detection equipment respectively, and they are integrated with the corresponding timestamps to obtain temperature sequence, humidity sequence and light sequence, thereby realizing a comprehensive quantitative evaluation of the effects of temperature, humidity and light environmental factors on textile performance, and thus improving the accuracy of textile quality inspection.
[0020] 3. The initial textile data is cleaned to obtain the first textile data, and then the first textile data is transformed to obtain textile data, and finally the textile data is integrated with the corresponding timestamp to obtain the textile time series, thereby ensuring the accuracy, completeness and consistency of the data, improving the comparability between data, and thus achieving the improvement of data quality as well as the efficiency and accuracy of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 A flow chart of a textile production quality prediction method based on data identification provided in an embodiment of the present application; Figure 2 A schematic diagram of changes in textile prediction results provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a textile production quality prediction system based on data recognition provided in an embodiment of the present application. DETAILED DESCRIPTION
[0022] The embodiments of the present application solve the problem of low accuracy of textile production quality prediction in the prior art by providing a textile production quality prediction method and system based on data recognition. Initial textile data is collected by a data acquisition device and preprocessed to obtain textile data. Then, performance testing of the textile is performed based on the textile data to obtain quality testing results. Finally, textile production quality prediction is performed based on the textile data and the quality testing results to obtain textile prediction results, thereby improving the accuracy of textile production quality prediction.
[0023] The technical solution in the embodiment of the present application is to solve the problem of low accuracy of the above-mentioned textile production quality prediction, and the overall idea is as follows: The obtained textile data is used to perform performance tests on textiles to obtain quality test results, and the textile production quality is predicted in combination with the textile data to obtain textile prediction results, thereby achieving the effect of improving the accuracy of textile production quality prediction.
[0024] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0025] like Figure 1As shown, it is a flow chart of a textile production quality prediction method based on data identification provided in an embodiment of the present application, and the method includes the following steps: S1, collecting initial textile data through a data acquisition device and preprocessing to obtain textile data, the textile data including textile material type data, production environment data and sensor data, the production environment data including temperature, humidity and light, and the sensor data including tension sensor data and pressure sensor data; S2, performing performance testing on the textile according to the textile data to obtain quality test results, the performance testing including temperature testing, humidity testing and light testing, the temperature testing is used to analyze the influence of temperature on textile quality during the production process, the humidity testing is used to analyze the influence of humidity on textile quality during the production process, and the light testing is used to analyze the influence of light on textile quality during the production process; S3, performing textile production quality prediction according to the textile data and the quality test results to obtain a textile prediction result, and the textile production quality prediction is used to predict the quality performance of the textile.
[0026] In this embodiment, the data acquisition equipment includes a tension sensor, a pressure sensor, a temperature sensor, a humidity sensor and a light sensor; the type of textile material represents the type information of the raw materials constituting the textile, including cotton, linen, wool, etc.; the production environment data represents the environmental conditions of the textile during the production process, including temperature, humidity and light; the sensor data can reflect the various physical states of the textile during the production process in real time; the performance test is used to analyze the impact of various environmental factors on the quality of the textile during the production process; the quality test result represents the evaluation result of the total quality of the textile obtained through the performance test, reflecting the performance of the textile under different environmental conditions; the textile production quality prediction represents the prediction of the quality performance of the textile in the future; and the accuracy of the textile production quality prediction is improved.
[0027] Furthermore, the specific steps of preprocessing are as follows: A1, data cleaning the initial textile data to obtain the first textile data, data cleaning includes missing value filling, outlier processing, duplicate value removal and noise data smoothing, missing value filling means identifying and processing the missing values in the initial textile data set, outlier processing means identifying and processing the outliers in the initial textile data set, duplicate value removal means identifying and processing the repeated data in the initial textile data set, and noise data smoothing means processing the errors or variations in the data; A2, data transformation is performed on the first textile data to obtain textile data, data transformation includes data format conversion and unit standardization, data format conversion means identifying and processing the data with inconsistent formats in the first textile data set, and unit standardization means identifying the first textile data, judging whether there are inconsistent units in the first textile data set, and if so, unifying the units of all numerical values into the same unit; A3, integrating the textile data with the corresponding timestamps to obtain the textile time series.
[0028] In this embodiment, missing value filling means filling using a suitable method, for example, filling using the mean or median; outlier means data that deviates significantly from the overall trend of the initial textile data. Generally, if a value exceeds the mean plus or minus 2 times the standard deviation, it is considered an outlier; duplicate value deletion means removing duplicate data records and retaining only one record; data format conversion means converting all data into the same format, for example, unifying all dates and converting all temperature units to degrees Celsius; unit standardization means unifying all values into the same unit for subsequent processing, for example, unifying all diameters into millimeters (mm); integrating timestamps means making one-to-one correspondence between textile data and corresponding timestamps to ensure that each piece of data has a corresponding timestamp to form a complete textile time series; the quality of data acquisition is improved.
[0029] Furthermore, the specific process of performance testing is as follows: performing temperature testing on the textile to obtain temperature influence values, obtaining a temperature sequence based on the obtained temperature influence values, the temperature influence values include a first tensile strength, a first fiber diameter and a first pilling number, and the temperature sequence represents a set of temperature influence values of the textile under temperature testing; performing humidity testing on the textile to obtain humidity influence values, obtaining a humidity sequence based on the obtained humidity influence values, the humidity influence values include a second tensile strength, a second fiber diameter and a second pilling number, and the humidity sequence represents a set of humidity influence values of the textile under humidity testing; performing light testing on the textile to obtain light influence values, obtaining a light sequence based on the obtained light influence values, the light influence values include a third tensile strength, a third fiber diameter and a third pilling number, and the light sequence represents a set of light influence values of the textile under light testing.
[0030] In this embodiment, temperature detection is used to analyze the impact of different temperature conditions on the quality of textiles during the production process; humidity detection is used to evaluate the impact of humidity changes on fibers or fabrics; light detection is used to analyze the impact of different light intensities on the quality of textiles during the production process; the temperature impact value indicates the performance change of textiles under different temperature conditions; tensile strength indicates the ability of textiles to resist breaking during stretching; fiber diameter reflects the thickness of the fiber, affecting the air permeability and feel of the textile; pilling number is used to evaluate the surface wear resistance and anti-pilling performance of textiles. The temperature sequence is used to analyze the law of textile performance changes with temperature; the humidity impact value indicates the performance change of textiles under different humidity conditions; the humidity sequence is used to analyze the law of textile performance changes with humidity; the light sequence is used to analyze the law of textile performance changes with light intensity; the light impact value indicates the performance change of textiles under different light conditions; and a comprehensive evaluation of the comprehensive performance of textiles is achieved.
[0031] Furthermore, a specific method for obtaining a temperature sequence is as follows: measuring a temperature influence value of a textile sample in a temperature simulation environment at a preset time point within a preset time period through a data detection device, integrating the temperature influence value with the corresponding timestamp to obtain a temperature sequence, the data detection device includes a micrometer, an image acquisition device, and a tensile testing machine, the temperature simulation environment is used to simulate the temperature detection of textile samples within different temperature ranges under a preset humidity and a preset light intensity, and the temperature sequence includes a first tensile strength sequence, a first fiber diameter sequence, and a first pilling quantity sequence.
[0032] In this embodiment, the preset time period represents a specific time range, during which data collection and detection are performed; the preset time point represents a specific time node within the preset time period, at which temperature detection is performed; the preset time period and the preset time point are obtained from a database; the preset time period is generally set to 9 am to 5 pm; the preset time point is generally set to 1 minute; the preset humidity represents a specific humidity set in a temperature simulation environment, which is used to simulate the humidity conditions of textiles in production; the preset humidity is obtained from a preset database, and the preset humidity is generally 50%; the preset light intensity represents a specific light intensity set in a temperature simulation environment, which is used to simulate the light intensity under sunlight conditions; the preset light intensity is obtained from a database, usually set according to the ISO 105-B02 standard; the timestamp represents the information recording the measurement time point, which is used to be integrated with the temperature influence value; a comprehensive evaluation of the impact of temperature on various properties of textiles is achieved.
[0033] Micrometer is a precision instrument used to measure fiber diameter; fiber samples are obtained from textile specimens and fiber cross sections are prepared and fiber diameters are measured using a micrometer.
[0034] Tensile testing machine is an instrument used to test the tensile strength of textiles; the textile sample is clamped in the clamp of the tensile testing machine, the tensile force is gradually increased, and the maximum tensile force when the textile sample breaks is recorded. The tensile strength is the ratio of the maximum tensile force to the original cross-sectional area of the textile sample.
[0035] The image acquisition device is used to take an image of the side of the textile sample; first, the image acquired by the image acquisition device is converted into a grayscale image, and the difference between the shadow area and the non-shadow area in the image is found through shadow elimination or shadow recognition technology, and the shadow difference image is enhanced to highlight the characteristics of the pilling area, and the pilling part is detected from the enhanced image using edge detection.
[0036] Furthermore, a specific method for obtaining a humidity sequence is as follows: measuring the humidity impact value of a textile sample in a humidity simulation environment at a preset time point within a preset time period through a data detection device, integrating the humidity impact value with the corresponding timestamp to obtain a humidity sequence, and the humidity simulation environment is used to simulate the humidity detection of textile samples within different humidity ranges under a preset temperature and a preset light intensity. The humidity sequence includes a second tensile strength sequence, a second fiber diameter sequence, and a second pilling number sequence of the textile, and the data detection device includes a micrometer, an image acquisition device, and a tensile testing machine.
[0037] In this embodiment, the preset time period represents a specific time range, during which data collection and detection are performed; the preset time point represents a specific time node within the preset time period, at which humidity detection is performed; the preset time period and the preset time point are obtained from a database; the preset time period is generally set to 9 am to 5 pm; the preset time point is generally set to 1 minute; the preset temperature represents a specific temperature set in a humidity simulation environment, which is used to simulate the temperature conditions of textiles in production, and the preset temperature is generally 25°C; the preset light intensity represents a specific light intensity set in a humidity simulation environment, which is used to simulate the light intensity under sunlight conditions; the preset light intensity is obtained from a database, usually set according to the ISO 105-B02 standard; the timestamp represents the information recording the measurement time point, which is used to be integrated with the humidity impact value; a comprehensive evaluation of the impact of humidity on various properties of textiles is achieved.
[0038] Furthermore, a specific method for obtaining a lighting sequence is as follows: measuring the lighting impact value of a textile sample in a lighting simulation environment at a preset time point within a preset time period through a data detection device, integrating the lighting impact value with the corresponding timestamp to obtain a lighting sequence, the lighting simulation environment is used to simulate the lighting detection of textile samples within different light intensity ranges at a preset temperature and a preset humidity, the lighting sequence includes a third tensile strength sequence, a third fiber diameter sequence and a third pilling number sequence of the textile, and the data detection device includes a micrometer, an image acquisition device and a tensile testing machine.
[0039] In this embodiment, the preset time point is used to perform light detection at a specific time node; the preset temperature represents a specific temperature set in the light simulation environment, which is used to simulate the temperature conditions of textiles in production; the preset humidity represents a specific humidity set in the light simulation environment, which is used to simulate the humidity conditions of textiles in production; the timestamp is used to integrate with the light impact numerical value; the third tensile strength sequence represents a collection of tensile strengths of textiles in light detection; the third fiber diameter sequence represents a collection of fiber diameters of textiles in light detection; the third pilling number sequence represents a collection of pilling numbers of textiles in light detection; and a comprehensive evaluation of the impact of light on various properties of textiles is achieved.
[0040] Furthermore, the specific method for obtaining the quality inspection result is as follows: the tensile strength coefficient, the fiber diameter coefficient and the pilling coefficient are obtained according to the temperature sequence, the humidity sequence and the light sequence, and the quality inspection result is obtained by combining the textile data and the detection weights obtained from the detection weight sequence in the preset database, the detection weight sequence includes a tensile strength weight sequence, a diameter weight sequence and a pilling weight sequence, the tensile strength weight sequence is an ordered set of tensile strength weights of different types of textile materials, the tensile strength weight is used to describe the influence of tensile strength on the quality inspection results of textiles of different textile materials, the diameter weight sequence is an ordered set of diameter weights of different types of textile materials, the diameter weight is used to describe the influence of diameter on the quality inspection results of textiles of different textile materials, the pilling weight sequence is an ordered set of pilling weights of different types of textile materials, the pilling weight is used to describe the influence of the number of pilling on the quality inspection results of textiles of different textile materials, the tensile strength coefficient represents the ratio of the tensile strength after the performance test to the initial tensile strength, the fiber diameter coefficient represents the ratio of the fiber diameter after the performance test to the initial fiber diameter, and the pilling coefficient represents the ratio of the number of pilling after the performance test to the initial number of pilling; the quality inspection result is calculated by the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the quality test result of the i-th performance test of the p-th textile sample, represents the tensile strength weight of the pth textile specimen, represents the diameter weight of the pth textile sample, represents the pilling weight of the pth textile sample, represents the tensile strength coefficient of the i-th property test of the p-th textile sample, represents the fiber diameter coefficient of the i-th property test of the p-th textile sample, It represents the pilling coefficient of the i-th performance test of the p-th textile sample.
[0041] In this embodiment, the tensile strength coefficient is used to measure the change in the tensile strength of the textile after performance testing relative to the initial tensile strength. The higher the tensile strength coefficient, the better the tensile performance of the textile is maintained after testing, which has a positive impact on the overall quality of the textile. The change in the fiber diameter coefficient can reflect the change in the fiber of the textile during processing or use. Pilling is the small spherical protrusions formed by fiber wear and entanglement during the use of textiles, and is an important indicator for evaluating the appearance quality of textiles. The lower the pilling coefficient, the better the anti-pilling performance of the textile. The detection weight sequence is used to describe the degree of influence of different influence values on the quality test results of different types of textile materials. The accuracy of quality assessment is achieved.
[0042] The quality test results represent a comprehensive evaluation of the quality of textiles based on the tensile strength, fiber diameter, and pilling quantity of each performance test; taking the performance test number as 1, the tensile strength weight as 0.5, the diameter weight as 0.3, and the pilling weight as 0.2 as an example, the statistical table of changes in the quality test results is shown in Table 1: Table 1 Statistics of changes in quality test results
[0043] It can be seen from Table 1 that with the increase of tensile strength coefficient, the quality inspection result decreases, and with the increase of fiber diameter coefficient and pilling coefficient, the quality inspection result increases; the quality inspection result is inversely proportional to the tensile strength coefficient, and the quality inspection result is proportional to the fiber diameter coefficient and pilling coefficient; the specific situation is compared based on the calculation results of the corresponding data, and the corresponding operating coefficient relationship cannot be estimated based solely on the size of a single independent variable.
[0044] The algorithm of this embodiment combines the "tensile strength coefficient of the ith performance test of the pth textile sample", "the fiber diameter coefficient of the ith performance test of the pth textile sample" and "the pilling coefficient of the ith performance test of the pth textile sample" to obtain the "quality test result of the ith performance test of the pth textile sample" through comprehensive analysis. In addition, each parameter is related to each other and does not exist independently. It is a whole. It is not possible to obtain the "quality test result of the ith performance test of the pth textile sample" by simple combination and addition. For example, the increase of the "tensile strength coefficient of the ith performance test of the pth textile sample" in this formula will lead to the increase of the "quality test result of the ith performance test of the pth textile sample". A decrease in the "quality test result of the ith performance test of the fabric sample" does not necessarily mean that the "quality test result of the ith performance test of the pth textile sample" is larger. At this time, it is also necessary to consider the "fiber diameter coefficient of the ith performance test of the pth textile sample" and the "pilling coefficient of the ith performance test of the pth textile sample". It can also determine the quality of the textile at the next moment. The correlation and mutual influence of various factors are considered in a quantitative way. The algorithm itself is obtained through comprehensive analysis, which can accurately and quantitatively analyze the quality of textiles, indicating the effect of accurately evaluating the quality test results of textiles. Therefore, precise analysis is carried out by establishing mathematical forms.
[0045] Furthermore, the specific steps for setting the detection weight are as follows: B1, arrange the influence of tensile strength on the types of textile materials in ascending order, and obtain the tensile strength weight of the textile material through the serial number of the textile material type and the total weight of tensile strength; B2, arrange the influence of fiber diameter on textile materials in ascending order, and obtain the fiber diameter weight of the textile material through the serial number of the textile material type and the total weight of fiber diameter; B3, arrange the influence of pilling number on textile materials in ascending order, and obtain the pilling weight of the textile material through the serial number of the textile material type and the total weight of pilling.
[0046] In this embodiment, tensile strength indicates the maximum tensile force that a textile can withstand when stretched to break; tensile strength data of textile material types are collected, and these data are sorted from smallest to largest according to the degree of their influence on the tensile strength of the textile material, and the influence degree after sorting is associated with the serial number of the textile material type; fiber diameter indicates the thickness of the fiber; fiber diameter data of textile material types are collected, and these data are sorted from smallest to largest according to the degree of their influence on the fiber diameter of the textile material, and the influence degree after sorting is associated with the serial number of the textile material type; pilling indicates small balls formed due to friction of fibers during use; pilling number data of textile material types are collected, and these data are sorted from smallest to largest according to the degree of their influence on the pilling number of the textile material, and the influence degree after sorting is associated with the serial number of the textile material type.
[0047] Specifically, the tensile strength weight indicates the influence of tensile strength on the quality of textile materials; the weight is searched from the sequence according to the corresponding material type; the tensile strength weight is calculated by the following formula: ; In the formula, m is the serial number of the textile material type, , M is the total number of textile material types, represents the tensile strength weight of the mth textile material, represents the total weight of tensile strength; Specifically, the total tensile strength weight is obtained from a preset database. In a specific embodiment, a fitting curve is obtained by fitting the relationship between the tensile strength coefficient and the quality inspection result by the minimum tensile strength and the maximum tensile strength in the tensile strength sequence, and the real-time tensile strength coefficient is obtained according to the fitting curve to determine the corresponding total tensile strength weight.
[0048] Specifically, the fiber diameter weight indicates the influence of the fiber diameter on the quality of the textile material; the weight is searched from the sequence according to the corresponding material type; the fiber diameter weight is calculated by the following formula: ; In the formula, represents the fiber diameter weight of the mth textile material, represents the total weight of fiber diameter; Specifically, the total fiber diameter weight is obtained from a preset database. In a specific embodiment, a fitting curve is obtained by fitting the relationship between the minimum fiber diameter and the maximum fiber diameter in the fiber diameter sequence and the quality detection result, and the real-time fiber diameter coefficient is obtained according to the fitting curve to determine the corresponding total fiber diameter weight.
[0049] Specifically, the pilling weight indicates the degree of influence of fiber pilling on the quality of textile materials; the weight is searched from the sequence according to the corresponding material type; the pilling weight is calculated by the following formula: ; In the formula, represents the pilling weight of the mth textile material, It indicates the total pilling weight; Specifically, the total pilling weight is obtained from a preset database. In a specific embodiment, the relationship between the pilling coefficient and the quality detection result is fitted by fitting the minimum pilling number and the maximum pilling number in the pilling sequence to obtain a fitting curve, and the real-time pilling coefficient is obtained according to the fitting curve to determine the corresponding total pilling weight.
[0050] Furthermore, the specific method for obtaining the textile prediction result is as follows: obtain the reference weight from the preset database, and obtain the textile prediction result in combination with the quality test result, the reference weight includes temperature weight, humidity weight and light weight, the temperature weight indicates the influence of temperature test on the textile quality, the humidity weight indicates the influence of humidity test on the textile quality, and the light weight indicates the influence of light test on the textile quality; the textile prediction result is calculated by the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the weight of the i-th performance test, represents the quality test result of the i-th performance test of the p-th textile sample, represents the textile prediction result of the p-th textile sample.
[0051] In this embodiment, if Figure 2 As shown, it is a schematic diagram of the change of the textile prediction result provided in the embodiment of the present application. It can be seen from the figure that the textile prediction result is proportional to the quality test results of each performance test. The textile prediction result is a comprehensive indicator used to predict the quality performance of the textile in the future; when the quality test results of each performance test increase, the textile prediction result increases.
[0052] The algorithm of this embodiment combines the "quality test results of the temperature test of the p-th textile sample", "the quality test results of the humidity test of the p-th textile sample" and "the quality test results of the illumination test of the p-th textile sample", and comprehensively analyzes to obtain the "textile prediction result", and the various parameters are related and do not exist independently. They are a whole, and the "textile prediction result" cannot be obtained by simple combination and addition. For example, the increase of the "quality test results of the temperature test of the p-th textile sample" in this formula will lead to an increase in the "textile prediction result", which does not necessarily mean that the "textile prediction result" is large. At this time, it is also necessary to consider that the "quality test results of the humidity test of the p-th textile sample" and the "quality test results of the illumination test of the p-th textile sample" can also determine the prediction result of the textile at the next moment. The correlation and mutual influence between various factors are considered in a quantitative way. The algorithm itself is obtained by comprehensive analysis, which can accurately and quantitatively predict the quality of future textiles, indicating the effect of accurately evaluating the prediction results of textiles. Therefore, accurate analysis is performed by establishing mathematical forms.
[0053] Specifically, the temperature weight is obtained from a preset database. In a specific embodiment, a fitting curve is obtained by fitting the relationship between the temperature impact value of the textile in the temperature detection and the textile data to the temperature quality detection result and the textile prediction result, and the corresponding temperature weight is determined according to the real-time temperature quality detection result obtained from the fitting curve.
[0054] Specifically, the humidity weight is obtained from a preset database. In a specific embodiment, a fitting curve is obtained by fitting the relationship between the humidity impact value of the textile in the humidity detection and the humidity quality detection result and the textile prediction result through the textile data, and the corresponding humidity weight is determined according to the real-time humidity quality detection result obtained from the fitting curve.
[0055] Specifically, the illumination weight is obtained from a preset database. In a specific embodiment, a fitting curve is obtained by fitting the relationship between the illumination quality detection result and the textile prediction result through the illumination impact value of the textile in the illumination detection and the textile data, and the corresponding illumination weight is determined based on the real-time illumination quality detection result obtained from the fitting curve.
[0056] An embodiment of the present application provides a textile production quality prediction system based on data identification, including: a data acquisition module, a performance detection module and a quality prediction module; wherein the data acquisition module is used to collect initial textile data through a data acquisition device and pre-process to obtain textile data, the textile data including textile material type data, production environment data and sensor data, the production environment data including temperature, humidity and light, and the sensor data including tension sensor data and pressure sensor data; the performance detection module is used to perform performance detection on the textile according to the textile data to obtain quality detection results, the performance detection includes temperature detection, humidity detection and light detection, the temperature detection is used to analyze the influence of temperature on textile quality during the production process, the humidity detection is used to analyze the influence of humidity on textile quality during the production process, and the light detection is used to analyze the influence of light on textile quality during the production process; the quality prediction module is used to perform textile production quality prediction according to the textile data and the quality detection results to obtain textile prediction results, and the textile production quality prediction is used to predict the quality performance of the textile.
[0057] In this embodiment, the purpose of the system is to predict the quality performance of textiles through data collection, performance testing and quality prediction; the data collection module ensures the accuracy and completeness of the collected verses, providing a basis for subsequent performance testing and quality prediction; the performance testing module conducts a comprehensive assessment of the quality of textiles by analyzing various parameters in the production environment; the quality prediction module predicts the quality performance of textiles in real time, discovers potential problems in advance, optimizes the production process, and improves product quality.
[0058] In summary, the embodiments of the present application perform performance testing on textiles through textile data to obtain quality test results, and perform textile production quality prediction based on the textile data and the quality test results to obtain textile prediction results, thereby achieving quantitative analysis of environmental factors and key production parameters, and further achieving improved accuracy in textile production quality prediction, effectively solving the problem of low accuracy in textile production quality prediction in the prior art.
[0059] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0060] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0061] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0062] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0063] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0064] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A textile production quality prediction method based on data recognition, characterized in that: The following steps are involved: S1, collecting initial textile data through a data collection device and preprocessing to obtain textile data, wherein the textile data includes textile material type data, production environment data and sensor data; S2, performing performance testing on the textile according to the textile data to obtain a quality test result, wherein the performance testing includes temperature testing, humidity testing and light testing, wherein the temperature testing is used to analyze the influence of temperature on the quality of the textile during the production process, the humidity testing is used to analyze the influence of humidity on the quality of the textile during the production process, and the light testing is used to analyze the influence of light on the quality of the textile during the production process; S3, performing textile production quality prediction according to the textile data and the quality test results to obtain a textile prediction result, wherein the textile production quality prediction is used to predict the quality performance of the textile.
2. The textile production quality prediction method based on data recognition as claimed in claim 1, characterized in that: The specific steps of the pretreatment are as follows: A1, performing data cleaning on the initial textile data to obtain first textile data, wherein the data cleaning includes missing value filling, outlier processing, duplicate value removal and noise data smoothing; A2, performing data transformation on the first textile data to obtain textile data, wherein the data transformation includes data format conversion and unit standardization; A3, integrate the textile data with the corresponding timestamps to obtain the textile time series.
3. The textile production quality prediction method based on data recognition as claimed in claim 1, characterized in that: The specific process of the performance test is as follows: Performing temperature detection on the textile to obtain temperature influence values, and obtaining a temperature sequence according to the obtained temperature influence values, wherein the temperature influence values include a first tensile strength, a first fiber diameter, and a first pilling number; Performing humidity detection on the textile to obtain humidity impact values, and obtaining a humidity sequence according to the obtained humidity impact values, wherein the humidity impact values include a second tensile strength, a second fiber diameter, and a second pilling number; The textile is subjected to light detection to obtain light impact values, and a light sequence is obtained according to the obtained light impact values, wherein the light impact values include a third tensile strength, a third fiber diameter, and a third pilling quantity.
4. The textile production quality prediction method based on data recognition as claimed in claim 3, characterized in that: The specific method for obtaining the temperature sequence is as follows: The temperature influence value of the textile sample in the temperature simulation environment is measured by a data detection device at a preset time point within a preset time period, and the temperature influence value is integrated with the corresponding time stamp to obtain a temperature sequence. The temperature simulation environment is used to simulate the temperature detection of the textile sample in different temperature ranges under a preset humidity and a preset light intensity. The temperature sequence includes a first tensile strength sequence, a first fiber diameter sequence and a first pilling number sequence.
5. The textile production quality prediction method based on data recognition as claimed in claim 3, characterized in that: The specific method for obtaining the humidity sequence is as follows: The humidity impact value of the textile sample in the humidity simulation environment is measured by a data detection device at a preset time point within a preset time period, and the humidity impact value is integrated with the corresponding timestamp to obtain a humidity sequence. The humidity simulation environment is used to simulate the humidity detection of textile samples in different humidity ranges under a preset temperature and a preset light intensity. The humidity sequence includes a second tensile strength sequence, a second fiber diameter sequence, and a second pilling number sequence of the textile.
6. The textile production quality prediction method based on data recognition as claimed in claim 3, characterized in that: The specific method for obtaining the illumination sequence is as follows: At a preset time point within a preset time period, a light impact value of a textile sample in a light simulation environment is measured by a data detection device, and the light impact value is integrated with the corresponding timestamp to obtain a light sequence. The light simulation environment is used to simulate the light detection of textile samples within different light intensity ranges under a preset temperature and a preset humidity. The light sequence includes a third tensile strength sequence, a third fiber diameter sequence, and a third pilling number sequence of the textile.
7. The textile production quality prediction method based on data recognition as claimed in claim 3, characterized in that: The specific method for obtaining the quality test results is as follows: The tensile strength coefficient, the fiber diameter coefficient and the pilling coefficient are obtained according to the temperature sequence, the humidity sequence and the light sequence, and the quality inspection result is obtained by combining the textile data and the detection weights obtained from the detection weight sequence in the preset database, wherein the detection weight sequence includes a tensile strength weight sequence, a diameter weight sequence and a pilling weight sequence, wherein the tensile strength weight sequence is an ordered set of tensile strength weights of different types of textile materials, the diameter weight sequence is an ordered set of diameter weights of different types of textile materials, and the pilling weight sequence is an ordered set of pilling weights of different types of textile materials; The quality test results are calculated using the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the quality test result of the ith performance test of the pth textile sample, represents the tensile strength weight of the pth textile specimen, represents the diameter weight of the pth textile sample, represents the pilling weight of the pth textile sample, represents the tensile strength coefficient of the i-th property test of the p-th textile sample, represents the fiber diameter coefficient of the i-th property test of the p-th textile sample, It represents the pilling coefficient of the i-th performance test of the p-th textile sample.
8. The textile production quality prediction method based on data recognition as claimed in claim 7, characterized in that: The specific steps for setting the detection weight are as follows: B1, the influence of tensile strength on the types of textile materials is arranged in ascending order, and the tensile strength weight of the textile material is obtained by the serial number of the textile material type and the total tensile strength weight; B2, arranging the influence of fiber diameter on textile materials in ascending order, and obtaining the fiber diameter weight of textile materials by the serial number of textile material types and the total weight of fiber diameter; B3, the influence of the pilling quantity on the textile materials is arranged in ascending order, and the pilling weight of the textile material is obtained by the serial number of the textile material type and the total pilling weight.
9. The textile production quality prediction method based on data recognition as claimed in claim 3, characterized in that: The specific method of obtaining the textile prediction result is as follows: Obtaining reference weights from a preset database and combining them with the quality inspection results to obtain textile prediction results, wherein the reference weights include temperature weights, humidity weights, and light weights; The textile prediction results are calculated using the following formula: ; Where p is the number of the textile sample, , N is the total number of textile samples, i is the number of performance tests, , Indicates temperature detection, Indicates humidity detection, Indicates light detection. represents the weight of the i-th performance test, represents the quality test result of the i-th performance test of the p-th textile sample, represents the textile prediction result of the p-th textile sample.
10. A textile production quality prediction system based on data recognition, characterized in that: The textile production quality prediction system based on data recognition includes a data acquisition module, a performance detection module and a quality prediction module; The data acquisition module is used to collect initial textile data through a data acquisition device and perform preprocessing to obtain textile data, wherein the textile data includes textile material type data, production environment data and sensor data; The performance detection module is used to perform performance detection on the textile according to the textile data to obtain a quality detection result, wherein the performance detection includes temperature detection, humidity detection and light detection, wherein the temperature detection is used to analyze the influence of temperature on the quality of the textile during the production process, the humidity detection is used to analyze the influence of humidity on the quality of the textile during the production process, and the light detection is used to analyze the influence of light on the quality of the textile during the production process; The quality prediction module is used to predict the production quality of textiles according to the textile data and the quality test results to obtain a textile prediction result, and the textile production quality prediction is used to predict the quality performance of the textiles.
Citation Information
Patent Citations
Anti-pilling performance detection method for spinning production and processing
CN116342583A
Textile production quality prediction method based on data identification
CN117273554A
Method and device for testing dry and wet wear resistance of textile product
CN118424936A
Intelligent management system for textile handicrafts
CN118864464A
Cited By
Intelligent manufacturing quality management method, system and equipment based on edge cloud collaboration
CN121091795A
Textile dyeing production quality prediction method based on data identification
CN122453239A
Method for predicting production quality of textile dyeing based on data recognition
CN122453239B
