Method, device and equipment for setting production parameters of warp knitting machine

Abstract

The invention relates to the technical field of intelligent manufacturing, and provides a warp knitting machine production parameter setting method, device and equipment, and the method comprises the steps: constructing a plurality of groups of typical process parameters, enabling a warp knitting machine to work and measure under each group of typical process parameters to obtain dynamic tension data under each group of typical process parameters, obtaining fabric image data by using a camera, wherein the fabric image data comprises normal fabric image data and defect image data; constructing a variational auto-encoder, and respectively inputting the normal fabric image data and the defect image data into the variational auto-encoder to obtain hidden variable codes; generating a fabric production quality state code; deriving a code for predicting the production quality state of the fabric; calculating tension dynamic measurement values under all the process parameters and calculating dynamic tension data index vectors under all the process parameters; and calculating the predicted fabric production quality state codes under all the process parameters, and selecting the process parameter corresponding to the minimum value of the production quality state codes as the final production process parameter.

Description

Technical Field

[0001] The embodiments of the present invention generally relate to the field of intelligent manufacturing technology, and particularly to a method, apparatus and equipment for setting production parameters of a warp knitting machine. Background Technology

[0002] Horizontal stripe defects are a common quality problem in warp knitting machine production, especially irregular horizontal stripe defects. These defects have complex causes and are diverse in type, making them difficult to control. Due to the lack of specific control standards, it is difficult to accurately grasp their occurrence patterns and root causes, seriously affecting the appearance quality of the fabric and the economic benefits of the enterprise.

[0003] Currently, the control of horizontal stripe defects mainly relies on the experience and judgment of process engineers. They judge yarn tension by touch, observe the fabric surface, and qualitatively adjust machine and process parameters. However, this experience-based adjustment method has significant limitations. First, the experience-based judgment of process engineers is highly subjective and uncertain; different process engineers have varying levels of experience and skill, and may adopt different adjustment methods when facing the same problem, thus affecting the consistency and stability of production results. Second, experience-based judgment lacks systematicity and scientific rigor, making it difficult to establish standardized operating procedures and hindering the transfer and accumulation of knowledge. Finally, when dealing with complex and challenging fabrics, the experience-based adjustment method involves numerous parameters and variables with complex interactions and influences. Process engineers need to spend considerable time and effort on repeated adjustments and experiments to find suitable parameter combinations. Therefore, there is an urgent need to develop more scientific, systematic, and efficient methods for setting production parameters to improve production efficiency and product quality, reduce production costs, and promote the further development of warp knitting machine production technology. Summary of the Invention

[0004] To address the above issues, this invention uses a variational autoencoder to obtain the latent variables of the fabric image, thereby obtaining the fabric production quality status code. Then, through measured dynamic tension data, the predicted fabric production quality status code is derived. When setting production parameters, all production parameters are traversed, the predicted fabric production quality status code under all production parameters is calculated, and the optimal production parameter corresponding to the minimum value of the predicted fabric production quality status code is selected for production, which greatly improves the efficiency of setting production parameters.

[0005] According to embodiments of the present invention, a method, apparatus, and equipment for setting production parameters of a warp knitting machine are provided.

[0006] In a first aspect of the invention, a method for setting production parameters of a warp knitting machine is provided. The method includes: Step S01: Set typical values ​​for warp feed, machine speed, and tension speed, construct several sets of typical process parameters, and make the warp knitting machine work under each set of typical process parameters to measure the dynamic tension data under each set of typical process parameters. Use a camera to obtain fabric image data, including normal fabric image data and defect image data. Step S02: Construct a variational autoencoder, and input the normal fabric image data and defect image data into the variational autoencoder to obtain the latent variable encoding of the normal fabric image and the latent variable encoding of the defect fabric image, respectively. Step S03: Generate fabric production quality status code based on the latent variable encoding of normal fabric images and the latent variable encoding of defective fabric images. ; Step S04: Code the fabric production quality status Derive the code for predicting fabric production quality status. ; Step S05: Set the range of process parameters according to the performance and pattern requirements of the fabric, traverse all the process parameters that can be set in all parameter ranges, obtain the dynamic tension measurement values ​​under all process parameters, and calculate the dynamic tension data index vector under all process parameters. Step S06: Calculate the predicted fabric production quality status code under all process parameters. The process parameters corresponding to the minimum value of the production quality status code are selected as the final production process parameters.

[0007] Furthermore, the dynamic tension data mentioned in step S01 includes: kurtosis, skewness, standard deviation, average slope, and average of the maximum value of tension in a single cycle.

[0008] Furthermore, the dynamic tension data mentioned in step S01 is obtained by measuring the dynamic tension of warp-knitted yarn using a dynamic tension measurement system, and the specific formula is as follows: Kudo : ; Skewness : ; Standard deviation : ; average slope : First, calculate the instantaneous slope at each point. ,in, For the first Tension value at each point For the first Tension value at each point; The sampling time interval; ; Average value of the maximum value of tension in a single cycle : Segmentation cycle: Based on the encoder pulse signal, the continuous tension signal stream is divided into independent data segments that correspond perfectly to the machine cycle. Let the segmentation result in... One cycle; For each period after segmentation Find the maximum value among all tension data points within that cycle. ; The above Sum the maximum values ​​of each period, then divide by the number of periods. That is, to obtain the average value of the maximum value in a single period. : ; In the formula, For sample size; The standard deviation is the sample standard deviation. For each data point in the sample; The sample mean is calculated using the following formula: .

[0009] Furthermore, the fabric production quality status coding described in step S03 Its formula is: , In the formula, For the first Number of images of normal fabrics under typical process parameters; Encoding latent variables for normal fabric images. , From normal fabric image data Obtained through the encoder; for Number of defective fabric images under typical process parameters; Encoding latent variables for images of defective fabrics. , From defective fabric image data Obtained through the encoder; For the first The fabric production quality status codes under typical process parameters are as follows: Typical process parameters, ; For the first The distance between the latent variable encoding of a normal fabric image and the mean vector of the latent variable encoding of a normal fabric image. For the first The distance between the latent variable encoding of a defective fabric image and the mean vector of the latent variable encoding of a normal fabric image is the distance between the latent variable encodings. The lower the defect rate, the smaller the fabric production quality status code; the higher the defect rate, the larger the fabric production quality status code.

[0010] Furthermore, the predicted fabric production quality status coding described in step S04 Its formula is: , In the formula, This is a vector of dynamic tension data indicators under the current process parameters. For the first A vector of dynamic tension data indicators under typical process parameters. For the first Fabric production quality status coding under typical process parameters. This represents the number of typical process parameter groups.

[0011] Furthermore, the step S05, which involves setting the range of process parameters based on the fabric's properties and pattern requirements, comprises: dividing the warp feed parameter range into equal parts. Duan got The parameters, including the speed parameter range, are divided into several categories. Duan got The parameters, including the traction speed parameter range, are divided into several categories. Duan got Each parameter is used to iterate through all settable process parameters within a given range to obtain... Group parameters refer to all process parameters.

[0012] In a second aspect of the invention, an apparatus for setting production parameters of a warp knitting machine is provided. The apparatus includes: Data acquisition module: Used to set typical values ​​for warp feed, machine speed, and tensioning speed, construct several sets of typical process parameters, and make the warp knitting machine work under each set of typical process parameters to measure the dynamic tension data under each set of typical process parameters. It also uses a camera to obtain fabric image data, including normal fabric image data and defect image data. Latent variable acquisition module: used to construct a variational autoencoder, and input normal fabric image data and defect image data into the variational autoencoder to obtain the latent variable encoding of normal fabric image and the latent variable encoding of defective fabric image respectively; Production quality status coding module: used to generate fabric production quality status codes based on latent variable coding of normal fabric images and latent variable coding of defective fabric images. ; Predictive model building module: used for coding based on fabric production quality status. Derive the code for predicting fabric production quality status. ; Process parameter traversal module: Used to set the range of process parameters according to the performance and pattern requirements of the fabric, traverse all settable process parameters in all parameter ranges, obtain the dynamic tension measurement value under all process parameters, and calculate the dynamic tension data index vector under all process parameters. Final process parameter selection module: used to calculate the predicted fabric production quality status code under all process parameters. The process parameters corresponding to the minimum value of the production quality status code are selected as the final production process parameters.

[0013] In a third aspect of the invention, an electronic device is provided. The electronic device includes a memory and a processor, the memory storing a computer program, the processor executing the program to implement the method according to a first aspect of the invention.

[0014] In a fourth aspect of the invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method according to a first aspect of the invention.

[0015] This invention uses a variational autoencoder to obtain the latent variables of the fabric image, thereby obtaining the fabric production quality status code. Then, through the measured dynamic tension data, the predicted fabric production quality status code is derived. When setting production parameters, all production parameters are traversed, the predicted fabric production quality status code under all production parameters is calculated, and the optimal production parameter corresponding to the minimum value of the predicted fabric production quality status code is selected for production, which greatly improves the efficiency of setting production parameters.

[0016] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description.

[0017] Beneficial effects: 1. By calculating the predicted fabric production quality status code under all production parameters and selecting the optimal production parameters corresponding to the minimum value of the predicted fabric production quality status code for production, the reliance on the process engineer's experience is reduced. This not only speeds up the parameter setting and debugging process but also significantly improves the stability and consistency of the production process, reducing production fluctuations and quality problems caused by human factors. 2. For fabrics with complex structures and high difficulty, the optimal parameter combination can be found quickly, significantly shortening the debugging time, reducing manpower and time costs, and thus effectively reducing the overall production cost; 3. Establish standardized operating guidelines and parameter libraries to facilitate the learning and application of new process engineers, and promote the inheritance and continuous improvement of production technology. Attached Figure Description

[0018] The above and other features, advantages, and aspects of the various embodiments of the present invention will become more apparent from the accompanying drawings and the following detailed description. Wherein: Figure 1 A flowchart illustrating a method for setting production parameters of a warp knitting machine according to an embodiment of the present invention is shown; Figure 2 A block diagram of a warp-knitted yarn dynamic tension measurement system according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of a variational autoencoder according to an embodiment of the present invention is shown; Figure 4 A block diagram of a device for setting production parameters of a warp knitting machine according to an embodiment of the present invention is shown; Figure 5 A schematic diagram of the equipment for setting production parameters of a warp knitting machine according to an embodiment of the present invention is shown. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] According to an embodiment of the present invention, a method, apparatus, and equipment for setting production parameters of a warp knitting machine are proposed. The method uses a variational autoencoder to obtain the latent variables of the fabric image based on the fabric image, thereby obtaining the fabric production quality status code. Then, the predicted fabric production quality status code is derived by measuring dynamic tension data. When debugging and setting production parameters, all production parameters are traversed, the predicted fabric production quality status code under all production parameters is calculated, and the optimal production parameter corresponding to the minimum value of the predicted fabric production quality status code is selected for production, which greatly improves the efficiency of setting production parameters.

[0021] The principles and spirit of the present invention will be explained in detail below with reference to several representative embodiments.

[0022] Figure 1 This is a schematic flowchart illustrating a method for setting production parameters of a warp knitting machine according to an embodiment of the present invention. The method includes: Step S01: Set typical values ​​for warp feed, machine speed, and tension speed, construct several sets of typical process parameters, and make the warp knitting machine work under each set of typical process parameters to measure the dynamic tension data under each set of typical process parameters. Use a camera to obtain fabric image data, including normal fabric image data and defect image data. Step S02: Construct a variational autoencoder, and input the normal fabric image data and defect image data into the variational autoencoder to obtain the latent variable encoding of the normal fabric image and the latent variable encoding of the defect fabric image, respectively. Step S03: Generate fabric production quality status code based on the latent variable encoding of normal fabric images and the latent variable encoding of defective fabric images. ; Step S04: Code the fabric production quality status Derive the code for predicting fabric production quality status. ; Step S05: Set the range of process parameters according to the performance and pattern requirements of the fabric, traverse all the process parameters that can be set in all parameter ranges, obtain the dynamic tension measurement values ​​under all process parameters, and calculate the dynamic tension data index vector under all process parameters. Step S06: Calculate the predicted fabric production quality status code under all process parameters. The process parameters corresponding to the minimum value of the production quality status code are selected as the final production process parameters.

[0023] It should be noted that although the operation of the method of the present invention has been described in a specific order in the above embodiments and figures, this does not require or imply that the operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0024] To provide a clearer explanation of the method for setting production parameters of the warp knitting machine, a specific embodiment will be used for illustration below. However, it is worth noting that this embodiment is only for better illustrating the present invention and does not constitute an improper limitation of the present invention.

[0025] The following specific example will further illustrate the method for setting production parameters for warp knitting machines in more detail: Step S01: Set typical values ​​for warp feed, machine speed, and tension speed, construct several sets of typical process parameters, and measure the dynamic tension data under each set of typical process parameters by making the warp knitting machine work under each set of typical process parameters. Use a camera to obtain fabric image data, including normal fabric image data and defect image data.

[0026] Specifically, the dynamic tension data is obtained by a warp-knitted yarn dynamic tension measurement system, such as... Figure 2 As shown. The electronic tension meter uses a contact measurement method. When the middle measuring wheel is deformed under force, the deformation is converted into a resistance value, which is then converted into an electrical signal output by the corresponding measuring circuit. During measurement, the yarn is wound around the two guide wheels and the middle measuring wheel of the tension meter. The angle of the tension meter is adjusted so that the yarn follows its original path as closely as possible. To make the test yarn more representative and better reflect the actual yarn tension, the yarn in the middle position is selected, and the measurement is performed between the tension bar and the looping component.

[0027] The system's DSP processor receives Z and A signals from the spindle encoder, processes and stores the relevant data, and finally transmits the data to a cloud server for in-depth analysis. After data processing, multiple dynamic tension data points for each set of process parameters can be obtained, including: kurtosis, skewness, standard deviation, average slope, and the average of the maximum tension value per cycle. This data provides an important foundation for the subsequent construction of process optimization models.

[0028] In this embodiment, the typical values ​​for the feed rate, machine speed, and traction speed are set as shown in Table 1: Table 1

[0029] The system operates for 3 minutes under each set of typical process parameters, and the dynamic tension data under each set of typical process parameters is measured. After processing, the dynamic tension data under each set of typical process parameters are finally obtained: kurtosis, skewness, standard deviation, average slope, and average value of the maximum tension value in a single cycle.

[0030] Specifically, the formula for calculating dynamic tension data is as follows.

[0031] 1. Kurtosis : .

[0032] 2. Skewness : .

[0033] 3. Standard deviation : .

[0034] 4. Average slope : First, calculate the instantaneous slope at each point. ,in, For the first Tension value at each point For the first Tension value at each point; The sampling time interval; .

[0035] 5. Average value of the maximum tension value in a single cycle : Segmentation cycle: Based on the encoder pulse signal, the continuous tension signal stream is divided into independent data segments that correspond perfectly to the machine cycle. Let the segmentation result in... One cycle.

[0036] For each period after segmentation Find the maximum value among all tension data points within that cycle. .

[0037] The above Sum the maximum values ​​of each period, then divide by the number of periods. That is, to obtain the average value of the maximum value in a single period. : .

[0038] In the above formula, For sample size; The standard deviation is the sample standard deviation. For each data point in the sample; The sample mean is calculated using the following formula: .

[0039] In this embodiment, the parameter for obtaining the corresponding dynamic tension data from the second set of typical values ​​is: kurtosis. skewness Standard deviation average slope The average value of the maximum value of tension in a single cycle .

[0040] Step S02: Construct a variational autoencoder, and input the normal fabric image data and defect image data into the variational autoencoder to obtain the latent variable encoding of the normal fabric image and the latent variable encoding of the defect fabric image, respectively.

[0041] like Figure 3 The diagram shown is a schematic of a variational autoencoder. : , This is a normal fabric image data; To reconstruct the data; For fabric image data The mean vector obtained after passing through the encoding network; For fabric image data The standard deviation vector obtained after passing through the encoding network; For fabric image data The latent variable encoding obtained after passing through the encoding network; encoder This is a convolutional neural network, which can be ResNet, VGG, etc., or a self-constructed convolutional neural network. These are the parameters of the convolutional neural network; the decoder. It is a convolutional neural network.

[0042] The loss function is: , In the formula, This represents the expected value, used to calculate the average. Represents data from fabric images Inference of latent variable encoding Distribution; Indicates encoding from hidden variables Generate fabric image data Log-likelihood; This represents the KL divergence, used to measure the difference between two probability distributions; Represents data from fabric images The distribution of the latent variable encoding Z inferred from the middle. Represents data from fabric images Latent variable encoding in inference The prior distributions are given by KL divergence, which is used to measure the difference between the two distributions.

[0043] Simplifying the loss function and minimizing it using gradient descent, we obtain: , Finally, the encoder is obtained. The parameters of the corresponding convolutional neural network The latent variable code can be output from the fabric image after passing through the encoder.

[0044] In this embodiment, the latent variable encoding of the normal fabric image obtained by the variational autoencoder is: [1.22, 1.15, -0.75, ..., -1.68]; [1.23, 1.13, -0.76, ..., -1.61];、 ...; [1.16, 1.15, -0.75, ..., -1.59].

[0045] The latent variable encoding of the defective fabric image obtained by the variational autoencoder is: [0.15, -0.31, -0.63, ..., 0.72]; [0.13, -0.33, -0.61, ..., 0.74].

[0046] Latent variable codes obtained from images of the same category are highly similar, while images of different categories show significant differences.

[0047] Step S03: Generate fabric production quality status code based on the latent variable encoding of normal fabric images and the latent variable encoding of defective fabric images. Its formula is: , In the formula, For the first Number of images of normal fabrics under typical process parameters; Encoding latent variables for normal fabric images. , From normal fabric image data Obtained through the encoder; for Number of defective fabric images under typical process parameters; Encoding latent variables for images of defective fabrics. , From defective fabric image data Obtained through the encoder; For the first The fabric production quality status codes under typical process parameters are as follows: Typical process parameters, ; For the first The distance between the latent variable encoding of a normal fabric image and the mean vector of the latent variable encoding of a normal fabric image. For the first The distance between the latent variable encoding of a defective fabric image and the mean vector of the latent variable encoding of a normal fabric image is the distance between the latent variable encodings. The lower the defect rate, the smaller the fabric production quality status code; the higher the defect rate, the larger the fabric production quality status code.

[0048] In this embodiment, 498 images of normal fabric and 2 images of defective fabric were obtained under the second set of typical process parameters.

[0049] The mean vector of the latent variable encoding for normal fabric images is obtained by averaging the latent variable encodings of 498 normal fabric images. [1.206, 1.138, -0.755, ..., -1.637].

[0050] Under these typical process parameters, the fabric production quality status code is:

[0051]

[0052] .

[0053] Following this logic, the fabric production quality status codes under five typical process parameters are as follows: , , , , .

[0054] Step S04: Code the fabric production quality status Derive the code for predicting fabric production quality status. , , In the formula, This is a vector of dynamic tension data indicators under the current process parameters. For the first A vector of dynamic tension data indicators under typical process parameters. For the first Fabric production quality status coding under typical process parameters. This represents the number of typical process parameter groups.

[0055] Step S05: Set the range of process parameters according to the performance and pattern requirements of the fabric, traverse all the process parameters that can be set in all parameter ranges, obtain the dynamic tension measurement values ​​under all process parameters, and calculate the dynamic tension data index vector under all process parameters.

[0056] The range of process parameters is set according to the fabric's properties and pattern requirements. The warp feed rate parameter range is set as follows: The speed parameter range is The range of traction speed parameters is: Divide the delivery volume parameter range into equal parts. The segments and speed parameter ranges are divided into sections, etc. The segments, including the range of traction speed parameters, are divided into equal parts. Specifically, when high fabric performance requirements are needed, the segments can be more densely spaced; when performance requirements are lower, the segments can be more sparsely spaced. By iterating through all settable process parameters within all parameter ranges, the corresponding dynamic tension measurement values ​​are obtained, and then the corresponding dynamic tension data indicators are calculated.

[0057] Specifically, the current process parameters during the traversal process. : , In the formula, This represents the current volume of mail delivered. Current machine speed This represents the current traction speed.

[0058] The dynamic tension measurement value under the current process parameters is obtained, and then the dynamic tension data index vector under the current process parameters is calculated. for: , In the formula, This represents the kurtosis value of the dynamic tension data under the current process parameters. This represents the skewness value of the dynamic tension data under the current process parameters. The standard deviation of the dynamic tension data under the current process parameters. This represents the average slope of the dynamic tension data under the current process parameters. This is the average of the maximum tension values ​​in all single cycles of the dynamic tension data under the current process parameters.

[0059] In this embodiment, the range of the warp feed parameter is set to [1400, 1600], the range of the machine speed parameter is set to [1200, 1400], and the range of the tension density parameter is set to [22, 24]. The range of the warp feed parameter is divided into 4 equal segments to obtain 5 parameters, the range of the machine speed parameter is divided into 2 equal segments to obtain 3 parameters, and the range of the tension density parameter is divided into 2 equal segments to obtain 3 parameters.

[0060] Iterate through all settable process parameters within all parameter ranges; obtain the dynamic tension measurement values ​​under all process parameters, and then calculate the corresponding dynamic tension data indicators: Warp feed rate: 1400, machine speed: 1200, drawing density: 22; Warp feed rate: 1400, machine speed: 1200, drawing density: 23; Warp feed rate: 1400, machine speed: 1200, drawing density: 24; Warp feed rate: 1400, machine speed: 1300, drawing density: 22; Warp feed rate: 1400, machine speed: 1300, drawing density: 23; Warp feed rate: 1400, machine speed: 1300, drawing density: 24; … Warp feed rate: 1450, machine speed: 1200, drawing density: 22; Warp feed rate: 1450, machine speed: 1200, drawing density: 23; Warp feed rate: 1450, machine speed: 1200, drawing density: 24; … Warp feed rate: 1600, machine speed: 1400, drawing density: 22; Warp feed rate: 1600, machine speed: 1400, drawing density: 23; Warp feed rate: 1600, machine speed: 1400, drawing density: 24; A total of 5 × 3 × 3 = 45 process parameters were obtained; the dynamic tension measurement values ​​and dynamic tension data indicators under all process parameters were obtained: For example, given the current process parameters of 1450 mm feed rate, 1200 mm machine speed, and 23 mm draw density: [1450, 1200, 23], the corresponding dynamic tension data index is calculated as: [0.18, 0.33, 0.90, 4.8, 0.16].

[0061] In this embodiment, the 45 sets of current process parameters are: [1400, 1200, 22]; [1400, 1200, 23]; [1400, 1200, 24]; [1400, 1300, 22]; [1400, 1300, 23]; [1400, 1300, 24]; … [1450, 1200, 22]; [1450, 1200, 23]; [1450, 1200, 24]; … [1600, 1400, 22]; [1600, 1400, 23]; [1600, 1400, 24].

[0062] The corresponding dynamic tension data indicators obtained from the calculation are as follows: [0.25, 0.41, 1.10, 6.8, 0.18]; [0.27, 0.43, 1.20, 6.2, 0.17]; [0.27, 0.43, 1.35, 6.1, 0.17]; [0.23, 0.39, 1.20, 6.1, 0.21]; [0.23, 0.35, 1.35, 5.8, 0.19]; [0.21, 0.35, 1.45, 6.1, 0.19]; … [0.21, 0.35, 0.92, 5.0, 0.15]; [0.18, 0.33, 0.90, 4.8, 0.16]; [0.20, 0.35, 1.05, 4.5, 0.15]; … [0.18, 0.32, 0.75, 4.2, 0.12]; [0.17, 0.29, 0.82, 4.3, 0.13]; [0.16, 0.28, 0.91, 4.3, 0.12].

[0063] Step S06: Calculate the predicted fabric production quality status code under all process parameters. The process parameters corresponding to the minimum value of the production quality status code are selected as the final production process parameters.

[0064] Calculate the predicted values ​​of fabric production quality status codes under 45 sets of process parameters: [1.36, 3.13, 4.21, 3.75, 2.12, 0.75, ..., 0.82, 0.51, 0.97, ..., 1.18, 1.23, 2.65].

[0065] Finally, 45 groups were selected. The minimum corresponding process parameters are used as the final production process parameters, which are [1450, 1200, 23], that is, the feed rate is 1450, the machine speed is 1200, and the drawing density is 23.

[0066] Based on the same inventive concept, this invention also proposes a device for setting production parameters of a warp knitting machine. The implementation of this device is similar to the implementation of the method described above, and repeated details will not be elaborated further. Figure 4 As shown, the device 100 includes: Data acquisition module 101: Used to set typical values ​​for warp feed amount, machine speed, and tensioning speed, construct several sets of typical process parameters, enable the warp knitting machine to work under each set of typical process parameters, measure the dynamic tension data under each set of typical process parameters, and use a camera to obtain fabric image data, including normal fabric image data and defect image data. Latent variable acquisition module 102: used to construct a variational autoencoder, and input normal fabric image data and defect image data into the variational autoencoder to obtain the latent variable encoding of normal fabric image and the latent variable encoding of defect fabric image respectively. Production quality status coding module 103: Used to generate fabric production quality status codes based on the latent variable coding of normal fabric images and the latent variable coding of defective fabric images. ; Predictive model building module 104: used for coding based on fabric production quality status Derive the code for predicting fabric production quality status. ; Process parameter traversal module 105: Used to set the range of process parameters according to the performance and pattern requirements of the fabric, traverse all settable process parameters in all parameter ranges, obtain the dynamic tension measurement value under all process parameters, and calculate the dynamic tension data index vector under all process parameters. Final process parameter selection module 106: Used to calculate the predicted fabric production quality status code under all process parameters. The process parameters corresponding to the minimum value of the production quality status code are selected as the final production process parameters.

[0067] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0068] like Figure 5 As shown, the device includes a central processing unit (CPU), which can perform various appropriate actions and processes based on computer program instructions stored in read-only memory (ROM) or loaded from storage units into random access memory (RAM). The RAM can also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0069] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0070] The processing unit executes the various methods and processes described above, such as method steps S01 to S06. For example, in some embodiments, method steps S01 to S06 may be implemented as a computer software program tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of method steps S01 to S06 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute method steps S01 to S06 by any other suitable means (e.g., by means of firmware).

[0071] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0072] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0073] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0074] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the invention. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0075] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method of setting production parameters for a warp knitting machine, characterized in that, The method comprises: Step S01: setting typical values of the warp feed amount, the machine speed, and the drawing speed, constructing a plurality of groups of typical process parameters, making the warp knitting machine work under each group of typical process parameters to measure dynamic tension data under each group of typical process parameters, and using a camera to obtain fabric image data, including normal fabric image data and defect image data; Step S02: constructing a variational autoencoder and inputting the normal fabric image data and the defect image data into the variational autoencoder to obtain normal fabric image hidden variable encoding and defect fabric image hidden variable encoding; Step S03: generating fabric production quality state code according to normal fabric image latent variable code and defect fabric image latent variable code ; Step S04: Deriving a predicted fabric production quality state code from the fabric production quality state code deriving a predicted fabric production quality state code ; Step S05: setting a process parameter interval range according to the performance of the fabric and the pattern requirement, traversing all process parameters that can be set in the parameter interval, obtaining dynamic tension measurement values under all process parameters, and calculating a dynamic tension data index vector under all process parameters; Step S06: Calculate the predicted fabric production quality state code under all process parameters And select the process parameters corresponding to the minimum production quality state code as the final production process parameters.

2. A method of setting the production parameters of a warp knitting machine according to claim 1, characterized in that, The dynamic tension data in step S01 includes kurtosis, skewness, standard deviation, average slope, and average maximum value of the single-period tension.

3. A method of setting the production parameters of a warp knitting machine according to claim 2, characterized in that, The dynamic tension data in step S01 is measured by a warp yarn dynamic tension measurement system, and the specific formula is: kurtosis : ； Skewness : ； Standard deviation : ； Slope average value : First, the instantaneous slope of each point is calculated wherein is the tension value of the point, is the tension value of the point; is the sampling time interval; ； average of the maximum values of the tension for one cycle : Segmentation cycle: according to the encoder pulse signal, the continuous tension signal stream is cut into individual data segments that completely correspond to the machine cycle, and the segmentation results in a number of cycles; For each cycle after segmentation , find the maximum value among all the tension data points in the cycle ; Sum the maximum values of the above periods and divide by the number of periods to obtain the average of the maximum values of the individual periods : ； wherein is the sample size; is the sample standard deviation; is each data point in the sample; is the sample mean, calculated as .

4. A method of setting the production parameters of a warp knitting machine according to claim 1, characterized in that, The fabric production quality state code described in step S03 The formula is: ， In the formula, For the first Number of images of normal fabrics under typical process parameters; Encoding latent variables for normal fabric images. , From normal fabric image data Obtained through the encoder; for Number of defective fabric images under typical process parameters; Encoding latent variables for images of defective fabrics. , From defective fabric image data Obtained through the encoder; For the first The fabric production quality status codes under typical process parameters are as follows: Typical process parameters, .

5. The method of setting the parameters of a warp knitting machine according to claim 1, characterized in that, The predicted fabric production quality state code in step S04 The formula is: ， In the formula, This is a vector of dynamic tension data indicators under the current process parameters. For the first A vector of dynamic tension data indicators under typical process parameters. For the first Fabric production quality status coding under typical process parameters. This represents the number of typical process parameter groups.

6. A method of setting the production parameters of a warp knitting machine according to claim 1, characterized in that, The step of setting the process parameter interval range according to the performance of the fabric and the pattern requirement in step S05 is that the delivery amount parameter interval is equally divided into segments to obtain parameters, the machine speed parameter interval is equally divided into segments to obtain parameters, the drawing speed parameter interval is equally divided into segments to obtain parameters, and all the process parameters that can be set in all the parameter intervals are traversed to obtain groups of parameters, that is, all the process parameters.

7. A device for setting production parameters of a warp knitting machine, characterized in that, The device implements the method in any one of claims 1-6, comprising: a data acquisition module configured to set typical values of the warp feed amount, the machine speed, and the drawing speed, construct a plurality of groups of typical process parameters, make the warp knitting machine work under each group of typical process parameters to measure dynamic tension data under each group of typical process parameters, and use a camera to obtain fabric image data, including normal fabric image data and defect image data; a hidden variable acquisition module configured to construct a variational autoencoder and input the normal fabric image data and the defect image data into the variational autoencoder to obtain normal fabric image hidden variable encoding and defect fabric image hidden variable encoding; a production quality state encoding module for generating fabric production quality state encoding according to normal fabric image latent variable encoding and defect fabric image latent variable encoding ; a prediction model construction module for constructing a prediction model based on the fabric production quality state code deriving a prediction fabric production quality state code ; a process parameter traversal module configured to set a process parameter interval range according to the performance of the fabric and the pattern requirement, traverse all process parameters that can be set in the parameter interval, obtain dynamic tension measurement values under all process parameters, and calculate a dynamic tension data index vector under all process parameters; Final process parameter selection module: for calculating the predicted fabric production quality state code under all process parameters and selecting the process parameter corresponding to the minimum production quality state code as the final production process parameter.

8. An electronic device comprising a memory and a processor, said memory having stored thereon a computer program, characterized in that, The processor executes the program to implement the method in any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method in any one of claims 1-6.

Images