A method for monitoring and regulating the production of textile yarns

Abstract

The present invention relates to a method for monitoring and regulating the production of textile yarns, which relates to the field of intelligent control of production lines and includes: based on predetermined yarn characteristics and predetermined spinning parameters, respectively performing differential characteristic analysis on multiple similar devices in multiple device domains to determine multiple high-low difference device domains; according to the predetermined yarn characteristics, respectively performing failure probability trend analysis on multiple high-low difference device domains, obtaining multiple sets of probability trend coefficients to compensate multiple reference failure rates, and obtaining multiple sets of adapted failure rates; configuring multiple sets of adapted monitoring parameters according to multiple sets of adapted failure rates, and performing production monitoring and regulation on multiple high-low difference device domains. Through the present application, the technical problem that the traditional method lacks the ability of dynamic adjustment and cannot set an adapted monitoring and regulation scheme according to the actual state of the device and the type of yarn product under the condition of limited monitoring resources, resulting in poor monitoring and regulation effects can be solved, the monitoring resource allocation can be optimized, and the monitoring and regulation effect of yarn production can be significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent production control of production lines, and particularly to a method for monitoring and regulating textile yarn production. Background Art

[0002] Traditional methods for monitoring and regulating textile yarn production usually adopt a full-coverage monitoring strategy, that is, each device in the workshop is monitored in real time. This method ignores the differences in actual production tasks and the changes in the operating status of devices. All devices are at the same monitoring level, resulting in some devices being over-monitored, while some key devices do not receive sufficient attention. Due to limited monitoring resources, this monitoring method often leads to low monitoring efficiency and fails to effectively detect potential quality problems or equipment failures in the production process.

[0003] Currently, with the expansion of production scale and the diversification of textile product types, the deficiencies of traditional monitoring methods have gradually emerged. Especially when facing a complex production environment, the lack of dynamic adjustment ability is particularly prominent.

[0004] In summary, traditional methods for monitoring and regulating yarn production lack dynamic adjustment ability and cannot set an appropriate monitoring and regulation scheme according to the actual state of devices and the types of yarn products under the condition of limited monitoring resources, resulting in the technical problem of poor monitoring and regulation effects. Summary of the Invention

[0005] In view of the technical problem that traditional methods for monitoring and regulating textile yarn production lack dynamic adjustment ability and cannot set an appropriate monitoring and regulation scheme according to the actual state of devices and the types of yarn products under the condition of limited monitoring resources, resulting in poor monitoring and regulation effects, the present invention provides a method for monitoring and regulating textile yarn production to solve this problem.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] In a first aspect, the present invention provides a method for monitoring and regulating textile yarn production, including: clustering spinning devices in a textile workshop according to predetermined device characteristics to determine multiple device domains, and respectively performing failure rate statistics on the multiple device domains to determine multiple baseline failure rates; performing differential feature analysis on multiple similar devices in the multiple device domains based on predetermined yarn characteristics and predetermined spinning parameters to determine multiple high-difference device domains; according to the predetermined yarn characteristics, respectively performing failure probability trend analysis on the multiple high-difference device domains to obtain multiple sets of probability trend coefficients, compensating the multiple baseline failure rates with the multiple sets of probability trend coefficients to obtain multiple sets of adapted failure rates; configuring multiple sets of adapted monitoring parameters according to the multiple sets of adapted failure rates, and based on the multiple sets of adapted monitoring parameters, performing production monitoring on the multiple high-difference device domains, and regulating abnormal monitoring devices according to a predetermined feedback regulation mechanism.

[0008] In a second aspect, the present invention further provides an electronic device, comprising:

[0009] At least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the steps of the method according to any one of the above first aspects.

[0010] The beneficial effects of the present invention are as follows: By clustering the spinning devices in the textile workshop according to the predetermined device characteristics, a plurality of device domains are determined, and the failure rates of the plurality of device domains are respectively statistically analyzed to determine a plurality of reference failure rates; then, based on the predetermined yarn characteristics and predetermined spinning parameters, differential characteristic analysis is respectively performed on a plurality of similar devices in the plurality of device domains to determine a plurality of high-difference device domains; then, according to the predetermined yarn characteristics, failure probability trend analysis is respectively performed on the plurality of high-difference device domains to obtain a plurality of probability trend coefficient sets to compensate the plurality of reference failure rates, thereby obtaining a plurality of adapted failure rate sets; finally, a plurality of adapted monitoring parameter sets are configured according to the plurality of adapted failure rate sets, and based on the plurality of adapted monitoring parameter sets, production monitoring is performed on the plurality of high-difference device domains, and abnormal monitoring devices are regulated according to a predetermined feedback regulation mechanism; that is to say, by introducing an intelligent monitoring and regulation mechanism, the monitoring scheme can be adaptively adjusted according to the real-time state of the device, production tasks, and yarn product characteristics, optimizing the allocation of monitoring resources, so that under the condition of limited monitoring resources, the monitoring and regulation effect of yarn production can be significantly improved, ensuring the high quality and consistency of yarn products. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a schematic flowchart of a method for monitoring and regulating textile yarn production provided by the present invention;

[0012] Figure 2 It is a schematic flowchart of determining a plurality of device domains in a method for monitoring and regulating textile yarn production provided by the present invention;

[0013] Figure 3 It is a schematic structural diagram of the electronic device provided by the present invention.

[0014] In the drawings, the components represented by the reference numerals are as follows:

[0015] Electronic device 500, memory 510, processor 520, first computer program 511. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0017] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0018] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0019] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides a method for monitoring and regulating the production of textile yarns, specifically including the following steps:

[0020] S100: Cluster the spinning equipment in the textile workshop according to the predetermined equipment characteristics, determine a plurality of equipment domains, and respectively perform failure rate statistics on the plurality of equipment domains to determine a plurality of reference failure rates.

[0021] In one embodiment, as Figure 2 shown, step S100 of this application further includes:

[0022] S110: Obtain predetermined device characteristics, where the predetermined device characteristics include device model, device service life, and device wear characteristics; S120: Perform a first clustering on the spinning devices in the textile workshop according to the device model to obtain a first clustering result; S130: Perform a second clustering on the first clustering result according to the device service life to obtain a second clustering result; S140: Perform a third clustering on the second clustering result according to the device wear characteristics to obtain the multiple device domains.

[0023] Specifically, first, obtain predetermined device characteristics. Among them, the predetermined device characteristics include device model, device service life, and device wear characteristics. The device model refers to the specific model or category of each spinning device. The device model usually determines the performance parameters, applicable range, and production capacity of the device and is an important feature for distinguishing different devices; the device service life refers to the cumulative working time or usage cycle elapsed since the device was put into use. The service life of the device can reflect the degree of aging of the device and the possible failure risks. Long-term use of the device may result in problems such as decreased efficiency and increased failure rate. Among them, the longer the service life of the device, the usually higher the potential failure probability. Therefore, as the device usage years increase, the required monitoring resources should also increase accordingly to ensure the normal operation of the device and timely detection of failures; the device wear characteristics refer to the wear conditions of the device during long-term operation due to factors such as friction and collision. These wears are usually reflected in the performance degradation of key components, such as the transmission system, mechanical gears, cutting tools, etc. Among them, the more severe the device wear, the more monitoring resources are required.

[0024] Then perform a first clustering on the spinning devices in the textile workshop according to the device model. At this stage, all devices are divided into several groups according to similar device models. Each group of devices has similar working principles and performance characteristics, and a first clustering result is obtained. Further perform a second clustering on the first clustering result according to the device service life, that is, by grouping the service life of the devices, the devices can be divided according to different aging stages (such as new devices, nearly new devices). For example, devices with a service life of 0 to 1000 hours are classified as new devices, and devices with a service life of 1000 to 3000 hours are classified as nearly new devices, etc., to obtain a second clustering result. Then perform a third clustering on the second clustering result according to the device wear characteristics. By analyzing the wear degree of each device, the device groups are further subdivided. The wear characteristics include the wear degree of mechanical components, the performance degradation of the transmission system, the wear of gears, etc., which all affect the normal operation of the device. Multiple device domains are obtained. Among them, within the same device domain, the devices have similar device models, functional characteristics, and performance performances. By uniformly managing the devices within the device domain, monitoring resources can be configured more efficiently, and the monitoring and control effect of yarn production can be improved.

[0025] Through multi-level clustering analysis, the equipment in the textile workshop can be more precisely grouped according to model, service life, and degree of wear, thereby providing a basis for subsequent production monitoring and fault prediction.

[0026] In one embodiment, step S100 of the present application further includes:

[0027] S150: Randomly select a first equipment domain, query the equipment operation logs of multiple devices in the first equipment domain, and obtain multiple fault counts and multiple operation durations of the multiple devices within a predetermined period; S160: Set the ratio of the sum of the multiple fault counts to the sum of the multiple operation durations as the first failure rate, perform normalization processing to obtain the first reference failure rate, and add it to the multiple reference failure rates.

[0028] Specifically, randomly select any one of the multiple equipment domains as the first equipment domain. Then, query the equipment operation logs of multiple devices in the first equipment domain to obtain multiple fault counts and multiple operation durations of the multiple devices within a predetermined period. The predetermined period can be set according to actual needs. For example, query the equipment operation logs in the past month. The fault count refers to the number of faults that occurred to the equipment within the predetermined period, and the operation duration refers to the total operation duration of the equipment within the predetermined period. Then, perform a summation calculation on the multiple fault counts and multiple operation durations to obtain the total fault count and total operation duration, and set the ratio of the total fault count to the total operation duration as the first failure rate. Further, for the convenience of comparison and analysis, perform normalization processing on the calculated first failure rate. The purpose of normalization is to scale the failure rate values between different equipment domains to a unified range. The failure rate after normalization processing is called the first reference failure rate, which represents the overall fault trend of the equipment within the first equipment domain, and add the first reference failure rate to the multiple reference failure rates.

[0029] By calculating the multiple reference failure rates of multiple equipment domains, it provides a basis for subsequent targeted monitoring of the equipment, and can customize appropriate monitoring strategies for each equipment domain. For example, equipment domains with higher failure rates may require more frequent monitoring and early warning, while equipment domains with lower failure rates can appropriately reduce the monitoring frequency. This dynamic adjustment method based on the failure rate can effectively optimize the allocation of monitoring resources and avoid resource waste.

[0030] S200: Based on predetermined yarn characteristics and predetermined spinning parameters, perform differential feature analysis on multiple similar devices in the multiple equipment domains respectively to determine multiple high-low difference equipment domains.

[0031] In one embodiment, step S200 of the present application further includes:

[0032] S210: Randomly select a first device domain, and read the predetermined yarn feature set and the predetermined spinning parameter set of multiple spinning devices in the first device domain within a future period. Among them, the yarn features include type, material, fineness, tensile strength, raw material cost, and spinning complexity, and the spinning parameters include twist, tension, spinning speed, and spinning temperature; S220: Randomly select a first spinning device from multiple spinning devices, where the first spinning device has a first yarn feature and a first spinning parameter; S230: Perform a similarity comparison between the first yarn feature and other yarn features in the predetermined yarn feature set, and count the number of features with a comprehensive similarity less than a predetermined similarity threshold, which is set as the first difference degree; S240: If the first difference degree is greater than a predetermined difference threshold, then perform a similarity comparison between the first spinning parameter and other spinning parameters in the predetermined spinning parameter set, and count the number of features with a comprehensive similarity less than a predetermined similarity threshold, which is set as the second difference degree; S250: If the second difference degree is greater than a predetermined difference threshold, then add the first spinning device to the first high-difference device domain; S260: Determine whether the number of devices in the first high-difference device domain is less than a predetermined scalar. If it is greater than or equal to, adjust the predetermined difference threshold for re-screening, where the predetermined scalar is one-tenth of the number of spinning devices in the first device domain; S270: If it is less, add the first high-difference device domain to the multiple high-difference device domains.

[0033] Specifically, randomly select a first device domain from the multiple device domains, and then read the predetermined yarn feature set and the predetermined spinning parameter set of multiple spinning devices in the first device domain within a future period (such as the next 24 hours). Among them, the yarn features include type (such as cotton yarn, wool yarn, chemical fiber yarn, etc.), material (such as cotton, wool, polyester, etc.), fineness (such as 20s, 40s, etc.), tensile strength, raw material cost, and spinning complexity, and the spinning parameters include twist, tension, spinning speed, and spinning temperature.

[0034] Then randomly select any device in the multiple spinning devices of the first device domain as the first spinning device, where the first spinning device has a first yarn feature and a first spinning parameter. Then perform a similarity comparison between the first yarn feature and other yarn features in the predetermined yarn feature set to obtain multiple comprehensive similarities. Among them, the comprehensive similarity refers to the average value of the similarity of multiple attribute features (type, material, fineness, tensile strength, raw material cost, and spinning complexity) in the yarn feature; further count the number of features with a comprehensive similarity less than a predetermined similarity threshold, which is set as the first difference degree, and the predetermined similarity threshold can be set according to the actual scenario, such as setting the predetermined similarity threshold to 50%.

[0035] Judge the first difference degree according to a predetermined difference threshold. If the first difference degree is greater than the predetermined difference threshold, compare the first spinning parameter with other spinning parameters in the predetermined spinning parameter set, and count the number of features whose comprehensive similarity of spinning parameters is less than the predetermined similarity threshold, which is set as the second difference degree. If the second difference degree is greater than the predetermined difference threshold, add the first spinning device to the first high-difference device domain, where a high-difference device refers to a device with a large overall difference in yarn characteristics and spinning parameters compared with the same type of devices in the device domain. Further judge whether the number of devices in the first high-difference device domain is less than a predetermined scalar, where the predetermined scalar is one-tenth of the number of spinning devices in the first device domain. If the number of devices in the first high-difference device domain is greater than or equal to the predetermined scalar, it is necessary to appropriately increase the predetermined difference threshold and perform re-screening; if the number of devices in the first high-difference device domain is less than the predetermined scalar, add the first high-difference device domain to the multiple high-difference device domains.

[0036] S300: According to the predetermined yarn characteristics, perform a failure probability trend analysis on the multiple high-difference device domains respectively, obtain multiple probability trend coefficient sets to compensate the multiple reference failure rates, and obtain multiple adapted failure rate sets.

[0037] In one embodiment, step S300 of the present application further includes:

[0038] S310: Randomly select a first high-difference device domain, and select any device in the first high-difference device domain as the first high-difference device.

[0039] Specifically, randomly select any one of the multiple high-difference device domains as the first high-difference device domain, and select any device in the first high-difference device domain as the first high-difference device.

[0040] S320: Obtain the first high-difference yarn characteristics of the first high-difference device, input the first high-difference yarn characteristics into a failure probability predictor for failure probability trend analysis, output a first probability trend coefficient, and add it to the first probability trend coefficient set.

[0041] In one embodiment, step S320 of the present application further includes:

[0042] S321: Query the historical spinning operation records, obtain a sample yarn characteristic set, and count the average failure probabilities of multiple spinning devices for different sample yarn characteristics within a predetermined historical period, which are set as sample probability trend coefficients, to obtain a sample probability trend coefficient set; S322: Use the sample yarn characteristics as the input and the sample probability trend coefficients as the output, and use the sample yarn characteristic set and the sample probability trend coefficient set as training data to perform supervised training on the BP neural network until convergence to obtain the failure probability predictor.

[0043] Specifically, first, query the historical spinning operation records, extract the yarn characteristic data in different production cycles from the historical spinning records, and obtain the sample yarn characteristic set; then, count the average values of the failure probabilities of multiple spinning devices for different sample yarn characteristics within a predetermined historical period, that is, for each sample yarn characteristic, count the failure probabilities of different spinning devices within the predetermined historical period, and calculate the average value of multiple failure probabilities, which is set as the sample probability trend coefficient, to obtain the sample probability trend coefficient set, where the sample yarn characteristics and the sample probability trend coefficients correspond one by one.

[0044] Next, construct a failure probability predictor based on the BP neural network. The failure probability predictor is a neural network model in machine learning that can be iteratively optimized, including an input layer, a hidden layer, and an output layer. The input data of the input layer is the sample yarn characteristics, and the output data is the sample probability trend coefficient; then, using the sample yarn characteristics as the input and the sample probability trend coefficient as the output, and using the sample yarn characteristic set and the sample probability trend coefficient set as the training data, perform supervised training on the failure probability predictor. In each training, input the sample yarn characteristics in the training set into the neural network. After the calculations of the input layer and the hidden layer, the output layer generates a predicted value (i.e., the predicted value of the trend coefficient of the failure probability); then use the mean square error (MSE) or other appropriate loss function to calculate the error between the network prediction output and the actual sample probability trend coefficient; the backpropagation algorithm calculates the gradient of each parameter (weight and bias) according to the loss function, uses the chain rule to gradually propagate the error from the output layer to the input layer, calculates the gradient of each layer, and updates the weights and biases; use gradient descent to update the weights and biases in the network. The update process continuously adjusts the parameters of the network, making the prediction error gradually decrease. Through multiple rounds of iteration, gradually optimize the network parameters to reduce the loss function value and improve the prediction accuracy, and continuously perform iterative training until reaching the predetermined number of iterations or the loss function converges to a certain threshold, to obtain the trained failure probability predictor.

[0045] Obtain the first high-difference yarn characteristics of the first high-difference device, where the first high-difference yarn characteristics include the first type, the first material, the first fineness, the first tensile strength, the first raw material cost, and the first spinning complexity; then input the first high-difference yarn characteristics into the failure probability predictor for failure probability trend analysis, output the first probability trend coefficient, and add it to the first probability trend coefficient set. By constructing a failure probability predictor based on the BP neural network for failure probability trend analysis, the efficiency and accuracy of the probability trend can be improved.

[0046] In one embodiment, step S300 of the present application further includes:

[0047] S330: Use 1 plus the probability trend coefficient as the compensation coefficient, and calculate multiple sets of compensation coefficients based on the multiple sets of probability trend coefficients; S340: Perform a product operation on the multiple benchmark failure rates according to the multiple sets of compensation coefficients to obtain the multiple sets of adapted failure rates.

[0048] Specifically, first, use 1 plus the probability trend coefficient as the compensation coefficient, perform a summation operation on the multiple sets of probability trend coefficients and 1 to obtain multiple sets of compensation coefficients. Then perform a product operation on the multiple benchmark failure rates according to the multiple sets of compensation coefficients, and use the product of the two as the adapted failure rate to obtain multiple sets of adapted failure rates. By performing a failure probability trend analysis based on the sample yarn characteristics and compensating the benchmark failure rate according to the probability trend coefficient to obtain the set of adapted failure rates, the scientificity, accuracy, and reliability of the adapted failure rate setting can be improved, thereby providing a more accurate and scientific basis for subsequent production monitoring, and helping to improve the accuracy of equipment failure prediction and the utilization efficiency of monitoring resources.

[0049] S400: Configure multiple sets of adapted monitoring parameters according to the multiple sets of adapted failure rates, and based on the multiple sets of adapted monitoring parameters, perform production monitoring on the multiple height difference equipment domains, and regulate the abnormal monitoring equipment according to a predetermined feedback regulation mechanism.

[0050] In one embodiment, step S400 of the present application further includes:

[0051] S410: Obtain the initial monitoring parameters of the spinning equipment, where the initial monitoring parameters include the initial monitoring frequency, the initial warning threshold, and the initial monitoring model complexity. The monitoring model is a machine learning model for abnormal monitoring of the operation process of the spinning equipment, and the monitoring model complexity is the number of analysis units of the monitoring model; S420: According to a predetermined adjustment plan, perform proportional optimization adjustment on the initial monitoring frequency, the initial warning threshold, and the initial monitoring model complexity according to the multiple sets of adapted failure rates to obtain the multiple sets of adapted monitoring parameters, where the adapted failure rate is positively correlated with the adapted monitoring frequency, the adapted warning threshold, and the adapted monitoring model complexity.

[0052] Specifically, first, obtain the initial monitoring parameters of the spinning equipment. The initial monitoring parameters include the initial monitoring frequency, the initial warning threshold, and the initial monitoring model complexity. The initial monitoring frequency refers to the frequency of monitoring data collection and analysis during the operation of the spinning equipment. It determines how often the system collects data for analysis within a certain time interval to promptly detect abnormalities in the equipment operation and can be set according to the actual scenario. For example, set the initial monitoring frequency to once per minute. The initial warning threshold means that during the operation of the equipment, if the monitoring data exceeds this threshold, the system will issue a warning indicating that the equipment may have a fault or abnormality. For example, the quality warning threshold of the yarn or the warning threshold of the spinning parameters. The quality warning threshold of the yarn is used to monitor and control the key quality indicators of the yarn (such as fineness, tensile strength, uniformity, etc.) to prevent yarns that do not meet the quality standards from entering the production process, thereby ensuring the qualification rate of the yarn products.

[0053] Among them, the monitoring model is a machine learning model for abnormal monitoring of the operation process of the spinning equipment, such as a yarn quality monitoring model or a spinning parameter monitoring model. Multiple machine learning models can be selected for abnormal monitoring of the spinning equipment, such as decision trees, random forests, neural networks, etc. Which model to choose specifically depends on the characteristics of the equipment data and the monitoring requirements. The monitoring model complexity is the number of analysis units of the monitoring model. Among them, the more the number of analysis units, the higher the monitoring model complexity. The higher the model complexity, the more equipment operation characteristics can be captured and analyzed, and the higher the accuracy may be, but the corresponding computing and storage requirements will also increase.

[0054] According to the predetermined adjustment plan, proportionally optimize and adjust the initial monitoring frequency, the initial warning threshold, and the initial monitoring model complexity according to the multiple sets of adapted failure rates. Among them, there is a positive correlation between the adapted failure rate and the monitoring parameters (monitoring frequency, warning threshold, monitoring model complexity), that is, as the failure rate increases, the corresponding monitoring frequency, warning threshold, and monitoring model complexity also need to increase accordingly to ensure more accurate and timely monitoring of the equipment. The predetermined adjustment plan includes the step size of each adjustment, that is, the proportional amplitude of each adjustment, including the monitoring frequency step size, the warning threshold step size, and the model complexity step size, to obtain multiple sets of adjusted adapted monitoring parameters. The adapted monitoring parameters include the adapted monitoring frequency, the adapted warning threshold, and the adapted monitoring model complexity. Finally, based on the multiple sets of adapted monitoring parameters, conduct production monitoring on the multiple height difference equipment domains.

[0055] Through a predetermined adjustment scheme, combined with a step size control mechanism, it is possible to dynamically optimize the monitoring frequency, warning threshold, and monitoring model complexity by reasonably adjusting the step size under the condition of limited monitoring resources. This can not only improve the monitoring efficiency of spinning equipment, reduce unnecessary resource waste, but also enhance the early warning ability for faults, thus ensuring the efficient and stable operation of the equipment.

[0056] In one embodiment, step S400 of the present application further includes:

[0057] S430: Establish a first mapping between the device domain and the elevation difference device domain; S440: Randomly select a first elevation difference device domain, and based on the first mapping, obtain the first mapped device domain of the first elevation difference device domain; S450: Count the abnormal types and abnormal proportions of abnormal monitoring devices in the first elevation difference device domain within a predetermined time window. If the abnormal proportion of the same abnormal type is greater than the first predetermined threshold, then regulate all spinning devices in the first mapped device domain according to a predetermined feedback regulation mechanism.

[0058] Specifically, first, establish a first mapping relationship between the device domain and the elevation difference device domain; then randomly select a first elevation difference device domain from multiple elevation difference device domains, and based on the first mapping, obtain the first mapped device domain of the first elevation difference device domain. Then, during the production monitoring of the multiple elevation difference device domains, within the selected time window, count the abnormal types and their occurrence proportions of all abnormal monitoring devices in the first elevation difference device domain. The abnormal types may include equipment failures, performance degradation, process abnormalities, etc. When the abnormal proportion of the same abnormal type is greater than the first predetermined threshold (such as 50%), it indicates that this abnormal type has a high occurrence frequency in this device domain, and then regulate all spinning devices in the first mapped device domain according to a predetermined feedback regulation mechanism. Among them, the predetermined feedback regulation mechanism includes various regulation schemes for abnormal states, that is, when an abnormality occurs in the production process, corresponding adjustment and processing schemes are made according to preset rules, which can be set according to historical abnormal maintenance records.

[0059] A textile yarn production monitoring and regulation method provided by an embodiment of the present invention has at least the following technical effects:

[0060] By clustering the spinning equipment in the textile workshop according to the predetermined equipment characteristics, multiple equipment domains are determined, and the failure rates of the multiple equipment domains are respectively counted to determine multiple baseline failure rates; then, based on the predetermined yarn characteristics and predetermined spinning parameters, differential characteristic analysis is respectively performed on multiple similar equipment in the multiple equipment domains to determine multiple height-difference equipment domains; then, according to the predetermined yarn characteristics, failure probability trend analysis is respectively performed on the multiple height-difference equipment domains to obtain multiple probability trend coefficient sets to compensate the multiple baseline failure rates, and multiple adapted failure rate sets are obtained; finally, multiple adapted monitoring parameter sets are configured according to the multiple adapted failure rate sets, and based on the multiple adapted monitoring parameter sets, production monitoring is performed on the multiple height-difference equipment domains, and abnormal monitoring equipment is regulated according to the predetermined feedback regulation mechanism; that is to say, by introducing an intelligent monitoring and regulation mechanism, the monitoring scheme can be adaptively adjusted according to the real-time state of the equipment, production tasks, and yarn product characteristics, and the allocation of monitoring resources can be optimized, so that under the condition of limited monitoring resources, the monitoring and regulation effect of yarn production can be significantly improved, and the high quality and consistency of yarn products can be ensured.

[0061] Embodiment 2. Please refer to Figure 3 , Figure 3 which is a schematic diagram of an embodiment of the electronic device provided by the embodiment of the present invention. As Figure 3 shown, the embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a first computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the first computer program 511, the following steps are implemented: clustering the spinning equipment in the textile workshop according to the predetermined equipment characteristics, determining multiple equipment domains, and respectively counting the failure rates of the multiple equipment domains to determine multiple baseline failure rates; based on the predetermined yarn characteristics and predetermined spinning parameters, performing differential characteristic analysis on multiple similar equipment in the multiple equipment domains respectively to determine multiple height-difference equipment domains; according to the predetermined yarn characteristics, performing failure probability trend analysis on the multiple height-difference equipment domains respectively to obtain multiple probability trend coefficient sets to compensate the multiple baseline failure rates, and obtaining multiple adapted failure rate sets; configuring multiple adapted monitoring parameter sets according to the multiple adapted failure rate sets, and based on the multiple adapted monitoring parameter sets, performing production monitoring on the multiple height-difference equipment domains, and regulating abnormal monitoring equipment according to the predetermined feedback regulation mechanism.

[0062] It should be noted that in the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0063] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0064] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one or more of the processes Figure 1 or blocks.

[0065] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realizes the functions specified in Figure 1 one or more of the processes Figure 1 or blocks.

[0066] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one or more of the processes Figure 1 or blocks.

[0067] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic inventive concept.

[0068] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for monitoring and controlling textile yarn production, characterized in that: Methods include: Clustering the spinning equipment in the textile workshop according to predetermined equipment characteristics to determine multiple equipment domains, and performing failure rate statistics on the multiple equipment domains respectively to determine multiple benchmark failure rates; Based on the predetermined yarn characteristics and the predetermined spinning parameters, respectively performing difference characteristic analysis on a plurality of similar devices in the plurality of device domains to determine a plurality of high-difference device domains; According to the predetermined yarn characteristics, respectively performing fault probability trend analysis on the multiple height difference equipment domains, obtaining multiple probability trend coefficient sets to compensate the multiple reference fault rates, and obtaining multiple adaptation fault rate sets; configuring a plurality of adaptive monitoring parameter sets according to the plurality of adaptive failure rate sets, performing production monitoring on the plurality of height difference equipment domains based on the plurality of adaptive monitoring parameter sets, and regulating the abnormal monitoring equipment according to a predetermined feedback regulation mechanism; According to the predetermined yarn characteristics, the multiple height difference equipment domains are respectively subjected to fault probability trend analysis, and multiple probability trend coefficient sets are obtained to compensate the multiple benchmark failure rates, thereby obtaining multiple adaptation failure rate sets, including: Randomly select a first height difference device domain, and select any device in the first height difference device domain as the first height difference device; Acquire a first height difference yarn feature of the first height difference device, input the first height difference yarn feature into a fault probability predictor to perform fault probability trend analysis, output a first probability trend coefficient, and add the coefficient to a first probability trend coefficient set; Taking 1 plus the probability trend coefficient as the compensation coefficient, a plurality of compensation coefficient sets are calculated according to the plurality of probability trend coefficient sets; Performing a product operation on the multiple reference failure rates according to the multiple compensation coefficient sets to obtain the multiple adaptation failure rate sets; Among them, building a fault probability predictor includes: Query historical spinning operation records to obtain a sample yarn feature set, and statistically calculate the mean value of the failure probability of multiple spinning equipment with different sample yarn features within a predetermined historical period as a sample probability trend coefficient to obtain a sample probability trend coefficient set; The sample yarn features are used as input, the sample probability trend coefficient is used as output, the sample yarn feature set and the sample probability trend coefficient set are used as training data, and the BP neural network is supervised and trained until convergence, so as to obtain the fault probability predictor.

2. A textile yarn production monitoring and control method according to claim 1, characterized in that: The spinning equipment in the textile workshop is clustered according to the predetermined equipment characteristics to determine multiple equipment domains, including: Acquiring predetermined equipment characteristics, wherein the predetermined equipment characteristics include equipment model, equipment service life, and equipment wear characteristics; Clustering the spinning equipment in the textile workshop according to the equipment model to obtain a clustering result; Performing secondary clustering on the primary clustering result according to the service life of the equipment to obtain a secondary clustering result; The secondary clustering result is clustered thirdly according to the equipment wear characteristics to obtain the multiple equipment domains.

3. A textile yarn production monitoring and control method according to claim 1, characterized in that: Performing failure rate statistics on the multiple device domains respectively to determine multiple benchmark failure rates includes: Randomly selecting a first device domain, querying device operation logs of multiple devices in the first device domain, and obtaining multiple fault numbers and multiple operation durations of the multiple devices within a predetermined period; The ratio of the sum of the multiple fault numbers to the sum of the multiple operating time periods is set as a first failure rate, and a first reference failure rate is obtained after normalization processing, and added to the multiple reference failure rates.

4. A textile yarn production monitoring and control method according to claim 1, characterized in that: Based on the predetermined yarn characteristics and the predetermined spinning parameters, a plurality of similar devices in the plurality of device domains are respectively analyzed for difference characteristics to determine a plurality of high-difference device domains, including: Randomly select a first device domain, and read a predetermined yarn feature set and a predetermined spinning parameter set of multiple spinning devices in the first device domain in a future period, wherein the yarn features include type, material, fineness, tensile strength, raw material cost and spinning complexity, and the spinning parameters include twist, tension, spinning speed and spinning temperature; Randomly selecting a first spinning device from a plurality of spinning devices, wherein the first spinning device has a first yarn characteristic and a first spinning parameter; Performing a similarity comparison between the first yarn feature and other yarn features in the predetermined yarn feature set, and counting the number of features whose comprehensive similarity is less than a predetermined similarity threshold, as a first difference degree; If the first difference is greater than a predetermined difference threshold, the first spinning parameter is compared with other spinning parameters in the predetermined spinning parameter set for similarity, and the number of features whose comprehensive similarity is less than the predetermined similarity threshold is counted and set as a second difference; If the second difference is greater than a predetermined difference threshold, adding the first spinning device to the first height difference device domain; Determine whether the number of devices in the first height difference device domain is less than a predetermined scalar, and if it is greater than or equal to, adjust the predetermined difference threshold to perform another screening, wherein the predetermined scalar is one tenth of the number of spinning devices in the first device domain; If it is less, the first height difference device domain is added to the multiple height difference device domains.

5. A textile yarn production monitoring and control method according to claim 1, characterized in that: Configuring multiple adaptation monitoring parameter sets according to the multiple adaptation failure rate sets includes: Obtaining initial monitoring parameters of the spinning equipment, wherein the initial monitoring parameters include an initial monitoring frequency, an initial warning threshold, and an initial monitoring model complexity, the monitoring model is a machine learning model for abnormal monitoring of the operation process of the spinning equipment, and the monitoring model complexity is the number of analysis units of the monitoring model; According to a predetermined adjustment plan, the initial monitoring frequency, initial warning threshold and initial monitoring model complexity are proportionally optimized and adjusted according to the multiple adaptation failure rate sets to obtain the multiple adaptation monitoring parameter sets, wherein the adaptation failure rate and adaptation monitoring frequency, the adaptation warning threshold and adaptation monitoring model complexity are positively correlated.

6. A textile yarn production monitoring and control method according to claim 1, characterized in that: The abnormal monitoring equipment is regulated according to the predetermined feedback control mechanism, including: Establishing a first mapping between the device domain and the height difference device domain; Randomly select a first height difference device domain, and based on the first mapping, obtain a first mapping device domain of the first height difference device domain; The abnormal types and abnormal proportions of abnormal monitoring equipment in the first height difference equipment domain within a predetermined time window are counted; if the abnormal proportion of the same abnormal type is greater than a first predetermined threshold, all spinning equipment in the first mapping equipment domain is regulated according to a predetermined feedback control mechanism.

7. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the steps of a textile yarn production monitoring and control method as described in any one of claims 1 to 6.

Images