Method for tracing quality of high polymer material in down garment based on block chain

Abstract

The invention provides a method for tracing the quality of a high polymer material in down garments based on a block chain. The method belongs to the technical field of clothing quality tracing and information management. The method comprises the following steps: performing Internet of Things tagging on all high polymer raw materials, creating digital assets for each batch of raw materials by utilizing a block chain technology, and uploading key data to a block chain to form a raw material file; related data of the production line are monitored in real time, and the related data are uploaded to the cloud platform in real time through the Internet of Things gateway; a unique logistics label is distributed for finished down garments through RFID and GPS, logistics information is recorded, and the logistics information is linked with raw material archives and production records on a block chain to form a complete supply chain view. Through the block chain technology, the whole process of all high polymer raw materials from purchasing to finished down garments is recorded on the block chain, and a data chain which cannot be tampered is formed.

Description

Technical Field

[0001] The present invention proposes a quality tracing method for polymer materials in down garments based on blockchain, which belongs to the technical field of garment quality tracing and information management. Background Art

[0002] Traditional down clothing quality traceability systems mostly rely on centralized databases, which have problems such as data silos, easy tampering, and low traceability efficiency. With consumers paying more and more attention to product quality, safety, and environmental protection attributes, it is particularly urgent to build a system that can both guarantee data authenticity and provide in-depth traceability information. At the same time, the existing technology lacks detailed information on environmental parameters, energy consumption levels, and other details of polymer materials during the production process, which limits the comprehensive evaluation of product quality. Summary of the invention

[0003] The present invention provides a quality tracing method for polymer materials in down garments based on blockchain to solve the problems mentioned in the above background technology: The present invention proposes a blockchain-based high polymer material quality tracing method for down clothing, the method comprising: S1. Label all polymer raw materials with the Internet of Things, use blockchain technology to create digital assets for each batch of raw materials, upload key data to the blockchain, and form a raw material archive; S2, real-time monitoring of production line related data, and uploading the relevant data to the cloud platform through the IoT gateway; S3. Assign unique logistics tags to finished down garments through RFID and GPS, record logistics information, and link the logistics information with the raw material archives and production records on the blockchain to form a complete supply chain view; S4. Access the blockchain platform through the QR code on the clothing or enter the unique ID to view the complete traceability chain; S5. Based on big data analysis technology, potential risk points in the supply chain are explored, and possible quality problems that may arise in the future are predicted based on historical data through machine learning models.

[0004] Beneficial effects of the present invention: Through blockchain technology, the entire process of all polymer raw materials from procurement to finished down clothing is recorded on the blockchain, forming an unalterable data chain. This makes every link of the supply chain highly transparent, and consumers and regulators can easily access complete product traceability information by scanning a QR code or entering a unique ID. This transparent traceability system helps to enhance consumer trust and reduce the circulation of counterfeit and shoddy products. IoT sensors and smart contracts are used to monitor various data of production lines and logistics links in real time, and machine learning models are combined to perform data analysis and risk warning. Once an abnormal situation is detected, the system will automatically trigger the early warning mechanism and lock the relevant batches to ensure that the problem can be discovered and handled in time, thereby greatly improving the quality monitoring level of raw materials and products. Through smart contracts and big data analysis technology, abnormal patterns in the production process can be automatically identified, quality problems that may occur in the future can be predicted, and corresponding emergency response plans can be initiated. This not only helps to reduce waste and losses in the production process, but also improves the production efficiency and management level of the enterprise. At the same time, the path optimization algorithm based on historical data and real-time information can automatically plan the optimal logistics path, reduce transportation costs, and improve logistics efficiency. By embedding augmented reality technology in the QR code, consumers can view detailed information about the clothing, including the production process, the source of raw materials, etc., through the mobile phone camera. This not only enhances the consumer's shopping experience, but also provides more transparency of product information. In addition, through the interactive traceability platform, consumers can gain an in-depth understanding of the complete traceability chain of clothing, increasing the interaction and trust between brands and consumers. The technical solution is based on a multi-party collaboration mechanism, allowing upstream and downstream enterprises in the supply chain to jointly participate in the formulation and execution of smart contracts. This collaborative mechanism not only promotes the cooperation of all parties in the supply chain, but also integrates the resources of all parties to jointly respond to risks and challenges in the supply chain, and enhances the stability and risk resistance of the entire supply chain. Through the self-learning and optimization mechanism of the machine learning model, the system can continuously train and optimize the model based on the newly collected data, improve the prediction accuracy and decision support capabilities. At the same time, a feedback mechanism is established to collect data feedback and user evaluations during the actual transportation process, continuously optimize the path planning algorithm, and ensure the efficiency and reliability of the logistics process. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 This is a step diagram of the method described in the present invention. DETAILED DESCRIPTION

[0006] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0007] One embodiment of the present invention, as Figure 1As shown, a quality traceability method for polymer materials in down clothing based on blockchain includes: S1. All polymer raw materials are labeled by IoT. Each label contains basic information such as unique identifier (UID), material type, origin, batch number, etc.; using blockchain technology, a digital asset is created for each batch of raw materials, and key data is uploaded to the blockchain. The key data includes basic information and the first quality inspection report; forming an unalterable raw material archive; S2. Deploy sensors and smart devices on the production line to monitor production line-related data in real time, including production environment (such as temperature, humidity), equipment operation status, energy consumption, etc., and upload the relevant data to the cloud platform in real time through the IoT gateway; the cloud platform pre-processes the collected data, and associates the timestamps of key production events (such as startup, shutdown, material change, etc.), operation records, and quality inspection results of the production stage with the raw material archives, and uploads them to the blockchain together; S3. Assign unique logistics tags to finished down garments through RFID and GPS, and record logistics information, including entry and exit time, location changes, and temperature and humidity conditions; link the logistics information with the raw material archives and production records on the blockchain to form a complete supply chain view; S4. Consumers scan the QR code on the clothing or enter the unique ID to access the blockchain platform and view the complete traceability chain from raw materials to finished products; S5. Based on big data analysis technology, potential risk points in the supply chain are explored, including quality fluctuation trends and abnormal energy consumption patterns. Through machine learning models, quality problems that may occur in the future are predicted based on historical data, and measures are taken in advance to achieve intelligent quality management. Smart contracts are deployed on the blockchain to define the rules for handling quality issues, including automatically triggering the return process and compensation mechanism.

[0008] The working principle of the above technical solution is: each batch of polymer raw materials is subjected to IoT labeling, and the polymer materials include bio-based materials. Each label contains basic information such as a unique identifier (UID), material type, origin, batch number, etc. These labels are digital proof of the identity of the raw materials; blockchain technology is used to create a corresponding digital asset for each batch of raw materials. This asset contains the basic information of the raw materials and key data such as the first quality inspection report. The characteristics of blockchain ensure that these data cannot be tampered with, thus forming a "digital ID card" for the raw materials; sensors and smart devices are deployed on the production line to monitor key data such as the production environment (such as temperature, humidity), equipment operation status, energy consumption, etc. in real time; and the real-time monitored data is uploaded to the cloud platform through the IoT gateway. The cloud platform pre-processes these data, extracts the timestamps of key production events (such as startup, shutdown, material change, etc.), operation records, and quality inspection results of the production stage; the processed key production event information is associated with the raw material archive, and uploaded to the blockchain together. In this way, the production process of each batch of raw materials has a complete digital record; through RFID and GPS technology, a unique logistics tag is assigned to the finished down garment, and logistics information is recorded, including the time of entry and exit, location changes, and temperature and humidity conditions; the logistics information is linked with the raw material archives and production records on the blockchain to form a complete traceability chain from raw materials to finished products. This provides consumers with a convenient way to inquire about the source and production process of products; consumers can access the blockchain platform by scanning the QR code on the clothing or entering the unique ID; on the blockchain platform, consumers can view the complete traceability chain information from raw material procurement, production process to logistics distribution to ensure the transparency and credibility of the product; based on big data analysis technology, the data in the supply chain is deeply mined to discover potential risk points such as quality fluctuation trends and abnormal energy consumption patterns; through machine learning models, historical data is used to predict possible quality problems in the future, and measures are taken in advance to prevent them, including the evaluation of material quality under multi-factor coupling conditions such as humidity and sun exposure; smart contracts are deployed on the blockchain to define the rules for handling quality problems. When quality problems occur, smart contracts can automatically trigger the return process or compensation mechanism to ensure that the problems are handled promptly and fairly.

[0009] The effect of the above technical solution is: through the application of advanced technologies such as IoT labeling, blockchain technology, RFID and GPS, the whole process traceability from raw materials to finished down clothing is realized. Consumers can easily access the blockchain platform by scanning the QR code or entering the unique ID, view the complete traceability chain of the product, and realize the quality traceability of the whole life cycle based on the standard dimension. This not only improves the transparency of the supply chain, but also enhances consumers' trust in the product; the distributed ledger and encryption algorithm of blockchain technology are used to ensure the authenticity and non-tamperability of the data uploaded to the blockchain. Once the data is recorded on the blockchain, it cannot be modified or deleted, thus ensuring the accuracy and reliability of data such as raw material archives, production records, and logistics information; through the application of big data analysis technology and machine learning models, potential risk points in the supply chain can be explored, quality problems that may occur in the future can be predicted, and measures can be taken in advance to intervene. This not only improves the efficiency of quality management, but also reduces the losses caused by quality problems. At the same time, smart contracts are deployed on the blockchain to define the rules for handling quality issues, realize the automated handling of quality issues, and further improve the intelligent level of quality management; real-time monitoring of production line-related data, including production environment, equipment operating status, energy consumption, etc., can help companies promptly discover problems and bottlenecks in the production process, optimize production processes, and improve production efficiency. At the same time, through big data analysis technology, abnormal energy consumption patterns can also be discovered to help companies formulate energy-saving and emission reduction measures, reduce production costs, and achieve sustainable development; the implementation of quality traceability methods can also be based on the analysis and tracking of hazardous substances. Quality assessment will not only help improve product quality and safety, but also demonstrate the company's responsible attitude towards consumers and strict control of product quality. This will help enhance the company's brand image and market competitiveness and attract more attention and trust from consumers.

[0010] In one embodiment of the present invention, the S1 includes: S11. Based on the polymer raw material information collection standard, RFID technology is used in combination with QR codes to record tag information; RFID tags have built-in UIDs, and QR codes contain visual information for quick identification and verification; S12. Introduce blockchain hash algorithm to encrypt raw material information and generate a unique hash value as the label anti-counterfeiting mark; combine the decentralized characteristics of blockchain to generate a unique digital asset ID for each batch of raw materials. S13. Deploy smart contracts on the blockchain. Based on the preset thresholds of raw material quality standards, if quality problems are detected, the early warning mechanism will be automatically triggered and the relevant batches will be locked; S14. Combine multi-source data, including IoT sensors and third-party testing agencies, to cross-verify raw material information. Based on the dynamic update mechanism of raw material archives, submit update applications under specific conditions (such as batch changes, quality improvements, etc.), and after verification by the blockchain consensus mechanism, perform seamless information updates.

[0011] The working principle of the above technical solution is as follows: Based on the polymer raw material information collection standard, the key information of the material is comprehensively collected, such as material composition, environmental protection level, manufacturer qualification, etc. This information is the basis for ensuring the quality of raw materials; RFID technology is used in combination with QR code to record the collected tag information. The RFID tag has a built-in unique identifier (UID) for rapid identification of raw materials in the IoT environment. The QR code contains visual information, such as the basic properties of the raw materials, manufacturer information, etc., which is convenient for manual rapid identification and verification; the blockchain hash algorithm is introduced to encrypt the raw material information and generate a unique hash value. This hash value is used as a label anti-counterfeiting mark to effectively prevent the information from being tampered with or forged; combined with the decentralized characteristics of the blockchain, a unique digital asset ID is generated for each batch of raw materials. This ID is mapped one-to-one with the physical raw materials, ensuring the uniqueness and traceability of the assets on the blockchain; the first quality inspection report not only contains traditional test results, but also introduces the digital signature of a third-party inspection agency. The digital signature ensures the authority and immutability of the data, making the quality inspection report more credible; smart contracts are deployed on the blockchain to preset the threshold of the raw material quality standard. These thresholds are set according to industry standards or enterprise requirements to determine whether the raw materials are qualified; if it is detected that the quality of the raw materials does not meet the preset standards, the smart contract will automatically trigger the early warning mechanism. This includes locking the relevant batches to prevent them from flowing into the production line and sending alarm information to relevant personnel; combining IoT sensors and multi-source data such as third-party testing agencies to cross-verify the raw material information. This can ensure the accuracy and completeness of the information and improve the reliability of the raw material archives; based on the dynamic update mechanism of the raw material archives, update applications are submitted under specific conditions (such as batch replacement, quality improvement, etc.). These applications will be verified by the blockchain consensus mechanism to ensure the legitimacy and validity of the update; once the update application is verified, the blockchain will automatically update the information seamlessly. This ensures that the raw material archives are always consistent with the actual situation, providing a reliable basis for subsequent traceability and quality management.

[0012] The effect of the above technical solution is: based on the polymer raw material information collection standard, it covers key information such as material composition, environmental protection level, and manufacturer qualifications, ensuring the comprehensiveness and accuracy of raw material information. The use of RFID technology combined with QR code for label recording not only improves the efficiency and convenience of information reading, but also realizes the rapid identification and verification of raw materials through the UID of the RFID tag and the visual information of the QR code, effectively avoiding information omissions or errors; the blockchain hash algorithm is introduced to encrypt the raw material information and generate a unique hash value as the label anti-counterfeiting mark, which greatly enhances the anti-counterfeiting ability of the raw materials. At the same time, combined with the decentralized characteristics of the blockchain, a unique digital asset ID is generated for each batch of raw materials, ensuring the one-to-one mapping of assets and physical raw materials, and achieving full traceability from raw materials to finished products. This not only helps to combat counterfeit and shoddy products, but also provides consumers with more transparent and reliable product information; the first quality inspection report not only contains traditional test results, but also introduces the digital signature of a third-party testing agency. The introduction of digital signatures ensures the authority and immutability of quality inspection reports, allowing consumers and relevant departments to trust and rely on these reports more. This helps to enhance consumers' trust in products and also provides strong support for product quality supervision. Smart contracts are deployed on the blockchain, and an intelligent quality early warning mechanism is implemented based on the preset threshold of raw material quality standards. Once a quality problem is detected, the smart contract will automatically trigger an early warning and lock the relevant batch to prevent it from flowing into the production line. This not only improves the efficiency and accuracy of quality management, but also effectively avoids the expansion of quality problems, protects the reputation of the company and the rights of consumers; combined with multi-source data, including IoT sensors and third-party testing agencies, cross-verification of raw material information ensures the accuracy and reliability of the information. At the same time, based on the dynamic update mechanism of raw material archives, update applications are submitted under specific conditions, and seamless updates are carried out after verification by the blockchain consensus mechanism. This not only ensures the timeliness and accuracy of raw material information, but also improves the flexibility and convenience of information management.

[0013] In one embodiment of the present invention, the S13 includes: According to the preset raw material quality standards (including but not limited to ingredient ratio, environmental protection level, physical properties, etc.), the logic rules of the smart contract are constructed; the rules are used as the benchmark for raw material quality monitoring, and once the data deviates from the preset threshold, an early warning is triggered; the constructed logic rules are encoded into a smart contract using the smart contract programming language; Before deployment, the smart contract is tested, including unit testing, integration testing and security auditing; the smart contract that passes the test is deployed on the blockchain network and linked to the raw material information management system; Through the deployment of IoT sensors at key links, key quality parameters of raw materials are collected in real time, pre-processed, and the pre-processed data is uploaded to the blockchain; Smart contracts analyze uploaded quality data in real time according to preset threshold rules; once data anomalies are detected, the early warning mechanism is immediately triggered and early warning information is sent to relevant personnel; After the early warning is triggered, the smart contract automatically locks the batch of raw materials involved in the quality problem; at the same time, through the immutability of the blockchain, the information of the problematic batch is permanently recorded; Based on the type and severity of the warning, the smart contract automatically initiates the corresponding emergency response plan; the smart contract automatically generates a raw material quality monitoring report based on monitoring data and warning records; and the quality monitoring report is shared with all relevant parties (including manufacturers, suppliers, customers, etc.) through the blockchain network.

[0014] The working principle of the above technical solution is as follows: construct the logical rules of the smart contract according to the preset raw material quality standards (such as ingredient ratio, environmental protection level, physical properties, etc.); use the smart contract programming language (such as Solidity) to encode the logical rules into the smart contract; before deployment, conduct a comprehensive test on the smart contract, including unit testing, integration testing and security auditing, to ensure its correctness and security; deploy the tested smart contract on the blockchain network and link it with the raw material information management system to realize real-time data interaction; deploy IoT sensors in key links such as raw material storage and processing to collect key quality parameters of raw materials (such as temperature, humidity, weight, etc.) in real time; pre-process the collected data to meet the input requirements of the smart contract; upload the pre-processed data to the blockchain for access and analysis by the smart contract; the smart contract performs real-time analysis on the uploaded quality data according to the preset threshold rules; once data abnormalities are detected (such as exceeding the preset threshold), the early warning mechanism is immediately triggered to send early warning information to relevant personnel; early warning The information includes detailed information of abnormal data, possible impact and recommended emergency response measures; after the warning is triggered, the smart contract automatically locks the batch of raw materials involved in quality problems to prevent them from flowing into the production line; through the immutability of blockchain, the information of the problem batch is permanently recorded, including batch number, abnormal data, warning time, etc.; it is convenient to trace and analyze the problem batch later to determine the cause and solution of the problem; according to the type and severity of the warning, the smart contract automatically initiates the corresponding emergency response plan; the emergency response plan includes notifying the third-party testing agency to conduct re-inspection, initiating the raw material recall procedure and adjusting the production plan; the smart contract automatically generates a raw material quality monitoring report based on the monitoring data and warning records; the report includes quality data trend analysis, warning event details, emergency response measures, etc., to provide comprehensive quality monitoring information for relevant personnel; the quality monitoring report is shared with all relevant parties (including manufacturers, suppliers, customers, etc.) through the blockchain network; the transparency and traceability of the supply chain are improved, and consumers' trust in the product is enhanced.

[0015] The effect of the above technical solution is: by building smart contracts, the automation and intelligence of raw material quality monitoring is realized. Smart contracts can analyze the uploaded quality data in real time according to the preset raw material quality standards, and immediately trigger the early warning mechanism once the data abnormality is detected. This not only greatly improves the response speed of quality management, but also reduces human intervention and reduces the risk of human error. Smart contracts are based on blockchain technology and have the characteristics of being tamper-proof. This means that once the data is uploaded to the blockchain, it cannot be modified or deleted. Therefore, smart contracts can ensure the accuracy and reliability of raw material quality monitoring data, and provide strong evidence support for the traceability of quality problems; through smart contracts, the quality data of raw materials is uploaded to the blockchain in real time and can be accessed by all relevant parties. This greatly improves the transparency and traceability of the supply chain, allowing manufacturers, suppliers and customers to understand the quality status of raw materials in real time, and enhances the trust and cooperation of the supply chain; smart contracts can automatically start the corresponding emergency response plan according to the type and severity of the early warning. This includes notifying third-party testing agencies to conduct re-inspections, initiating raw material recall procedures, and adjusting production plans. By optimizing the emergency response process, companies can respond to quality issues more quickly, reduce quality risks, and protect consumer rights; smart contracts can automatically generate raw material quality monitoring reports based on monitoring data and warning records. The report includes quality data trend analysis, warning event details, emergency response measures, etc., providing comprehensive quality monitoring information for relevant personnel. This information helps companies better understand the quality status of raw materials, support decision-making, and improve product quality and competitiveness; this technical solution applies blockchain technology and smart contracts to raw material quality management, providing new ideas and directions for the application and development of blockchain technology in the field of quality management. This will not only help promote the popularization and application of blockchain technology, but also promote innovation and development in the field of quality management.

[0016] In one embodiment of the present invention, the S2 includes: S21. Using monitoring equipment, including high-precision temperature sensors, humidity sensors, and energy consumption monitoring equipment, to monitor subtle changes in the production environment in real time, and using a health management system equipped with smart devices to monitor the operating status of the equipment in real time; S22. Based on the event association algorithm, intelligently match production events with related data to form a complete event chain; based on the machine learning algorithm, build a production anomaly detection model to automatically identify abnormal patterns in the production process; S23. Based on the preset warning response level, the corresponding emergency handling process is automatically triggered according to the degree of abnormality, such as suspending production, adjusting process parameters, etc.

[0017] The working principle of the above technical solution is as follows: high-precision temperature sensors, humidity sensors and energy consumption monitoring equipment are deployed in the production environment to monitor the subtle changes in the production environment (such as small fluctuations in temperature and humidity) and the operating status of the equipment (such as energy consumption, vibration, etc.) in real time; the health management system equipped with the smart equipment receives data from the monitoring equipment and performs real-time analysis to monitor the health status of the equipment; through data analysis, the health management system can predict and prevent potential equipment failures, such as predicting motor failures by monitoring abnormal increases in energy consumption, or predicting mechanical wear by analyzing vibration data; the algorithm intelligently matches production events (such as equipment downtime, quality anomalies, etc.) with related data to form a complete event chain. Related data include raw material archives (such as batches, ingredients, sources, etc.) and quality inspection reports (such as test results, abnormal records, etc.); based on a large amount of historical data, a production anomaly detection model is built. The model can automatically identify abnormal patterns in the production process, such as identifying potential quality problems or reduced production efficiency by analyzing abnormal fluctuations in production data; different warning response levels are preset according to the degree of abnormality, such as minor abnormalities, moderate abnormalities, and severe abnormalities; once the abnormality detection model identifies abnormal patterns in the production process and determines the degree of abnormality according to the warning response level, the corresponding emergency response process will be automatically triggered. For example, for minor abnormalities, it may only be necessary to adjust process parameters or perform simple equipment maintenance; for moderate abnormalities, it may be necessary to suspend production for further inspection; for severe abnormalities, it may be necessary to immediately stop production and start a comprehensive troubleshooting and repair process.

[0018] The effect of the above technical solution is: through monitoring equipment such as high-precision temperature sensors, humidity sensors and energy consumption monitoring equipment, it is possible to monitor subtle changes in the production environment and the operating status of the equipment in real time. This real-time monitoring not only helps to timely discover potential problems in the production environment, such as abnormal fluctuations in temperature or humidity, but also can predict and prevent potential failures of the equipment through the health management system equipped with smart devices. This preventive maintenance measure can significantly improve the stability and efficiency of the production line and reduce production interruptions caused by equipment failure. Based on the event association algorithm, production events are intelligently matched with related data to form a complete event chain. This intelligent matching can reveal the inherent connection between production events and provide a more comprehensive perspective for problem identification. At the same time, the production anomaly detection model constructed by the machine learning algorithm can automatically identify abnormal patterns in the production process and improve the accuracy and timeliness of problem identification. This intelligent detection capability helps enterprises to discover problems faster, analyze problems and take measures to solve problems. According to the preset early warning response level, the corresponding emergency handling process is automatically triggered according to the degree of abnormality. This automated emergency handling mechanism can quickly respond to abnormal situations in the production process and reduce the delay and uncertainty of manual intervention. By timely suspending production, adjusting process parameters and other measures, enterprises can effectively control the spread of abnormal situations and reduce losses caused by production abnormalities. The above technical solutions not only help to improve the current production management level, but also provide strong support for the company's transformation to intelligence. By introducing high-precision monitoring equipment, intelligent algorithms and automated emergency response mechanisms, companies can gradually build an intelligent production management system and lay a solid foundation for future sustainable development.

[0019] In one embodiment of the present invention, the S22 includes: S221. Preprocess a large amount of raw data collected from monitoring equipment and equipment health management systems, and use feature engineering techniques to extract key event features from the preprocessed data, such as temperature fluctuation range, humidity change trend, abnormal energy consumption peak, etc.; S222. Combine raw material files (such as batch number, ingredient ratio, supplier information, etc.) and quality inspection reports (defect type, occurrence frequency, impact degree, etc.) to build a multi-dimensional event feature set; S223. Based on the event association algorithm, the potential causal relationship and time series correlation between different data sources are automatically identified, and various events in production (such as equipment failure warning, raw material quality fluctuation, environmental parameter abnormality, etc.) are intelligently matched with relevant data to form a continuous and complete event chain; S224. Use unsupervised learning methods to train the integrated multi-dimensional event feature set, learn the data distribution characteristics under normal production conditions, and automatically identify data points that deviate from the normal pattern, i.e., potential production anomalies; S225. For the abnormal patterns identified by the model, cluster analysis is used to further analyze their internal laws and influencing factors, and on this basis, a production anomaly knowledge graph is constructed.

[0020] The working principle of the above technical solution is as follows: First, a large amount of raw data collected from monitoring equipment (such as high-precision temperature sensors, humidity sensors, energy consumption monitoring equipment, etc.) and equipment health management systems is preprocessed, including data cleaning, denoising, normalization and other operations to ensure data quality and consistency. Then, feature engineering technology is used to extract key event features from the preprocessed data, such as temperature fluctuation range, humidity change trend, abnormal energy consumption peak, etc. These features can reflect the operating status of the production environment and equipment; combined with raw material archives (such as batch number, component ratio, supplier information, etc.) and quality inspection reports (defect type, occurrence frequency, impact degree, etc.), key event features are combined with raw material information and quality inspection results to construct a multi-dimensional event feature set. This feature set not only contains the status information of the production environment and equipment, but also covers the quality and defects of raw materials, providing a comprehensive data basis for subsequent anomaly detection; based on the event association algorithm, the potential causal relationship and time series correlation between different data sources are automatically identified. By intelligently matching various events in production (such as equipment failure warning, raw material quality fluctuation, environmental parameter abnormality, etc.) with relevant data, a continuous and complete event chain is formed. This event chain can reveal the inherent connection between production events and help understand the background and causes of abnormalities. The integrated multi-dimensional event feature set is trained using an unsupervised learning method. By learning the data distribution characteristics under normal production conditions, the model can automatically identify data points that deviate from the normal pattern, that is, potential production anomalies. This method does not require the prior annotation of abnormal data, so it is more flexible and efficient in practical applications. For the abnormal patterns identified by the model, cluster analysis is used to further analyze their internal laws and influencing factors. Cluster analysis can classify similar abnormal patterns into one category, thereby revealing the commonalities and differences between abnormalities. On this basis, a production anomaly knowledge graph is constructed to intuitively display the type of anomaly, frequency of occurrence, potential causes and scope of impact. The knowledge graph not only provides a visual representation of the abnormal pattern, but also provides deep insights for decision makers, helping them better understand the abnormal situations in the production process and formulate corresponding response strategies.

[0021] The effect of the above technical solution is: by preprocessing and extracting features from the data collected by the monitoring equipment, key parameters in the production process, such as temperature, humidity and energy consumption, can be monitored in real time. Once these parameters fluctuate abnormally, the system can immediately issue an early warning, enabling the production team to respond quickly and take measures to avoid potential production failures and quality problems; combined with raw material archives and quality inspection reports, a multi-dimensional event feature set can be constructed, and the causal relationship and time series correlation between different data sources can be automatically identified through event association algorithms. This helps to accurately locate abnormal links in production, reduce troubleshooting time, and improve problem solving efficiency; unsupervised learning methods are used to train the integrated multi-dimensional event feature set to learn the data distribution characteristics under normal production conditions. By automatically identifying data points that deviate from normal patterns, the system can provide data support for decision makers, enabling them to make more informed production decisions based on facts; through cluster analysis of abnormal patterns and knowledge graph construction, bottlenecks and potential risks in the production process can be revealed. This helps enterprises optimize resource allocation, such as adjusting production plans, improving process flow or strengthening quality control, thereby improving overall production efficiency and profitability; the implementation of the above technical solutions helps enterprises improve their intelligence level and realize automation, digitization and intelligent management of the production process. This can not only improve production efficiency and quality, but also reduce labor costs and safety risks; through real-time monitoring and early warning, it helps enterprises to promptly discover and solve environmental problems in the production process, such as abnormal energy consumption. This helps enterprises reduce energy consumption, reduce emissions, and improve resource utilization efficiency, thereby enhancing sustainable development capabilities; through comprehensive monitoring of the production process and abnormal early warning, it helps enterprises improve product quality, reduce defective product rates, and improve customer satisfaction and loyalty; high-quality product quality and efficient production management help enterprises shape a good brand image and improve market competitiveness. This can not only attract more customers, but also promote the long-term development of enterprises.

[0022] In one embodiment of the present invention, the S223 includes: Integrate information from multiple data sources; use natural language processing (NLP) and machine learning techniques to perform preliminary event identification on the integrated data; Use time series analysis technology to identify the time series characteristics of events in different data sources, and use causal relationship mining algorithms to automatically explore the potential causal relationship between events and build causal chains between events; Combine key event features extracted from raw data with multi-dimensional information from raw material archives and quality inspection reports to fuse event features; Through feature selection, dimensionality reduction and fusion algorithms, a high-dimensional, complex and comprehensive multi-dimensional event feature set is constructed; Based on the event feature set and causal chain constructed above, a matching algorithm is used to automatically and intelligently match various events in production with relevant data; at the same time, the association optimization technology is used to further optimize and adjust the event chain; Visualize the constructed event chain.

[0023] The working principle of the above technical solution is as follows: First, the information from multiple data sources such as monitoring equipment, equipment health management system, raw material archive system and quality inspection report is integrated. In order to ensure the consistency and comparability of data, data standardization technologies such as data cleaning, format conversion, missing value processing, etc. are used to provide a high-quality data foundation for subsequent analysis; natural language processing (NLP) and machine learning technologies are used to perform preliminary event identification on the integrated data. By defining a set of event keywords and rules, key event information such as equipment failure type, raw material quality problem description, etc. are automatically extracted from text data, and these events are preliminarily classified to provide a clear framework for subsequent correlation analysis; time series analysis techniques such as autoregressive moving average model (ARIMA) and long short-term memory network (LSTM) are used to identify the time series characteristics of events in different data sources, including periodicity, trend and seasonality. At the same time, causal relationship mining algorithms, such as Granger causality test and causal reasoning based on Bayesian network, are used to automatically explore the potential causal relationship between events and build causal chains between events; event features are fused by combining key event features extracted from raw data (such as temperature fluctuation range, humidity change trend, etc.) and multi-dimensional information in raw material archives and quality inspection reports. Through feature selection, dimensionality reduction and fusion algorithms, such as principal component analysis (PCA), linear discriminant analysis (LDA) and deep learning models, a high-dimensional, complex and comprehensive multi-dimensional event feature set is constructed to provide a rich information foundation for subsequent anomaly detection and pattern recognition; based on the above-constructed event feature set and causal relationship chain, advanced matching algorithms, such as similarity-based matching and graph-based matching, are used to automatically and intelligently match various events in production with related data. At the same time, we use association optimization techniques, such as local optimization based on greedy algorithms and global optimization based on genetic algorithms, to further optimize and adjust the event chain to ensure the continuity and integrity of the event chain; finally, we visualize the constructed event chain and intuitively present the association between events through timeline, causal chain, event nodes, etc. At the same time, we introduce explanatory machine learning technology to provide interpretability support for the event chain, helping users better understand the logic and basis of event association.

[0024] The effects of the above technical solutions are as follows: by integrating information from multiple data sources, data silos are broken and comprehensive data convergence is achieved. This not only improves the utilization rate of data, but also provides a rich data foundation for subsequent event identification and association analysis; natural language processing (NLP) and machine learning technology are used to perform preliminary event identification on the integrated data, greatly improving the intelligence level of event identification. At the same time, by training and optimizing the model, various key events in production can be accurately identified, providing accurate event information for subsequent analysis; time series analysis technology is used to identify the time series characteristics of events in different data sources, which helps to understand the dynamic change laws and trends of events. At the same time, the causal relationship mining algorithm is used to automatically explore the potential causal relationship between events, and the causal chain between events is constructed, which provides strong support for in-depth understanding of event associations in the production process; the event features are fused by combining the key event features extracted from the original data and the multi-dimensional information in the raw material archives and quality inspection reports. This not only enriches the content of event features, but also improves the comprehensiveness and complexity of the features, providing a more comprehensive information foundation for subsequent pattern recognition and anomaly detection; through feature selection, dimensionality reduction and fusion algorithms, a high-dimensional, complex and comprehensive multi-dimensional event feature set is constructed. This not only improves the quality of the feature set, but also provides an efficient basis for subsequent intelligent matching. At the same time, the matching algorithm is used to automatically match various events in production with relevant data, greatly improving the efficiency and accuracy of matching; the association optimization technology is used to further optimize and adjust the event chain to ensure the accuracy and consistency of the event chain. At the same time, the constructed event chain is visualized to make the association relationship between events more intuitive and easy to understand, providing strong support for the monitoring and management of the production process.

[0025] In one embodiment of the present invention, S3 includes: S31, encrypt RFID tags and GPS trackers, record the corresponding information, and deploy edge computing devices at logistics nodes to process logistics data in real time; S32. Based on big data visualization technology, the supply chain view is presented in the form of dynamic charts, 3D maps and other visualizations; S33. Develop a path optimization algorithm based on historical logistics data and real-time traffic information to automatically plan the optimal logistics path; S34. Combined with blockchain smart contracts, set risk warning thresholds in the logistics process. If a warning is triggered, the preset response strategy will be automatically executed.

[0026] The working principle of the above technical solution is as follows: encrypt the sensitive information in RFID tags and GPS trackers to ensure the security and privacy of data in the logistics process; encryption technology may include symmetric encryption algorithms (such as AES) or asymmetric encryption algorithms (such as RSA), as well as dynamic encryption technology, etc., to enhance the security of data transmission and storage; record basic in and out time, location changes, temperature and humidity, vibration and other environmental parameters, which provide an important basis for quality control in the logistics process; ensure the accuracy and timeliness of data through real-time monitoring of RFID tags and GPS trackers; deploy edge computing devices at logistics nodes to process and analyze logistics data in real time; edge computing devices can reduce data transmission delays, improve data processing efficiency, and reduce the load on central servers; use big data visualization technology to present supply chain views in visual forms such as dynamic charts and 3D maps; this visual presentation method helps managers intuitively understand the operating status of the supply chain, discover potential problems in a timely manner, and make corresponding decision adjustments; based on historical logistics data and real-time Communication information, collect and analyze relevant data on logistics routes; these data may include road congestion, traffic control information, vehicle performance parameters, etc.; develop path optimization algorithms to automatically plan the optimal logistics route; the algorithm may use intelligent algorithms (such as Dijkstra algorithm, ant colony algorithm, etc.), comprehensively consider multiple factors (such as distance, time, cost, etc.) to find the optimal path; arrange the path of logistics transportation according to the optimal path planned by the algorithm; during the transportation process, dynamically adjust the path according to real-time traffic information to ensure the smooth transportation; combine blockchain smart contract technology to set risk warning thresholds in the logistics process; smart contracts can automatically monitor logistics data and execute preset response strategies when the warning threshold is triggered; when logistics data reaches or exceeds the warning threshold, smart contracts automatically execute preset response strategies; these response strategies may include adjusting logistics plans, activating spare warehouses, notifying relevant personnel, etc.; through the automatic execution of smart contracts, effective management and control of risks in the logistics process can be achieved; it helps to reduce uncertainty in the logistics process and improve logistics efficiency and safety.

[0027] The effect of the above technical solution is: by encrypting RFID tags and GPS trackers, logistics data can be effectively prevented from being illegally intercepted or tampered with during transmission, thereby ensuring the integrity and authenticity of the data. This encryption measure provides a strong guarantee for information security in the logistics process; edge computing devices are deployed at logistics nodes to process and analyze logistics data in real time. Compared with the traditional centralized data processing method, edge computing can reduce data transmission delays and improve data processing efficiency, thereby responding to various needs in the logistics process more quickly; based on big data visualization technology, the supply chain view is presented in a visual form such as dynamic charts and 3D maps, so that managers can more intuitively understand the overall situation and operation status of the supply chain. This visual management method helps managers quickly discover problems, make decisions, and optimize the management process of the supply chain; through visualization technology, managers can obtain various data and information of the supply chain in real time, such as inventory status, transportation progress, environmental parameters, etc. These data and information provide managers with rich decision-making support, which helps to improve the efficiency and accuracy of decision-making; the path optimization algorithm developed based on historical logistics data and real-time traffic information can automatically plan the optimal logistics path. The application of this algorithm helps to reduce transportation time, reduce transportation costs, and improve logistics efficiency; the path optimization algorithm can continuously adjust and optimize the logistics path based on real-time traffic information and historical data, so as to adapt to various complex and changing logistics environments; combined with blockchain smart contract technology, risk warning thresholds in the logistics process can be set. When actual data triggers these warning thresholds, the smart contract will automatically execute the preset response strategy, such as adjusting the logistics plan, enabling spare warehouses, etc.; the automated execution characteristics of the smart contract make risk response faster and more accurate. This helps to reduce delays and errors caused by human intervention and improve the stability and reliability of the logistics process.

[0028] In one embodiment of the present invention, the S33 includes: S331, integrating multi-source data and pre-processing the integrated multi-source data, wherein the multi-source data includes historical logistics data (such as transportation time, distance, cost, historical traffic conditions, etc.), real-time traffic information (such as road conditions, congestion, traffic accidents, etc.), weather forecast data (such as rainfall, strong winds, snow conditions, etc., weather factors that may affect transportation), geographic information (such as road grades, bridge and tunnel restrictions, restricted areas, etc.), and vehicle status data (such as fuel consumption, tire wear, load conditions, etc.); S332. Based on historical logistics data, a path planning algorithm is used to generate a preliminary optimal path set; and in combination with real-time traffic information and vehicle status data, the preliminary path is adjusted in real time through a machine learning model; S333, incorporating multi-objective optimization strategies into path planning, and using genetic algorithms to find the optimal solution set through multi-objective optimization functions; S334. Use logistics simulation software to simulate the optimized path and test its performance in different scenarios; verify the effectiveness and reliability of the path optimization algorithm by comparing the simulation results with the actual situation; S335. Establish a feedback mechanism to collect data feedback and user evaluation during the actual transportation process and continuously optimize the path planning algorithm; S336. Combine real-time traffic information and vehicle status data to identify potential transportation risks, such as severe congestion, traffic accidents, and bad weather. Based on the risk warning results, automatically trigger preset emergency response strategies, such as adjusting logistics plans, activating backup warehouses, and changing transportation methods. At the same time, use machine learning models to evaluate the effects of different strategies in real time and select the optimal emergency plan.

[0029] The working principle of the above technical solution is as follows: First, the system will collect data from multiple sources, including historical logistics data, real-time traffic information, weather forecast data, geographic information, and vehicle status data; the collected multi-source data will be pre-processed, including data cleaning, format conversion, outlier processing, etc., to ensure the accuracy and consistency of the data; based on historical logistics data, the system uses path planning algorithms (such as Dijkstra algorithm) to generate a preliminary set of optimal paths. These path sets take into account factors such as transportation time, distance, and cost; combined with real-time traffic information and vehicle status data, the system uses machine learning models to make real-time adjustments to the preliminary paths. The model can predict future changes in road conditions, evaluate the potential risks and costs of different paths, and dynamically select the optimal path; in path planning, the system incorporates multi-objective optimization strategies, while considering multiple goals such as transportation cost, time efficiency, and carbon emissions; through multi-objective optimization functions, the system uses genetic algorithms and other solution methods to find the optimal solution set that meets multiple goals from multiple possible paths; using logistics simulation software, the system simulates the optimized path to test its performance in different scenarios; by comparing the simulation results with the actual situation, the system verifies the effectiveness and reliability of the path optimization algorithm. According to the verification results, the system will further optimize and adjust the algorithm; the system establishes a feedback mechanism to collect data feedback and user evaluations during the actual transportation process; based on the collected feedback and evaluations, the system continuously optimizes the path planning algorithm to improve the accuracy and practicality of the algorithm; combined with real-time traffic information and vehicle status data, the system can identify potential transportation risks, such as severe congestion, traffic accidents, and bad weather; based on the risk warning results, the system automatically triggers the preset emergency response strategy, such as adjusting the logistics plan, activating spare warehouses, changing the mode of transportation, etc. At the same time, the system uses machine learning models to evaluate the effects of different strategies in real time and select the best emergency plan.

[0030] The effects of the above technical solutions are as follows: through the integration and preprocessing of multi-source data, the application of path planning algorithms and multi-objective optimization strategies, a more scientific and reasonable logistics path can be generated, and the waiting time, number of transfers and transportation distance in the transportation process can be reduced, thereby improving logistics efficiency; the optimized logistics path can reduce fuel consumption, vehicle wear and tear, labor costs and other expenses, thereby reducing transportation costs. At the same time, through the application of risk warning and emergency response mechanisms, additional cost expenditures caused by emergencies can be avoided; on-time delivery of goods and reliable logistics services can improve customer satisfaction. Through the application of path planning algorithms and real-time adjustment mechanisms, it can ensure that the goods are delivered within the time required by the customer, and improve the accuracy and reliability of delivery; through the application of logistics path planning and optimization systems, enterprises can improve logistics efficiency, reduce costs and improve customer satisfaction, thereby enhancing the brand image and market competitiveness of enterprises; by incorporating environmental protection indicators such as carbon emissions into path planning, and through the application of multi-objective optimization strategies, carbon emissions can be reduced while meeting transportation needs, promoting the sustainable development of enterprises.

[0031] In one embodiment of the present invention, the S332 includes: Extracting transportation data from a historical logistics database, using the transportation data as input to a path planning algorithm, and using the path planning algorithm to generate a series of preliminary optimal path sets; Obtain real-time traffic information and vehicle status data from external data sources through API interfaces or IoT technology, integrate them in real time, and update them to the route planning system in real time; Using machine learning models, the preliminary set of routes is dynamically adjusted; possible future traffic conditions are predicted, and routes are dynamically adjusted based on real-time data; In the process of route adjustment, the impact of vehicle status data is fully considered; by building an evaluation model of vehicle status and route adaptability, the adjusted route is quantitatively evaluated; The dynamically adjusted path planning results are fed back to the transportation management system in real time for reference by the dispatcher or the autonomous driving system. At the same time, a real-time feedback mechanism for path optimization results is established to collect actual data during the transportation process; these actual data are used to iteratively optimize the path planning algorithm.

[0032] The working principle of the above technical solution is as follows: extract rich transportation data from the historical logistics database, including but not limited to transportation time, distance, cost and historical traffic conditions. These data will be used as input to the path planning algorithm, and a series of preliminary optimal path sets will be generated using advanced path planning algorithms (such as Dijkstra algorithm). These path sets are based on historical data and reflect the optimal choice under normal traffic conditions; real-time traffic information (such as current road conditions, congestion, traffic accident reports, etc.) and vehicle status data (such as fuel consumption, tire wear, load limit, etc.) are integrated in real time. This information is obtained from external data sources through API interfaces or Internet of Things technology and updated to the path planning system in real time; the preliminary path set is dynamically adjusted using machine learning models (such as deep learning networks, reinforcement learning models, etc.). These models can learn the changing patterns of traffic conditions, predict possible future traffic conditions, and dynamically adjust the path based on real-time data to avoid congested sections or potential risk areas; in the process of path adjustment, the impact of vehicle status data is fully considered. For example, according to the fuel consumption of the vehicle, choose a route that can reduce fuel consumption; according to the tire wear, avoid choosing a road section with large tire wear; according to the load limit, choose a path that can safely carry goods and comply with traffic regulations; by building an evaluation model of vehicle status and path adaptability, quantitatively evaluate the adjusted path to ensure that the path is consistent with the actual status of the vehicle and meets the efficiency and safety requirements of transportation; the dynamically adjusted path planning results are fed back to the transportation management system in real time for reference by the dispatcher or the automatic driving system. At the same time, a real-time feedback mechanism for path optimization results is established to collect actual data during transportation (such as actual driving time, fuel consumption, vehicle status changes, etc.); use these actual data to iteratively optimize the path planning algorithm to continuously improve the accuracy and adaptability of the path planning algorithm. Through continuous learning and improvement, the path planning algorithm can better adapt to complex and changing traffic environments and vehicle status changes.

[0033] The effect of the above technical solution is: by extracting rich transportation data from the historical logistics database and using advanced path planning algorithms, a series of preliminary optimal path sets can be generated. These path sets are based on historical data and take into account multiple factors (such as transportation time, distance, cost, etc.), providing a reliable basis for subsequent dynamic adjustments; traffic information and vehicle status data are obtained from external data sources in real time through API interfaces or Internet of Things technology, and integrated into the path planning system. The integration of this real-time information enables path planning to dynamically adapt to changes in traffic conditions, improving the practicality and accuracy of the path; the preliminary path is dynamically adjusted using a machine learning model to predict possible future traffic conditions and adjust the path based on real-time data. This prediction and adjustment mechanism helps to avoid congested sections, reduce transportation time, and improve transportation efficiency; in the process of path adjustment, the impact of vehicle status data, such as fuel consumption, tire wear, load limit, etc., is fully considered. By constructing an evaluation model for vehicle status and path adaptability, the adjusted path is quantitatively evaluated to ensure that the path is consistent with the actual state of the vehicle and meets the efficiency and safety requirements of transportation; the dynamically adjusted path planning results are fed back to the transportation management system in real time for reference by the dispatcher or the autonomous driving system. This real-time feedback mechanism helps dispatchers or systems make timely decisions and optimize transportation plans; establish a real-time feedback mechanism for path optimization results, collect actual data during transportation, and use this data to iteratively optimize the path planning algorithm. This continuous optimization mechanism helps improve the accuracy and adaptability of the path planning algorithm, enabling it to better adapt to complex and changing traffic environments and vehicle status changes; overall, this technical solution significantly improves transportation efficiency and safety through the integration of real-time information, dynamic adjustment, vehicle status and path adaptability evaluation, real-time feedback and iterative optimization. It helps to reduce transportation time, reduce costs, improve customer satisfaction, and bring greater economic and social benefits to enterprises.

[0034] In one embodiment of the present invention, the S4 includes: S41. Embed augmented reality technology in the QR code, so that consumers can view the clothing information of the clothing through the mobile phone camera after scanning, and provide multi-language support for the QR code and unique ID; S42. Consumers understand the complete traceability chain of clothing through operations including clicking and dragging based on an interactive traceability platform; the traceability chain includes environmental certification of polymer materials, transparency of production process and logistics path.

[0035] The working principle of the above technical solution is: embedding augmented reality technology in the QR code of the clothing, which means that the QR code is no longer just a simple link or information carrier, but can trigger an AR experience on the mobile phone camera; when consumers use their mobile phones to scan this QR code, the mobile phone camera will capture the QR code information and start the AR application; after starting the AR application, consumers can view the 3D model of the clothing through the mobile phone camera. This 3D model is based on the actual size and style of the clothing, which can provide consumers with an approximate fitting effect. In addition to the 3D model, consumers can also view detailed information about the clothing, such as material, size, washing instructions, etc. The QR code and unique ID provide multi-language support, which means that no matter which country or region the consumer comes from, they can use their own language to view clothing information and fitting effects. This greatly increases the convenience and satisfaction of consumers, and also enhances the international image of the brand. Consumers can understand the complete traceability chain of clothing on the interactive traceability platform through operations such as clicking and dragging. This traceability chain includes every link of clothing from raw material procurement to production and then to logistics and transportation. The traceability chain first shows whether the polymer material used in the clothing has been environmentally certified, which helps consumers understand the environmental performance and sustainability of clothing. Then, the traceability chain shows the production process of clothing, including production technology, quality control and other links. This helps consumers understand the quality and manufacturing process of clothing. Finally, the traceability chain shows the logistics path of clothing from the factory to the hands of consumers, including delivery, transportation, customs clearance and other links. This helps consumers understand the shipping status and estimated arrival time of clothing; through the interactive traceability platform, consumers can freely click and drag elements on the page to gain in-depth insights into the aspects they are interested in; this interactive experience not only increases consumer engagement, but also improves their trust and loyalty to the brand.

[0036] The effect of the above technical solution is: by embedding augmented reality technology in the QR code, consumers can view the 3D model and fitting effect of the clothing through the mobile phone camera by simply scanning the QR code. This intuitive and vivid display method can greatly enhance the shopping experience of consumers, allowing them to better understand the appearance and effect of the product before purchasing; the 3D model and fitting effect provide consumers with an approximate fitting experience, which helps to reduce the return rate due to inappropriate size or unsatisfactory style. At the same time, this novel display method can also attract more consumers' attention and interest, thereby promoting sales conversion; providing multilingual support for QR codes and unique IDs can meet the needs of consumers in different countries and regions. This international design enables brands to reach global consumers more widely, enhance brand influence and market share; through the interactive traceability platform, consumers can understand the complete traceability chain of clothing, including environmental certification of polymer materials, transparency of production process, and logistics path information. This transparency helps to enhance consumers' trust in the brand and make them more confident in purchasing products; providing complete traceability chain information not only demonstrates the brand's commitment to product quality and environmental performance, but also reflects the brand's social responsibility and integrity. This helps to enhance the brand image, strengthen consumer loyalty to the brand and spread word-of-mouth; by showing the environmental certification of polymer materials and the transparency of the production process, brands can guide consumers to pay more attention to environmental protection and sustainable development issues. This guidance helps to drive the entire industry towards a more environmentally friendly and sustainable direction; although this is not directly mentioned in the technical solution, the complete traceability chain information can also provide brands with valuable data on inventory management and supply chain optimization. By analyzing this data, brands can more accurately predict demand, optimize inventory levels and reduce operating costs.

[0037] In one embodiment of the present invention, S5 includes: S51. Based on deep learning algorithms, build complex models such as quality prediction and fault warning; combine upstream and downstream data of the supply chain to conduct multi-dimensional analysis to identify potential quality problems and risk points; and continuously train and optimize the model based on the newly collected data according to the model self-learning and optimization mechanism; S52. Combine machine learning models with intelligent decision-making systems to provide managers with intelligent decision-making suggestions, introduce complex event processing algorithms into smart contracts, and handle complex events and abnormal situations in the supply chain; S53. Based on a multi-party collaboration mechanism, upstream and downstream enterprises in the supply chain are allowed to jointly participate in the formulation and execution of smart contracts.

[0038] The working principle of the above technical solution is: using deep learning algorithms to build complex models such as quality prediction and fault warning. These models can process and analyze a large amount of supply chain data to identify potential quality problems and risk points; combine data from upstream and downstream of the supply chain to conduct multi-dimensional analysis. This includes data from various links such as raw material suppliers, manufacturers, and distributors to ensure a comprehensive understanding of the supply chain; by analyzing this data, the model can identify key factors and potential risk points that may cause quality problems; based on newly collected data, the model can continuously self-learn and optimize. This means that over time, the model's prediction and warning capabilities will gradually improve and more accurately reflect the actual situation of the supply chain; combine machine learning models with intelligent decision-making systems to provide managers with intelligent decision-making suggestions. These suggestions are based on the model's in-depth analysis of supply chain data, aiming to help managers make more informed and efficient decisions; introduce complex event processing algorithms in smart contracts to handle complex events and abnormal situations in the supply chain. These algorithms can monitor the operating status of the supply chain in real time, and once an abnormal or complex event is found, the preset response mechanism will be automatically triggered; based on a multi-party collaboration mechanism, upstream and downstream enterprises in the supply chain are allowed to jointly participate in the formulation and execution of smart contracts. This means that all parties involved have the opportunity to comment on the content and execution of smart contracts to ensure their fairness and feasibility; once a smart contract is formulated, it will automatically execute the preset rules and processes. This includes data sharing, transaction confirmation, quality issues and other aspects. The execution of smart contracts will greatly improve the transparency and efficiency of the supply chain.

[0039] The effects of the above technical solutions are as follows: the quality prediction and fault warning model constructed by deep learning algorithms can accurately identify potential quality problems and risk points in the supply chain. This helps enterprises take measures in advance to avoid or reduce losses caused by quality problems; multi-dimensional analysis combined with upstream and downstream data of the supply chain helps enterprises to fully understand the operation status of the supply chain and discover possible problems and hidden dangers. This provides strong support for enterprises to formulate targeted improvement measures; the model can continuously self-learn and optimize according to the newly collected data, thereby continuously improving the accuracy of prediction and warning. This means that over time, the supply chain quality management and risk control capabilities of enterprises will continue to improve; combining machine learning models with intelligent decision-making systems to provide managers with intelligent decision-making suggestions. This helps managers make smart decisions quickly and improve the response speed and flexibility of the supply chain; introducing complex event processing algorithms in smart contracts can automatically handle complex events and abnormal situations in the supply chain. This reduces the need for manual intervention and improves the level of automation processing in the supply chain; based on a multi-party collaboration mechanism, upstream and downstream enterprises in the supply chain are allowed to jointly participate in the formulation and execution of smart contracts. This helps to enhance trust and cooperation among all parties involved in the supply chain and promote information sharing and collaborative work; the introduction and execution of smart contracts make the operation status and transaction information of the supply chain more transparent. This helps companies to identify and solve problems in a timely manner and reduce the risk of information asymmetry in the supply chain; through accurate prediction, early warning and automated processing, companies can reduce costs such as rework and returns caused by quality problems. At the same time, intelligent decision-making and multi-party collaboration can also help reduce the operating costs and communication costs of companies; automated processing and intelligent decision-making can significantly improve the response speed and operating efficiency of the supply chain. This helps companies meet market demand more quickly, improve customer satisfaction and competitiveness; by optimizing supply chain management and improving resource utilization efficiency, companies can reduce environmental impact and enhance the sustainability of the supply chain. This helps companies fulfill their social responsibilities and enhance their brand image and reputation.

[0040] In one embodiment of the present invention, the S51 includes: Collect multi-source heterogeneous data from upstream and downstream enterprises in the supply chain, including raw material quality, production process parameters, logistics information, and market demand feedback; and pre-process the collected multi-source heterogeneous data, including noise removal, missing value filling, and data standardization; Based on domain knowledge and statistical methods, we extract characteristic variables that are crucial for quality prediction and fault warning, such as time series trends, periodic fluctuations, and outlier detection indicators. We use the long short-term memory network architecture to process time series data and predict the quality trends of products at different production stages. Combine convolutional neural networks and recurrent neural networks to build a hybrid model to comprehensively analyze image data and sensor data to identify potential failure modes; Combine supply chain network graph analysis to identify key nodes and vulnerable links and assess supply chain disruption risks; use unsupervised learning algorithms to continuously monitor supply chain data and dynamically identify new risk points and potential quality issues; Based on the feedback mechanism between the model prediction results and the actual quality conditions, the model parameters are continuously adjusted through reinforcement learning.

[0041] The working principle of the above technical solution is as follows: widely collect multi-source heterogeneous data from upstream and downstream enterprises in the supply chain, including raw material quality data, production process parameters, logistics information, and market demand feedback; these data come from different systems, equipment and platforms, and are diverse, complex and heterogeneous; pre-process the collected multi-source heterogeneous data to remove noise, fill in missing values ​​and standardize the data; the pre-processing step ensures the quality and consistency of the data, and provides a reliable foundation for subsequent data analysis; based on domain knowledge and statistical methods, extract characteristic variables that are crucial to quality prediction and fault warning; these characteristic variables may include time series trends, periodic fluctuations, outlier detection indicators, etc.; use the long short-term memory network (LSTM) architecture to process time series data and predict the quality trend of products at different production stages; the LSTM network can capture long-term dependencies in time series data and is suitable for complex quality prediction tasks; combined with volume The convolution neural network (CNN) and recurrent neural network (RNN) are used to build a hybrid model to conduct comprehensive analysis of image data (such as device monitoring video screenshots) and sensor data; CNN is used to extract local features in image data, while RNN is used to capture time dependencies in time series data; through the comprehensive analysis of the hybrid model, potential failure modes are identified and early warning signals are issued; combined with supply chain network graph analysis, key nodes and vulnerable links are identified, and supply chain disruption risks are assessed; it helps companies take measures in advance to deal with potential risks and challenges; unsupervised learning algorithms (such as cluster analysis and anomaly detection) are used to continuously monitor supply chain data; through these algorithms, new risk points and potential quality problems are dynamically identified to ensure the robust operation of the supply chain; based on the feedback mechanism between the model prediction results and the actual quality situation, the model parameters are continuously adjusted through reinforcement learning; reinforcement learning enables the model to self-optimize according to real-time feedback to improve prediction accuracy and robustness.

[0042] The effects of the above technical solutions are: widely collect multi-source heterogeneous data from upstream and downstream enterprises in the supply chain, covering raw material quality, production process parameters, logistics information, market demand feedback and other aspects. This comprehensive data collection method helps enterprises to have a deeper understanding of each link of the supply chain, and provides a solid foundation for subsequent data analysis and decision support; pre-process the collected multi-source heterogeneous data, including steps such as noise removal, missing value filling and data standardization. These pre-processing operations can significantly improve the quality and availability of data, and provide a reliable data foundation for subsequent data analysis and model training; extract characteristic variables that are crucial for quality prediction and fault warning based on domain knowledge and statistical methods. These characteristic variables can reflect the key information and potential risks of the supply chain, and provide targeted input for subsequent data analysis and model training; use the long short-term memory network (LSTM) architecture to process time series data and predict the quality trend of products at different production stages. The LSTM network can capture long-term dependencies in time series data and improve the accuracy and stability of quality prediction; combine convolutional neural networks (CNN) and recurrent neural networks (RNN) to build a hybrid model to comprehensively analyze image data and sensor data and identify potential failure modes. This hybrid model can make full use of the advantages of different types of data to improve the accuracy and timeliness of fault warnings; combined with supply chain network graph analysis, it can identify key nodes and vulnerable links and assess the risk of supply chain disruptions. This analysis method helps companies to discover potential risk points in the supply chain in advance and take corresponding measures to prevent and respond; use unsupervised learning algorithms (such as cluster analysis and anomaly detection) to continuously monitor supply chain data and dynamically identify new risk points and potential quality problems. This monitoring method can reflect the operating status of the supply chain in real time, helping companies to discover and deal with potential problems in a timely manner; based on the feedback mechanism between the model prediction results and the actual quality situation, the model parameters are continuously adjusted through reinforcement learning. This feedback mechanism enables the model to gradually improve the prediction accuracy and robustness in continuous learning and optimization, so as to better adapt to the actual operation of the supply chain.

[0043] Obviously, those skilled in the art can make various changes and modifications 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 claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A quality traceability method for polymer materials in down clothing based on blockchain, characterized in that: The method comprises: S1. Label all polymer raw materials with the Internet of Things, use blockchain technology to create digital assets for each batch of raw materials, upload key data to the blockchain, and form a raw material archive; S2, real-time monitoring of production line related data, and uploading the relevant data to the cloud platform through the IoT gateway; S3. Assign unique logistics tags to finished down garments through RFID and GPS, record logistics information, and link the logistics information with the raw material archives and production records on the blockchain to form a complete supply chain view; S4. Access the blockchain platform through the QR code on the clothing or enter the unique ID to view the complete traceability chain; S5. Based on big data analysis technology, potential risk points in the supply chain are explored, and possible quality problems that may arise in the future are predicted based on historical data through machine learning models.

2. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 1 is characterized in that: Said S1 comprises: S11. Based on the polymer raw material information collection standard, RFID technology is used in combination with QR code to record the label information; S12. Introduce blockchain hash algorithm to encrypt raw material information and generate a unique hash value as a label anti-counterfeiting mark; combine the decentralized characteristics of blockchain to generate a unique digital asset ID for each batch of raw materials; S13. Deploy smart contracts on the blockchain. Based on the preset thresholds of raw material quality standards, if quality problems are detected, the early warning mechanism will be automatically triggered and the relevant batches will be locked; S14. Combine multi-source data to cross-verify raw material information. Based on the dynamic update mechanism of raw material archives, submit update applications under specific conditions. After verification by the blockchain consensus mechanism, perform seamless update of information.

3. The quality tracing method of polymer materials in down clothing based on blockchain according to claim 2 is characterized in that: The S13 comprises: Construct the logic rules of smart contracts according to the preset raw material quality standards; use the smart contract programming language to encode the constructed logic rules into smart contracts; Before deployment, the smart contract is tested, and the tested smart contract is deployed to the blockchain network and linked to the raw material information management system; Through the deployment of IoT sensors at key links, key quality parameters of raw materials are collected in real time, pre-processed, and the pre-processed data is uploaded to the blockchain; Smart contracts analyze uploaded quality data in real time according to preset threshold rules; once data anomalies are detected, the early warning mechanism is immediately triggered and early warning information is sent to relevant personnel; After the early warning is triggered, the smart contract automatically locks the batch of raw materials involved in the quality problem. At the same time, through the immutability of the blockchain, the information of the problematic batch is permanently recorded; Based on the type and severity of the warning, the smart contract automatically initiates the corresponding emergency response plan; the smart contract automatically generates a raw material quality monitoring report based on monitoring data and warning records; and the quality monitoring report is shared with all relevant parties through the blockchain network.

4. The quality tracing method of polymer materials in down clothing based on blockchain according to claim 1 is characterized in that: The S2 comprises: S21. Use monitoring equipment to monitor subtle changes in the production environment in real time, and use the health management system equipped with smart devices to monitor the operating status of equipment in real time; S22. Based on the event association algorithm, intelligently match production events with related data to form a complete event chain; based on the machine learning algorithm, build a production anomaly detection model to automatically identify abnormal patterns in the production process; S23. Based on the preset warning response level, the corresponding emergency handling process is automatically triggered according to the degree of abnormality.

5. The quality tracing method of polymer materials in down clothing based on blockchain according to claim 4 is characterized in that: The S22 comprises: Preprocess the raw data collected from monitoring equipment and equipment health management systems, and use feature engineering techniques to extract key event features from the preprocessed data; Combine raw material archives and quality inspection reports to build a multi-dimensional event feature set; Based on the event association algorithm, it automatically identifies the potential causal relationship and time series correlation between different data sources, intelligently matches various events in production with related data, and forms a continuous and complete event chain; Adopt unsupervised learning methods to train the integrated multi-dimensional event feature set, learn the data distribution characteristics under normal production conditions, and automatically identify data points that deviate from normal patterns; For the abnormal patterns identified by the model, cluster analysis is used to further analyze their internal laws and influencing factors, and on this basis, a production anomaly knowledge graph is constructed.

6. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 1 is characterized in that: The S3 includes: S31, encrypt RFID tags and GPS trackers, record the corresponding information, and deploy edge computing devices at logistics nodes to process logistics data in real time; S32. Based on big data visualization technology, the supply chain view is visualized; S33. Develop a path optimization algorithm based on historical logistics data and real-time traffic information to automatically plan the optimal logistics path; S34. Combined with blockchain smart contracts, set risk warning thresholds in the logistics process. If a warning is triggered, the preset response strategy will be automatically executed.

7. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 6 is characterized in that: The S33 comprises: Integrate multi-source data and pre-process the integrated multi-source data; Based on historical logistics data, a path planning algorithm is used to generate a preliminary set of optimal paths. In combination with real-time traffic information and vehicle status data, the preliminary paths are adjusted in real time through a machine learning model. Incorporate multi-objective optimization strategies into path planning, and use genetic algorithms to find the optimal solution set through multi-objective optimization functions; Use logistics simulation software to simulate the optimized path and test its performance in different scenarios; verify the path optimization algorithm by comparing the simulation results with the actual situation; Establish a feedback mechanism to collect data feedback and user evaluation during the actual transportation process and continuously optimize the path planning algorithm; Combining real-time traffic information and vehicle status data, potential transportation risks are identified, and based on risk warning results, preset emergency response strategies are automatically triggered. At the same time, machine learning models are used to evaluate the effects of different strategies in real time and select the optimal emergency plan.

8. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 1 is characterized in that: The S4 comprises: S41. Embed augmented reality technology in the QR code, so that consumers can view the clothing information of the clothing through the mobile phone camera after scanning, and provide multi-language support for the QR code and unique ID; S42. Consumers can understand the complete traceability chain of clothing through operations and based on the interactive traceability platform.

9. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 1 is characterized in that: The S5 comprises: S51. Build complex models based on deep learning algorithms; combine upstream and downstream supply chain data to conduct multi-dimensional analysis to identify potential quality issues and risk points; and continuously train and optimize the model based on newly collected data according to the model's self-learning and optimization mechanism; S52. Combine machine learning models with intelligent decision-making systems to provide managers with intelligent decision-making suggestions, introduce complex event processing algorithms into smart contracts, and handle complex events and abnormal situations in the supply chain; S53. Based on a multi-party collaboration mechanism, upstream and downstream enterprises in the supply chain are allowed to jointly participate in the formulation and execution of smart contracts.

10. The method for tracing the quality of polymer materials in down clothing based on blockchain according to claim 8 is characterized in that: The S51 includes: Collecting multi-source heterogeneous data and preprocessing the collected multi-source heterogeneous data, wherein the preprocessing includes removing noise, filling missing values, and data standardization; Based on domain knowledge and statistical methods, we extract characteristic variables that are crucial for quality prediction and fault warning, and use the long short-term memory network architecture to process time series data to predict the quality trend of products at different production stages. Combine convolutional neural networks and recurrent neural networks to build a hybrid model to comprehensively analyze image data and sensor data to identify potential failure modes; Combine supply chain network graph analysis to identify key nodes and vulnerable links and assess supply chain disruption risks; use unsupervised learning algorithms to continuously monitor supply chain data and dynamically identify new risk points and potential quality issues; Based on the feedback mechanism between the model prediction results and the actual quality conditions, the model parameters are continuously adjusted through reinforcement learning.

Images