Industrial internet data management method and system based on edge computing

Abstract

The application provides an industrial internet data management method and system based on edge computing, and relates to the technical field of data management. The method comprises the following steps: obtaining industrial internet data to be managed and operation evaluation data of each edge node; then analyzing original protocol characteristics of the industrial internet data and preset system standard protocols to obtain protocol adaptation requirements, and analyzing the operation evaluation data to obtain node operation intervals; then obtaining target edge nodes and scheduling results by using a dynamic resource scheduling method based on the protocol adaptation requirements, the node operation intervals and energy efficiency data between the edge nodes; finally routing the industrial internet data to the target edge nodes according to the scheduling results, and completing protocol identification and conversion by using a dynamic adaptation method of a protocol cluster to generate standardized data. The application can realize full-process closed-loop management of energy efficiency perception and protocol collaboration of industrial internet data at the edge side.

Description

Technical Field

[0001] This application relates to the technical field of data management, and in particular to an industrial internet data management method and system based on edge computing. Background Technology

[0002] In the field of industrial internet, edge computing can effectively meet the core requirements of real-time response and low latency in industrial production by deploying data processing tasks at the network edge close to the data source, thereby providing key support for applications such as intelligent equipment monitoring and real-time process optimization.

[0003] Currently, existing edge data management solutions typically adopt a phased, independent processing approach. The common practice is to first perform protocol conversion or encapsulation on multi-source industrial data using pre-configured rules, and then distribute the processed data to edge nodes for computation and storage based on the real-time computing load of the nodes.

[0004] However, when dealing with data streams that are continuously generated in industrial settings, have diverse protocol types, and exhibit dynamic fluctuations in node energy consumption, the protocol conversion rules and resource scheduling strategies are often statically preset and isolated from each other. This can easily lead to high-efficiency nodes not being fully utilized, while low-efficiency nodes are overloaded. At the same time, the fixed protocol processing flow is difficult to adapt to the dynamic changes in the data path after scheduling, resulting in low overall system energy efficiency and difficulties in multi-protocol collaborative management. Summary of the Invention

[0005] The purpose of this application is to provide an industrial internet data management method and system based on edge computing, so as to solve the problem that energy efficiency optimization and multi-protocol compatibility are difficult to effectively coordinate in the existing technology of industrial internet edge data management.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an industrial internet data management method based on edge computing, comprising:

[0007] Acquire the industrial internet data to be managed, as well as the operational evaluation data of each edge node in the industrial internet;

[0008] Based on the original protocol characteristics of the industrial internet data and the preset system standard protocol, the protocol adaptation requirements corresponding to the industrial internet data are determined by matching and analyzing them.

[0009] A multi-dimensional coupled analysis is performed on the operational evaluation data to obtain the node operational range corresponding to the edge node;

[0010] Based on the node operating range and the protocol adaptation requirements, and combined with the energy efficiency data between edge nodes, an energy efficiency-aware dynamic resource scheduling method is used to allocate target edge nodes to the industrial internet data and generate scheduling results.

[0011] Based on the scheduling results, the industrial internet data is routed to the corresponding target edge nodes, so that the target edge nodes can use the dynamic adaptation method of protocol families to perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements, so as to obtain standardized data and store it locally.

[0012] Optionally, based on the node's operating range and the protocol adaptation requirements, and combined with energy efficiency data between edge nodes, an energy efficiency-aware dynamic resource scheduling method is used to allocate target edge nodes to the industrial internet data and generate scheduling results, including:

[0013] Based on the data processing complexity and node operating range indicated by the protocol adaptation requirements, the industrial internet data is divided into energy efficiency-sensing data streams to obtain real-time data streams and batch data streams.

[0014] Energy efficiency data between edge nodes is acquired, and a state tracking mechanism is used to monitor and process the real-time operating status of the edge nodes to obtain a state view;

[0015] Based on the state view and the energy efficiency data, dynamic resource scheduling processing is performed on the real-time data stream and the batch data stream to obtain the scheduling results of the data stream and edge nodes.

[0016] Optionally, the step of performing dynamic resource scheduling processing on the real-time data stream and the batch data stream based on the state view and the energy efficiency data to obtain the scheduling result of the data stream and edge nodes includes:

[0017] The real-time running status of each edge node in the state view is used as the node attribute. Based on the communication efficiency and collaborative energy consumption between edge nodes in the energy efficiency data, the connection relationship and connection weight between nodes are determined, and a navigation graph is constructed.

[0018] Using the navigation map, differentiated node matching is performed on the real-time data stream and the batch data stream to obtain a node matching scheme;

[0019] Based on the node matching scheme, the scheduling results of the data stream and edge nodes are generated.

[0020] Optionally, the step of using the navigation map to perform differentiated node matching on the real-time data stream and the batch data stream to obtain a node matching scheme includes:

[0021] Based on the protocol adaptation requirements, delay constraint parameters and energy efficiency constraint parameters are extracted for the real-time data stream and the batch data stream, respectively.

[0022] Based on the node attributes and connection weights of each node in the navigation graph, and combined with the latency constraint parameters and energy efficiency constraint parameters respectively, a first matching evaluation function is constructed for the real-time data stream with the optimization objective of minimizing end-to-end latency, and a second matching evaluation function is constructed for the batch data stream with the optimization objective of maximizing energy efficiency utilization.

[0023] Based on the first matching evaluation function and the second matching evaluation function, candidate node sets that satisfy their respective constraints are screened in parallel for the real-time data stream and the batch data stream in the navigation graph;

[0024] Conflict detection and elimination are performed on the candidate node sets of all data streams. Based on the results of conflict elimination, a unique target edge node is determined for each data stream from the corresponding candidate node set, thus obtaining a node matching scheme.

[0025] Optionally, generating the scheduling results of the data stream and edge nodes based on the node matching scheme includes:

[0026] Based on the node matching scheme, a transmission path that meets the corresponding quality of service requirements is selected for each data stream in the navigation map. During the selection process, the bandwidth resource allocation among multiple data streams is coordinated to obtain a path configuration scheme.

[0027] Based on the path configuration scheme and the node matching scheme, the scheduling results of the data stream and edge nodes are obtained.

[0028] Optionally, the step of routing the industrial internet data to the corresponding target edge node based on the scheduling result, so that the target edge node can use a dynamic adaptation method for protocol families, and perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements to obtain standardized data and store it locally, includes:

[0029] Based on the path configuration scheme and node matching scheme in the scheduling results, the industrial internet data is routed to the corresponding target edge node;

[0030] On the target edge node, a protocol mapping table is established based on the protocol adaptation type corresponding to the node's operating range;

[0031] According to the protocol adaptation requirements, a target protocol configuration is selected from the protocol mapping table, and the protocol characteristics of the industrial internet data are matched with the feature templates in the protocol cluster to obtain the source protocol type.

[0032] Based on the target protocol configuration and the source protocol type, the industrial internet data is converted from the source protocol format to the target protocol format, and the industrial internet data in the target protocol format is verified. Based on the verification results, standardized data conforming to the preset protocol specifications is generated and stored locally.

[0033] Optionally, the step of performing multi-dimensional coupled analysis on the operational evaluation data to obtain the node operational range corresponding to the edge node includes:

[0034] A feature set is obtained by performing coupled analysis on the computing resource utilization rate, energy consumption value and data processing throughput in the operation evaluation data;

[0035] Based on the feature set, the state space partitioning method is used to determine the running state boundary of each edge node. According to the running state boundary, the node running state is classified to obtain the interval partitioning result.

[0036] Based on the dynamic changing trend of the operational evaluation data, the boundary optimization process is performed on the interval division results to obtain the node operational interval.

[0037] Secondly, this application provides an industrial internet data management system based on edge computing, including:

[0038] The acquisition module is used to acquire industrial internet data to be managed, as well as operational evaluation data of each edge node in the industrial internet;

[0039] The first analysis module is used to perform matching analysis between the original protocol characteristics of the industrial internet data and the preset system standard protocol to determine the protocol adaptation requirements corresponding to the industrial internet data.

[0040] The second analysis module is used to perform multi-dimensional coupled analysis on the operation evaluation data to obtain the node operation range corresponding to the edge node;

[0041] The allocation module is used to allocate target edge nodes to the industrial internet data based on the node's operating range and the protocol adaptation requirements, combined with the energy efficiency data between edge nodes, using an energy efficiency-aware dynamic resource scheduling method, and generating scheduling results.

[0042] The identification module is used to route the industrial internet data to the corresponding target edge node based on the scheduling result, so that the target edge node can use the dynamic adaptation method of the protocol family to perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements, obtain standardized data and store it locally, so as to realize the full-process management of industrial internet data.

[0043] Thirdly, this application provides an electronic device, comprising:

[0044] Memory, used to store computer programs;

[0045] A processor, used to execute the computer program to implement the steps of the edge computing-based industrial internet data management method as described in the first aspect above.

[0046] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the industrial internet data management method based on edge computing as described in the first aspect above.

[0047] The industrial internet data management method based on edge computing provided in this application has the following beneficial effects:

[0048] This application first acquires the industrial internet data to be managed and the operational evaluation data of each edge node, which provides a comprehensive and reliable data foundation for subsequent intelligent decision-making and management. Then, it performs matching analysis based on the original protocol characteristics of the industrial data and the system's preset standards, which can accurately identify the protocol adaptation requirements of the data flow, thus providing clear guidance for subsequent compatibility processing. Then, it performs multi-dimensional coupling analysis on the operational evaluation data, which can scientifically distinguish the node operating ranges of different energy efficiency states, thereby providing a key basis for resource scheduling.

[0049] Then, based on the node's operating range and protocol adaptation requirements, and combined with real-time energy efficiency data between nodes, dynamic scheduling can be performed to achieve optimal matching between data tasks and edge node capabilities, thereby improving resource utilization efficiency and overall system energy efficiency. Finally, based on the scheduling results, data is routed to the target node and protocol conversion and local storage are completed, enabling closed-loop management of the entire process from data access and intelligent scheduling to standardized processing and implementation, thereby effectively improving the collaboration and efficiency of data processing at the industrial internet edge.

[0050] Furthermore, this application divides the data stream into energy-efficiency-aware segments and continuously monitors the node status to form a dynamic view. This enables subsequent dynamic resource scheduling to perform differentiated resource allocation based on the real-time processing needs of the data stream and the precise energy efficiency status of the nodes. This approach achieves refined and differentiated scheduling of real-time data streams and batch data streams, thereby maximizing the utilization of the energy efficiency potential of edge nodes while meeting real-time requirements, and further enhancing the balance and optimization effect of resources and energy efficiency. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of 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 only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A flowchart illustrating an industrial internet data management method based on edge computing, provided for an embodiment of this application;

[0053] Figure 2 A schematic diagram illustrating a specific implementation of an industrial internet data management method based on edge computing, provided in this application embodiment;

[0054] Figure 3 A schematic diagram of the structure of an industrial internet data management system based on edge computing, provided for an embodiment of this application;

[0055] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0056] In the field of industrial internet, the management of industrial internet data is crucial for improving industrial production efficiency. However, current common methods have significant shortcomings when dealing with complex industrial scenarios. These methods usually treat protocol conversion and resource scheduling as two separate steps: first, data from multiple industrial protocols are uniformly converted according to fixed rules, and then computing tasks are allocated based on simple node load indicators. In real industrial environments with a wide variety of protocols and dynamic changes in node energy consumption, this approach can easily lead to a mismatch between the protocol conversion strategy and the actual resource load. For example, it may result in high-energy-consuming nodes being assigned dense data flows, while low-energy-consuming nodes are idle, ultimately leading to low overall system energy efficiency and difficulty in flexibly adapting to complex situations where multiple protocols coexist.

[0057] To address this, this application proposes an industrial internet data management method based on edge computing. This method first analyzes the protocol characteristics of industrial data and the operational energy efficiency of edge nodes in parallel, determining the required protocol conversion type and the real-time performance range of the nodes. Then, it integrates the results of these two analyses, along with real-time energy efficiency data between nodes, to dynamically allocate suitable target edge nodes for each type of data stream. Finally, based on the allocation results, the data is routed to the target edge node, where precise protocol conversion and local storage are performed. This scheme achieves optimal matching between data tasks and edge computing resources by enabling collaborative decision-making between protocol adaptation requirements and node energy efficiency status, effectively solving the problems of low system energy efficiency and poor multi-protocol compatibility caused by the disconnect between resource scheduling and protocol management in existing technologies.

[0058] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] The core of this application is to provide an industrial internet data management method based on edge computing, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0060] S101. Obtain the industrial internet data to be managed, as well as the operation evaluation data of each edge node in the industrial internet.

[0061] Industrial Internet data refers to the collection of raw information generated by various industrial devices or information systems within the edge network managed by this application, which needs to be processed and circulated in the network.

[0062] Operational evaluation data refers to a set of core indicators used to quantify the working status and efficiency of edge computing nodes. Specifically, it includes the node's current computing resource utilization, real-time energy consumption when processing data, and the data throughput successfully processed per unit time.

[0063] In S101, firstly, raw industrial internet data streams are collected from various industrial equipment and information systems deployed in factory workshops or production lines and then connected to them for subsequent intelligent management.

[0064] Meanwhile, lightweight monitoring agents or interfaces deployed on each edge computing node continuously and in real time collect the node's operational evaluation data. This process is usually accomplished by reading information such as the node's system performance counters, energy consumption monitoring chip readings, and network interface traffic statistics, thereby providing objective and quantitative evidence of the node's own operational status for subsequent analysis.

[0065] S102. Based on the original protocol characteristics of the industrial internet data, perform matching analysis with the preset system standard protocol to determine the protocol adaptation requirements corresponding to the industrial internet data.

[0066] Among them, the original protocol features refer to the specific information parsed from industrial internet data that is used to identify its communication rules. The original protocol features include the data message structure, instruction set encoding method and interaction timing rules.

[0067] System standard protocols refer to a set or a few sets of baseline communication specifications that are pre-selected and set in order to achieve unified data understanding and exchange throughout the entire edge computing system. They can be understood as the "common language" defined within the system.

[0068] Protocol adaptation requirements refer to a list of specific requirements for how to "translate" or "convert" raw data into the system's standard protocol. These requirements clarify the source format, target format, and rule differences that need special attention during the conversion process.

[0069] In step S102, the protocol parsing engine deployed on the edge or central management unit is used to perform deep parsing of the industrial Internet data stream. The engine extracts the original protocol features that can uniquely identify its communication rules by scanning the message header structure of the data stream, identifying specific instruction code sequences, and reconstructing the interaction logic of its communication session.

[0070] Subsequently, these extracted raw features are compared and matched one by one with the feature templates of various system standard protocols in the pre-existing rule base. The preset system standard protocols are usually the benchmark protocols selected to achieve efficient interoperability between edge nodes. For example, a lightweight message queue telemetry transmission protocol can be selected as the standard for unified interaction within the system. The feature templates can be the message header structure features of "transaction identifier, protocol identifier, and length unit" predefined for industrial communication protocols based on TCP / IP networks, or the mapping relationship between frame ID and data slots predefined for real-time automation communication protocols based on industrial Ethernet.

[0071] Finally, based on the results of the above matching analysis, a clear protocol adaptation requirement is generated. This requirement specifically indicates which system standard protocol should be selected as the conversion target for the current data flow, as well as the core differences that need to be processed to complete this conversion, such as the mapping rules of data point addressing methods or the correspondence of service quality levels.

[0072] This application accurately identifies the protocol conversion targets and paths of multi-source heterogeneous industrial data through automated feature matching and analysis, thereby providing precise instructions for performing the correct protocol conversion at the correct node and laying the foundation for multi-protocol compatibility management.

[0073] S103. Perform multi-dimensional coupling analysis on the operation evaluation data to obtain the node operation range corresponding to the edge node.

[0074] The node operating range refers to the classification of edge nodes into different energy efficiency categories based on coupled analysis and dynamic learning of multi-dimensional operational evaluation data, such as edge node computing resource utilization, real-time energy consumption, and data throughput. For example, in an industrial edge computing cluster, edge nodes can be dynamically divided into: First, the high-efficiency operating range, where the computing load is typically at a medium-high level of 50%-80%, and the data processing efficiency per watt of energy consumption is higher than 8MB / s, demonstrating excellent energy efficiency and suitable for processing data streams with high real-time requirements; Second, the balanced operating range, where both resource utilization and energy efficiency are at a medium level; and Third, the inefficient or limited operating range, where the energy efficiency may be low due to hardware bottlenecks or excessively high basic energy consumption, such as a computing load consistently higher than 85% but low throughput. The boundaries of these ranges are not fixed but are continuously optimized based on the overall cluster load and historical performance data.

[0075] In one specific implementation, step S103 includes:

[0076] Step 1031: Perform coupled analysis on the computing resource utilization rate, energy consumption value and data processing throughput in the operation evaluation data to obtain a feature set.

[0077] Coupling analysis refers to linking three indicators—computing resource utilization, energy consumption, and data processing throughput—to analyze their mutual influence and correlation patterns.

[0078] In step 1031, the obtained operation evaluation data from each edge node is subjected to coupled analysis. Specifically, the computing resource utilization rate, energy consumption value and data processing throughput of each node at the same time are examined simultaneously, and the mutual influence relationship between these three core indicators is analyzed. It should be noted that the specific mathematical model or algorithm used to analyze the correlation between the indicators can refer to relevant data processing technology. The specific implementation process of this application embodiment will not be described in detail.

[0079] Next, the original indices that are related to each other are transformed and integrated into a set of composite features that can comprehensively characterize the energy efficiency level and processing capability of nodes, thereby constructing a feature set for deep state discrimination.

[0080] Step 1032: Based on the feature set, the state space partitioning method is used to determine the running state boundary of each edge node, and the node running state is classified according to the running state boundary to obtain the interval partitioning result.

[0081] The state space partitioning method is used to divide different regions in a multi-dimensional coordinate system based on the node states described by the feature set. Each region represents a typical operating state category, such as high-efficiency operating state, normal load state, resource idle state, high load state, and fault warning state. The operating state boundary refers to the dividing line or threshold used to distinguish the operating states of different categories of nodes after the state space is partitioned. These boundaries define which feature value combinations belong to the "high-efficiency operating range" and which belong to the "inefficient operating range".

[0082] In step 1032, based on the constructed feature set, a classification standard for node operating states is established using a state space partitioning method. This method places each node in a multi-dimensional feature space according to its feature value, and automatically calculates the operating state boundaries that distinguish different performance levels based on the distribution of all nodes in the space. Then, based on these initially determined operating state boundaries, the first classification is performed on all edge nodes, thereby obtaining the preliminary partitioning results of the node operating range.

[0083] It should be noted that the state space partitioning method can be a threshold-based partitioning method, a cluster analysis-based partitioning method, a classification model-based partitioning method, etc. As a known data classification technology, its specific partitioning algorithm can be referred to relevant existing technologies. The specific implementation process of this application will not be described in detail.

[0084] Step 1033: Based on the dynamic change trend of the operation evaluation data, perform boundary optimization processing on the interval division result to obtain the node operation interval.

[0085] Boundary optimization refers to fine-tuning the initially defined operational state boundaries based on the changing trends of node operation evaluation data over a period of time, so that the nodes can better adapt to the dynamically changing real environment.

[0086] In step 1033, the historical sequence of operational evaluation data of each node is continuously monitored and analyzed to identify dynamic patterns such as the continuous growth or decline trend of energy consumption and the periodic fluctuation pattern of throughput. Then, based on the identified dynamic trends, the initially divided operational state boundaries are re-evaluated and adjusted. For example, if it is found that the overall energy efficiency level of most nodes has improved over a period of time, the boundary conditions of the high-efficiency range may be relaxed accordingly. This optimization process makes the division of the node operating range no longer based on a static snapshot of a single point in time, but can absorb and reflect the performance evolution pattern of the node over a period of time, thereby outputting a more stable and accurate final node operating range.

[0087] This application scientifically and stably classifies edge nodes into different energy efficiency status categories by comprehensively analyzing multiple key operational indicators of nodes and considering their dynamic changes, thereby providing a reliable and adaptive judgment basis for the subsequent accurate matching between data tasks and node capabilities.

[0088] S104. Based on the node operating range and the protocol adaptation requirements, and combined with the energy efficiency data between edge nodes, an energy efficiency-aware dynamic resource scheduling method is used to allocate target edge nodes to the industrial internet data and generate scheduling results.

[0089] Among them, the energy efficiency-aware dynamic resource scheduling method refers to an automated scheduling method that considers not only the real-time computing power of nodes when making resource allocation decisions, but also the energy efficiency of nodes as a core trade-off factor.

[0090] In one specific implementation, such as Figure 2 As shown, step S104 includes:

[0091] Step 1041: Based on the data processing complexity and node operating range indicated by the protocol adaptation requirements, perform energy efficiency-aware data stream division on the industrial internet data to obtain real-time data stream and batch data stream.

[0092] Among them, data processing complexity refers to the relative computing resource intensity required to parse, transform or process the data based on protocol adaptation requirements; real-time data stream refers to a subset of industrial internet data that is highly sensitive to processing latency and needs to be responded to and processed quickly; batch data stream refers to a subset of industrial internet data that has lower requirements for processing timeliness and is allowed to be accumulated within a certain time window before being processed centrally.

[0093] In step 1041, the protocol adaptation requirements determined in step S102 are first analyzed, and the data processing complexity implied by the data to be processed is inferred from them. At the same time, the differences in energy efficiency and processing capacity of different node groups are grasped by combining the node operating range information obtained in step S103.

[0094] Next, the data streams are classified according to a preset partitioning strategy. This strategy is usually set as follows: data with strict requirements for processing latency and high processing complexity are classified as real-time data streams, in order to ensure the responsiveness of critical business operations; data that is not sensitive to latency or has a fixed processing mode and is suitable for accumulation and post-processing are classified as batch data streams, in order to pursue the optimization of overall processing efficiency.

[0095] For example, in a smart factory, two types of industrial internet data need to be processed: real-time vibration monitoring data of machine tools and daily energy consumption statistics reports of the workshop. The protocol adaptation requirements analysis shows that the vibration data adopts a binary stream protocol, which has high parsing complexity and requires millisecond-level response; the energy consumption report adopts JSON format, which is simple to parse and allows for hourly processing delays.

[0096] Meanwhile, it is known that there is node A in the high-efficiency operating range and node B in the low-efficiency operating range but idle; then, according to the preset strategy of "high complexity and high real-time requirement data are classified as real-time streams, and low complexity and delayable data are classified as batch streams", the vibration data stream is classified as a real-time data stream and the energy consumption report data stream is classified as a batch data stream.

[0097] Step 1042: Obtain energy efficiency data between edge nodes and use a state tracking mechanism to monitor and process the real-time operating status of edge nodes to obtain a state view.

[0098] Among them, the state tracking mechanism refers to a software process that continuously collects, filters, and aggregates raw operating indicators from edge nodes and generates a unified and stable state snapshot; the state view refers to a dynamic and comprehensive data snapshot, which is formed by aggregating operating status indicators collected in real time from various edge nodes, and is used to globally present the load and health status of the entire edge computing cluster at a certain moment.

[0099] In step 1042, firstly, energy efficiency data between edge nodes is obtained from the network management unit or the node self-reporting interface. This data quantifies the performance and energy consumption characteristics of the communication link between nodes. At the same time, core operating indicators of the nodes, such as the utilization rate of the central processing unit (CPU), memory usage rate, and current task queue depth, are collected at a fixed frequency through the monitoring agent deployed on each edge node.

[0100] The state tracking mechanism then smooths, thresholds, and performs aggregation calculations within the time window on these real-time reported raw data that may have instantaneous fluctuations, ultimately integrating them into a unified state view that reflects the global real-time load status of the cluster.

[0101] For example, the network management unit obtains that the transmission latency from node A to node C is 10ms, and the energy consumption per unit of data transmission is 0.5J / Mb; the transmission latency from node B to node C is 50ms, and the energy consumption is 0.8J / Mb. Meanwhile, the state tracking mechanism collects the state of each node once per second, and generates a state view after noise reduction and aggregation: it records that the CPU utilization of node A is 65% and stable, the CPU utilization of node B is 10% and idle, and the CPU utilization of node C is 80% and has an upward trend.

[0102] Step 1043: Based on the state view and the energy efficiency data, perform dynamic resource scheduling processing on the real-time data stream and the batch data stream to obtain the scheduling results of the data stream and edge nodes. Here, data stream refers to the real-time data stream and the batch data stream.

[0103] Step 1043 may specifically include the following steps:

[0104] Step a1: Use the real-time running status of each edge node in the status view as node attributes, determine the connection relationship and connection weight between nodes based on the communication efficiency and collaborative energy consumption between edge nodes in the energy efficiency data, and construct a navigation map.

[0105] The navigation graph refers to an abstract network model used to assist in scheduling decisions. In this model, each edge node is abstracted as a vertex, and its attributes are defined by the real-time running status in the state view, which are called node attributes. Node attributes include at least node identifier, current computing load level, node running range, and a list of supported protocol types. Furthermore, the data transmission links that may exist between nodes are abstracted as edges, and each edge is accompanied by a quantified connection weight to characterize the overall cost of using the link to transmit data.

[0106] For example, the running status of nodes A, B, and C in the status view is mapped to node attributes, and then a maximum tolerable delay benchmark is preset based on energy consumption data. =100ms, maximum transmission power consumption benchmark =1.0J / Mb, and set the delay factor. =0.7, energy consumption coefficient =0.3;

[0107] Then, for the link from node A to node C, its transmission delay d AC =10ms, unit energy consumption e AC=0.5J / Mb, and through the formula w AC Calculate the connection weight between A and C to obtain w. AC =0.22; Similarly, for the link from node B to node C, its transmission delay d BC =50ms, unit energy consumption e BC =0.8J / Mb, calculate its connection weight w BC =0.59, and finally the navigation map is constructed.

[0108] Step a2: Using the navigation map, perform differentiated node matching on the real-time data stream and the batch data stream to obtain a node matching scheme.

[0109] Step a2 may specifically include the following steps:

[0110] Step b1: Based on the protocol adaptation requirements, extract the delay constraint parameters and energy efficiency constraint parameters for the real-time data stream and the batch data stream, respectively.

[0111] For example, protocol adaptation requirements analysis shows that the real-time vibration data stream uses an industrial real-time protocol that is extremely sensitive to latency, corresponding to monitoring services that require millisecond-level response. Therefore, latency constraint parameters are set for the real-time data stream. =25ms and energy efficiency constraint parameters =1 record / second / watt; while the batch report data stream uses an offline-friendly protocol, corresponding to backend statistical business that allows for delayed processing, therefore, energy efficiency is a greater concern. Thus, energy efficiency constraints and latency constraints are set for the batch report data stream. =1 hour and energy efficiency constraint parameters =5 lines / second / watt.

[0112] Step b2: Based on the node attributes and connection weights of each node in the navigation graph, and combined with the latency constraint parameters and energy efficiency constraint parameters respectively, construct a first matching evaluation function for the real-time data stream with the optimization objective of minimizing end-to-end latency, and construct a second matching evaluation function for the batch data stream with the optimization objective of maximizing energy efficiency utilization.

[0113] For example, construct a first matching evaluation function for the real-time data stream, namely: for any node i, define a set of nodes that satisfy the real-time stream constraints. ,in, This represents the total transmission delay of the path from the data source to node i, calculated based on the connection weights in the navigation graph. This represents the estimated processing energy efficiency evaluated based on the attributes of node i. The latency constraint parameter represents the real-time data stream. Let the energy efficiency constraint parameters of the real-time data stream represent the target of the first matching evaluation function. Select the node with the minimum transmission delay: A second matching evaluation function is constructed for the batch data stream, namely: for any node j, a set of nodes that satisfy the batch stream constraints is defined. ,in, This represents the total transmission delay of the path from the data source to node j, calculated based on the connection weights in the navigation graph. This represents the estimated energy efficiency based on the attributes of node j. The delay constraint parameter represents the batch data stream. Let the energy efficiency constraint parameters of the batch data stream represent the energy efficiency constraints. Then, the objective of the second matching evaluation function is to obtain the energy efficiency constraint parameters from the batch data stream. Select the node with the highest energy efficiency: In the above plan, " " represents the "logical AND" operation.

[0114] It should be noted that the first matching evaluation function takes time delay as the main optimization objective, but may also include other objectives. Similarly, the second matching evaluation function takes maximizing energy efficiency as the main optimization objective, but may also include other objectives. Therefore, the specific expressions of these two functions are not specifically limited in the embodiments of this application, and can be analyzed according to the actual situation.

[0115] In step b2, firstly, a first evaluation function for matching real-time data streams is constructed. The core of this function is to prioritize low latency while ensuring basic processing efficiency. Specifically, the function evaluates each candidate node:

[0116] First, check whether the node's real-time status allows it to receive new tasks and confirm that its estimated energy efficiency is not lower than the set minimum energy efficiency requirement. Only if these two prerequisites are met will the transmission latency cost from the data source to the node be further calculated. The transmission latency cost can be calculated based on the link weights on the path in the network. Finally, among all nodes whose latency costs do not exceed the set threshold, the node with the lowest latency is selected as the matching result. This function organically combines network link status, node real-time load and energy efficiency attributes, as well as the dual constraints of latency and energy efficiency, to form a complete decision model for real-time data streams.

[0117] Secondly, a second evaluation function is constructed to match batch data streams. This function primarily aims for high energy efficiency while allowing for relatively lenient transmission latency. Specifically, the function first checks whether each candidate node has sufficient resources and protocol compatibility, and calculates whether the cumulative transmission time from the data source to the node is within the maximum allowable latency range.

[0118] After the above conditions are met, the energy efficiency ratio of the node for processing the batch tasks is estimated based on the node's historical performance data and operating status range. Finally, among all nodes with an energy efficiency ratio not lower than the set energy efficiency benchmark, the node with the highest energy efficiency is selected as the matching result. This function also integrates node attributes, network link weights, and dual constraints of energy efficiency and latency to construct a screening and selection mechanism suitable for batch data streams.

[0119] Step b3: Based on the first matching evaluation function and the second matching evaluation function, in the navigation graph, candidate node sets that meet their respective constraints are screened in parallel for the real-time data stream and the batch data stream.

[0120] In step b3, for the real-time data stream, this application traverses all nodes in the navigation graph according to the rules defined by the first matching evaluation function. For each node, it first determines whether its node attributes indicate that it can accept new tasks and the estimated processing efficiency meets the basic requirements. If it does, it calculates the transmission delay cost based on the connection weight on the path between the node and the data source, and determines whether the cost does not exceed the set delay constraint threshold. All nodes that meet all the above conditions are included in the candidate node set of the real-time data stream.

[0121] For batch data streams: This application synchronously traverses all nodes in the navigation graph according to the rules defined by the second matching evaluation function. For each node, it first determines whether its node attributes indicate sufficient resources and protocol compatibility, and calculates whether the cumulative transmission time is within the maximum allowable latency range based on the path connection weight. If satisfied, it calculates or obtains the estimated energy efficiency ratio for processing this type of batch task based on the node attributes, and determines whether the energy efficiency ratio is not lower than the set energy efficiency constraint threshold. All nodes that meet all the above conditions are included in the candidate node set of the batch data stream.

[0122] Step b4: Perform conflict detection and elimination processing on the candidate node set of all data streams, and based on the results of conflict elimination, determine a unique target edge node from the corresponding candidate node set for each data stream to obtain a node matching scheme.

[0123] In step b4, firstly, conflict detection is performed on the candidate node sets of real-time data stream and batch data stream, that is, to check whether the two sets contain the same node, or to determine whether the relevant nodes have a competitive relationship in terms of computing, bandwidth and other resources.

[0124] If a conflict is detected, it is eliminated according to a preset conflict resolution strategy. For example, real-time data streams are given a higher scheduling priority to ensure their node allocation, while batch data streams are removed from their candidate set and re-selected; or the conflicting nodes are evaluated to see if they can serve both types of data streams simultaneously through resource allocation; if there is no conflict, the selection proceeds directly to the next step.

[0125] Based on conflict resolution, a unique target node is determined from the candidate node set corresponding to each data stream: for real-time data streams, the node with the minimum transmission latency is selected; for batch data streams, the node with the highest estimated processing energy efficiency is selected. Finally, a definite mapping relationship is established between each data stream and its corresponding target node, generating a complete node matching scheme.

[0126] For example, if the two candidate node sets do not overlap after conflict detection, the node matching scheme is determined as follows: the real-time vibration data stream is assigned to node A, and the batch energy consumption report data stream is assigned to node B.

[0127] Step a3: Based on the node matching scheme, generate the scheduling results of the data stream and edge nodes.

[0128] Step a3 may specifically include the following steps:

[0129] Step c1: Based on the node matching scheme, select a transmission path that meets the corresponding quality of service requirements for each data stream in the navigation map. During the selection process, coordinate the bandwidth resource allocation among multiple data streams to obtain a path configuration scheme.

[0130] For example, based on the node matching scheme, a high-priority queue is configured and bandwidth is reserved for the path of real-time vibration flow to node A, and a standard priority is configured for the path of batch report flow to node B.

[0131] Step c2: Based on the path configuration scheme and the node matching scheme, obtain the scheduling results of the data stream and edge nodes.

[0132] For example, based on the above results, the final scheduling instructions are generated: send the real-time vibration data to node A for processing via a high-priority path; and send the batch report data to node B for processing via a standard path.

[0133] This application achieves dynamic and optimal matching of industrial data tasks and edge computing resources based on energy efficiency through refined data flow partitioning, real-time cluster status awareness, and differentiated resource matching and path planning. This ensures the performance and service quality of real-time critical businesses, while guiding non-urgent tasks to nodes with better energy efficiency. Thus, while meeting the diverse needs of complex industrial scenarios, it further improves the resource utilization efficiency and system energy efficiency of the entire edge computing cluster.

[0134] S105. Based on the scheduling result, the industrial internet data is routed to the corresponding target edge node, so that the target edge node can use the dynamic adaptation method of the protocol suite to perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements, obtain standardized data and store it locally, so as to realize the full-process management of industrial internet data.

[0135] Among them, the protocol suite refers to a component library that is pre-integrated into the edge node software and contains the ability to parse and generate various common industrial communication protocols.

[0136] In one specific implementation, step S105 includes:

[0137] Step 1051: Based on the path configuration scheme and node matching scheme in the scheduling result, route the industrial internet data to the corresponding target edge node.

[0138] In step 1051, the scheduling result generated in step S104 is first parsed. The result explicitly includes a node matching scheme and a path configuration scheme. The node matching scheme indicates the target edge node corresponding to each data stream, while the path configuration scheme defines the specific network path, priority, and resource reservation information to reach these target nodes.

[0139] Next, based on the path configuration scheme, the forwarding devices in the network configure corresponding quality of service policies and forwarding rules for different data streams; finally, the industrial internet data to be managed is injected into the network, and the data packets are accurately classified and forwarded according to the configured rules, and finally arrive at their respective target edge nodes along the predetermined path, thus completing the closed loop from logical scheduling to physical transmission.

[0140] For example, if the scheduling result indicates that "vibration data flows to node A with high path priority", the network switch will configure a low-latency queue for data packets destined for node A and matching a specific port to ensure that they are forwarded with priority.

[0141] Step 1052: On the target edge node, establish a protocol mapping table based on the protocol adaptation type corresponding to the node's operating range.

[0142] In step 1052, after the data successfully arrives at the target edge node, the node first queries its own status identifier and obtains the node operating range determined in step S103; then, based on a pre-loaded or configuration policy obtained from the management unit, it searches for the protocol adaptation type bound to the operating range.

[0143] Finally, based on the found protocol type, the corresponding detailed configuration parameters are initialized or loaded in memory, thus forming a protocol mapping table that can be quickly queried in subsequent steps. This table clarifies which protocol data from different runtime contexts should ultimately be converted to on this node.

[0144] For example, a node marked as being in a "high-efficiency operating range" has its protocol adaptation type set to "Message Queuing Telemetry Transport (MQTT) protocol" according to a pre-defined policy; the node then builds a protocol mapping table based on this, which includes specific configurations such as the MQTT broker server address, client authentication information, and topic namespace.

[0145] Step 1053: Based on the protocol adaptation requirements, select the target protocol configuration from the protocol mapping table, match the protocol features of the industrial internet data with the feature templates in the protocol cluster, and obtain the source protocol type.

[0146] Among them, the feature template refers to the key feature fingerprint predefined for each known protocol in the protocol family for rapid identification; the source protocol type refers to the specific communication protocol variety followed by the original industrial Internet data, which is identified through feature matching.

[0147] In step 1053, firstly, based on the protocol adaptation requirements attached to the data stream or obtained from the scheduling signaling, a specific target protocol configuration is selected from the established protocol mapping table to determine the target format that the data needs to be converted into. At the same time, the built-in protocol cluster analysis engine is called to parse the received raw data packets to extract key features such as message structure and instruction code.

[0148] Then, these extracted original protocol features are quickly compared with various industrial protocol feature templates pre-stored in the protocol cluster. By matching, the communication protocol originally used by the data packet is accurately determined, that is, its source protocol type is determined. Here, the industrial protocol feature template is the same as the feature template explained in S102.

[0149] Step 1054: Based on the target protocol configuration and the source protocol type, convert the industrial internet data from the source protocol format to the target protocol format, perform verification processing on the industrial internet data in the target protocol format, and generate standardized data that conforms to the preset protocol specifications and store it locally according to the verification results.

[0150] Among them, the source protocol format and the target protocol format refer to the two different and specific message organization rules and data encoding methods followed by industrial Internet data before and after conversion; verification processing refers to the process of checking the rule compliance of the converted data format to ensure that the generated data fully complies with the specifications of the target protocol.

[0151] In step 1054, the target edge node activates the corresponding converter in the protocol cluster based on the determined source protocol type and target protocol configuration. The converter unpacks, semantically parses, extracts and reassembles the original data according to predefined mapping rules, thereby completing the translation from the source protocol format to the target protocol format.

[0152] Then, after the data is converted and generated, a validation process is performed to check whether its syntax structure, field integrity, and business logic conform to the preset protocol specifications. For example, if the target protocol is MQTT, the validation process includes checking whether the generated data packet contains a fixed MQTT message header, whether the control message type is valid, whether the remaining length field is calculated correctly, and whether the topic name format conforms to the specifications for published messages. The data that passes the validation is then confirmed as valid standardized data.

[0153] Finally, this batch of standardized data is written to persistent storage devices according to the local storage strategy, thus completing the final transformation from raw heterogeneous data to unified and standardized data within the system.

[0154] This application delivers data to designated nodes through precise routing control, and completes a fully automated processing flow on the nodes, from intelligent protocol identification and on-demand configuration conversion to rigorous verification and implementation. Ultimately, it achieves the standardization and persistence of multi-source heterogeneous industrial data at the edge, thereby providing a high-quality, unified format data source for upper-layer applications and further connecting the last link in the entire process management of industrial internet data.

[0155] Example 2

[0156] Figure 3 This is a schematic diagram illustrating a specific implementation of an industrial internet data management system based on edge computing, as provided in this application embodiment. (Refer to...) Figure 3 The system may include:

[0157] The acquisition module 31 is used to acquire industrial internet data to be managed, as well as operational evaluation data of each edge node in the industrial internet.

[0158] The first analysis module 32 is used to perform matching analysis between the original protocol characteristics of the industrial internet data and the preset system standard protocol to determine the protocol adaptation requirements corresponding to the industrial internet data.

[0159] A second analysis module 33, configured to perform multi-dimensional coupling analysis on the operation evaluation data to obtain a node operation interval corresponding to the edge node.

[0160] An allocation module 34, configured to, based on the node operation interval and the protocol adaptation requirement, combine the energy efficiency data among the edge nodes, and use an energy efficiency-aware dynamic resource scheduling method to allocate target edge nodes for the industrial Internet data, and generate a scheduling result.

[0161] An identification module 35, configured to route the industrial Internet data to the corresponding target edge node based on the scheduling result, so that the target edge node uses a dynamic adaptation method of a protocol cluster to perform protocol identification and adaptation conversion processing on the industrial Internet data according to the protocol adaptation requirement, obtain standardized data and store it locally, so as to implement the full-process management of the industrial Internet data.

[0162] The industrial Internet data management system based on edge computing provided by the embodiments of the present application is used to implement the foregoing industrial Internet data management method based on edge computing. Therefore, the specific implementation manners in the industrial Internet data management system based on edge computing can be seen in the embodiment part of the industrial Internet data management method based on edge computing in the foregoing text. The specific implementation manners can refer to the descriptions of the corresponding various part embodiments and will not be elaborated herein.

[0163] As Figure 4 shown, the present application further provides an electronic device, including: a memory 41, configured to store a computer program; a processor 42, configured to implement the steps of any one of the foregoing industrial Internet data management methods based on edge computing when executing the computer program.

[0164] The present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the foregoing industrial Internet data management methods are implemented.

[0165] In an exemplary embodiment, the foregoing computer-readable storage medium may include, but is not limited to: various media such as a USB flash drive, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk, or an optical disc that can store a computer program.

[0166] The embodiments of the present invention further provide a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any one of the embodiments of the foregoing industrial Internet data management method based on edge computing are implemented.

[0167] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0168] The above provides a detailed description of the industrial internet data management method and system based on edge computing provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. An industrial internet data management method based on edge computing, characterized in that, include: Acquire the industrial internet data to be managed, as well as the operational evaluation data of each edge node in the industrial internet; Based on the original protocol characteristics of the industrial internet data, a matching analysis is performed with the preset system standard protocol to determine the protocol adaptation requirements corresponding to the industrial internet data. A multi-dimensional coupled analysis is performed on the operational evaluation data to obtain the node operational range corresponding to the edge node; Based on the node operating range and the protocol adaptation requirements, and combined with the energy efficiency data between edge nodes, an energy efficiency-aware dynamic resource scheduling method is used to allocate target edge nodes to the industrial internet data and generate scheduling results. Based on the scheduling results, the industrial internet data is routed to the corresponding target edge nodes, so that the target edge nodes can use the dynamic adaptation method of protocol families to perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements, so as to obtain standardized data and store it locally.

2. The method according to claim 1, characterized in that, Based on the node operating range and the protocol adaptation requirements, and combined with energy efficiency data between edge nodes, an energy efficiency-aware dynamic resource scheduling method is used to allocate target edge nodes to the industrial internet data, generating scheduling results, including: Based on the data processing complexity and node operating range indicated by the protocol adaptation requirements, the industrial internet data is divided into energy efficiency-sensing data streams to obtain real-time data streams and batch data streams. Energy efficiency data between edge nodes is acquired, and a state tracking mechanism is used to monitor and process the real-time operating status of the edge nodes to obtain a state view; Based on the state view and the energy efficiency data, dynamic resource scheduling processing is performed on the real-time data stream and the batch data stream to obtain the scheduling results of the data stream and edge nodes.

3. The method according to claim 2, characterized in that, The process of performing dynamic resource scheduling on the real-time data stream and the batch data stream based on the state view and the energy efficiency data to obtain the scheduling results of the data stream and edge nodes includes: The real-time running status of each edge node in the state view is used as the node attribute. Based on the communication efficiency and collaborative energy consumption between edge nodes in the energy efficiency data, the connection relationship and connection weight between nodes are determined, and a navigation graph is constructed. Using the navigation map, differentiated node matching is performed on the real-time data stream and the batch data stream to obtain a node matching scheme; Based on the node matching scheme, the scheduling results of the data stream and edge nodes are generated.

4. The method according to claim 3, characterized in that, The step of using the navigation map to perform differentiated node matching on the real-time data stream and the batch data stream to obtain a node matching scheme includes: Based on the protocol adaptation requirements, delay constraint parameters and energy efficiency constraint parameters are extracted for the real-time data stream and the batch data stream, respectively. Based on the node attributes and connection weights of each node in the navigation graph, and combined with the latency constraint parameters and energy efficiency constraint parameters respectively, a first matching evaluation function is constructed for the real-time data stream with the optimization objective of minimizing end-to-end latency, and a second matching evaluation function is constructed for the batch data stream with the optimization objective of maximizing energy efficiency utilization. Based on the first matching evaluation function and the second matching evaluation function, candidate node sets that satisfy their respective constraints are screened in parallel for the real-time data stream and the batch data stream in the navigation graph; Conflict detection and elimination are performed on the candidate node sets of all data streams. Based on the results of conflict elimination, a unique target edge node is determined for each data stream from the corresponding candidate node set, thus obtaining a node matching scheme.

5. The method according to claim 3, characterized in that, The generation of data stream and edge node scheduling results based on the node matching scheme includes: Based on the node matching scheme, a transmission path that meets the corresponding quality of service requirements is selected for each data stream in the navigation map. During the selection process, the bandwidth resource allocation among multiple data streams is coordinated to obtain a path configuration scheme. Based on the path configuration scheme and the node matching scheme, the scheduling results of the data stream and edge nodes are obtained.

6. The method according to claim 1, characterized in that, The process of routing the industrial internet data to the corresponding target edge node based on the scheduling result, so that the target edge node can use a dynamic adaptation method for protocol families, and perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements to obtain standardized data and store it locally, includes: Based on the path configuration scheme and node matching scheme in the scheduling results, the industrial internet data is routed to the corresponding target edge node; On the target edge node, a protocol mapping table is established based on the protocol adaptation type corresponding to the node's operating range; According to the protocol adaptation requirements, a target protocol configuration is selected from the protocol mapping table, and the protocol characteristics of the industrial internet data are matched with the feature templates in the protocol cluster to obtain the source protocol type. Based on the target protocol configuration and the source protocol type, the industrial internet data is converted from the source protocol format to the target protocol format, and the industrial internet data in the target protocol format is verified. Based on the verification results, standardized data conforming to the preset protocol specifications is generated and stored locally.

7. The method according to claim 1, characterized in that, The multi-dimensional coupled analysis of the operational evaluation data to obtain the node operational range corresponding to the edge node includes: A feature set is obtained by performing coupled analysis on the computing resource utilization rate, energy consumption value and data processing throughput in the operation evaluation data; Based on the feature set, the state space partitioning method is used to determine the running state boundary of each edge node. According to the running state boundary, the node running state is classified to obtain the interval partitioning result. Based on the dynamic changing trend of the operational evaluation data, the boundary optimization processing of the interval division results is performed to obtain the node operational interval.

8. An industrial internet data management system based on edge computing, characterized in that, include: The acquisition module is used to acquire industrial internet data to be managed, as well as operational evaluation data of each edge node in the industrial internet. The first analysis module is used to perform matching analysis between the original protocol characteristics of the industrial internet data and the preset system standard protocol to determine the protocol adaptation requirements corresponding to the industrial internet data. The second analysis module is used to perform multi-dimensional coupled analysis on the operation evaluation data to obtain the node operation range corresponding to the edge node; The allocation module is used to allocate target edge nodes to the industrial internet data based on the node's operating range and the protocol adaptation requirements, combined with the energy efficiency data between edge nodes, using an energy efficiency-aware dynamic resource scheduling method, and generating scheduling results. The identification module is used to route the industrial internet data to the corresponding target edge node based on the scheduling result, so that the target edge node can use the dynamic adaptation method of the protocol family to perform protocol identification and adaptation conversion processing on the industrial internet data according to the protocol adaptation requirements, so as to obtain standardized data and store it locally.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the edge computing-based industrial internet data management method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of the industrial internet data management method based on edge computing as described in any one of claims 1 to 7.

Images