A yarn quality monitoring system

Abstract

This invention relates to the field of yarn quality technology, specifically to a yarn quality monitoring system, comprising a sensing layer, a processing layer, and an application layer. The sensing layer collects yarn data; the processing layer evaluates yarn quality through yarn data analysis and optimizes yarn process parameters by training a process parameter optimization model using historical data; the application layer generates process adjustment suggestions and records the execution results. This system achieves closed-loop monitoring and feedback throughout the entire process, improving production efficiency, reducing costs, ensuring product stability, and promoting the intelligent transformation of the textile industry.

Description

Technical Field

[0001] This invention relates to the field of yarn quality technology, specifically to a yarn quality monitoring system. Background Technology

[0002] The variety of yarn quality parameters and data sources makes yarn quality monitoring dynamic, complex, and uncertain. This can easily lead to a lag between yarn quality data collection, yarn quality assessment, and process adjustments, with the lag often ranging from hours to days. As a result, a large number of defective products have already entered the next process, causing waste.

[0003] Furthermore, traditional methods cannot achieve real-time recording and traceability of data throughout the entire process, making it difficult to systematically analyze the root causes of quality problems.

[0004] The yarn quality monitoring system provided in this solution enables closed-loop monitoring and feedback throughout the entire process, allowing for real-time detection of all samples, avoiding missed detections and shortening the feedback cycle. Summary of the Invention

[0005] (a) Technical issues The present invention aims to at least solve the problems existing in the prior art.

[0006] (II) Technical Content This solution provides a yarn quality monitoring system, which is achieved through the following specific technical means, including a sensing layer, a processing layer, and an application layer; The sensing layer is used to collect yarn data; The processing layer evaluates yarn quality through yarn data analysis and optimizes yarn process parameters by training a process parameter optimization model using historical data. The application layer is used to generate process adjustment suggestions and record the execution results.

[0007] Preferred technical solution 1: The sensing layer includes a yarn physical parameter acquisition module, a yarn appearance parameter acquisition module, and an environmental parameter acquisition module; The yarn physical parameter acquisition module is used to collect yarn diameter, tension, and twist. The yarn appearance parameter acquisition module is used to collect defect types (thickness, thinness, neps, fuzz, etc.), color difference, and surface texture. The environmental parameter acquisition module is used to collect information on workshop temperature and humidity, and workshop dust concentration.

[0008] Preferred technical solution 2: The processing layer contains a defect classification model, which is connected to the yarn appearance parameter acquisition module to perform image recognition on the acquired images and realize the classification of defects; The processing layer contains databases, including time-series databases and relational databases; Time series database: stores real-time parameters; Relational database: stores defect records; The processing layer also includes a yarn quality assessment model, which takes into account real-time parameters and parameter fluctuation trends within a set time period, and outputs a quality score.

[0009] Preferred technical solution three: The yarn quality assessment model is constructed by acquiring a multimodal fusion model, which fuses time-series data with image features.

[0010] Preferred technical solution four: The process parameter optimization model is based on random forest to simulate the virtual quality changes after process adjustment.

[0011] Preferred technical solution five: The application layer provides adjustment strategies and connects to the spinning equipment to control the spinning equipment parameters to execute the adjustment strategies.

[0012] Preferred technical solution six: The processing layer also includes a yarn quality prediction model, which takes yarn quality indicators, production process parameters and environmental data as input, and outputs yarn quality trend prediction and early warning.

[0013] (III) Technical Effects The above structure gives this solution the following advantages: 1. Improved efficiency: Reduced defect detection rate and shorter response time for process adjustments; 2. Cost savings: Reduced waste yarn rate, saving millions of dollars in production costs annually (based on a scale of 100,000 spindles). 3. Level of intelligence: Achieving a paradigm shift from "post-event inspection" to "pre-event prevention". Attached Figure Description

[0014] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is the logic block diagram of this solution; Figure 2 This is the system architecture diagram for this solution.

[0015] The layers are: 1. Perception layer, 2. Transmission layer, 3. Processing layer, and 4. Application layer. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0017] Please see Figures 1-2 The yarn quality monitoring system includes a sensing layer 1, a transmission layer 2, a processing layer 3, and an application layer 4 connected in sequence. Sensing layer 1 is used to collect yarn data; The transmission layer 2 contains a signal converter, which is connected to various data acquisition sensors in the perception layer 1. The signal converter unifies the signals from the various sensors and feeds them back to the processing layer 3. Processing layer 3 evaluates yarn quality through yarn data analysis and optimizes yarn process parameters by training a process parameter optimization model using historical data. Application layer 4 is used to generate adjustment strategies, connect to the spinning equipment, control the spinning equipment parameters to execute the adjustment plan, and record the execution results.

[0018] Please see Figure 2 The yarn quality monitoring system includes a yarn physical parameter acquisition module, a yarn appearance parameter acquisition module, and an environmental parameter acquisition module in the sensing layer 1. The yarn physical parameter acquisition module is used to acquire yarn diameter (laser diameter gauge), tension (strain gauge sensor), and twist (photoelectric encoder). The yarn appearance parameter acquisition module is used to collect defect types (thick spots, thin spots, neps, fuzz, etc.) (industrial camera), color difference (spectrometer), and surface texture (X-ray / infrared imaging). The environmental parameter acquisition module is used to collect workshop temperature and humidity (temperature and humidity sensor) and workshop dust concentration (PM2.5 sensor). Data acquisition method: Sensor arrays are deployed at key nodes of the spinning equipment (such as the drafting zone, twisting zone, and winding zone); high-speed industrial cameras (gigabit network interface) capture real-time images of the yarn surface with a resolution ≥0.01mm / pixel; data acquisition frequency: physical parameters ≥1kHz, image parameters ≥100Hz.

[0019] Please see Figure 2 The yarn quality monitoring system has a defect classification model in the processing layer 3, which is connected to the yarn appearance parameter acquisition module to perform image recognition on the acquired images and realize the classification of defects. The defect classification model is based on the YOLOv5 classification algorithm; The processing layer 3 contains databases, including time-series databases and relational databases; Time-series database: Stores real-time parameters (timestamp + value), that is, real-time collected yarn physical parameters, environmental parameters and yarn appearance parameters after image recognition are stored in the time-series database; Relational database: Stores defect records (location, type, grade, associated process parameters); The processing layer 3 also includes a yarn quality assessment model. Input real-time parameters and parameter fluctuation trends within a set time period, and output a quality score. The quality grading rules in the assessment are formulated according to the quality grade regulations for different yarn varieties. The yarn quality assessment model is constructed by collecting a multimodal fusion model that integrates time-series data with image features.

[0020] The process parameter optimization model is built based on random forest to simulate the virtual quality changes after process adjustment. Based on the sensitivity of yarn quality, sensitive parameters are adjusted first (the key process parameters involved include draft ratio, roller spacing, twist setting, traveler weight, and temperature and humidity settings). The processing layer 3 also includes a yarn quality prediction model. It takes yarn quality indicators, production process parameters and environmental data as input, and outputs yarn quality trend prediction and early warning (the prediction method includes short-term prediction and medium-to-long-term prediction, and the early warning threshold can be dynamically adjusted according to historical data, such as triggering an alarm when the evenness CV value exceeds the historical average +2σ; and dynamically corrects the prediction results in combination with the status of yarn production equipment). in: Basic indicators: yarn strength (single strength / breaking strength), yarn evenness (CV value), hairiness index, number of thick / thin / neps, twist unevenness, etc. Production process parameters: draw ratio, roller spacing, temperature and humidity, equipment speed, raw material batch characteristics (fiber length / strength), etc. Environmental data: workshop temperature and humidity, dust concentration, power stability, etc.; The input data for yarn quality prediction models needs to undergo data cleaning, which involves removing outliers (such as data during equipment failures), filling in missing values ​​(through interpolation or mean filling), and standardizing the data (such as Z-score normalization) to eliminate the influence of dimensions.

[0021] The parts not disclosed in this invention are all prior art, and their specific structures and working principles will not be described in detail.

[0022] Unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0023] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A yarn quality monitoring system, characterized in that: It includes a perception layer (1), a processing layer (3), and an application layer (4); The sensing layer (1) is used to collect yarn data; The processing layer (3) evaluates the yarn quality through yarn data analysis and optimizes the yarn process parameters by training a process parameter optimization model using historical data. The application layer (4) is used to generate process adjustment suggestions and record the execution results.

2. The yarn quality monitoring system according to claim 1, characterized in that: The sensing layer (1) includes a yarn physical parameter acquisition module, a yarn appearance parameter acquisition module, and an environmental parameter acquisition module; The yarn physical parameter acquisition module is used to acquire yarn diameter, tension, and twist. The yarn appearance parameter acquisition module is used to collect defect type, color difference, and surface texture; The environmental parameter acquisition module is used to collect information on workshop temperature and humidity, and workshop dust concentration.

3. The yarn quality monitoring system according to claim 2, characterized in that: The processing layer (3) contains a defect classification model, which is connected to the yarn appearance parameter acquisition module. It performs image recognition on the acquired images to classify defects. The processing layer (3) contains databases, including time-series databases and relational databases.

4. The yarn quality monitoring system according to claim 3, characterized in that: The processing layer (3) also includes a yarn quality assessment model, which takes into account real-time parameters and parameter fluctuation trends within a set time period, and outputs a quality score.

5. A yarn quality monitoring system according to claim 3, characterized in that: The process parameter optimization model simulates the virtual quality changes after process adjustments.

6. A yarn quality monitoring system according to claim 2, characterized in that: The processing layer (3) also includes a yarn quality prediction model, which takes yarn quality indicators, production process parameters and environmental data as input, and outputs yarn quality trend prediction and early warning.

7. A yarn quality monitoring system according to claim 6, characterized in that: The input data for the yarn quality prediction model needs to undergo data cleaning and standardization.

8. A yarn quality monitoring system according to claim 1, characterized in that: The application layer (4) provides adjustment strategies and connects to the spinning equipment to control the spinning equipment parameters to execute adjustment schemes.

9. A yarn quality monitoring system according to claim 2, characterized in that: It also includes a transmission layer (2), which contains a signal converter. The signal converter is connected to various data acquisition sensors in the perception layer (1) and feeds back the signals from the various sensors to the processing layer (3) after unifying them.

Images