An efficient and short-process yarn quality prediction analysis and optimization method

Abstract

The present invention belongs to the technical field of yarn quality prediction, analysis and optimization, and relates to an efficient and short-process yarn quality prediction, analysis and optimization method. By analyzing the quality of the fiber raw materials of each specified type of yarn, the present invention provides a more comprehensive understanding of the performance characteristics of different yarns, which helps to reduce the defective rate and improve product quality. By analyzing the influencing factors of each specified device corresponding to the quality of each specified type of yarn and analyzing the process quality of each specified device when weaving each specified type of yarn, the root causes of yarn quality problems are more accurately discovered, ensuring that these factors are adjusted and controlled in time during its production process, helping to ensure the stability of yarn quality. By predicting and optimizing the quality grade of each specified type of yarn, it helps to reduce the defective rate, reduce the waste of raw materials and energy, and also help to reduce quality fluctuations and production interruptions caused by equipment failures.

Description

Technical Field

[0001] The present invention belongs to the technical field of yarn quality analysis and optimization, and relates to a high-efficiency short-process yarn quality prediction analysis and optimization method. Background Art

[0002] Yarn, a slender, spun yarn made from processed textile fibers, is the basic unit of textiles. It can be made from a single fiber, such as pure cotton or wool, or a blend of multiple fibers, such as cotton-linen or polyester-cotton blends. As a raw material for textiles, yarn quality directly impacts production efficiency and scrap rates. High-quality yarn reduces problems such as broken ends and skipped yarns during the production process, thereby reducing production costs. Therefore, systems for analyzing and optimizing yarn quality play a crucial role.

[0003] In the prior art, there are also some related solutions involving the analysis of yarn quality. For example, the invention patent application of a spinning quality prediction method based on the Attention-GRU model with Chinese patent number 202211328257.0 uses the Attention-GRU model to aggregate and fully learn the information of the raw cotton performance indicators of the yarn, and analyze the importance of the raw cotton performance indicators at different times according to the size of the weight, effectively highlighting the key information in the factors affecting the yarn quality, and then outputting the predicted value of the yarn quality, thereby reducing the amount of calculation while ensuring the prediction accuracy and improving the prediction efficiency.

[0004] Another invention patent application with Chinese patent number 202211278000.9 is for a yarn quality prediction method based on meta-learning method. It uses the grey correlation method to analyze the weights of input parameters affecting yarn quality, improves the calculation speed and accuracy of the calculation results, and proposes a BiLSTM model that introduces confidence. It can deeply explore the relationship between input parameter characteristics and yarn quality, improve the prediction accuracy and efficiency of the model, and solve the problem of scarcity of yarn quality prediction samples based on the meta-machine learning method, thereby improving the prediction accuracy of the algorithm.

[0005] There is also an invention patent application with Chinese patent number 202410634643.5 for an intelligent optimization method and system for vortex spinning yarn preparation. It fully automatically monitors the yarn production process, predicts and optimizes the yarn preparation parameters, and monitors and adjusts them in real time, thereby improving the scientificity and accuracy of decision-making, reducing ineffective operations and resource consumption in the yarn production process, and further improving the stability and predictability of the production process.

[0006] Although the above schemes have proposed some solutions to yarn quality, they still have certain limitations: on the one hand, the existing schemes mainly predict and analyze a certain type of yarn, and lack the extensiveness of testing and analysis of various types of yarns, which leads to the single direction of yarn quality prediction, and thus it is impossible to have a more comprehensive understanding of the performance characteristics of different yarns, and to more accurately control the quality during the production process, reduce the defective rate, and improve product quality.

[0007] On the other hand, the existing solutions ignore the monitoring and analysis of textile equipment in the yarn production process, and thus cannot eliminate the impact of yarn textile equipment on its quality, and thus cannot more accurately discover the root causes of yarn quality problems, cannot adjust and control these factors in time during its production process, and cannot ensure the stability of yarn quality. Summary of the Invention

[0008] In view of this, in order to solve the problems raised in the above background technology, an efficient short-process yarn quality prediction analysis and optimization method is proposed.

[0009] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an efficient short-process yarn quality prediction analysis and optimization method, including: S1, basic information acquisition: several textile equipment for spinning yarns are recorded as designated equipment, and several types of yarns that can be spun by each designated equipment are recorded as designated types of yarns, and then the basic information of the fiber raw materials of each designated type of yarn and the basic information and basic process parameter information of each designated equipment are obtained.

[0010] S2. Basic information analysis: Based on the basic information of each designated equipment, analyze the influencing factors of the quality of each designated type of yarn corresponding to each designated equipment; based on the basic information of the fiber raw materials of each designated type of yarn, analyze the quality of the fiber raw materials of each designated type of yarn; based on the basic process parameter information of each designated equipment, analyze the process quality of each designated equipment when weaving each designated type of yarn.

[0011] S3. Yarn quality prediction: predict the quality grade of each designated type of yarn spun by each designated equipment.

[0012] S4. Yarn quality optimization: Optimize the quality of each specified type of yarn based on the predicted results of the quality grade of each specified type of yarn spun by each specified device.

[0013] Compared with the existing technology, the beneficial effects of the present invention are as follows: 1. The present invention provides supporting data for subsequent prediction analysis and optimization of yarn quality by obtaining basic information of fiber raw materials of each specified type of yarn and basic information and basic process parameter information of each specified equipment.

[0014] 2. By analyzing the quality of fiber raw materials of each specified type of yarn, the present invention provides a more comprehensive understanding of the performance characteristics of different yarns, more accurately controls the quality during the production process, reduces the defective rate, and improves product quality.

[0015] 3. The present invention fully considers the impact of textile equipment in the yarn production process by analyzing the influencing factors of each designated equipment corresponding to each designated type of yarn quality and analyzing the process quality of each designated equipment when weaving each designated type of yarn, thereby more accurately discovering the root cause of yarn quality problems, ensuring timely adjustment and control of these factors during its production process, and helping to ensure the stability of yarn quality.

[0016] 4. The present invention helps to reduce the defective product rate and the waste of raw materials and energy by predicting and optimizing the quality grade of each specified type of yarn. It also helps to reduce quality fluctuations and production interruptions caused by equipment failure, thereby reducing production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. 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 creative work.

[0018] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0019] Figure 2 The specific analysis model corresponds to the specific type of yarn of the present invention.

[0020] Figure Description: 1 is the input vector of the input layer in the analysis model corresponding to a specified type of yarn. 2 is the output vector of the hidden layer in the analysis model corresponding to a specified type of yarn. 3 is the output vector of the output layer in the analysis model corresponding to a specified type of yarn. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] See also Figure 1As shown, the present invention provides an efficient short-process yarn quality prediction analysis and optimization method, and the specific steps are as follows: S1. Basic information acquisition: several textile equipment for spinning yarns are recorded as designated equipment, and several types of yarns that can be spun by each designated equipment are recorded as designated types of yarns, and then the basic information of the fiber raw materials of each designated type of yarn and the basic information and basic process parameter information of each designated equipment are obtained.

[0023] As a preferred feasible embodiment, the basic information of the fiber raw materials of each specified type of yarn includes fiber type, fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate.

[0024] It should be further explained that the fiber maturity refers to the degree of thickening of the fiber cell wall, that is, the thickness of the fiber cell wall is taken as the fiber maturity.

[0025] It needs to be further explained that the specific method of obtaining the fiber types of each specified type of yarn is: using a high-definition camera to shoot the fiber raw materials of each specified type of yarn, obtaining pictures of the fiber raw materials of each specified type of yarn, and matching them with the fiber raw material pictures corresponding to each fiber type stored in the information storage library to obtain the fiber types of each specified type of yarn.

[0026] The specific method for obtaining the fiber length, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of each specified type of yarn is: using the HVI fiber testing system to test the fiber raw materials of each specified type of yarn to obtain the fiber length, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of each specified type of yarn.

[0027] Among them, the HVI fiber testing system is an analytical instrument that integrates multiple testing functions. It is used to conduct comprehensive testing of cotton fibers and can quickly and large-capacity test multiple physical properties of cotton fibers.

[0028] The specific method for obtaining the fiber fineness of each designated type of yarn is: directly measuring the fiber fineness of each designated type of yarn using a fiber fineness analyzer.

[0029] The basic information of each designated device includes the cleanliness, precision and stability of the device.

[0030] It should be further explained that the specific method for obtaining the cleanliness, precision and stability of each designated device is: directly extracting the cleanliness, precision and stability of each designated device from the maintenance record of each designated device.

[0031] The basic process parameter information of each designated device includes the draft ratio, spinning speed, tension and twist when weaving each designated type of yarn.

[0032] It needs to be further explained that the specific method of obtaining the draft ratio, spinning speed, tension and twist of each designated equipment when weaving each designated type of yarn is: directly obtaining the draft ratio, spinning speed, tension and twist of each designated equipment when weaving each designated type of yarn from the equipment process parameter settings of each designated equipment when weaving each designated type of yarn.

[0033] The present invention provides supporting data for subsequent prediction and analysis and optimization of yarn quality by obtaining basic information of fiber raw materials of each specified type of yarn and basic information and basic process parameter information of each specified equipment.

[0034] S2. Basic information analysis: Based on the basic information of each designated equipment, analyze the influencing factors of the quality of each designated type of yarn corresponding to each designated equipment; based on the basic information of the fiber raw materials of each designated type of yarn, analyze the quality of the fiber raw materials of each designated type of yarn; based on the basic process parameter information of each designated equipment, analyze the process quality of each designated equipment when weaving each designated type of yarn.

[0035] As a preferred feasible embodiment, the influencing factors of the quality of each designated device corresponding to each designated type of yarn are analyzed in detail as follows: extracting the cleanliness, precision and stability of each designated device, respectively denoted as ,in , Number each designated device and analyze the factors affecting the quality of each designated type of yarn

[0036] ,in

[0037] The first The specified type of yarn is The standard cleanliness, standard accuracy and standard stability required for a specified device, They are the permissible difference between the set cleanliness and the standard cleanliness, the permissible difference between the accuracy and the standard accuracy, and the permissible difference between the stability and the standard stability. , The number of each specified type of yarn.

[0038] As a preferred feasible embodiment, in the process of analyzing the quality of the fiber raw materials of each specified type of yarn, it is necessary to construct a quality evaluation index of the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of the fiber raw materials of each specified type of yarn. The specific analysis method includes: extracting the fiber types of each specified type of yarn, and matching them with the reference fiber length, reference fiber fineness, reference fiber strength, reference fiber maturity, reference fiber short fiber rate and reference fiber impurity rate of each fiber type stored in the information storage library to obtain the reference fiber length, reference fiber fineness, reference fiber strength, reference fiber maturity, reference fiber short fiber rate and reference fiber impurity rate of each specified type of yarn.

[0039] Extract fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber percentage and fiber trash content of each specified type of yarn.

[0040] The fiber length of each specified type of yarn is ratioed to its corresponding reference fiber length, and the obtained ratio is recorded as the fiber length quality evaluation index of the fiber raw material of each specified type of yarn. Similarly, the fiber strength and fiber maturity quality evaluation index of the fiber raw material of each specified type of yarn can be obtained.

[0041] The fiber fineness of each specified type of yarn is ratioed to its corresponding reference fiber fineness, and the obtained ratio is added by one and the reciprocal is taken. The value is recorded as the fiber fineness quality evaluation index of the fiber raw material of each specified type of yarn. Similarly, the fiber short fiber rate and fiber impurity content quality evaluation index of the fiber raw material of each specified type of yarn can be obtained.

[0042] By analyzing the quality of fiber raw materials of each specified type of yarn, the present invention provides a more comprehensive understanding of the performance characteristics of different yarns, more accurately controls the quality during the production process, reduces the defective rate, and improves product quality.

[0043] As a preferred feasible embodiment, in the process of analyzing the yarn spinning process quality of each designated device, it is necessary to construct the process quality evaluation coefficient of each designated type of yarn when spinning each designated type of yarn. The specific analysis method includes: extracting the draft multiple, spinning speed, tension and twist of each designated device when spinning each designated type of yarn, and recording them as .

[0044] Analyze the process quality evaluation coefficient of each specified equipment when spinning each specified type of yarn

[0045] ,in

[0046] The first The standard draft multiple, standard spinning speed, standard tension and standard twist of a specified type of yarn during spinning, They are the allowable difference between the set draft multiple and the standard draft multiple, the allowable difference between the spinning speed and the standard spinning speed, the allowable difference between the tension and the standard tension, and the allowable difference between the twist and the standard twist.

[0047] The process quality evaluation coefficient of each designated equipment when weaving each designated type of yarn is compared with the preset process quality evaluation coefficient threshold. If the process quality evaluation coefficient of a designated equipment when weaving a designated type of yarn is less than the process quality evaluation coefficient threshold, the process quality of the designated equipment when weaving the designated type of yarn is recorded as unqualified; otherwise, the process quality of the designated equipment when weaving the designated type of yarn is recorded as qualified, thereby obtaining the process quality of each designated equipment when weaving each designated type of yarn.

[0048] The present invention fully considers the impact of textile equipment in the yarn production process by analyzing the influencing factors of each designated equipment corresponding to each designated type of yarn quality and analyzing the process quality of each designated equipment when weaving each designated type of yarn, thereby more accurately discovering the root cause of yarn quality problems, ensuring timely adjustment and control of these factors during its production process, and helping to ensure the stability of yarn quality.

[0049] S3. Yarn quality prediction: predict the quality grade of each designated type of yarn spun by each designated equipment.

[0050] As a preferred feasible embodiment, when predicting the quality grade of each designated equipment in weaving each designated type of yarn, it is necessary to construct a comprehensive quality prediction coefficient of each designated equipment in weaving each designated type of yarn. The specific analysis method includes: extracting the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate quality evaluation index of the fiber raw material of each designated type of yarn, and based on this analysis, obtain the strength, elasticity and uniformity of each designated type of yarn, which are respectively recorded as .

[0051] Extract the process quality evaluation coefficient of each specified equipment when weaving each specified type of yarn.

[0052] According to the analytical formula Get the comprehensive prediction coefficient of the quality of each specified equipment in weaving each specified type of yarn ,in They are the weight factors of the comprehensive quality prediction coefficient corresponding to the set yarn quality and the process quality of the equipment when spinning yarn.

[0053] It should be further explained that the weight factors of the comprehensive quality prediction coefficient corresponding to the set yarn quality and the process quality of the equipment when spinning the yarn can be set to 0.7 and 0.3 respectively.

[0054] Yarn, as the fundamental raw material for textiles, has a direct impact on the performance, appearance, and comfort of the final product. Yarn quality plays a crucial role in predicting the overall quality of textiles. Therefore, a weighting factor of 0.7 has been assigned to the comprehensive quality prediction coefficient corresponding to yarn quality, reflecting the dominant role of yarn quality in textile quality control.

[0055] Yarn quality is fundamental, but the process quality of the equipment also plays a significant role in textile quality. Factors such as the equipment's operating status, accuracy, and stability directly impact yarn production efficiency and quality. Therefore, a weighting factor of 0.3 is assigned to the comprehensive quality prediction coefficient corresponding to the equipment's process quality during yarn spinning. This factor not only takes into account the importance of equipment process quality but also reflects its supporting role in the overall quality control system.

[0056] As a preferred feasible embodiment, the prediction of the quality grade of each designated device when weaving each designated type of yarn specifically includes: matching the comprehensive quality prediction coefficient of each designated device when weaving each designated type of yarn with the comprehensive quality prediction coefficient range corresponding to each yarn quality grade stored in the information reserve library, to obtain the yarn quality grade of each designated device when weaving each designated type of yarn.

[0057] As a preferred feasible embodiment, the strength, elasticity and uniformity of each specified type of yarn are specifically obtained by matching each specified type of yarn with the exclusive analysis model corresponding to each specified type of yarn belonging to the preset analysis model to obtain the exclusive analysis model corresponding to each specified type of yarn.

[0058] By inputting the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate quality evaluation index of the fiber raw materials of each specified type of yarn into its corresponding exclusive analysis model, the strength, elasticity and uniformity of each specified type of yarn can be obtained.

[0059] As a preferred feasible embodiment, the specific acquisition method of the preset analysis model includes: F1, training data acquisition: randomly selecting several groups of fiber raw materials of a specified type of yarn with known fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate, weaving them to obtain several groups of the specified type of yarn, and then testing them to obtain the strength, elasticity and uniformity of several groups of the specified type of yarn, and then using the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of several groups of fiber raw materials of the specified type of yarn and the strength, elasticity and uniformity of several groups of the specified type of yarn as training data.

[0060] F2. Establishing a training model: The training model consisting of an input layer, a hidden layer, and an output layer is recorded as a target training model, wherein the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate of several groups of fiber raw materials of the specified type of yarn are used as input vectors of the input layer of the target training model, and the strength, elasticity, and uniformity of several groups of the specified type of yarn are used as output vectors of the output layer of the target training model.

[0061] F3. Data training: Train the target training model to obtain the hidden parameter matrix in the target training model, and use it as the parameter matrix of the hidden layer in the exclusive analysis model corresponding to the specified type of yarn.

[0062] F4. Obtaining the analysis model: The trained target training model is used as the exclusive analysis model corresponding to the specified type of yarn. Similarly, the exclusive analysis model corresponding to each specified type of yarn can be obtained. The exclusive analysis models corresponding to each specified type of yarn are collectively referred to as analysis models.

[0063] It should be further explained that each designated type of yarn corresponds to an exclusive analysis model such as Figure 2 As shown, the exclusive analysis model corresponding to each specified type of yarn includes an input layer, a hidden layer and an output layer.

[0064] It should be further explained that the exclusive analysis model corresponding to each designated type of yarn is constructed by a neural network. In a neural network, the input layer, hidden layer and output layer are the basic components of the network structure.

[0065] The input layer is the beginning of the neural network, receiving input data from the outside world. Each neuron in the input layer represents an input feature. The input layer does not perform any computations; it simply passes the input data to the next layer, the hidden layer.

[0066] Specifically, the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of the fiber raw material of the specified type of yarn are the neurons of the input layer.

[0067] The hidden layer, located between the input and output layers, is the part of the neural network that actually performs computations. There can be one or more hidden layers, and each hidden layer can contain multiple neurons. These neurons are connected to neurons in the previous layer via weighted connections, or weights, and receive the output of the previous layer as input. Each neuron performs a weighted summation of its inputs and applies an activation function to produce an output. This output serves as the input for neurons in the next layer. The primary function of the hidden layer is to learn feature representations of the input data and capture the complex relationships between input and output.

[0068] The output layer is the final layer of a neural network. It receives the output from the last hidden layer and produces the network's final output. The number of neurons in the output layer depends on the specific task. Similar to neurons in the hidden layers, neurons in the output layer perform a weighted sum of their inputs and apply an activation function to produce an output. The output of the output layer can be compared with the true labels to calculate the loss function and used for backpropagation during training to update the weights.

[0069] Specifically, the strength, elasticity and uniformity of the specified type of yarn are the neurons of the output layer.

[0070] S4. Yarn quality optimization: Optimize the quality of each specified type of yarn based on the predicted results of the quality grade of each specified type of yarn spun by each specified device.

[0071] As a preferred feasible embodiment, the optimization of the quality of each specified type of yarn includes the following specific analysis methods: extracting the yarn quality grade of each specified device when weaving each specified type of yarn, and extracting the process quality of each specified device when weaving each specified type of yarn.

[0072] It should be further explained that the yarn quality grades include A-grade yarn, B-grade yarn, C-grade yarn and D-grade yarn.

[0073] Among them, grade A yarn is the highest quality yarn with good quality, high strength and beautiful appearance.

[0074] Grade B yarn is of medium quality. Compared with grade A yarn, it has some shortcomings, but it has stable quality and higher strength.

[0075] Grade C yarn is of average quality and has obvious defects, but if it is improved under corresponding technical conditions, satisfactory products can be obtained.

[0076] D-grade yarn is of poor quality and has many defects and shortcomings.

[0077] It is analyzed according to preset principles to obtain optimization suggestions for the quality of each specified type of yarn and provide feedback.

[0078] It needs to be further explained that the specific contents of the preset principles include: extracting the process quality of each designated equipment when weaving each designated type of yarn, and extracting the designated equipment whose process quality when weaving each designated type of yarn is unqualified, and recording them as unqualified equipment. Similarly, qualified equipment can be obtained.

[0079] From the yarn quality grades of each designated device when weaving each designated type of yarn, the yarn quality grades of each unqualified device when weaving each designated type of yarn and the yarn quality grades of each qualified device when weaving each designated type of yarn are extracted.

[0080] If the yarn quality grade of each qualified equipment when weaving each specified type of yarn is A-grade yarn or B-grade yarn, and the yarn quality grade of each qualified equipment when weaving each specified type of yarn is higher than the yarn quality grade of each unqualified equipment when weaving each specified type of yarn, then the optimization suggestion for the quality of each specified type of yarn is recorded as the optimized equipment.

[0081] If the yarn quality grades of each qualified device and each unqualified device when weaving each specified type of yarn are both C-grade yarn or D-grade yarn, the optimization suggestions for the quality of each specified type of yarn are recorded as optimized fiber raw materials.

[0082] The present invention helps to reduce the defective product rate and the waste of raw materials and energy by predicting and optimizing the quality grade of each specified type of yarn, and also helps to reduce quality fluctuations and production interruptions caused by equipment failure, thereby reducing production costs.

[0083] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. An efficient and short-process yarn quality prediction, analysis and optimization method, characterized by: include: S1. Basic Information Acquisition: Several textile machines for spinning yarns are designated as designated machines, and several types of yarns that can be spun by each designated machine are designated as designated yarn types. Basic information of the fiber raw materials of each designated yarn type is obtained, including fiber type, fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate. Obtain basic information of each designated equipment, including the cleanliness, accuracy and stability of the equipment; Obtain basic process parameter information for each designated equipment when spinning each designated type of yarn, including draft ratio, spinning speed, tension and twist; S2. Basic information analysis: Based on the cleanliness, precision and stability of each designated equipment, analyze the factors affecting the quality of each designated equipment for each designated type of yarn: Construct a fiber raw material quality evaluation index for each specified type of yarn, including: The fiber length, fiber strength, and fiber maturity are compared with their reference values ​​to generate quality evaluation indices for fiber length, fiber strength, and fiber maturity respectively; The fiber fineness, fiber short fiber rate, and fiber impurity rate are compared with their reference values, and then the inverse is added to generate the quality evaluation index of the corresponding fiber fineness, fiber short fiber rate, and fiber impurity rate. Analyze the process quality evaluation coefficient based on the process parameters of each specified equipment; S3. Yarn quality prediction: Input the fiber raw material quality evaluation index of each specified type of yarn into its corresponding dedicated analysis model, and output the yarn strength, elasticity and uniformity; the dedicated analysis model is trained through the following steps: For different yarn types, fiber raw material data and yarn performance data are collected as training data; Train the target training model; Based on the yarn performance data and process quality evaluation coefficient, the quality comprehensive prediction coefficient is analyzed, and the quality comprehensive prediction coefficient of each designated equipment when weaving each designated type of yarn is matched with the quality comprehensive prediction coefficient range corresponding to each yarn quality grade stored in the information reserve library, so as to obtain the quality grade of each designated type of yarn spun by each designated equipment; S4. Yarn quality optimization: Extract the yarn quality grade and process quality of each designated equipment when weaving each designated type of yarn and provide corresponding optimization suggestions; If the yarn quality grade of the qualified equipment is A / B and higher than that of the unqualified equipment, an "optimize equipment" recommendation will be generated; If the yarn quality grade produced by all equipment is C / D, an "Optimize Fiber Raw Materials" recommendation is generated; Feedback optimization suggestions to the production side.

2. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 1, characterized in that: The specific analysis method of the factors affecting the quality of yarns of each designated type corresponding to each designated device includes: Extract the cleanliness, precision and stability of each specified device and record them as ,in , Number each designated device and analyze the factors affecting the quality of each designated type of yarn ,in The first The specified type of yarn is The standard cleanliness, standard accuracy and standard stability required for a specified device, They are the permissible difference between the set cleanliness and the standard cleanliness, the permissible difference between the accuracy and the standard accuracy, and the permissible difference between the stability and the standard stability. , The number of each specified type of yarn.

3. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 2, characterized in that: The quality analysis of the fiber raw materials of each specified type of yarn requires the construction of quality evaluation indexes for fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate of the fiber raw materials of each specified type of yarn. The specific analysis methods include: Extracting the fiber types of each specified type of yarn, and matching them with the reference fiber length, reference fiber fineness, reference fiber strength, reference fiber maturity, reference fiber short fiber rate, and reference fiber trash content of each fiber type stored in the information storage library, to obtain the reference fiber length, reference fiber fineness, reference fiber strength, reference fiber maturity, reference fiber short fiber rate, and reference fiber trash content of each specified type of yarn; Extract the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of each specified type of yarn; The fiber length of each specified type of yarn is ratioed to its corresponding reference fiber length, and the obtained ratio is recorded as the fiber length quality evaluation index of the fiber raw material of each specified type of yarn. Similarly, the fiber strength and fiber maturity quality evaluation index of the fiber raw material of each specified type of yarn can be obtained; The fiber fineness of each specified type of yarn is ratioed to its corresponding reference fiber fineness, and the obtained ratio is added by one and the reciprocal is taken. The value is recorded as the fiber fineness quality evaluation index of the fiber raw material of each specified type of yarn. Similarly, the fiber short fiber rate and fiber impurity content quality evaluation index of the fiber raw material of each specified type of yarn can be obtained.

4. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 3, characterized in that: The specific analysis methods for the process quality of each designated equipment when spinning each designated type of yarn include: Extract the draft multiple, spinning speed, tension and twist of each specified equipment when spinning each specified type of yarn, and record them as ; Analyze the process quality evaluation coefficient of each specified equipment when spinning each specified type of yarn ,in The first The standard draft multiple, standard spinning speed, standard tension and standard twist of a specified type of yarn during spinning, They are the permissible difference between the set draft multiple and the standard draft multiple, the permissible difference between the spinning speed and the standard spinning speed, the permissible difference between the tension and the standard tension, and the permissible difference between the twist and the standard twist; The process quality evaluation coefficient of each designated equipment when weaving each designated type of yarn is compared with the preset process quality evaluation coefficient threshold. If the process quality evaluation coefficient of a designated equipment when weaving a designated type of yarn is less than the process quality evaluation coefficient threshold, the process quality of the designated equipment when weaving the designated type of yarn is recorded as unqualified; otherwise, the process quality of the designated equipment when weaving the designated type of yarn is recorded as qualified, thereby obtaining the process quality of each designated equipment when weaving each designated type of yarn.

5. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 4, characterized in that: When predicting the quality grade of each designated device when weaving each designated type of yarn, it is necessary to construct a comprehensive quality prediction coefficient of each designated device when weaving each designated type of yarn. The specific analysis method includes: The fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate of the fiber raw materials of each specified type of yarn are extracted, and the strength, elasticity and uniformity of each specified type of yarn are obtained based on the analysis, which are recorded as ; Extract the process quality evaluation coefficient of each designated equipment when spinning each designated type of yarn; According to the analytical formula Obtain the comprehensive prediction coefficient of the quality of each specified equipment when spinning each specified type of yarn ,in They are the weight factors of the comprehensive quality prediction coefficient corresponding to the set yarn quality and the process quality of the equipment when spinning yarn.

6. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 5, characterized in that: The prediction of the quality grade of each designated type of yarn in weaving by each designated device specifically includes: The comprehensive quality prediction coefficients of each designated device when weaving each designated type of yarn are matched with the comprehensive quality prediction coefficient ranges corresponding to each yarn quality grade stored in the information reserve library to obtain the yarn quality grade of each designated device when weaving each designated type of yarn.

7. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 5, characterized in that: The strength, elasticity and uniformity of each specified type of yarn are specifically obtained in the following manner: Matching each designated type of yarn with the exclusive analysis model corresponding to each designated type of yarn belonging to the preset analysis model to obtain the exclusive analysis model corresponding to each designated type of yarn; By inputting the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate and fiber impurity rate quality evaluation index of the fiber raw materials of each specified type of yarn into its corresponding exclusive analysis model, the strength, elasticity and uniformity of each specified type of yarn can be obtained.

8. The high-efficiency short-process yarn quality prediction, analysis and optimization method according to claim 7, characterized in that: The specific method of obtaining the preset analysis model includes: F1. Acquisition of training data: Randomly select several groups of fiber raw materials of a specified type of yarn with known fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate, and weave them to obtain several groups of the specified type of yarn. Then, test them to obtain the strength, elasticity, and uniformity of several groups of the specified type of yarn. Then, use the fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate of the several groups of fiber raw materials of the specified type of yarn and the strength, elasticity, and uniformity of the several groups of the specified type of yarn as training data; F2. Establishing a training model: A training model consisting of an input layer, a hidden layer, and an output layer is recorded as a target training model. The fiber length, fiber fineness, fiber strength, fiber maturity, fiber short fiber rate, and fiber impurity rate of several sets of fiber raw materials of the specified type of yarn are used as input vectors of the input layer of the target training model. The strength, elasticity, and uniformity of several sets of the specified type of yarn are used as output vectors of the output layer of the target training model. F3. Data training: Train the target training model to obtain the hidden parameter matrix in the target training model, and use it as the parameter matrix of the hidden layer in the exclusive analysis model corresponding to the specified type of yarn; F4. Obtaining the analysis model: The trained target training model is used as the exclusive analysis model corresponding to the specified type of yarn. Similarly, the exclusive analysis model corresponding to each specified type of yarn can be obtained. The exclusive analysis models corresponding to each specified type of yarn are collectively referred to as analysis models.

Images