A method and system for precise yarn control in tufted carpet production process

Abstract

The present invention discloses a method and system for precise control of yarn in a tufted carpet production process, the method comprising the following steps: S1, obtaining design parameters of the tufted carpet, determining a preset textile path and preset textile parameters according to the design parameters, and determining the type and length of the yarn in combination with image recognition technology; S2, performing textile processing according to the preset textile path and preset textile parameters, and using an ARIMA model to perform real-time prediction of the textile parameters in the textile process to obtain predicted textile parameters; S3, comparing the predicted textile parameters with the preset textile parameters, and analyzing the textile state at the next moment according to the comparison results; S4, correcting the abnormal textile parameters in the textile process in real time according to the analysis results, and performing textile processing according to the corrected textile parameters. The present invention greatly improves the textile efficiency of the tufted carpet by real-time monitoring of the preset textile parameters during the textile process of the tufted carpet.

Description

Technical Field

[0001] The invention relates to the field of yarn precision control, and in particular to a method and system for yarn precision control in a tufted carpet production process. Background Art

[0002] Tufted carpet is a kind of carpet with pile, usually made of synthetic fiber or natural fiber, and can provide good warmth effect, so it is very popular in cold winter. Tufted carpet is usually composed of many small piles, which can be released when stepped on, giving a soft and comfortable feeling. At the same time, it can also provide a certain sound insulation effect. Tufted carpets are available in many different styles and colors, which can be selected according to personal preferences and home decoration style.

[0003] The significance of precise yarn control is to ensure the quality and consistency of the fabric. By precisely controlling the thickness, tightness, twist and other parameters of the yarn, the texture, feel, strength, wear resistance and other aspects of the fabric can be optimized, thereby improving the quality and performance of the fabric. The fibers in the fabric can be arranged more closely, thereby improving the density and uniformity of the fabric. By controlling the twist and stretch of the yarn, the fibers in the fabric can be more closely interwoven to form a stronger structure, so that fabrics produced in different batches have the same quality and performance, thereby ensuring the stability and consistency of product quality. This is especially important for companies that need to produce large quantities of textiles of the same specifications and quality, and can enable companies to produce high-quality textiles more efficiently.

[0004] However, the existing method and system for precise yarn control in the production process of tufted carpets cannot predict the textile parameters during weaving in real time when they are used, resulting in the inability to make timely adjustments when a tufted carpet fails during weaving, which causes the textile texture of the tufted carpet to be abnormal during weaving, thereby affecting the overall aesthetics of the tufted carpet, making the effect of the tufted carpet during weaving unsatisfactory, greatly affecting the use efficiency of the method and system for precise yarn control in the production process of tufted carpets.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a method and system for accurately controlling yarns in a tufted carpet production process to overcome the above technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, a method for precise yarn control in a tufted carpet production process is provided, the method comprising the following steps:

[0009] S1. Obtaining design parameters of the tufted carpet, determining a preset weaving path and preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with image recognition technology;

[0010] S2, performing textile processing according to a preset textile path and preset textile parameters, and using an ARIMA model to perform real-time prediction of the textile parameters in the textile process to obtain predicted textile parameters;

[0011] S3, comparing the predicted textile parameters with the preset textile parameters, and analyzing the textile state at the next moment according to the comparison results;

[0012] S4. Correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

[0013] Preferably, obtaining design parameters of the tufted carpet, determining a preset weaving path and preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with image recognition technology comprises the following steps:

[0014] S11, obtaining design parameters of the tufted carpet, and analyzing the design parameters using image recognition technology to obtain preset textile parameters;

[0015] S12, calculating the total weight of the textile yarn and the weight of different types of yarn according to the preset textile parameters, and calculating the yarn length according to the yarn weight calculation results.

[0016] Preferably, calculating the total weight of the textile yarn and the weight of different types of yarn according to the preset textile parameters, and calculating the yarn length according to the yarn weight calculation results comprises the following steps:

[0017] S111, calculating the type of yarn by using image recognition technology combined with design parameters;

[0018] S112, using image recognition technology combined with design parameters to calculate the proportion of different types of yarn color blocks;

[0019] S113, calculating yarn parameters according to the calculation results of yarn type and color block proportion.

[0020] Preferably, weaving according to a preset weaving path and preset weaving parameters, and using an ARIMA model to perform real-time prediction of the preset weaving parameters in the weaving process includes the following steps:

[0021] S21, weaving according to a preset weaving path and preset weaving parameters, and monitoring the weaving parameters in real time during the weaving process to obtain real-time monitoring data;

[0022] S22, importing the textile parameters in the real-time monitoring data into the ARIMA model for analysis and judgment, and predicting the textile parameters based on the judgment results.

[0023] Preferably, the textile parameters in the real-time monitoring data are imported into the ARIMA model for analysis and judgment, and the textile parameter prediction is performed based on the judgment result, which includes the following steps:

[0024] S221, arranging the real-time monitoring data according to the observation time, and drawing a graph of the real-time monitoring data;

[0025] S222, performing textile parameter autoregression analysis on the graph drawing to obtain textile parameter autoregression coefficients;

[0026] S223. Construct an ARIMA model based on the autoregression coefficients of textile parameters, and use the ARIMA model to predict textile parameters.

[0027] The calculation formula of the preferred ARIMA model is:

[0028]

[0029] Among them, Y t for t Predicted textile parameters at the moment;

[0030] Y t-1 for t-1 Predicted textile parameters at the moment;

[0031] Y t-2 for t-2 Predicted textile parameters at the moment;

[0032] Y t-3 for t-3 Predicted textile parameters at the moment;

[0033] Y t-p is the initial textile parameter;

[0034] Φ represents the autoregressive coefficient;

[0035] U t is the error value range.

[0036] Preferably, correcting abnormal textile parameters in the textile process in real time according to the analysis results, and performing textile processing according to the corrected textile parameters comprises the following steps:

[0037] S41, classifying the predicted textile parameters according to parameter types, and training the classified parameters using a weak classifier to obtain original corrected parameters;

[0038] S42, allocating weights according to the modified parameters, and merging the modified parameters to obtain a strong classifier;

[0039] S43, training the classified parameters according to the strong classifier to obtain adjustment correction parameters;

[0040] S44, determine whether the adjusted correction parameters reach the correction index. If so, correct the abnormal textile parameters according to the adjusted correction parameters. If not, return to step S41 to continue training the correction parameters until the correction parameters meet the correction index after training, stop repeating, and correct the abnormal textile parameters.

[0041] Preferably, the predicted textile parameters are classified according to parameter types, and the classified parameters are trained using a weak classifier to obtain the original corrected parameters, including the following steps:

[0042] S411, calculating and predicting the occurrence frequency of textile parameters;

[0043] S412, converting the predicted textile parameter occurrence frequency into a feature vector, and calculating the sum of the feature vector in the posterior probability;

[0044] S413: Perform weighted averaging on the sum of the posterior probabilities according to the weights to obtain original correction parameters.

[0045] Preferably, training the classified parameters according to the strong classifier to obtain the adjustment correction parameters includes the following steps:

[0046] S431, extracting features of the predicted textile parameters to obtain feature parameters;

[0047] S432, using the original correction parameters to train the adopted SVM model;

[0048] S433, using characteristic parameters to evaluate and adjust the trained SVM model;

[0049] S433. Classify the predicted textile parameters through the adjusted SVM model.

[0050] According to another aspect of the present invention, a system for precise yarn control in a tufted carpet production process is provided, the system comprising:

[0051] An image data recognition module is used to obtain design parameters of the tufted carpet, determine a preset weaving path and preset weaving parameters according to the design parameters, and determine the type and length of the yarn in combination with image recognition technology;

[0052] A textile prediction module is used to perform textile processing according to a preset textile path and preset textile parameters, and use an ARIMA model to perform real-time prediction of textile parameters in the textile process to obtain predicted textile parameters;

[0053] The data comparison and analysis module is used to compare the predicted textile parameters with the preset textile parameters, and analyze the textile state at the next moment according to the comparison results;

[0054] The analysis and correction module is used to correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

[0055] The beneficial effects of the present invention are:

[0056] 1. The present invention monitors the preset textile parameters during the weaving of the tufted carpet in real time and makes real-time predictions based on the monitoring results, so that the tufted carpet can be accurately controlled during the weaving of the yarn, and the textile defects caused by the errors in the textile parameters during the weaving of the tufted carpet are reduced, thereby greatly improving the weaving efficiency of the tufted carpet.

[0057] 2. The present invention monitors the preset textile parameters during real-time weaving, and makes real-time predictions based on the monitoring results. By comparing the real-time prediction results with the textile parameters, timely adjustments are made when errors occur in the weaving, thereby improving the overall aesthetics of the tufted carpet when the textile texture errors occur during weaving. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 is a flow chart of a method for precise yarn control in a tufted carpet production process according to an embodiment of the present invention;

[0060] Figure 2 The figure is a principle block diagram of a system for precise yarn control in a tufted carpet production process according to an embodiment of the present invention.

[0061] In the figure:

[0062] 1. Image data recognition module; 2. Textile prediction module; 3. Data comparison and analysis module; 4. Analysis and correction module. DETAILED DESCRIPTION

[0063] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0064] According to an embodiment of the present invention, a method and system for accurately controlling yarns in a tufted carpet production process are provided.

[0065] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the method for precise yarn control in the tufted carpet production process according to an embodiment of the present invention, the method comprises the following steps:

[0066] S1. Obtaining design parameters of the tufted carpet, determining a preset weaving path and preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with image recognition technology;

[0067] Specifically, the step of obtaining the design parameters of the tufted carpet, determining the preset weaving path and preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with the image recognition technology includes the following steps:

[0068] S11, obtaining design parameters of the tufted carpet, and analyzing the design parameters using image recognition technology to obtain preset textile parameters;

[0069] Specifically, the design parameters of the tufted carpet include: obtaining the tufting density parameters of the tufted carpet by using a tufting density tester, obtaining the parameters of the length of the pile from the root to the tip and the width of the pile in the lateral direction by using a microscope, obtaining the parameters of the color and pattern design of the carpet by using a spectrometer, directly obtaining the length and width parameters of the carpet by measurement, measuring the tensile strength and wear resistance of the carpet by a tensile tester, and analyzing the chemical composition and physical properties of the carpet manufacturing materials by a material analyzer to judge the carpet manufacturing materials and obtain the standard parameters of the material.

[0070] S12, calculating the total weight of the textile yarn and the weight of different types of yarn according to the preset textile parameters, and calculating the yarn length according to the yarn weight calculation results;

[0071] Specifically, the tufted carpet is weighed multiple times using an electronic balance, and the data with the most identical weighing data is taken as the weight of the yarn.

[0072] S13, calculating the yarn length according to the yarn weight calculation result;

[0073] Specifically, when calculating the yarn length, the weight of the yarn is converted to grams per meter, the weight of the yarn per meter is determined, the required yarn length is determined, the required yarn length is multiplied by the weight of the yarn per meter to obtain the required total weight, and the required total weight is divided by the weight of the yarn per meter to obtain the required yarn length.

[0074] Specifically, calculating the total weight of textile yarns and the weights of different types of yarns according to preset textile parameters, and calculating the yarn length according to the yarn weight calculation results includes the following steps:

[0075] S111, calculating the type of yarn by using image recognition technology combined with design parameters;

[0076] Specifically, the design pictures and relevant parameter data of the tufted carpet are collected, the design pictures are preprocessed using image processing technology, and the preprocessed pictures are analyzed and calculated using image recognition technology to obtain parameters such as tufting density, pile length, and pile width. Specifically, the design pictures of the tufted carpet are photographed with a digital camera to obtain the details and colors of the tufted carpet, and then design sketches and detailed drawings are drawn, which will be used to convert the design sketches into digital drawings, and then modified and optimized. According to the obtained parameter data, combined with the physical properties of the yarn and the textile process, the appropriate yarn type is selected for calculation, and the design parameters are adjusted according to the calculation results.

[0077] S112, using image recognition technology combined with design parameters to calculate the proportion of different types of yarn color blocks;

[0078] Specifically, collect image data of the carpet, obtain photos of the carpet taken from different angles so that customers can fully understand the design and quality of the carpet, and use microscopes and other equipment to photograph and magnify the detailed structure of the carpet surface, and then use spectrometers and other equipment to take photos of the carpet color, and analyze and compare them to ensure that they meet the color standards required by customers. When the carpet has a specific pattern or pattern design, it is necessary to take corresponding photos for reference and confirmation, and mark them as different color areas so that each color area becomes a separate layer. Use image recognition technology to detect the boundaries and areas of each color area in order to calculate the proportion of each color, and for each color, compare its proportion with the design parameters to determine whether it meets the specification requirements.

[0079] S113, calculating yarn parameters according to the calculation results of yarn types and color block proportions;

[0080] Specifically, the calculation of yarn parameters includes: the linear density of the yarn, the maximum tensile force that the yarn can withstand when stretched, the shrinkage per unit length when the yarn is stretched to break, the diameter of the fibers in the yarn, the bending angle formed when the yarn is twisted or bent, the yarn structure, etc.

[0081] S2, performing textile processing according to a preset textile path and preset textile parameters, and using an ARIMA model to perform real-time prediction of the textile parameters in the textile process to obtain predicted textile parameters;

[0082] Specifically, the weaving according to the preset weaving path and preset weaving parameters, and using the ARIMA model to perform real-time prediction of the preset weaving parameters in the weaving process includes the following steps:

[0083] S21, weaving according to a preset weaving path and preset weaving parameters, and monitoring the weaving parameters in real time during the weaving process to obtain real-time monitoring data;

[0084] Specifically, sensors are installed under the carpet or in the surrounding environment to monitor parameters such as temperature, humidity, and pressure. Cameras are installed above the carpet or in the surrounding environment to capture the condition and changes on the carpet surface. Environmental device detection devices are installed indoors or outdoors to measure ambient temperature, humidity, air pressure and other parameters. Data collectors are then connected to sensors, cameras and other devices to collect and transmit real-time monitoring data. Finally, a remote monitoring system is used to connect to each device via the Internet to achieve remote real-time monitoring and control.

[0085] S22, importing the textile parameters in the real-time monitoring data into the ARIMA model for analysis and judgment, and predicting the textile parameters based on the judgment results.

[0086] Specifically, the textile parameters in the real-time monitoring data are imported into the ARIMA model for analysis and judgment, and the textile parameter prediction is performed based on the judgment results, including the following steps:

[0087] S221, arranging the real-time monitoring data according to the observation time, and drawing a graph of the real-time monitoring data;

[0088] Specifically, define a data structure to store the observation time and value of each parameter. For example, you can use a dictionary to store this data, read real-time data from sensors or monitoring devices, and store it in a data structure, sort the data, sort it in chronological order or according to other rules;

[0089] The acquired parameters need to be subjected to operations such as noise removal and filtering to ensure the accuracy and stability of the data. The processed data needs to be stored in memory or disk for subsequent analysis and visualization. The stored data can then be analyzed using statistical methods, machine learning algorithms and other techniques to extract useful information and patterns. The analysis results can be presented in the form of graphics, creating charts, curve graphs, scatter plots, etc. Finally, a real-time update function needs to be implemented, that is, the graphics need to be updated and new results displayed in a timely manner when the data changes.

[0090] S222, performing textile parameter autoregression analysis on the graph drawing to obtain textile parameter autoregression coefficients;

[0091] Specifically, outliers are removed and standardized on textile parameters, and autoregression analysis is performed on the data using time series analysis methods to obtain autoregression coefficients of textile parameters. The time series data is converted into a data set suitable for autoregression analysis, and the autocorrelation function and partial autocorrelation function are used to determine the optimal lag order. According to the confirmed lag order, an autoregression model is fitted, and the order of the autoregression model, that is, the lag order between the independent variable and the dependent variable, is determined. For each time point and the correlation coefficient between the value of the independent variable at the time point to be calculated, the coefficient in the autoregression equation is calculated for each time point, that is, the degree of influence of the value of the independent variable on the dependent variable, and the obtained autoregression coefficient is used to fit the autoregression model, that is, the value of the dependent variable at the current time point and the corresponding autoregression coefficient are used to predict the future value of the dependent variable, and the quality of the model is evaluated using residual analysis and other statistical tests;

[0092] Then use the sliding average method, select the lag order and use the known data point sequence to fit an autoregressive model. Starting from the initial moment, new data points are continuously added, and the value of the next point is predicted based on the information of the previous data points. After each new data point is added, the autoregressive coefficient is re-estimated, and the moving average of the most recent autoregressive coefficient is calculated. The new autoregressive coefficient and moving average are used for the next prediction. Repeat the steps to calculate the data points until all data points have been processed, perform moving average processing on the autoregressive coefficient, and optimize and adjust the textile process parameters based on the analysis results.

[0093] S223, constructing an ARIMA model according to the autoregression coefficients of textile parameters, and predicting textile parameters through the ARIMA model;

[0094] Specifically, the calculation formula of the ARIMA model is:

[0095] Among them, Y t for t Predicted textile parameters at the moment;

[0096] Y t-1 for t-1 Predicted textile parameters at the moment;

[0097] Y t-2 for t-2 Predicted textile parameters at the moment;

[0098] Y t-3 for t-3Predicted textile parameters at the moment;

[0099] Y t-p is the initial textile parameter;

[0100] Φ represents the autoregressive coefficient;

[0101] U t is the error value range.

[0102] Specifically, the error value range is selected according to the specific requirements of carpet production, such as the allowable error value during weaving.

[0103] Specifically, it is obtained through real-time monitoring during detection. For example, at time t, the predicted value at time t cannot be known, while the predicted textile parameters at the previous moment have been obtained by real-time correction of abnormal textile parameters in the textile process through analysis results, and the predicted textile parameters at the first moment are equal to the predicted textile parameters.

[0104] S3, comparing the predicted textile parameters with the preset textile parameters, and analyzing the textile state at the next moment according to the comparison results;

[0105] Specifically, the data of the predicted textile parameters are compared with the data of the preset textile parameters, and the textile state is analyzed based on the comparison result. If they are the same, there is no problem with the textile state; if they are different, there is no problem with the textile state.

[0106] S4. Correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

[0107] Specifically, the method of correcting abnormal textile parameters in the textile process in real time according to the analysis results and performing textile processing according to the corrected textile parameters includes the following steps:

[0108] S41, classifying the predicted textile parameters according to parameter types, and training the classified predicted textile parameters using a weak classifier to obtain original correction parameters;

[0109] Specifically, the predicted textile parameters are classified according to the types of textile parameters, such as needle and thread parameters, yarn parameters, and the effective parameters that need to be repaired first are judged according to the weights of the parameters.

[0110] Specifically, the method of classifying the predicted textile parameters according to parameter types, and training the classified parameters using a weak classifier to obtain the original corrected parameters includes the following steps:

[0111] S411, calculating and predicting the occurrence frequency of textile parameters;

[0112] Specifically, the frequency of parameter occurrence includes the following steps: the length of the initialization parameter should be equal to the length of the array to be sorted plus one, the array to be sorted is traversed, and for each element, the number of occurrences is accumulated to the corresponding counting array position, and the array is rearranged using the values ​​in the counting array, starting from the first element, and each element is compared with the value at the corresponding position in the counting array in turn. If they are equal, the element is output, otherwise the value at the corresponding position in the counting array is increased by one, the counting array is cleared, and the operation is continued on the next element until all elements have been processed.

[0113] S412, converting the predicted textile parameter occurrence frequency into a feature vector, and calculating the sum of the feature vector in the posterior probability;

[0114] Specifically, the value of each textile parameter is decomposed into several separate words, and a vocabulary containing all the words is established. The entire data set is traversed, the frequency of each word of each textile parameter in the data set is counted, and it is recorded in a word frequency matrix. For each textile parameter, its corresponding word frequency vector is used as the feature vector, and the feature vector is input into the machine learning model for training and prediction.

[0115] S413, performing weighted averaging on the sum of the posterior probabilities according to the weights to obtain the original correction parameters;

[0116] Specifically, the weak classifier model calculates the prediction probability of the model for each category, uses the label of the category to which each sample in the training data set belongs to calculate the prediction probability of the model for the category, and counts the number of samples of each category predicted by the model as the same category in all training data sets, thereby obtaining the frequency of occurrence of the category in the model. For a new sample, it is converted into a feature vector, and the sum of the posterior probabilities of the vector appearing in each category is calculated. The new sample is converted into a feature vector, and the label of the category to which each sample in the training data set belongs is used as a feature. The new sample is also mapped to the corresponding feature space, and the sum of the posterior probabilities of the vector appearing in each category is calculated. The average value of the vector in each category is calculated, and then the posterior probability of each category is calculated based on the average value.

[0117] S42, allocating weights according to the modified parameters, and merging the modified parameters to obtain a strong classifier;

[0118] Specifically, the weights are allocated according to the modified parameters such as the tuft density, the pile length, the pile width and the like during yarn weaving, and the weights are allocated according to the obtained parameter data in combination with the physical properties of the yarn and the weaving process.

[0119] S43, training the classified parameters according to the strong classifier to obtain adjustment correction parameters;

[0120] Specifically, the training of the classified parameters according to the strong classifier to obtain the adjustment correction parameters includes the following steps:

[0121] S431, extracting features of the predicted textile parameters to obtain feature parameters;

[0122] Specifically, the statistical characteristics of the classified parameters include: mean, variance, standard deviation, maximum value, minimum value, etc., and time series features and image features are extracted at the same time. The image feature extraction includes: for image data, deep learning methods such as convolutional neural networks are used to extract image features.

[0123] S432, using the original correction parameters to train the adopted SVM model;

[0124] Specifically, the original correction parameters are divided into training sets and test sets according to a certain ratio, using 70% training sets and 30% test sets. At the same time, the training sets and test sets need to be preprocessed, such as normalization and standardization, and appropriate SVM models are selected according to specific problems and data types, such as linear SVM, nonlinear SVM, kernel function SVM, etc. For the SVM model, some parameters need to be initialized, such as C value, penalty coefficient, etc., and each SVM model is trained using the training data set to obtain the optimal parameter combination.

[0125] S433, using characteristic parameters to evaluate and adjust the trained SVM model;

[0126] Specifically, the trained model is evaluated using the test data set, and indicators such as prediction accuracy, recall rate, and F1 value are calculated to determine the performance of the model.

[0127] S433. Classify the predicted textile parameters through the adjusted SVM model.

[0128] S44, determine whether the adjusted correction parameters reach the correction index. If so, correct the abnormal textile parameters according to the adjusted correction parameters. If not, return to step S41 to continue training the correction parameters until the correction parameters meet the correction index after training, stop repeating, and correct the abnormal textile parameters.

[0129] According to another embodiment of the present invention, Figure 2 As shown, a system for precise yarn control in a tufted carpet production process is provided, the system comprising:

[0130] An image data recognition module is used to obtain design parameters of the tufted carpet, determine a preset weaving path and preset weaving parameters according to the design parameters, and determine the type and length of the yarn in combination with image recognition technology;

[0131] A textile prediction module is used to perform textile processing according to a preset textile path and preset textile parameters, and use an ARIMA model to perform real-time prediction of textile parameters in the textile process to obtain predicted textile parameters;

[0132] The data comparison and analysis module is used to compare the predicted textile parameters with the preset textile parameters, and analyze the textile state at the next moment according to the comparison results;

[0133] The analysis and correction module is used to correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

[0134] To sum up, with the help of the above-mentioned technical scheme of the present invention, the present invention monitors the preset textile parameters during the weaving of the tufted carpet in real time, and performs real-time prediction based on the monitoring results, so that the tufted carpet can be easily controlled during the weaving of the yarn, and the textile defects caused by the errors in the weaving parameters of the tufted carpet are reduced, thereby greatly improving the weaving efficiency of the tufted carpet.

[0135] In addition, the present invention monitors preset textile parameters during real-time weaving, performs real-time prediction based on the monitoring results, and compares the real-time prediction results with the textile parameters to make timely adjustments when errors occur in weaving, thereby improving the overall aesthetics of the tufted carpet when errors in the textile texture occur during weaving.

[0136] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for precise yarn control in a tufted carpet production process, characterized in that: The method comprises the following steps: S1. Obtaining design parameters of the tufted carpet, determining a preset weaving path and preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with image recognition technology; S2, performing textile processing according to a preset textile path and preset textile parameters, and using an ARIMA model to perform real-time prediction of the textile parameters in the textile process to obtain predicted textile parameters; The weaving according to the preset weaving path and preset weaving parameters, and using the ARIMA model to perform real-time prediction of the preset weaving parameters in the weaving process includes the following steps: S21, weaving according to a preset weaving path and preset weaving parameters, and monitoring the weaving parameters in real time during the weaving process to obtain real-time monitoring data; S22, importing the textile parameters in the real-time monitoring data into the ARIMA model for analysis and judgment, and predicting the textile parameters based on the judgment results; The step of importing the textile parameters in the real-time monitoring data into the ARIMA model for analysis and judgment, and predicting the textile parameters based on the judgment results comprises the following steps: S221, arranging the real-time monitoring data according to the observation time, and drawing a graph of the real-time monitoring data; S222, performing textile parameter autoregression analysis on the graph drawing to obtain textile parameter autoregression coefficients; S223, constructing an ARIMA model according to the autoregression coefficients of textile parameters, and predicting textile parameters through the ARIMA model; The calculation formula of the ARIMA model is: Y t =φY t-1 +φY t-2 +φY t-3 …+φY t-p +U t Among them, Y t is the predicted textile parameter at time t; Y t-1 is the predicted textile parameter at time t-1; Y t-2 is the predicted textile parameter at time t-2; Y t-3 is the predicted textile parameter at time t-3; Y t-p is the initial textile parameter; φ represents the autoregressive coefficient; U t is the error value range; S3, comparing the predicted textile parameters with the preset textile parameters, and analyzing the textile state at the next moment according to the comparison results; S4. Correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

2. The method for precise yarn control in the tufted carpet production process according to claim 1, characterized in that: The method of obtaining the design parameters of the tufted carpet, determining the preset weaving path and the preset weaving parameters according to the design parameters, and determining the type and length of the yarn in combination with the image recognition technology comprises the following steps: S11, obtaining design parameters of the tufted carpet, and analyzing the design parameters using image recognition technology to obtain preset textile parameters; S12, calculating the total weight of the textile yarn and the weight of different types of yarn according to the preset textile parameters, and calculating the yarn length according to the yarn weight calculation results.

3. The method for precise yarn control in the tufted carpet production process according to claim 2, characterized in that: The method of calculating the total weight of the textile yarn and the weight of different types of yarns according to the preset textile parameters, and calculating the yarn length according to the yarn weight calculation results comprises the following steps: S111, calculating the type of yarn by using image recognition technology combined with design parameters; S112, using image recognition technology combined with design parameters to calculate the proportion of different types of yarn color blocks; S113, calculating yarn parameters according to the calculation results of yarn type and color block proportion.

4. A system for precise yarn control in a tufted carpet production process, used to implement the steps of the method for precise yarn control in a tufted carpet production process according to any one of claims 1 to 3, characterized in that: The system includes: An image data recognition module is used to obtain design parameters of the tufted carpet, determine a preset weaving path and preset weaving parameters according to the design parameters, and determine the type and length of the yarn in combination with image recognition technology; A textile prediction module is used to perform textile processing according to a preset textile path and preset textile parameters, and use an ARIMA model to perform real-time prediction of textile parameters in the textile process to obtain predicted textile parameters; The data comparison and analysis module is used to compare the predicted textile parameters with the preset textile parameters, and analyze the textile state at the next moment according to the comparison results; The analysis and correction module is used to correct abnormal textile parameters in the textile process in real time according to the analysis results, and perform textile processing according to the corrected textile parameters.

Images