A method and system for processing industrial production data based on blockchain

Abstract

The present invention discloses a method and system for industrial production data processing based on blockchain, which relates to data processing related fields. The method includes: pre-building an industrial production cloud-edge architecture; obtaining multiple device IDs, obtaining multiple device fault identification modules; synchronizing to multiple edge fault identification nodes, identifying multiple edge fault identification nodes; pre-building a data management private chain; analyzing equipment operation data for fault identification, and performing real-time storage based on the results; pre-setting a data storage frequency threshold, extracting blockchain data in the data management private chain, and obtaining multiple operation fault data sets; optimizing the industrial production cloud-edge architecture, and performing dynamic identification of operation faults. The method solves the technical problems of insufficient data reliability in existing industrial production data processing and affecting the accuracy of fault identification, and achieves the technical effect of ensuring the authenticity and integrity of industrial production data through the non-tamperability of blockchain, thereby improving the accuracy and reliability of fault identification.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a blockchain-based industrial production data processing method and system. Background Art

[0002] With the rapid development of the industrial Internet, industrial production has increasingly higher requirements for the accuracy and reliability of fault identification. The reliability of fault identification is directly related to the stable operation of the production line, product quality and overall production efficiency. However, when dealing with massive and complex industrial data, existing fault identification methods are difficult to ensure the authenticity and integrity of data. Data is easily tampered or lost during transmission and storage, resulting in low data credibility and identification results being easily disturbed, thus affecting the accuracy of fault identification.

[0003] In the current related technologies, industrial production data processing has the technical problem of insufficient data reliability, which affects the accuracy of fault identification. Summary of the invention

[0004] The present application provides an industrial production data processing method and system based on blockchain, adopts the construction of industrial production cloud-edge architecture, synchronizes multiple equipment fault identification modules to multiple edge fault identification nodes, builds a data management private chain, adopts multiple edge fault identification nodes for fault identification, performs real-time storage of the data management private chain based on the fault identification results, extracts multiple operation fault data sets from the blockchain, adopts multiple operation fault data sets to optimize the industrial production cloud-edge architecture, performs dynamic identification of operation faults of multiple industrial production equipment, and other technical means, thereby achieving the technical effect of ensuring the authenticity and integrity of industrial production data through the tamper-proof nature of blockchain, thereby improving the accuracy and reliability of fault identification.

[0005] This application provides an industrial production data processing method based on blockchain, including:

[0006] Pre-build an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes; interactively obtain multiple device IDs of multiple industrial production equipment in the production environment, and call and obtain multiple device fault identification modules based on the multiple device IDs; after synchronizing the multiple device fault identification modules to the multiple edge fault identification nodes, use the multiple device IDs as multiple device unique identifiers to identify the multiple edge fault identification nodes, and complete the localization update of the industrial production cloud-edge architecture; pre-build a data management private chain; during the operation of the multiple industrial production equipment, use the multiple edge fault identification nodes to analyze the real-time uploaded device operation data of the multiple industrial production equipment for fault identification, and store the operation data in real time in the data management private chain based on the fault identification results; preset a data storage frequency threshold, extract blockchain data in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers, and obtain multiple operation fault data sets; use the multiple operation fault data sets to optimize the industrial production cloud-edge architecture, and use the optimized industrial production cloud-edge architecture to perform dynamic identification of operation faults of the multiple industrial production equipment.

[0007] In a possible implementation, multiple device IDs of multiple industrial production devices in a production environment are interactively obtained, and multiple device fault identification modules are called according to the multiple device IDs to perform the following processing:

[0008] Matching equipment purchase logs is performed according to the multiple equipment IDs to obtain multiple equipment model information; network data collection is performed according to the multiple equipment model information to obtain multiple sample fault data sets; data structure integration is performed on the multiple sample fault data sets to obtain multiple standard fault data structures, and the multiple standard fault data structures are used to perform data cleaning on the multiple sample fault data sets to obtain multiple optimized fault data sets; standard operators are obtained from an operator library, and the multiple optimized fault data sets are used to train the standard operators to obtain multiple equipment fault identification modules.

[0009] In a possible implementation, a data management private chain is pre-built to perform the following processing:

[0010] A data management smart contract is written and generated according to the multiple device unique identifiers; the multiple standard fault data structures and the data management smart contract are deployed on a pre-selected blockchain platform to complete the construction of the data management private chain.

[0011] In a possible implementation, during the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the real-time uploaded equipment operation data of the plurality of industrial production equipment for fault identification, and the operation data is stored in real time in the data management private chain based on the fault identification result, and the following processing is performed:

[0012] During the operation of the multiple industrial production equipment, the first industrial production equipment uploads the first equipment operation data to the first edge fault identification node in real time to perform fault identification and obtain a first identification result, wherein the first industrial production equipment is any one of the multiple industrial production equipment, and the first edge fault identification node uses the first equipment ID of the first industrial production equipment as the first equipment unique identifier; a fault time window is preset, and if the first identification result is a fault, the first identification time is extracted according to the first identification result, and the first identification time and the fault time window are used as constraints to call the local operation log of the first industrial production equipment to obtain the first operation data; the first operation data is stored in real time in the data management private chain.

[0013] In a possible implementation, the following processing is performed:

[0014] If the first identification result is not a fault, the local operation log of the first industrial production equipment is formatted.

[0015] In a possible implementation, a data storage frequency threshold is preset, blockchain data is extracted in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers, multiple operation fault data sets are obtained, and the following processing is performed:

[0016] Interact with the first edge fault identification node to obtain a first fault identification frequency; if the first fault identification frequency meets the data storage frequency threshold, run the data management smart contract according to the first device unique identifier to perform data-oriented extraction in the data management private chain to obtain a first operation fault data set; and so on, to obtain the multiple operation fault data sets.

[0017] In a possible implementation, the industrial production cloud-edge architecture is optimized using the multiple operation fault data sets, and the following processing is performed:

[0018] Preset a training data reorganization ratio; use the multiple operating fault data sets to partially replace the multiple optimized fault data sets with the training data reorganization ratio as a constraint to obtain multiple updated fault data sets; use the multiple updated fault data sets as training data to update the model parameters of the multiple equipment fault identification modules.

[0019] In a possible implementation, the following processing is performed:

[0020] If the first identification result is a fault, the first identification time and the first device ID are associated and stored in the fault record cloud.

[0021] In a possible implementation, the following processing is performed:

[0022] Preset a fault correlation analysis window; use the fault correlation analysis window as a constraint, interact with the fault record cloud, and obtain multiple identification times of the multiple device IDs; perform correlation analysis on the multiple identification times to obtain multiple joint maintenance device groups; if the first identification result is a fault, traverse the multiple joint maintenance device groups according to the first device ID to obtain a first joint maintenance device group; send the first joint maintenance device group as an operation and maintenance signal to the operation and maintenance terminal.

[0023] This application also provides an industrial production data processing system based on blockchain, including:

[0024] An industrial production cloud-edge architecture construction module, the industrial production cloud-edge architecture construction module is used to pre-build an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes; an equipment fault identification calling module, the equipment fault identification calling module is used to interactively obtain multiple device IDs of multiple industrial production equipment in a production environment, and obtain multiple device fault identification modules based on the multiple device IDs; a localization update module, the localization update module is used to synchronize the multiple device fault identification modules to the multiple edge fault identification nodes, use the multiple device IDs as multiple device unique identifiers, identify the multiple edge fault identification nodes, and complete the localization update of the industrial production cloud-edge architecture; a data management private chain construction module, the data management private chain construction module is used to pre-build a data management private chain; A fault identification module, wherein the fault identification module is used to analyze the real-time uploaded equipment operation data of the multiple industrial production equipment using the multiple edge fault identification nodes to perform fault identification during the operation of the multiple industrial production equipment, and to perform real-time storage of the operation data in the data management private chain based on the fault identification results; an operation fault data set extraction module, wherein the operation fault data set extraction module is used to preset a data storage frequency threshold, and to perform blockchain data extraction in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers to obtain multiple operation fault data sets; an operation fault dynamic identification module, wherein the operation fault dynamic identification module is used to optimize the industrial production cloud-edge architecture using the multiple operation fault data sets, and to perform dynamic identification of operation faults of the multiple industrial production equipment using the optimized industrial production cloud-edge architecture.

[0025] The blockchain-based industrial production data processing method and system proposed in this application first pre-builds an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes, and then interactively obtains multiple device IDs of multiple industrial production equipment in the production environment, and obtains multiple device fault identification modules based on the multiple device IDs. Then, after synchronizing the multiple device fault identification modules to multiple edge fault identification nodes, the multiple device IDs are used as multiple device unique identifiers to identify multiple edge fault identification nodes, and the localized update of the industrial production cloud-edge architecture is completed. Then, a data management private chain is pre-built, and then during the operation of multiple industrial production equipment, multiple An edge fault identification node analyzes the real-time uploaded equipment operation data of multiple industrial production equipment for fault identification, and stores the operation data in real time in the data management private chain based on the fault identification results. Then, a data storage frequency threshold is preset, and blockchain data is extracted in the data management private chain according to the data storage frequency threshold and multiple device IDs to obtain multiple operation fault data sets. Finally, multiple operation fault data sets are used to optimize the industrial production cloud-edge architecture, and the optimized industrial production cloud-edge architecture is used to perform dynamic identification of operation faults of multiple industrial production equipment, achieving the technical effect of ensuring the authenticity and integrity of industrial production data through the tamper-proof nature of blockchain, thereby improving the accuracy and reliability of fault identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0027] Figure 1 A flowchart of a blockchain-based industrial production data processing method provided in an embodiment of the present application;

[0028] Figure 2 A schematic diagram of the structure of an industrial production data processing system based on blockchain provided in an embodiment of the present application.

[0029] Explanation of the accompanying drawings: industrial production cloud-edge architecture building module 10, equipment fault identification and calling module 20, localization update module 30, data management private chain building module 40, fault identification module 50, operation fault data set extraction module 60, operation fault dynamic identification module 70. DETAILED DESCRIPTION

[0030] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0031] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0032] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0033] The present application embodiment provides a method for processing industrial production data based on blockchain, such as Figure 1 As shown, the method includes:

[0034] Step S100, pre-build an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes. Specifically, determine the specific needs of the industrial production scenario, including the number of equipment, data type, fault identification accuracy requirements, real-time requirements, etc., and determine the basic functions and performance indicators of the industrial production cloud-edge architecture according to the needs. Among them, the industrial production cloud-edge architecture is a system architecture that combines the advantages of cloud computing and edge computing, and is used to support data processing and fault identification in industrial production. The industrial production cloud-edge architecture provides powerful computing power and data storage capabilities through the cloud, and provides real-time data processing and preliminary fault identification capabilities through edge nodes. Design the overall framework of the industrial production cloud-edge architecture, determine the roles and responsibilities of the cloud and edge nodes, wherein the fault recording cloud is a central server or cloud service platform used in the industrial production cloud-edge architecture to centrally store and process fault data and equipment operation data from edge nodes; the edge fault identification node is a computing device or intelligent device deployed at the production site, which can collect equipment operation data in real time, perform preliminary data processing and fault identification, and upload the results to the cloud. Deploy and configure the fault recording cloud and edge fault identification nodes, select a cloud service platform or self-built data center as the cloud infrastructure, and deploy the fault recording system in the cloud, including databases, data processing engines, fault analysis algorithms, etc.; according to the specific conditions of the production site, select a suitable location to deploy the edge fault identification node, and configure the network connection of the edge node to ensure smooth communication with the cloud and other edge nodes.

[0035] Step S200, interactively obtain multiple device IDs of multiple industrial production equipment in the production environment, and call and obtain multiple device fault identification modules based on the multiple device IDs. Specifically, use network scanning, Internet of Things protocols (such as MQTT, CoAP) or equipment management systems (such as SCADA systems) and other technologies to discover and identify industrial production equipment in the production environment. Extract the unique identifier of each industrial production equipment from the registered device information (device ID, which is a unique serial number, MAC address or custom identifier assigned by the manufacturer). According to the device ID, query and obtain the corresponding device fault identification module from the cloud or local database, wherein the device fault identification module is a software or algorithm component for identifying and analyzing industrial production equipment faults.

[0036] In a possible implementation, multiple device IDs of multiple industrial production devices in a production environment are interactively obtained, and multiple device fault identification modules are called according to the multiple device IDs. Step S200 further includes:

[0037] Step S210, matching the equipment purchase log according to the multiple equipment IDs to obtain multiple equipment model information. Specifically, the equipment IDs of multiple industrial production equipment in the production environment are obtained through human-computer interaction or other methods (such as API calls), and these equipment IDs are used as query conditions to query the corresponding equipment purchase logs in the enterprise resource planning system, equipment management system or procurement database, and extract the model information of each equipment from the equipment purchase log, including manufacturer, model code, serial number, etc. Step S220, network data collection is performed according to the multiple equipment model information to obtain multiple sample fault data sets. Specifically, according to the equipment model information, the data collection source of each industrial production equipment is determined, including the equipment's own sensors, remote monitoring systems, historical fault records, etc., and the operation data and fault records of the equipment are collected from the determined data sources through network connections to form sample fault data sets. Step S230, data structure integration is performed on the multiple sample fault data sets to obtain multiple standard fault data structures, and the multiple sample fault data sets are cleaned using the multiple standard fault data structures to obtain multiple optimized fault data sets. Specifically, the data structures of sample fault data sets of different devices are analyzed to find out the commonalities and differences. According to the analysis results, one or more standard fault data structures are defined. The standard fault data structure is a unified format or template that is compatible with the fault data of different devices. The sample fault data set is converted and cleaned using the standard fault data structure to remove invalid data, outliers and duplicates to ensure the accuracy and consistency of the data. Step S240, a standard operator is obtained by calling from an operator library, and the standard operator is trained using the multiple optimized fault data sets to obtain multiple equipment fault identification modules. Specifically, according to the requirements of the fault identification task, a suitable standard operator (a specific algorithm or model in the operator library, used to perform specific data processing or analysis tasks) is selected from an operator library (a library containing various algorithms and models). These operators can be machine learning algorithms, deep learning models, etc. The optimized fault data set is used as training data to train the selected standard operator, adjust its parameters to optimize performance, evaluate the performance of the trained model through cross-validation or other evaluation methods, and select the model with the best performance as the equipment fault identification module. This implementation method quickly builds fault identification modules for different industrial production equipment by calling standard operators in the operator library and using optimized fault data sets for training, achieving the technical effect of improving development efficiency and ensuring the quality and accuracy of the model.

[0038] Step S300, after synchronizing the multiple device fault identification modules to the multiple edge fault identification nodes, use the multiple device IDs as multiple device unique identifiers to identify the multiple edge fault identification nodes, and complete the localized update of the industrial production cloud edge architecture. Specifically, based on the device ID and the geographical location, processing capacity and other factors of the node, determine which device fault identification modules are required for each edge fault identification node, and use a secure file transfer protocol (such as SFTP, HTTPS) to transfer the device fault identification module files (such as algorithm codes, model files, etc.) to the corresponding edge fault identification node for installation and configuration. On the edge fault identification node, assign a corresponding device ID as a unique identifier to each synchronized device fault identification module, update the configuration file or database of the edge fault identification node, record the mapping relationship between each device ID and its corresponding fault identification module, and ensure that the edge fault identification node can correctly call the corresponding fault identification module for processing according to the device ID in the received device operation data. After completing the synchronization and configuration of the device fault identification module on all edge fault identification nodes, a localized update check of the industrial production cloud-edge architecture is performed to verify whether each edge fault identification node can correctly identify and process the operating data from the corresponding equipment and generate accurate fault identification results. If any problems or inconsistencies are found, necessary debugging and corrections are performed.

[0039] Step S400, pre-build a data management private chain. Specifically, select a suitable blockchain platform (such as Ethereum, Hyperledger, etc.) according to the needs, define the data structure stored on the chain, including equipment operation data, fault records, etc., deploy the blockchain network on the cloud or a specific node, and initialize the relevant configuration. Among them, the data management private chain is a data management platform based on blockchain technology, which is used to manage data within a specific organization or institution, and realize the decentralization, immutability and traceability of data.

[0040] In a possible implementation, the data management private chain is pre-built, and step S400 further includes:

[0041] Step S410, write and generate a data management smart contract based on the multiple device unique identifiers. Specifically, determine the functions that the data management smart contract needs to implement, such as device data storage, access control, data integrity verification, etc., and at the same time, associate these functions with the device unique identifier (device ID). Based on the results of the demand analysis, design the architecture and interface of the data management smart contract, including defining variables (such as device ID list, fault data structure template, etc.), functions (such as data upload, data query, permission verification, etc.) and events (such as data update notification) in the contract. Use smart contract programming languages ​​such as Solidity and Vyper to write the code of the data management smart contract according to the design document. In the code, each device ID is accurately mapped to the corresponding data structure and operation permission. Test the data management smart contract in the development environment to verify whether its function meets expectations, including unit testing, integration testing and testing of simulated production environments. Step S420, deploy the multiple standard fault data structures and data management smart contracts on the pre-selected blockchain platform to complete the construction of the data management private chain. Specifically, select a blockchain platform such as Ethereum, Hyperledger Fabric, etc. according to actual needs (such as performance, security, cost, etc.). Create the necessary environment on the selected blockchain platform, including node configuration, network settings, key management, etc., use the tools or APIs provided by the blockchain platform to deploy the written data management smart contract to the blockchain network, upload multiple standard fault data structures to the blockchain in the form of JSON, XML, etc., and associate them with the data structure template in the data management smart contract. Verify the deployment of the data management smart contract on the blockchain network to ensure that all functions are available. At the same time, carry out necessary tests to ensure that the data can be uploaded, stored and queried correctly. This implementation method achieves seamless integration and management of data by deploying the standard fault data structure and the data management smart contract on the same blockchain platform, achieving the technical effect of improving the overall performance and efficiency of the system.

[0042] Step S500, during the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the real-time uploaded equipment operation data of the plurality of industrial production equipment for fault identification, and the operation data is stored in real time in the data management private chain based on the fault identification result. Specifically, the edge fault identification node receives the operation data from the industrial production equipment in real time, and the operation data is transmitted through a predetermined communication protocol (such as MQTT, CoAP, etc.). The received operation data is preprocessed, including data cleaning, format conversion, timestamp alignment, etc., and the preprocessed data is identified by using the equipment fault identification module, and the equipment fault identification module performs pattern matching or anomaly detection based on historical fault data and equipment operation status characteristics. Whether the industrial production equipment has a fault is determined according to the result output by the equipment fault identification module, and the fault level and type can be determined by setting a threshold or rule. The identification result and the related equipment operation data are packaged into a specific data format (including equipment ID, timestamp, fault type, fault level, etc.), and uploaded to the data management private chain in real time, and the data management private chain verifies and stores the data according to the consensus mechanism.

[0043] In a possible implementation, during the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the equipment operation data uploaded in real time by the plurality of industrial production equipment to perform fault identification, and the operation data is stored in real time in the data management private chain based on the fault identification result, and step S500 further includes:

[0044] Step S510, during the operation of the plurality of industrial production equipment, the first industrial production equipment uploads the first equipment operation data to the first edge fault identification node in real time to perform fault identification and obtain a first identification result, wherein the first industrial production equipment is any one of the plurality of industrial production equipment, and the first edge fault identification node uses the first equipment ID of the first industrial production equipment as the first equipment unique identifier. Specifically, the first industrial production equipment will generate operation data such as temperature, pressure, vibration frequency, etc. in real time during its operation, and these data are collected and converted into digital signals through sensors and other equipment. The collected first equipment operation data is transmitted to the first edge fault identification node connected thereto in real time via wired or wireless means. After receiving the data, the first edge fault identification node uses the built-in equipment fault identification module to perform real-time analysis on the first equipment operation data to identify whether there is a fault. After the fault identification is completed, the first edge fault identification node outputs the first identification result, including information such as whether a fault is detected, the type of fault, and the degree of the fault. Step S520, preset the fault time window. If the first identification result is a fault, extract the first identification time according to the first identification result, and use the first identification time and the fault time window as constraints to call the local operation log of the first industrial production equipment to obtain the first operation data. Specifically, a time window is pre-set in the system configuration to determine the time range that needs to be traced back after the fault identification result appears. When the first edge fault identification node outputs the fault identification result, the timestamp of this moment is recorded as the first identification time. With the first identification time and the fault time window as constraints, call the local operation log within the corresponding time period from the first industrial production equipment or related system. These logs record detailed information on the equipment operation status, including but not limited to various sensor data, operation records, etc. Extract the first operation data related to the fault from the called local operation log. Step S530, store the first operation data in real time in the data management private chain. Specifically, the extracted first operation data is formatted and converted into a standard fault data structure to meet the storage requirements of the data management private chain, call the data management smart contract deployed on the data management private chain, and upload the formatted first operation data to the blockchain network. This implementation method uses the operating data before and after the fault occurs to fully understand the process and cause of the fault, thereby achieving the technical effect of facilitating subsequent fault analysis and processing and improving the precision and accuracy of fault diagnosis.

[0045] In a possible implementation, a fault time window is preset, and if the first recognition result is a fault, a first recognition time is extracted according to the first recognition result, and the first recognition time and the fault time window are used as constraints to call a local operation log of the first industrial production equipment to obtain first operation data, and step S520 further includes:

[0046] Step S521, if the first identification result is a fault, the first identification time and the first device ID are associated and stored in the fault record cloud. Specifically, if the first identification result output by the first edge fault identification node is a fault, the first identification time (i.e., the specific time point when the fault is detected) and the first device ID (used to uniquely identify the device where the fault occurs) are obtained from the first edge fault identification node, and the first identification time and the first device ID are used as a pair of associated data and transmitted to the fault record cloud through the network. This implementation method facilitates the subsequent rapid positioning of the specific faulty device and the time when the fault occurs by storing the fault identification time and the device ID in the cloud, thereby achieving the technical effect of accelerating the fault response and processing speed.

[0047] In a possible implementation, if the first recognition result is a fault, the first recognition time and the first device ID are associated and stored in the fault record cloud, and step S521 further includes:

[0048] Step S5211, preset a fault association analysis window. Specifically, according to actual needs, a fault association analysis window is preset in the system configuration, and the fault association analysis window is a preset time range, which is used to find and associate the fault records of other devices within the time period when a fault occurs. Step S5212, using the fault association analysis window as a constraint, interact with the fault record cloud to obtain multiple identification times of the multiple device IDs. Specifically, according to the identification time (the specific time point when the equipment fault is detected) and the preset fault association analysis window, send a query request to the fault record cloud. According to the request, the fault record cloud retrieves the fault identification time of all industrial production equipment (or industrial production equipment within a specific range) within the fault association analysis window, and obtains the retrieved multiple device IDs and their corresponding multiple identification times. Step S5213, perform association analysis on the multiple identification times to obtain multiple joint maintenance equipment groups. Specifically, according to multiple identification times, equipment failure records that overlap or are close in time are found, and according to the result of time matching, industrial production equipment with similar failure times are grouped into the same group to form multiple joint maintenance equipment groups. The equipment in the joint maintenance equipment group has related reasons when the failure occurs, and is suitable for joint maintenance. Step S5214: If the first identification result is a failure, the multiple joint maintenance equipment groups are traversed according to the first equipment ID to obtain the first joint maintenance equipment group. Specifically, the first equipment ID is used as a keyword to traverse all joint maintenance equipment groups, and the group containing the equipment ID is found. After the group containing the first equipment ID is found, it is determined as the first joint maintenance equipment group. Step S5215: The first joint maintenance equipment group is sent to the operation and maintenance terminal as an operation and maintenance signal. Specifically, the information of the first joint maintenance equipment group (including the equipment ID list, the cause of the failure, etc.) is encapsulated into an operation and maintenance signal, and the operation and maintenance signal is sent to the operation and maintenance terminal or the mobile device of the operation and maintenance personnel through a network or other communication methods. This implementation method discovers potential equipment failure correlations by correlating and analyzing the failure times of multiple devices, thereby identifying groups of equipment that may fail due to common causes. This achieves the technical effect of enabling operation and maintenance personnel to more accurately locate the source of the problem and take more effective maintenance measures.

[0049] In a possible implementation, a fault time window is preset, and if the first recognition result is a fault, a first recognition time is extracted according to the first recognition result, and the first recognition time and the fault time window are used as constraints to call a local operation log of the first industrial production equipment to obtain first operation data, and step S520 further includes:

[0050] Step S522, if the first identification result is not a fault, the local operation log of the first industrial production equipment is formatted. Specifically, if the first identification result is not a fault, it indicates that the first industrial production equipment is operating normally, and the local operation log of the first industrial production equipment is located. These logs record normal data such as the operation history and operation records of the first industrial production equipment. The collected local operation logs are formatted to remove data without fault information. This implementation method removes redundant information and reduces the size of the log file through formatting, thereby achieving the technical effect of saving storage space.

[0051] Step S600, preset a data storage frequency threshold, extract blockchain data from the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers, and obtain multiple operation fault data sets. Specifically, analyze the fault data storage frequency of each industrial production equipment in the past period of time, determine the normal fluctuation range, and set the data storage frequency threshold for different equipment or equipment types based on the results of historical data analysis. The data storage frequency threshold is used to distinguish between data fluctuations under normal conditions and high-frequency data storage under abnormal conditions. Use the device unique identifier as a retrieval condition to retrieve the fault identification result and storage timestamp of each industrial production equipment from the data management private chain. For each industrial production equipment, calculate its fault data storage frequency within a specific time period (such as the last 24 hours, 7 days, etc.). Compare the calculated frequency with the preset threshold (data storage frequency threshold). If the storage frequency of a certain industrial production equipment exceeds the threshold, it will be regarded as a potential high-risk device, and its related fault data will be extracted to form an operation fault data set.

[0052] In a possible implementation, a data storage frequency threshold is preset, and blockchain data is extracted in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers to obtain multiple operation fault data sets. Step S600 further includes:

[0053] Step S610, interact with the first edge fault identification node to obtain the first fault identification frequency. Specifically, establish a communication connection with the first edge fault identification node through a network interface, send a request to the first edge fault identification node, request to obtain the fault frequency identified by the node, and after the first edge fault identification node processes the request, return the first fault identification frequency as a response. The first fault identification frequency indicates the number of times the first edge fault identification node identifies the fault of the first industrial production equipment within a specific time period. Step S620, if the first fault identification frequency meets the data storage frequency threshold, run the data management smart contract according to the first device unique identifier to perform data directional extraction in the data management private chain to obtain a first operation fault data set. Specifically, compare the first fault identification frequency with the preset data storage frequency threshold. If the first fault identification frequency meets (for example, is greater than or equal to) the data storage frequency threshold, it means that the first industrial production equipment has frequent faults. At this time, according to the first device unique identifier, trigger the execution of the data management smart contract, and the data management smart contract defines the logic of extracting the first industrial production equipment operation fault data from the data management private chain. The data management smart contract is executed on the data management private chain, and the relevant operation fault data set is extracted according to the first device unique identifier to obtain the first operation fault data set, which includes information such as fault type, occurrence time, and impact range. Step S630, and so on, to obtain the multiple operation fault data sets. Specifically, for other device unique identifiers, the process of steps S610 to S620 is repeated, including interacting with the corresponding edge fault identification node, obtaining the fault identification frequency, judging whether the data storage frequency threshold is met, and triggering the data management smart contract to extract data accordingly. Finally, the operation fault data sets extracted from multiple industrial production equipment are integrated to form a complete operation fault data set. This implementation method screens out industrial production equipment that frequently fails and needs to be focused on by presetting the data storage frequency threshold, and extracts its operation fault data set in a targeted manner, thereby reducing unnecessary data storage and processing burdens and improving the targetedness and efficiency of data extraction.

[0054] Step S700, using the multiple operation fault data sets to optimize the industrial production cloud edge architecture, and using the optimized industrial production cloud edge architecture to perform dynamic identification of operation faults of the multiple industrial production equipment. Specifically, extract key features from multiple operation fault data sets, use machine learning or deep learning algorithms to analyze the extracted features, identify different fault modes, and establish a classification model based on the identified fault modes to assign new fault data to corresponding fault categories. According to the results of fault mode identification and classification, analyze the problems and bottlenecks existing in the industrial production cloud edge architecture, such as data transmission delay, insufficient processing capacity, etc., and design an optimization plan for the industrial production cloud edge architecture for the analyzed problems, including adjusting the distribution of edge fault identification nodes, upgrading equipment fault identification modules, etc. According to the optimization plan, configure the optimized industrial production cloud edge architecture, including network settings, security policies, data processing procedures, etc. Collect operation data in real time through sensors on industrial production equipment, and use the optimized industrial production cloud edge architecture to transmit the collected data to the edge fault identification node in real time for real-time dynamic identification of operation faults. The embodiment of the present application adopts technical means such as building an industrial production cloud-edge architecture, synchronizing multiple equipment fault identification modules to multiple edge fault identification nodes, building a data management private chain, using multiple edge fault identification nodes for fault identification, real-time storage of the data management private chain based on the fault identification results, extracting multiple operation fault data sets from the blockchain, using multiple operation fault data sets to optimize the industrial production cloud-edge architecture, and performing dynamic identification of operation faults of multiple industrial production equipment. The technical effect is achieved by ensuring the authenticity and integrity of industrial production data through the tamper-proof nature of the blockchain, thereby improving the accuracy and reliability of fault identification.

[0055] In a possible implementation, the industrial production cloud-edge architecture is optimized using the multiple operation fault data sets, and step S700 further includes:

[0056] Step S710, preset the training data reorganization ratio. Specifically, based on the training effect of the balance model and the consumption of computing resources, preset a training data reorganization ratio, and the training data reorganization ratio determines how much proportion of data in the existing training data set will be replaced by the new operation fault data set. Step S720, using the training data reorganization ratio as a constraint to partially replace the multiple optimized fault data sets with the multiple operation fault data sets to obtain multiple updated fault data sets. Specifically, according to the preset training data reorganization ratio, a part of the data is randomly or according to a specific rule selected from the existing training data set (multiple optimized fault data sets), and this part of the data is replaced with the new operation fault data set (the multiple operation fault data sets). After the replacement is completed, multiple updated fault data sets are generated, and the updated fault data sets contain both the original training data and the new and more timely operation fault data. Step S730, using the multiple updated fault data sets as training data, the model parameters of the multiple equipment fault identification modules are updated. Specifically, multiple updated fault data sets are loaded into the training environment as input data for model training, and the equipment fault identification module is trained using the loaded updated fault data sets. During the training process, hyperparameters such as learning rate and optimization algorithm are dynamically adjusted according to the feedback and performance indicators of the model. After the training is completed, the updated model parameters are saved in the equipment fault identification module to replace the original parameters. This implementation method enriches the training data of the model by introducing new and more timely operating fault data sets, making it closer to the actual industrial production environment, and achieving the technical effect of improving the accuracy of the model in fault identification tasks.

[0057] In the above, refer to Figure 1 A method for processing industrial production data based on blockchain according to an embodiment of the present invention is described in detail. Figure 2 An industrial production data processing system based on blockchain according to an embodiment of the present invention is described.

[0058] According to an embodiment of the present invention, a blockchain-based industrial production data processing system is used to solve the technical problem of insufficient data reliability in existing industrial production data processing, which affects the accuracy of fault identification, and achieves the technical effect of ensuring the authenticity and integrity of industrial production data through the immutability of blockchain, thereby improving the accuracy and reliability of fault identification. A blockchain-based industrial production data processing system includes: an industrial production cloud-edge architecture construction module 10, an equipment fault identification call module 20, a localization update module 30, a data management private chain construction module 40, a fault identification module 50, an operation fault data set extraction module 60, and an operation fault dynamic identification module 70.

[0059] The industrial production cloud-edge architecture construction module 10 is used to pre-build the industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes; the equipment fault identification calling module 20 is used to interactively obtain multiple equipment IDs of multiple industrial production equipment in the production environment, and call and obtain multiple equipment fault identification modules according to the multiple equipment IDs; the localization update module 30 is used to synchronize the multiple equipment fault identification modules to the multiple edge fault identification nodes, and then use the multiple equipment IDs as multiple equipment unique identifiers to identify the multiple edge fault identification nodes, and complete the localization update of the industrial production cloud-edge architecture; the data management private chain construction module 40 is used to pre-build the data management private chain; the fault identification module 5 0 is used to analyze the real-time uploaded equipment operation data of the multiple industrial production equipment by using the multiple edge fault identification nodes to perform fault identification during the operation of the multiple industrial production equipment, and to perform real-time storage of the operation data in the data management private chain based on the fault identification results; the operation fault data set extraction module 60 is used to preset a data storage frequency threshold, and to perform blockchain data extraction in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers to obtain multiple operation fault data sets; the operation fault dynamic identification module 70 is used to optimize the industrial production cloud-edge architecture by using the multiple operation fault data sets, and to perform dynamic identification of operation faults of the multiple industrial production equipment by using the optimized industrial production cloud-edge architecture.

[0060] The specific configuration of the equipment fault identification calling module 20 will be described in detail below. As described above, multiple equipment IDs of multiple industrial production equipment in the production environment are interactively obtained, and multiple equipment fault identification modules are called according to the multiple equipment IDs. The equipment fault identification calling module 20 may further include: an equipment procurement log matching unit is used to match equipment procurement logs according to the multiple equipment IDs to obtain multiple equipment model information; a networked data collection unit is used to perform networked data collection according to the multiple equipment model information to obtain multiple sample fault data sets; multiple optimized fault data set acquisition units are used to integrate the data structures of the multiple sample fault data sets to obtain multiple standard fault data structures, and use the multiple standard fault data structures to clean the multiple sample fault data sets to obtain multiple optimized fault data sets; a standard operator training unit is used to call and obtain a standard operator from an operator library, and use the multiple optimized fault data sets to train the standard operator to obtain multiple equipment fault identification modules.

[0061] The specific configuration of the data management private chain construction module 40 will be described in detail below. As described above, the data management private chain is pre-built, and the data management private chain construction module 40 may further include: a data management smart contract writing unit for writing and generating a data management smart contract according to the multiple device unique identifiers; a blockchain platform deployment unit for deploying the multiple standard fault data structures and the data management smart contract on a pre-selected blockchain platform to complete the construction of the data management private chain.

[0062] The specific configuration of the fault identification module 50 will be described in detail below. As described above, during the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the real-time uploaded equipment operation data of the plurality of industrial production equipment for fault identification, and the operation data is stored in real time in the data management private chain based on the fault identification result. The fault identification module 50 may further include: a first identification result acquisition unit is used for, during the operation of the plurality of industrial production equipment, the first industrial production equipment uploads the first equipment operation data to the first edge fault identification node in real time to perform fault identification and obtain a first identification result, wherein the first industrial production equipment is any one of the plurality of industrial production equipment, and the first edge fault identification node uses the first equipment ID of the first industrial production equipment as the first equipment unique identifier; a first operation data retrieval unit is used to preset a fault time window, if the first identification result is a fault, the first identification time is extracted according to the first identification result, and the first identification time and the fault time window are constrained to perform a local operation log call on the first industrial production equipment to obtain the first operation data; a data storage unit is used to store the first operation data in real time in the data management private chain.

[0063] The first operation data retrieval unit may further include: a local operation log formatting subunit configured to format the local operation log of the first industrial production equipment if the first identification result is not a fault.

[0064] The specific configuration of the operation fault data set extraction module 60 will be described in detail below. As described above, a data storage frequency threshold is preset, and blockchain data is extracted in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers to obtain multiple operation fault data sets. The operation fault data set extraction module 60 may further include: a first fault identification frequency acquisition unit for interacting with the first edge fault identification node to obtain a first fault identification frequency; a data directional extraction unit for executing the data management smart contract in the data management private chain according to the first device unique identifier to perform data directional extraction if the first fault identification frequency meets the data storage frequency threshold, and obtaining a first operation fault data set; multiple operation fault data set acquisition units are used to obtain the multiple operation fault data sets by analogy.

[0065] The specific configuration of the operation fault dynamic identification module 70 will be described in detail below. As described above, the industrial production cloud-edge architecture is optimized using the multiple operation fault data sets, and the operation fault dynamic identification module 70 may further include: a local replacement unit is used to preset the training data reorganization ratio, and the multiple operation fault data sets are used to partially replace the multiple optimized fault data sets with the training data reorganization ratio as a constraint to obtain multiple updated fault data sets; a model parameter updating unit is used to use the multiple updated fault data sets as training data to update the model parameters of the multiple equipment fault identification modules.

[0066] Among them, the first operation data retrieval unit may further include: a fault record storage subunit is used to associate the first identification time and the first device ID and store them in the fault record cloud if the first identification result is a fault.

[0067] Among them, the fault record storage subunit may further include: multiple identification time acquisition micro-units for pre-setting a fault correlation analysis window, using the fault correlation analysis window as a constraint, interacting with the fault record cloud, and obtaining multiple identification times of the multiple device IDs; an association analysis micro-unit for performing association analysis on the multiple identification times to obtain multiple joint maintenance equipment groups; a first joint maintenance equipment group acquisition micro-unit for traversing the multiple joint maintenance equipment groups according to the first device ID to obtain the first joint maintenance equipment group if the first identification result is a fault; an operation and maintenance signal sending micro-unit for sending the first joint maintenance equipment group as an operation and maintenance signal to the operation and maintenance terminal.

[0068] An industrial production data processing system based on blockchain provided by an embodiment of the present invention can execute an industrial production data processing method based on blockchain provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0069] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0070] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for processing industrial production data based on blockchain, characterized in that: The method comprises: Pre-constructing an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes; Interactively obtain multiple device IDs of multiple industrial production equipment in a production environment, and call and obtain multiple device fault identification modules according to the multiple device IDs; After synchronizing the multiple device fault identification modules to the multiple edge fault identification nodes, the multiple device IDs are used as multiple device unique identifiers to identify the multiple edge fault identification nodes, thereby completing the localized update of the industrial production cloud-edge architecture; Pre-built data management private chain; During the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the equipment operation data uploaded in real time by the plurality of industrial production equipment to identify faults, and based on the fault identification results, the operation data is stored in real time in the data management private chain; Preset a data storage frequency threshold, and extract blockchain data from the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers to obtain multiple operation fault data sets; The multiple operation fault data sets are used to optimize the industrial production cloud-edge architecture, and the optimized industrial production cloud-edge architecture is used to perform dynamic identification of operation faults of the multiple industrial production equipment.

2. The method for processing industrial production data based on blockchain according to claim 1, characterized in that: Interactively obtaining multiple device IDs of multiple industrial production devices in a production environment, and calling and obtaining multiple device fault identification modules according to the multiple device IDs, the method further includes: Matching device purchase logs according to the multiple device IDs to obtain multiple device model information; Perform network data collection according to the multiple device model information to obtain multiple sample fault data sets; Performing data structure integration on the multiple sample fault data sets to obtain multiple standard fault data structures, and performing data cleaning on the multiple sample fault data sets using the multiple standard fault data structures to obtain multiple optimized fault data sets; A standard operator is obtained by calling from an operator library, and the standard operator is trained by using the multiple optimized fault data sets to obtain multiple equipment fault identification modules.

3. The method for processing industrial production data based on blockchain as claimed in claim 2, characterized in that: Pre-building a data management private chain, the method further includes: Compile and generate a data management smart contract based on the multiple device unique identifiers; The multiple standard fault data structures and data management smart contracts are deployed on a pre-selected blockchain platform to complete the construction of the data management private chain.

4. The method for processing industrial production data based on blockchain as claimed in claim 3, characterized in that: During the operation of the plurality of industrial production equipment, the plurality of edge fault identification nodes are used to analyze the equipment operation data uploaded in real time by the plurality of industrial production equipment to identify faults, and the operation data is stored in real time in the data management private chain based on the fault identification results, and the method further includes: During the operation of the multiple industrial production equipment, the first industrial production equipment uploads the first equipment operation data to the first edge fault identification node in real time to perform fault identification and obtain a first identification result, wherein the first industrial production equipment is any one of the multiple industrial production equipment, and the first edge fault identification node uses the first equipment ID of the first industrial production equipment as the first equipment unique identifier; A fault time window is preset, and if the first recognition result is a fault, a first recognition time is obtained according to the first recognition result, and a local operation log call is performed on the first industrial production equipment with the first recognition time and the fault time window as constraints to obtain first operation data; The first operation data is stored in real time in the data management private chain.

5. The method for processing industrial production data based on blockchain as claimed in claim 4, characterized in that: If the first identification result is not a fault, the local operation log of the first industrial production equipment is formatted.

6. The method for processing industrial production data based on blockchain as claimed in claim 4, characterized in that: Presetting a data storage frequency threshold, extracting blockchain data in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers, and obtaining multiple operation fault data sets, the method further includes: Interacting with the first edge fault identification node to obtain a first fault identification frequency; If the first fault identification frequency meets the data storage frequency threshold, executing the data management smart contract according to the first device unique identifier to perform data directional extraction in the data management private chain to obtain a first operation fault data set; And so on, the multiple operation fault data sets are obtained.

7. The method for processing industrial production data based on blockchain as claimed in claim 6, characterized in that: The method further comprises: optimizing the industrial production cloud-edge architecture by using the multiple operation fault data sets; Preset the training data reorganization ratio; Using the multiple operating fault data sets as a constraint to partially replace the multiple optimized fault data sets, to obtain multiple updated fault data sets; The multiple updated fault data sets are used as training data to update model parameters of the multiple equipment fault identification modules.

8. The method for processing industrial production data based on blockchain as claimed in claim 4, characterized in that: The method further comprises: If the first identification result is a fault, the first identification time and the first device ID are associated and stored in the fault record cloud.

9. The method for processing industrial production data based on blockchain as claimed in claim 8, characterized in that: The method further comprises: Preset fault correlation analysis window; Taking the fault correlation analysis window as a constraint, interacting with the fault record cloud to obtain multiple identification times of the multiple device IDs; Performing correlation analysis on the multiple identification times to obtain multiple joint maintenance equipment groups; If the first identification result is a fault, traverse the multiple joint maintenance device groups according to the first device ID to obtain a first joint maintenance device group; The first joint maintenance equipment group is sent as an operation and maintenance signal to an operation and maintenance terminal.

10. An industrial production data processing system based on blockchain, characterized in that: The system is used to implement the blockchain-based industrial production data processing method according to any one of claims 1 to 9, and the system includes: An industrial production cloud-edge architecture building module, wherein the industrial production cloud-edge architecture building module is used to pre-build an industrial production cloud-edge architecture, wherein the industrial production cloud-edge architecture includes a fault recording cloud and multiple edge fault identification nodes; An equipment fault identification calling module, the equipment fault identification calling module is used to interactively obtain multiple equipment IDs of multiple industrial production equipment in a production environment, and call and obtain multiple equipment fault identification modules according to the multiple equipment IDs; A localized update module, wherein the localized update module is used to synchronize the multiple device fault identification modules to the multiple edge fault identification nodes, use the multiple device IDs as multiple device unique identifiers, identify the multiple edge fault identification nodes, and complete the localized update of the industrial production cloud-edge architecture; A data management private chain construction module, wherein the data management private chain construction module is used to pre-construct a data management private chain; A fault identification module, wherein the fault identification module is used to analyze the real-time uploaded equipment operation data of the multiple industrial production equipment by the multiple edge fault identification nodes during the operation of the multiple industrial production equipment to identify faults, and store the operation data in real time in the data management private chain based on the fault identification results; An operation fault data set extraction module, the operation fault data set extraction module is used to preset a data storage frequency threshold, perform blockchain data extraction in the data management private chain according to the data storage frequency threshold and the multiple device unique identifiers, and obtain multiple operation fault data sets; An operation fault dynamic identification module is used to optimize the industrial production cloud-edge architecture using the multiple operation fault data sets, and use the optimized industrial production cloud-edge architecture to perform dynamic identification of operation faults of the multiple industrial production equipment.

Images