A cold-chain storage multi-node environment anomaly monitoring method and system

Abstract

The application provides a cold-chain storage multi-node environment anomaly monitoring method and system, relates to the technical field of storage environment anomaly monitoring, and comprises the following steps: collecting multi-source heterogeneous data in real time in a cold-chain storage; preprocessing the multi-source heterogeneous data, performing time-space alignment and data fusion, and generating time-space distribution field data; inputting the time-space distribution field data into a preset anomaly detection model, identifying an environment anomaly mode in combination with time-space correlation analysis, and generating an anomaly judgment result; in response to the anomaly judgment result, triggering a linkage control strategy, and encrypting and storing multi-source heterogeneous data, the anomaly judgment result, and strategy disposal records of the linkage control strategy to obtain uplink data; and generating an environment quality traceability report for the whole life cycle based on the uplink data. The application can solve the technical problem that existing technologies cannot accurately identify abnormal data in a cold-chain storage environment, and achieve the technical effect of accurately identifying abnormal data and responding in a timely manner.

Description

Technical Field

[0001] This application relates to the field of abnormal monitoring technology for warehousing environment, and in particular to a method and system for monitoring abnormal environmental conditions at multiple nodes in cold chain warehousing. Background Technology

[0002] With the rapid development of the cold chain logistics industry, environmental monitoring during cold chain warehousing and transportation is receiving increasing attention, especially for commodities such as food and pharmaceuticals that have strict temperature and humidity requirements. Environmental quality control and traceability are therefore crucial. To ensure the quality of cold chain transportation, many cold chain warehousing systems have begun to rely on sensor networks to monitor environmental changes in real time.

[0003] Currently, existing cold chain monitoring systems use independent sensors for data acquisition and centralized computing systems for data analysis and processing. While this approach provides some environmental monitoring data, it faces several limitations. First, existing technologies lack effective fusion and spatiotemporal consistency analysis for the acquisition and processing of multi-source heterogeneous data. This leads to problems such as inaccurate data synchronization, data loss, or failure to detect anomalies in a timely manner in multi-sensor networks due to the large number of sensor nodes and the complexity of data types. Second, traditional data storage and transmission methods are not flexible enough for real-time identification and processing of abnormal data, and cannot effectively coordinate with other control systems for timely response when environmental anomalies occur.

[0004] In summary, existing technologies suffer from the technical problem of failing to accurately identify abnormal data in cold chain storage environments due to the lack of real-time processing and spatiotemporal consistency analysis mechanisms for multi-source heterogeneous data, which further affects real-time monitoring and anomaly handling of environmental quality. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for monitoring environmental anomalies in a multi-node cold chain warehouse, in order to solve the technical problem in the prior art that the lack of a real-time processing and spatiotemporal consistency analysis mechanism for multi-source heterogeneous data makes it impossible to accurately identify abnormal data in the cold chain warehouse environment, which further affects the real-time monitoring and anomaly handling of environmental quality.

[0006] In view of the above problems, this application provides a method and system for monitoring environmental anomalies at multiple nodes in cold chain warehousing.

[0007] Firstly, this application provides a method for monitoring environmental anomalies at multiple nodes in cold chain warehousing, implemented through a system for monitoring environmental anomalies at multiple nodes in cold chain warehousing. The method includes: deploying a multi-dimensional sensor network within the cold chain warehouse to collect multi-source heterogeneous data in real time; preprocessing the multi-source heterogeneous data via an edge computing gateway, performing spatiotemporal alignment and data fusion to generate spatiotemporal distribution field data of the warehousing environment; inputting the spatiotemporal distribution field data into a pre-set anomaly detection model, combining it with spatiotemporal correlation analysis to identify environmental anomaly patterns, and generating anomaly determination results; responding to the anomaly determination results, triggering a linkage control strategy, and encrypting and storing the multi-source heterogeneous data, the anomaly determination results, and the strategy handling records of the linkage control strategy on the blockchain to obtain blockchain data; and generating a full lifecycle environmental quality traceability report based on the blockchain data.

[0008] Preferably, the method for monitoring anomalies in a multi-node environment of a cold chain warehouse further includes: receiving the multi-source heterogeneous data through the edge computing gateway, mapping the data of each sensor node to a unified time axis using a time interpolation algorithm; and estimating the missing area data in the data of each sensor node after unifying the time axis based on the spatial coordinate position of each sensor node, thereby constructing spatiotemporal distribution field data covering the warehouse space.

[0009] Preferably, the method for monitoring environmental anomalies at multiple nodes in cold chain warehousing further includes: constructing a time series prediction model based on a long short-term memory network, training the time series prediction model using historical spatiotemporal distribution field data; inputting the spatiotemporal distribution field data at the current moment into the trained time series prediction model to predict the environmental parameter prediction value at the next moment; calculating the residual between the actual observed value and the predicted environmental parameter value, and identifying an environmental anomaly pattern when the residual exceeds a dynamic threshold.

[0010] Preferably, the method for monitoring multi-node environmental anomalies in cold chain warehousing further includes: inputting the spatiotemporal distribution field data into a preset anomaly detection model to generate single-node data anomalies and extracting the anomaly trend of the single-node data anomalies; for the single-node data anomalies, simultaneously extracting neighboring node data within the neighborhood space; if the neighboring node data synchronously shows an anomaly trend, it is determined to be a regional environmental anomaly; if the neighboring node data is normal, it is determined to be a single-node sensor failure of the single-node data; combining the regional environmental anomaly and the single-node sensor failure, generating the anomaly determination result.

[0011] Preferably, the method for monitoring multi-node environmental anomalies in cold chain warehousing further includes: if the neighboring node data synchronously shows an abnormal trend, determining whether the neighborhood anomaly trend of the neighboring node data is in the same direction as the abnormal trend of the single node data; if it is determined to be in the same direction, combining the neighboring node data with the single node data to obtain a regional environmental anomaly; if it is determined to be in opposite directions, using the neighboring node data as the second node data, extracting the neighboring node data in the neighborhood space based on the second node data anomaly traversal of the second node data, and adding the anomaly determination result corresponding to the neighboring node data to the anomaly determination result.

[0012] Preferably, the method for monitoring multi-node environmental anomalies in cold chain warehousing further includes: comparing the magnitude of the anomaly trend of the single node data anomaly with the magnitude of the anomaly trend of the second node data anomaly; taking the node data anomaly with the larger magnitude of the anomaly trend as the regional environmental anomaly trend, until the neighborhood space is traversed.

[0013] Preferably, the method for monitoring anomalies in a multi-node environment of cold chain warehousing further includes: packaging the multi-source heterogeneous data, the anomaly judgment result, and the policy handling record and corresponding hash value of the linkage control strategy to obtain block data; using each cold chain participant as a consortium chain node to perform consensus verification on the block data; after verification, distributing and storing the block data on each sensor node to obtain the on-chain data.

[0014] Preferably, the method for monitoring anomalies in a multi-node environment of cold chain warehousing further includes: automatically verifying, based on a smart contract, whether the on-chain data during storage exceeds a preset security threshold range; if compliant, automatically generating a qualified traceability certificate containing a blockchain electronic signature; if anomalies exist but have been handled in compliance, automatically generating a compliance exception report containing anomaly descriptions and policy handling records; and combining the qualified traceability certificate and the compliance exception report to generate the environmental quality traceability report.

[0015] Preferably, the method for monitoring environmental anomalies at multiple nodes in cold chain warehousing further includes: deploying fixed environmental sensor nodes at the shelf layers, entrances and exits, and return air vents of refrigeration equipment in the cold chain warehouse to obtain the multi-dimensional sensor network; integrating mobile RFID tags with built-in temperature-sensitive elements on the packaging of stored goods; collecting micro-environmental data of the goods based on the multi-dimensional sensor network; and polling and reading the micro-environmental data of the goods according to a preset period to generate the multi-source heterogeneous data.

[0016] Secondly, this application also provides a multi-node environmental anomaly monitoring system for cold chain warehousing, used to execute a multi-node environmental anomaly monitoring method for cold chain warehousing as described in the first aspect, comprising: a multi-source heterogeneous data acquisition module, used to deploy a multi-dimensional sensor network within the cold chain warehouse to acquire multi-source heterogeneous data in real time; a spatiotemporal distribution field data generation module, used to preprocess the multi-source heterogeneous data through an edge computing gateway, and perform spatiotemporal alignment and data fusion to generate spatiotemporal distribution field data of the warehousing environment; an anomaly judgment result generation module, used to input the spatiotemporal distribution field data into a preset anomaly detection model, combine spatiotemporal correlation analysis to identify environmental anomaly patterns, and generate an anomaly judgment result; an on-chain data acquisition module, used to respond to the anomaly judgment result, trigger a linkage control strategy, and encrypt and store the multi-source heterogeneous data, the anomaly judgment result, and the strategy handling record of the linkage control strategy on the blockchain to obtain on-chain data; and an environmental quality traceability report generation module, used to generate a full lifecycle environmental quality traceability report based on the on-chain data.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by realizing the efficient fusion and real-time monitoring of multi-source data, it achieves the technical effect of accurately identifying abnormal data and responding in a timely manner.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a multi-node environmental anomaly monitoring method for cold chain warehousing according to this application.

[0021] Figure 2 This is a schematic diagram of the structure of a multi-node environmental anomaly monitoring system for cold chain warehousing according to this application.

[0022] Figure labeling: 1. Multi-source heterogeneous data acquisition module; 2. Spatiotemporal distribution field data generation module; 3. Anomaly judgment result generation module; 4. On-chain data acquisition module; 5. Environmental quality traceability report generation module. Detailed Implementation

[0023] This application provides a method and system for monitoring environmental anomalies at multiple nodes in cold chain warehousing. It solves the technical problem in existing technologies where the lack of real-time processing and spatiotemporal consistency analysis mechanisms for multi-source heterogeneous data leads to the inability to accurately identify abnormal data in the cold chain warehousing environment, further impacting real-time monitoring and anomaly handling of environmental quality. The method achieves efficient fusion and real-time monitoring of multi-source data, resulting in accurate identification of abnormal data and timely response.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a method for monitoring environmental anomalies at multiple nodes in cold chain warehousing, which is applied to a system for monitoring environmental anomalies at multiple nodes in cold chain warehousing. The method specifically includes the following steps: Deploy a multi-dimensional sensor network within the cold chain warehouse to collect multi-source heterogeneous data in real time.

[0026] Furthermore, this application also includes: deploying fixed environmental sensor nodes on the shelf layers, entrances and exits, and return air vents of refrigeration equipment in cold chain warehouses to obtain the multi-dimensional sensor network; integrating mobile RFID tags with built-in temperature-sensitive elements on the packaging of stored goods, collecting micro-environmental data of the goods based on the multi-dimensional sensor network; and polling and reading the micro-environmental data of the goods according to a preset period to generate the multi-source heterogeneous data.

[0027] Specifically, fixed environmental sensor nodes are deployed at the shelving levels, entrances / exits, and return air vents of refrigeration equipment in cold chain warehouses, forming a multi-dimensional sensing network covering the warehouse space. These environmental sensor nodes can collect key parameters such as temperature, humidity, airflow, and gas concentration in the warehouse environment in real time, ensuring comprehensive monitoring of the cold chain storage environment. By deploying environmental sensor nodes in different locations within the warehouse, environmental changes in different areas can be effectively captured, providing fundamental data support for subsequent data analysis and anomaly detection.

[0028] Mobile RFID tags with built-in temperature-sensitive elements are integrated into the packaging of stored goods to collect data on the microscopic environment surrounding the goods. The temperature-sensitive elements monitor real-time temperature changes by being embedded within the packaging, while the RFID tags wirelessly interact with sensor nodes within the warehouse. This structure enables environmental monitoring at the goods level, providing a basis for accurately identifying and analyzing potential temperature and humidity changes that may occur during storage.

[0029] The system polls and reads micro-environmental data of goods according to a preset cycle, generating multi-source heterogeneous data. By periodically acquiring micro-environmental data and synchronously transmitting it to the warehouse management system, changes in the packaging and surrounding environment of goods can be monitored in real time. This facilitates the analysis of trends in the cold chain environment and provides multi-dimensional data support for subsequent anomaly detection and quality traceability.

[0030] The multi-source heterogeneous data is preprocessed by the edge computing gateway, and spatiotemporal alignment and data fusion are performed to generate spatiotemporal distribution field data of the warehousing environment.

[0031] Furthermore, this application also includes: receiving the multi-source heterogeneous data through the edge computing gateway, mapping the data of each sensor node to a unified time axis using a time interpolation algorithm; and estimating the missing region data in the data of each sensor node after unifying the time axis based on the spatial coordinate position of each sensor node, using a spatial interpolation algorithm to construct spatiotemporal distribution field data covering the warehouse space.

[0032] Specifically, edge computing gateways receive heterogeneous data from multiple sources. By processing and transforming data from different sensor nodes, they use time interpolation algorithms to map the collected data from each sensor node to a unified time axis. The time interpolation algorithm standardizes sensor data at different points in time, ensuring that data from different sensor nodes can be compared and analyzed within the same time frame, laying the foundation for subsequent data fusion and spatiotemporal analysis.

[0033] When performing spatiotemporal data fusion, the spatial coordinates of each sensor node are considered, and a spatial interpolation algorithm is used to estimate the values ​​of missing regions in the sensor data after unifying the time axis. The spatial interpolation algorithm analyzes the spatial distribution relationships between existing data points, infers reasonable values ​​for missing data points, and constructs spatiotemporal distribution field data covering the entire warehouse space. This allows for the filling of data gaps caused by different sensor locations or data loss, achieving complete spatiotemporal data reconstruction.

[0034] The spatiotemporal distribution field data is input into a preset anomaly detection model, and spatiotemporal correlation analysis is used to identify environmental anomaly patterns and generate anomaly judgment results.

[0035] Furthermore, this application also includes: constructing a time series prediction model based on a long short-term memory network, training the time series prediction model using historical spatiotemporal distribution field data; inputting the spatiotemporal distribution field data of the current moment into the trained time series prediction model to predict the environmental parameter prediction value of the next moment; calculating the residual between the actual observed value and the predicted environmental parameter value, and identifying an abnormal environmental pattern when the residual exceeds a dynamic threshold.

[0036] Furthermore, this application also includes: inputting the spatiotemporal distribution field data into a preset anomaly detection model to generate single-node data anomalies and extracting the anomaly trend of the single-node data anomalies; for the single-node data anomalies, simultaneously extracting neighboring node data within the neighborhood space; if the neighboring node data synchronously shows an anomaly trend, it is determined to be a regional environmental anomaly; if the neighboring node data is normal, it is determined to be a single-node sensor failure of the single-node data; combining the regional environmental anomaly and the single-node sensor failure, generating the anomaly determination result.

[0037] Furthermore, this application also includes: if the neighboring node data synchronously shows an abnormal trend, determining whether the neighborhood abnormal trend of the neighboring node data is in the same direction as the abnormal trend of the single node data; if it is determined to be in the same direction, combining the neighboring node data with the single node data to obtain a regional environmental anomaly; if it is determined to be in opposite directions, using the neighboring node data as the second node data, extracting the neighboring node data in the neighborhood space according to the second node data anomaly traversal of the second node data, and adding the anomaly determination result corresponding to the neighboring node data to the anomaly determination result.

[0038] Furthermore, this application also includes: comparing the abnormal trend magnitude of the single node data anomaly with the abnormal trend magnitude of the second node data anomaly; taking the node data anomaly with the larger abnormal trend magnitude as the regional environmental anomaly trend, until the neighborhood space is traversed.

[0039] Specifically, a time series prediction model based on a Long Short-Term Memory (LSTM) network is constructed and trained using historical spatiotemporal distribution data. LSTM is a special type of recurrent neural network that effectively captures long-term dependencies in time series data. By training the model on historical spatiotemporal distribution data, it learns the regularity of environmental parameters changing over time, thereby enabling accurate predictions of future environmental changes.

[0040] After model training is complete, the spatiotemporal distribution data of the current moment is input into the trained time series prediction model to predict environmental parameters for the next moment. By analyzing the current environmental state using the model trained with historical data, the values ​​of various environmental parameters in future moments are predicted to determine whether the environmental parameters are normal and to provide a basis for subsequent anomaly detection.

[0041] The residual between the actual observed values ​​and the predicted values ​​is calculated. When the residual exceeds a dynamic threshold, it is identified as an abnormal environmental pattern. The residual is the difference between the actual observed values ​​and the model's predicted values. By calculating this difference, it is possible to detect whether the environmental data deviates from expectations. When the residual exceeds a preset dynamic threshold, it indicates that the current environmental state may be abnormal, thereby triggering the anomaly detection mechanism and taking corresponding control and response actions.

[0042] Spatiotemporal distribution data is input into a pre-defined anomaly detection model to generate single-node data anomalies and extract the anomaly trend of these single-node anomalies. The anomaly detection model processes the input spatiotemporal distribution data, analyzes the anomaly behavior of each sensor node at different times and spatial locations, and identifies the anomaly of a single node at a specific moment. By analyzing the trend of the anomaly data, the anomaly development trajectory of the node over a future period can be extracted, providing a basis for further anomaly assessment.

[0043] For data anomalies at a single node, neighboring node data within the neighborhood space are extracted simultaneously. To further confirm the nature of the anomaly, neighborhood analysis is used to obtain data from other sensor nodes related to the single node. By comparing data within the neighborhood space, the correlation between different nodes can be comprehensively considered to determine whether there are broader anomaly patterns or fault conditions.

[0044] If neighboring node data synchronously exhibits an abnormal trend, it is determined to be a regional environmental anomaly. In this case, if multiple neighboring node data show similar abnormal trends, it indicates that the problem may not be caused by a single node, but rather that the entire region's environment is experiencing issues. By identifying these interrelated abnormal data, the scope of the abnormal event can be determined, and appropriate regional control measures can be taken.

[0045] If the data from neighboring nodes is normal, the problem is determined to be a single-node sensor malfunction. If no anomalies are shown in the neighboring data, it can be determined that the anomaly in the single node is not due to environmental changes, but rather a malfunction in the node's own sensor. Therefore, by comparing and analyzing the neighboring data, the location and nature of the problem can be accurately pinpointed.

[0046] By combining regional environmental anomalies and single-node sensor failures, anomaly determination results are generated. Through comprehensive analysis of the detection results of regional environmental anomalies and single-node sensor failures, a comprehensive assessment of abnormal events can be made, the root cause of the problem can be identified, and data support can be provided for subsequent control strategies.

[0047] If neighboring node data synchronously exhibits an abnormal trend, determine whether the abnormal trend of the neighboring node data is in the same direction as the abnormal trend of the single node data. Analyze the neighboring node data to determine if its abnormal trend is consistent with that of the single node data. If the abnormal changes in the two data sources are in the same direction, it indicates that the same environmental factors or fault sources may have affected the two nodes, leading to a correlation in their abnormal trends.

[0048] If the trend is determined to be the same, the data from neighboring nodes is combined with the data from a single node to obtain a regional environmental anomaly. When the anomaly trends of multiple neighboring nodes' data are consistent with those of a single node's data, it means that the overall environmental condition of that area may be problematic. By combining the data from neighboring nodes with the data from a single node, a regional environmental anomaly pattern can be formed, thereby identifying environmental anomalies affecting a wider area and further improving the accuracy and responsiveness of environmental monitoring.

[0049] If an anomaly is identified, the neighboring node data is used as the second node data. Based on the anomaly of the second node data, neighboring node data within the neighborhood space is extracted, and the anomaly detection results corresponding to the neighboring node data are added to the overall anomaly detection result. If the anomaly trend of the neighboring node data is opposite to the anomaly trend of the single node data, it is considered that the anomaly may be caused by different sources. In this case, through anomaly analysis of the second node data, data within its neighborhood is extracted, and anomaly detection is performed again to obtain a more comprehensive anomaly identification result. This additional anomaly detection information is added to the original anomaly detection to ultimately form a more accurate judgment.

[0050] The magnitude of the abnormal trend in data from a single node is compared with the magnitude of the abnormal trend in data from a second node. By comparing the magnitudes of the abnormal trends in the data from the single node and the second node, it is possible to determine which node's abnormal trend is more significant. The magnitude of the abnormal trend refers to the degree of abnormal change in the data; a larger magnitude indicates a more severe anomaly. Therefore, by comparing the magnitudes of the abnormal trends in these two nodes, it is possible to determine which node's anomaly has a greater impact on the environment, thus providing a basis for identifying anomaly patterns.

[0051] Nodes exhibiting larger anomalous trend amplitudes are identified as regional environmental anomalies, and this process is repeated across the surrounding data space. If a node's anomalous trend amplitude is found to be large, it means that the anomaly reflected by that node may have an impact on a wider area of ​​the environment. Therefore, the anomalous trend of that node is considered a regional environmental anomaly trend, and this is further extended to the surrounding data space. By traversing the data of surrounding neighboring nodes and comprehensively considering the anomalies of multiple nodes, it is ultimately determined whether a large-scale environmental anomaly exists.

[0052] In response to the anomaly determination result, a linkage control strategy is triggered, and the multi-source heterogeneous data, the anomaly determination result, and the strategy handling record of the linkage control strategy are encrypted and stored on the blockchain to obtain on-chain data.

[0053] Furthermore, this application also includes: packaging the multi-source heterogeneous data, the anomaly determination result, and the policy handling record and corresponding hash value of the linkage control strategy to obtain block data; using each cold chain participant as a consortium chain node to perform consensus verification on the block data; after verification, distributing and storing the block data on each sensor node to obtain the on-chain data.

[0054] Specifically, in response to the anomaly detection result, a linkage control strategy is triggered. This involves packaging multi-source heterogeneous data, the anomaly detection result, the strategy handling record of the linkage control strategy, and the corresponding hash value into a block of data. After anomaly detection is completed, the linkage control strategy is initiated based on the anomaly detection result, executing relevant control measures and recording detailed information about the strategy execution. Multi-source heterogeneous data refers to data from different types of sensors, the anomaly detection result is the detected anomaly information, the strategy handling record of the linkage control strategy documents the specific execution process of the strategy, and the hash value is used to ensure data integrity and consistency.

[0055] Each participant in the cold chain acts as a node in the consortium blockchain, performing consensus verification of block data. After block data is generated, each participant in the cold chain system becomes a node in the consortium blockchain and verifies the block data. Through a consensus mechanism, different nodes ensure the accuracy and reliability of the data by reviewing and confirming the block data. The verification process ensures data consistency and prevents data tampering or errors, thus guaranteeing the security and trustworthiness of the blockchain.

[0056] After successful verification, the block data is distributed and stored across various sensor nodes, resulting in on-chain data. Block data verified through consensus will be distributed and stored across various sensor nodes in the network to ensure data redundancy and security. By storing data on multiple nodes, high availability and fault tolerance can be achieved, ensuring that even if some nodes fail, data can still be retrieved on other nodes, ultimately forming on-chain data with traceability and immutability.

[0057] An environmental quality traceability report covering the entire lifecycle is generated based on the on-chain data.

[0058] Furthermore, this application also includes: automatically verifying whether the on-chain data during storage exceeds a preset security threshold range based on the smart contract; if compliant, automatically generating a qualified traceability certificate containing a blockchain electronic signature; if there is an anomaly but it has been handled in compliance, automatically generating a compliance exception report containing an anomaly description and policy handling records; and combining the qualified traceability certificate and the compliance exception report to generate the environmental quality traceability report.

[0059] Specifically, smart contracts automatically verify whether on-chain data stored during the storage period exceeds preset security threshold ranges. A smart contract is a self-executing contract that can automatically execute and verify conditions according to preset rules. During data storage, smart contracts automatically check whether the on-chain data exceeds pre-defined security threshold ranges, which may involve upper and lower limits of environmental parameters such as temperature and humidity. Through the automatic verification mechanism of smart contracts, data compliance can be ensured, and warnings or actions can be triggered promptly when anomalies are detected.

[0060] If compliant, a valid traceability certificate containing a blockchain digital signature is automatically generated. When the stored data meets security threshold requirements, the system will automatically generate a valid traceability certificate according to a preset process. This certificate contains a blockchain digital signature, proving that the data has not been tampered with during storage and meets predetermined security standards. The blockchain digital signature ensures the immutability and reliability of the traceability certificate, providing legal validity and data verification basis for subsequent quality traceability.

[0061] If an anomaly is detected but has been handled in compliance with regulations, a compliance exception report will be automatically generated, including an explanation of the anomaly and records of the appropriate handling strategy. When a data anomaly is detected, if it is promptly identified and handled in compliance with regulations, the system will generate a compliance exception report. The report will record the specific details of the anomaly, including its nature, the time of occurrence, and the handling strategy adopted. This report provides documentation support for the compliant handling of anomaly events and ensures that all anomalies are handled according to prescribed procedures.

[0062] By combining the valid traceability certificate and the compliance exception report, an environmental quality traceability report is generated. Ultimately, the system will combine the valid traceability certificate and the compliance exception report to generate a comprehensive environmental quality traceability report. This report not only proves the compliance of the data but also records the handling process of all abnormal events.

[0063] In summary, the multi-node environmental anomaly monitoring method for cold chain warehousing provided in this application has the following technical effects: by realizing the efficient fusion and real-time monitoring of multi-source data, it achieves the technical effect of accurately identifying abnormal data and responding in a timely manner.

[0064] Example 2: Based on the same inventive concept as the multi-node environmental anomaly monitoring method for cold chain warehousing in the foregoing examples, this application also provides a multi-node environmental anomaly monitoring system for cold chain warehousing. Please refer to the appendix. Figure 2 The system includes: a multi-source heterogeneous data acquisition module 1, used to deploy a multi-dimensional sensor network within the cold chain warehouse to collect multi-source heterogeneous data in real time; a spatiotemporal distribution field data generation module 2, used to preprocess the multi-source heterogeneous data through an edge computing gateway, and perform spatiotemporal alignment and data fusion to generate spatiotemporal distribution field data of the warehouse environment; an anomaly judgment result generation module 3, used to input the spatiotemporal distribution field data into a preset anomaly detection model, combine spatiotemporal correlation analysis to identify environmental anomaly patterns, and generate anomaly judgment results; an on-chain data acquisition module 4, used to respond to the anomaly judgment results, trigger a linkage control strategy, and encrypt and store the multi-source heterogeneous data, the anomaly judgment results, and the strategy handling records of the linkage control strategy on the blockchain to obtain on-chain data; and an environmental quality traceability report generation module 5, used to generate a full lifecycle environmental quality traceability report based on the on-chain data.

[0065] Furthermore, the aforementioned multi-node environmental anomaly monitoring system for cold chain warehousing is also used to: receive the multi-source heterogeneous data through the edge computing gateway, map the data of each sensor node to a unified time axis using a time interpolation algorithm; and, based on the spatial coordinates of each sensor node, estimate the missing area data in the data of each sensor node after unifying the time axis using a spatial interpolation algorithm to construct spatiotemporal distribution field data covering the warehousing space.

[0066] Furthermore, the aforementioned multi-node environmental anomaly monitoring system for cold chain warehousing is also used for: constructing a time series prediction model based on a long short-term memory network, training the time series prediction model using historical spatiotemporal distribution field data; inputting the spatiotemporal distribution field data of the current moment into the trained time series prediction model to predict the environmental parameter prediction value of the next moment; calculating the residual between the actual observed value and the predicted environmental parameter value, and identifying an environmental anomaly mode when the residual exceeds a dynamic threshold.

[0067] Furthermore, the aforementioned multi-node environmental anomaly monitoring system for cold chain warehousing is also used for: inputting the spatiotemporal distribution field data into a preset anomaly detection model, generating single-node data anomalies, and extracting the anomaly trend of the single-node data anomalies; for the single-node data anomalies, simultaneously extracting neighboring node data within the neighborhood space; if the neighboring node data synchronously shows an anomaly trend, it is determined to be a regional environmental anomaly; if the neighboring node data is normal, it is determined to be a single-node sensor failure of the single-node data; combining the regional environmental anomaly and the single-node sensor failure, generating the anomaly determination result.

[0068] Furthermore, the cold chain warehousing multi-node environmental anomaly monitoring system is also used for: if the neighboring node data synchronously shows an abnormal trend, determining whether the neighborhood anomaly trend of the neighboring node data is in the same direction as the abnormal trend of the single node data; if it is determined to be in the same direction, combining the neighboring node data with the single node data to obtain a regional environmental anomaly; if it is determined to be in opposite directions, using the neighboring node data as the second node data, extracting the neighboring node data in the neighborhood space based on the second node data anomaly traversal of the second node data, and adding the anomaly determination result corresponding to the neighboring node data to the anomaly determination result.

[0069] Furthermore, the cold chain warehousing multi-node environmental anomaly monitoring system is also used to: compare the abnormal trend magnitude of the single node data anomaly with the abnormal trend magnitude of the second node data anomaly; and take the node data anomaly with the larger abnormal trend magnitude as the regional environmental anomaly trend, until the neighborhood space is traversed.

[0070] Furthermore, the aforementioned cold chain warehousing multi-node environmental anomaly monitoring system is also used to: package the multi-source heterogeneous data, the anomaly judgment results, and the policy handling records and corresponding hash values ​​of the linkage control strategy to obtain block data; use each cold chain participant as a consortium chain node to perform consensus verification on the block data; after verification, distribute and store the block data on each sensor node to obtain the on-chain data.

[0071] Furthermore, the aforementioned multi-node environmental anomaly monitoring system for cold chain warehousing is also used for: automatically verifying, based on smart contracts, whether the on-chain data during storage exceeds a preset security threshold range; if compliant, automatically generating a qualified traceability certificate containing a blockchain electronic signature; if an anomaly exists but has been handled in compliance, automatically generating a compliance exception report containing an anomaly description and policy handling records; and combining the qualified traceability certificate and the compliance exception report to generate the environmental quality traceability report.

[0072] Furthermore, the aforementioned multi-node environmental anomaly monitoring system for cold chain warehousing is also used for: deploying fixed environmental sensor nodes on the shelf layers, entrances and exits, and return air vents of refrigeration equipment in the cold chain warehouse to obtain the multi-dimensional sensor network; integrating mobile RFID tags with built-in temperature-sensitive elements on the packaging of stored goods, and collecting micro-environmental data of the goods based on the multi-dimensional sensor network; and polling and reading the micro-environmental data of the goods according to a preset period to generate the multi-source heterogeneous data.

[0073] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The method and specific examples for monitoring multi-node environmental anomalies in cold chain warehousing described in the foregoing embodiment one are also applicable to the system for monitoring multi-node environmental anomalies in cold chain warehousing in this embodiment. Through the foregoing detailed description of the method for monitoring multi-node environmental anomalies in cold chain warehousing, those skilled in the art can clearly understand the system for monitoring multi-node environmental anomalies in cold chain warehousing in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0074] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0075] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for monitoring environmental anomalies at multiple nodes in cold chain warehousing, characterized in that, include: Deploy a multi-dimensional sensor network within cold chain warehouses to collect multi-source heterogeneous data in real time; The multi-source heterogeneous data is preprocessed through an edge computing gateway, and spatiotemporal alignment and data fusion are performed to generate spatiotemporal distribution field data of the warehousing environment. The spatiotemporal distribution field data is input into a preset anomaly detection model, and combined with spatiotemporal correlation analysis, anomaly patterns in the environment are identified to generate anomaly judgment results. In response to the anomaly determination result, a linkage control strategy is triggered, and the multi-source heterogeneous data, the anomaly determination result, and the strategy handling record of the linkage control strategy are encrypted and stored on the blockchain to obtain on-chain data; An environmental quality traceability report covering the entire lifecycle is generated based on the on-chain data.

2. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 1, characterized in that, The multi-source heterogeneous data is preprocessed through an edge computing gateway, and spatiotemporal alignment and data fusion are performed to generate spatiotemporal distribution field data of the warehouse environment, including: The edge computing gateway receives the multi-source heterogeneous data and uses a time interpolation algorithm to map the data of each sensor node to a unified time axis. Based on the spatial coordinates of each sensor node, a spatial interpolation algorithm is used to estimate the missing data in the data of each sensor node after unifying the time axis, thereby constructing a spatiotemporal distribution field data covering the warehouse space.

3. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 1, characterized in that, Constructing an anomaly detection model, including: A time series prediction model based on a long short-term memory network is constructed, and the model is trained using historical spatiotemporal distribution data. Input the spatiotemporal distribution field data at the current moment into the trained time series prediction model to predict the environmental parameter values ​​at the next moment. Calculate the residual between the actual observed value and the predicted value of the environmental parameter. When the residual exceeds a dynamic threshold, it is identified as an abnormal environmental pattern.

4. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 1, characterized in that, The spatiotemporal distribution field data is input into a preset anomaly detection model, and combined with spatiotemporal correlation analysis to identify environmental anomaly patterns and generate anomaly judgment results, including: The spatiotemporal distribution field data is input into a preset anomaly detection model to generate single-node data anomalies and extract the anomaly trend of the single-node data anomalies. In response to the anomaly in the data of a single node, the data of neighboring nodes in the neighborhood space are extracted simultaneously. If the data synchronization of the neighboring nodes shows an abnormal trend, it is determined to be a regional environmental anomaly. If the neighboring node data is normal, it is determined that the single-node sensor is faulty. The anomaly determination result is generated by combining the regional environmental anomaly and the single-node sensor failure.

5. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 4, characterized in that, The process of inputting the spatiotemporal distribution field data into a pre-set anomaly detection model, combining it with spatiotemporal correlation analysis to identify environmental anomaly patterns, and generating anomaly determination results also includes: If the neighboring node data synchronously shows an abnormal trend, determine whether the abnormal trend of the neighboring node data is in the same direction as the abnormal trend of the single node data. If the direction is determined to be the same, the neighboring node data is combined with the single node data to obtain the regional environmental anomaly. If it is determined to be an anomaly, the neighboring node data is used as the second node data. The neighboring node data in the neighborhood space is extracted by abnormal traversal of the second node data based on the second node data, and the anomaly determination result corresponding to the neighboring node data is added to the anomaly determination result.

6. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 5, characterized in that, The process of inputting the spatiotemporal distribution field data into a pre-set anomaly detection model, combining it with spatiotemporal correlation analysis to identify environmental anomaly patterns, and generating anomaly determination results also includes: Compare the magnitude of the abnormal trend of the single node data anomaly with the magnitude of the abnormal trend of the second node data anomaly; Node data anomalies with larger abnormal trend magnitudes are identified as regional environmental anomalies, and this process continues until the entire neighborhood space is traversed.

7. The method for monitoring multi-node environmental anomalies in cold chain warehousing as described in claim 1, characterized in that, The multi-source heterogeneous data, the anomaly determination results, and the policy handling records of the linkage control strategy are encrypted and stored on the blockchain to obtain on-chain data, including: The multi-source heterogeneous data, the anomaly determination results, and the policy handling records and corresponding hash values ​​of the linkage control strategy are packaged to obtain block data; Each cold chain participant acts as a consortium blockchain node to perform consensus verification on the block data; After successful verification, the block data is distributed and stored on each sensor node to obtain the on-chain data.

8. The method for monitoring environmental anomalies at multiple nodes in cold chain warehousing as described in claim 1, characterized in that, Based on the on-chain data, a full lifecycle environmental quality traceability report is generated, including: The smart contract automatically verifies whether the on-chain data stored during the storage period exceeds a preset security threshold range. If compliant, a qualified traceability certificate containing a blockchain electronic signature will be automatically generated; If an anomaly exists but has been handled in compliance with regulations, a compliance exception report containing an explanation of the anomaly and records of the policy handling will be automatically generated. The environmental quality traceability report is generated by combining the qualified traceability certificate and the compliance exception report.

9. The method for monitoring environmental anomalies at multiple nodes in cold chain warehousing as described in claim 1, characterized in that, Deploying a multi-dimensional sensor network within cold chain warehouses to collect multi-source heterogeneous data in real time, including: Fixed environmental sensor nodes are deployed on the shelf layers, entrances and exits, and return air vents of refrigeration equipment in cold chain warehouses to obtain the multidimensional sensor network. Mobile RFID tags with built-in temperature-sensitive elements are integrated into the packaging of stored goods, and the micro-environmental data of the goods are collected based on the multi-dimensional sensor network. The cargo micro-environment data is read in a polling cycle according to a preset period to generate the multi-source heterogeneous data.

10. A multi-node environmental anomaly monitoring system for cold chain warehousing, characterized in that, The steps for implementing the multi-node environmental anomaly monitoring method for cold chain warehousing according to any one of claims 1 to 9 include: The multi-source heterogeneous data acquisition module is used to deploy a multi-dimensional sensor network in cold chain warehouses to collect multi-source heterogeneous data in real time. The spatiotemporal distribution field data generation module is used to preprocess the multi-source heterogeneous data through the edge computing gateway, and perform spatiotemporal alignment and data fusion to generate spatiotemporal distribution field data of the warehousing environment. The anomaly determination result generation module is used to input the spatiotemporal distribution field data into a preset anomaly detection model, combine it with spatiotemporal correlation analysis to identify environmental anomaly patterns, and generate anomaly determination results. The on-chain data acquisition module is used to respond to the anomaly determination result, trigger the linkage control strategy, and encrypt and store the multi-source heterogeneous data, the anomaly determination result, and the strategy handling record of the linkage control strategy on the blockchain to obtain on-chain data. The environmental quality traceability report generation module is used to generate a full lifecycle environmental quality traceability report based on the on-chain data.

Images