Power edge cloud collaborative management method based on swan OS

Abstract

The invention relates to a power edge cloud cooperative management method based on a swan OS, and the method comprises the following steps: S1, deploying intelligent edge nodes, setting an embedded AI acceleration unit, operating the swan OS on all the intelligent edge nodes, and achieving the seamless cooperation of equipment through the distributed characteristics of the swan OS; and S2, realizing self-organizing connection of edge devices by adopting a Mesh network technology, designing a direct communication path between the edge devices, and realizing rapid processing of local problems. S3, aiming at the real-time requirement of a power system, optimizing an MQTT / CoAP protocol stack, improving the efficiency and reliability of message transmission, and allocating special network resources for different types of power applications by utilizing the 5G network slice characteristics, S4, constructing a security architecture based on a zero-trust principle, enhancing the equipment access control and data transmission security, and S5, establishing a security architecture based on a zero-trust principle. And identity authentication and data integrity protection are performed by using a block chain technology, so that data tampering and counterfeiting are prevented. According to the method, the local distributed decision-making capability and the global cooperation capability are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of edge computing technology, and in particular to a power edge cloud collaborative management method based on Hongmeng OS. Background Art

[0002] As the power system rapidly develops towards intelligence, the traditional centralized cloud computing model can no longer meet the power grid's demand for low latency, high reliability and high security communications. The widespread application of smart grids, such as distributed energy access, dynamic energy management, and real-time fault protection, requires edge devices to have stronger independent processing capabilities to cope with emergencies and massive data processing. On the other hand, the complexity and diversity of the power system also puts forward the demand for differentiated resource allocation and precise security control of the communication network. At the same time, with the rise of emerging technologies such as edge computing, blockchain, and 5G network slicing, in order to further enhance the collaborative management capabilities of the power system, the collaborative management model based on edge cloud has become a better choice. Summary of the invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide a power edge cloud collaborative management method based on Hongmeng OS, which effectively improves local distributed decision-making capabilities and global collaborative capabilities.

[0004] To achieve the above object, the present invention adopts the following technical solutions: A power edge cloud collaborative management method based on Hongmeng OS, comprising the following steps S1: Deploy intelligent edge nodes, set up embedded AI acceleration units, and run Hongmeng OS on all intelligent edge nodes, using its distributed characteristics to achieve seamless collaboration of devices; S2: Use Mesh network technology to achieve self-organizing connection of edge devices and improve the robustness and self-healing ability of the system; reduce dependence on the cloud, design direct communication paths between edge devices, and achieve rapid processing of local problems.

[0005] S3: In response to the real-time requirements of the power system, the MQTT / CoAP protocol stack is optimized to improve the efficiency and reliability of message transmission. The network slicing feature of 5G is used to allocate dedicated network resources for different types of power applications to improve communication quality.

[0006] S4: Build a security architecture based on the zero-trust principle, enhance device access control and data transmission security, use blockchain technology for identity authentication and data integrity protection, and prevent data tampering and forgery.

[0007] Further, S1 is as follows: S11: Each intelligent edge node runs Hongmeng OS and acts as an intelligent terminal, with edge storage, computing power, distributed service capabilities, and supports AI reasoning. Through the broadcast discovery mechanism of the soft bus in the Hongmeng OS distributed architecture, dynamic networking between edge nodes is completed, a collaborative working environment is established, and storage, computing and task management resource sharing of distributed devices is realized. Through the Hongmeng OS distributed communication framework, direct communication between edge nodes is prioritized, and task linkage management is carried out in conjunction with the cloud: S12: Based on the distributed soft bus of Hongmeng OS, node devices automatically discover and complete spatial perception and connection; edge node resources are shared, including storage and computing power, and multiple nodes collaborate to perform large tasks; S13: Through the Hongmeng OS distributed communication framework, point-to-point data exchange between edge devices is prioritized to reduce dependence on the cloud.

[0008] Furthermore, based on the distributed soft bus of Hongmeng OS, node devices automatically discover and complete spatial perception and connection, as follows: In the Hongmeng OS environment, the distributed soft bus supports the use of broadcast discovery protocol to complete spatial perception of multiple nodes through broadcast access or peripheral scanning; Based on the node identifier, each device automatically authenticates and securely joins the group; Perceive the relative distance and topological relationship between nodes, and generate collaborative communication network topology based on the network quality and hardware capabilities of the devices; Each node contributes its local storage space to the collaborative group: Use Hongmeng OS's distributed file system to achieve file storage distribution and shared access; multiple nodes collaborate to write front-end data collection records and dynamically synchronize them to other edge nodes; Cold data and hot data are extracted from different nodes through hierarchical management to improve access speed.

[0009] In the collaborative group, tasks are dynamically divided based on the size of computing tasks and node load conditions: CPU / NPU-intensive tasks are preferentially distributed to high-performance nodes; storage-intensive tasks are distributed and managed in nodes with larger storage space; and the distributed soft bus is used to uniformly schedule requests to call the resources of other idle nodes to execute large-scale analysis tasks.

[0010] Further, S2 is as follows: Deploy Mesh network technology, and intelligent edge devices automatically join the Mesh network using the broadcast discovery mechanism after they are powered on; Establish network topology and generate adjacency matrix. Edge nodes periodically send node information based on the distributed soft bus of Hongmeng OS, including node ID and cost information. The cost information includes link delay, RSSI, load, and remaining bandwidth. The information of directly connected nodes is stored through the adjacency table: A dj (i) = {j∈Nodes | j is directly connected to i} The adjacency matrix is ​​expressed as:

[0011] w ij represents the link weight; Dynamically plan the optimal communication path between nodes based on link quality evaluation; Data is transmitted between devices through multi-hop routing, while node status is detected in real time, and failed nodes are automatically bypassed. After each routing adjustment is completed, the network topology is synchronously updated to the entire network; Based on the optimal communication path between nodes, a point-to-point direct communication link is established without cloud intervention; the routing table is updated regularly and the communication path is adjusted dynamically; Through edge layer collaboration, problems can be quickly detected. If a node failure or temperature abnormality is found, the adjacent node will directly take over the corresponding task, and local problems can be quickly handled through data collaboration and control logic transmission. Through dynamic task distribution and linkage between edge nodes, fast scheduling of distributed logic can be achieved; With the distributed architecture, Hongmeng OS decomposes the computing, data sharing and control logic of the device to the edge layer.

[0012] Furthermore, based on link quality evaluation (RSSI, bandwidth), the optimal communication path between nodes is dynamically planned as follows: Periodic monitoring link l ij Quality indicators, including RSSI, bandwidth and load; Update link weight w ij And store it in the adjacency matrix; ; Among them, α, β, γ are weight coefficients; RSSI ij Link l ij Signal strength; BW ij Link l ij The remaining available bandwidth of C j is the current load of the receiving node; Defined in the network graph G=(N,A), find the optimal path P from the source node s to the target node t s→t, so that the total path weight is minimized; N is the set of intelligent edge nodes in the network; A is the set of communication links in the network; Use Dijkstra algorithm to generate the optimal path to the target node in real time based on the dynamic adjacency matrix. Initialize the shortest distance d[i]=∞ of the current node i, except d[s]=0 for i∈N; Update the distance of neighbor nodes: ; Continue iterating unvisited nodes until the target node t is visited; Total path weight:

[0013] When link l ij The RSSI or bandwidth changes, and the weight w is adjusted in real time ij , recalculate path P; if RSSI ij Too low or BW ij = 0, marking the weight as infinity (∞) and triggering replanning.

[0014] Furthermore, based on the optimal communication path between nodes, a point-to-point direct communication link is established without cloud intervention; the routing table is updated regularly and the communication path is adjusted dynamically, as follows: Each node independently maintains a dynamic routing table to record all target nodes and their communication paths. The routing table structure of node i is : ; Among them, NextHop is the next hop node to the target node; Each edge node searches its routing table according to the target node t and obtains the next hop node NextHop: NextHop=R i [t].NextHop; If NextHop is the target node t, then communicate directly; R i [t] is the routing table obtained according to the target node t; Otherwise, the data is forwarded to the next hop node until it reaches the target node; Data transmission is forwarded hop by hop through the nodes in the target path according to the following process: The source node generates path P according to the routing table s→t ; The data packet is forwarded hop by hop according to each node in the path; The target node t receives and processes the data; In the Hongmeng OS environment, its distributed soft bus is used to support direct communication between different edge devices, eliminating intermediate nodes and cloud intervention.

[0015] Furthermore, in view of the real-time requirements of the power system, the MQTT / CoAP protocol stack is optimized to improve the efficiency and reliability of message transmission, as follows: MQTT uses a lightweight version of MQTT-SN to adapt to the situation where edge devices have limited resources: UDP-based transmission is used to avoid complex connection management and improve real-time performance; CoAP uses the Non-confirmable mode to process low-priority messages and reduce handshake delays. It optimizes ACK messages and combines UDP's fast timeout retransmission mechanism to ensure reliable transmission of critical data. Use efficient compression to enable CBOR encoding for MQTT / CoAP message content, replacing traditional JSON to reduce data volume; enable data merging and compression in scenarios with high link load, bundle multiple data for transmission, and reduce message-level overhead; Improved QoS policy: MQTT QoS mechanism: QoS 0: low priority monitoring signal, used for non-critical monitoring data, using UDP transmission to reduce the overhead of establishing connections; QoS 1: Medium priority power collection signal, optimized retransmission strategy, set dynamic retransmission timeout Tretry(i'), adjust the message retransmission interval according to network delay and packet loss rate:

[0016] in, For the Number of retries, is the basic retransmission time, is the maximum retransmission time; QoS 2: Key protection control signal: With 5G URLLC slicing, delays caused by retransmission are avoided. In CoAP, the timeout assumption model is improved: Dynamically update the estimated timeout RTT value based on the particularity of the collection environment: ; Where a is the weight; R(s') refers to the actual network return time measured at the current time s'; Use EWMA to smoothly measure RTT and reduce unnecessary retries, especially in unstable networks; Integrate a PTP / NTP-based time synchronization module into the protocol stack to ensure strict timestamp accuracy for power protection signals and event messages; Add priority identifiers to MQTT / CoAP messages and process them through a dual-queue scheduling mechanism.

[0017] Furthermore, the network slicing feature of 5G is used to allocate dedicated network resources for different types of power applications to improve communication quality, as follows: According to the different business requirements of the power system, three types of slices are divided: URLLC slice: used for protection and control communications, such as line protection, fault detection, and automation control signals; Configure high-priority resource scheduling to ensure bandwidth and reliability (such as preventing data packet loss through HARQ (Hybrid ARQ)); eMBB (enhanced mobile broadband) slice: used to transmit data that requires high bandwidth, such as power status monitoring and video monitoring; The configuration throughput is greater than the threshold and is adapted to the intelligent monitoring system to transmit big data waveforms, event records or analysis results; mMTC (massive machine type communication) slicing is used to collect data from wide-area power terminal devices (such as electricity meters and distribution network equipment) to optimize terminal energy consumption and signaling overhead.

[0018] Network slice resource allocation is based on the actual load demand of the power system and is adjusted dynamically: Slice resource allocation function: Slice resource Rs is allocated by the ratio Decisions are made and dynamically adjusted according to task priority and latency requirements: ; Among them, P s is the slice priority; Q s Real-time load for slicing; R total is the total network bandwidth; For all n' The weighted sum of the priority and real-time load factor of each slice. The subscript i'' is used to traverse all slices. Based on the real-time monitoring of the power system, the network slice orchestrator is used to dynamically adjust the weight of each slice. The QoS flow mechanism of 5G is utilized, combined with the priority of MQTT / CoAP messages, to provide the most suitable slices for power communications of different priorities.

[0019] Furthermore, S4 is specifically: The power system's identity authentication and permission management are designed based on the zero-trust principle. A unique identifier is configured for each edge node, and a PKI public key system is used to establish secure communication between devices. Real-time identity authentication is implemented when nodes are connected, accessed or communicated to continuously verify the reliability of the equipment.

[0020] Introduce blockchain in data exchange and exception handling: Use blockchain to record device behavior logs (such as operating status or instructions) to ensure that data cannot be tampered with; provide on-chain evidence storage capabilities to achieve data auditing and tracking.

[0021] Use lightweight AES encryption algorithm to encrypt all edge node communication data; UDP-based TLS protects CoAP transmission.

[0022] A power edge cloud collaborative management system based on Hongmeng OS, comprising a processor, a memory and a computer program stored on the memory. When the processor executes the computer program, it specifically executes the steps in the power edge cloud collaborative management method based on Hongmeng OS as described above.

[0023] The present invention has the following beneficial effects: 1. Based on the distributed characteristics of Hongmeng OS and combined with an embedded AI acceleration unit, the present invention can realize the intelligence and efficient collaboration of power edge devices, reduce dependence on cloud computing, and significantly improve the real-time and stability of edge communications by introducing Mesh network technology and optimized MQTT / CoAP protocol stack design; 2. The present invention utilizes the network slicing characteristics of 5G to provide customized communication resources for different types of power business scenarios (such as protection control, status monitoring, and massive data collection) to meet diverse needs. It also provides reliable identity authentication and data integrity protection for the power system through the integration of zero-trust architecture and blockchain technology, thereby comprehensively improving the security and defense capabilities of the power grid. 3. The present invention utilizes Hongmeng OS and distributed architecture to give full play to edge computing capabilities, reduce dependence on the cloud, and improve the system's autonomy and local problem-solving capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0025] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: refer to Figure 1 In the present invention, a power edge cloud collaborative management method based on Hongmeng OS is provided, comprising the following steps: S1: Deploy intelligent edge nodes, set up embedded AI acceleration units, and run Hongmeng OS on all intelligent edge nodes, using its distributed characteristics to achieve seamless collaboration of devices; S2: Use Mesh network technology to achieve self-organizing connection of edge devices and improve the robustness and self-healing ability of the system; reduce dependence on the cloud, design direct communication paths between edge devices, and achieve rapid processing of local problems; S3: In view of the real-time requirements of the power system, the MQTT / CoAP protocol stack is optimized to improve the efficiency and reliability of message transmission. The network slicing feature of 5G is used to allocate dedicated network resources for different types of power applications to improve communication quality. S4: Build a security architecture based on the zero-trust principle, enhance device access control and data transmission security, use blockchain technology for identity authentication and data integrity protection, and prevent data tampering and forgery.

[0026] In this embodiment, S1 is specifically as follows: S11: Each intelligent edge node runs Hongmeng OS and acts as an intelligent terminal, with edge storage, computing power, distributed service capabilities, and supports AI reasoning. Through the broadcast discovery mechanism of the soft bus in the Hongmeng OS distributed architecture, dynamic networking between edge nodes is completed, a collaborative working environment is established, and storage, computing and task management resource sharing of distributed devices is realized. Through the Hongmeng OS distributed communication framework, direct communication between edge nodes is prioritized, and task linkage management is carried out in conjunction with the cloud: S12: Based on the distributed soft bus of Hongmeng OS, node devices automatically discover and complete spatial perception and connection; edge node resources are shared, including storage and computing power, and multiple nodes collaborate to perform large tasks; S13: Through the Hongmeng OS distributed communication framework, point-to-point data exchange between edge devices is prioritized to reduce dependence on the cloud.

[0027] In this embodiment, based on the distributed soft bus of Hongmeng OS, the node devices automatically discover and complete space perception and connection, as follows: In the Hongmeng OS environment, the distributed soft bus supports the use of broadcast discovery protocol to complete spatial perception of multiple nodes through broadcast access or peripheral scanning; Based on node identifiers (such as MAC addresses, device IDs), each device automatically authenticates and securely joins the group; Perceive the relative distance and topological relationship between nodes (such as distance perception technology: ultrasonic sensors or signal strength evaluation), and generate collaborative communication network topology based on the network quality and hardware capabilities of the device (such as NPU computing power and memory size); Each node contributes its local storage space to the collaborative group: Use the Harmony Distributed File System (HDFS) of Hongmeng OS to achieve file storage distribution and shared access; multiple nodes collaborate to write front-end data collection records (such as power monitoring data such as voltage and current), and dynamically synchronize them to other edge nodes; Cold data and hot data are extracted from different nodes through hierarchical management to improve access speed.

[0028] In the collaborative group, tasks are dynamically divided based on the size of computing tasks and node load conditions: CPU / NPU-intensive tasks (such as real-time assessment of power equipment failures) are preferentially distributed to high-performance nodes; storage-intensive tasks (such as aggregation of historical data for load forecasting) are distributed and managed in nodes with larger storage space; and the distributed soft bus is used to uniformly schedule requests to call on the resources of other idle nodes to perform large-scale analysis tasks.

[0029] In this embodiment, S2 is as follows: Deploy Mesh network technology, and intelligent edge devices automatically join the Mesh network using the broadcast discovery mechanism after they are powered on; Establish network topology and generate adjacency matrix. Edge nodes periodically send node information based on the distributed soft bus of Hongmeng OS, including node ID and cost information. The cost information includes link delay, RSSI (signal strength), load, and remaining bandwidth. The information of directly connected nodes is stored through the adjacency table: A dj (i) = {j∈Nodes | j is directly connected to i} The adjacency matrix is ​​expressed as:

[0030] w ij represents the link weight; Dynamically plan the optimal communication path between nodes based on link quality evaluation; Data is transmitted between devices through multi-hop routing, while node status is detected in real time, and failed nodes are automatically bypassed. After each routing adjustment is completed, the network topology is synchronously updated to the entire network; Based on the optimal communication path between nodes, a point-to-point direct communication link is established without cloud intervention; the routing table is updated regularly and the communication path is adjusted dynamically; Through edge layer collaboration, problems can be quickly detected. If a node failure or temperature abnormality is found, the adjacent node will directly take over the corresponding task, and local problems can be quickly handled through data collaboration and control logic transmission. Through dynamic task distribution and linkage between edge nodes, fast scheduling of distributed logic can be achieved; With the distributed architecture, Hongmeng OS decomposes the computing, data sharing and control logic of the device to the edge layer.

[0031] In this embodiment, based on link quality evaluation (RSSI, bandwidth), the optimal communication path between nodes is dynamically planned, as follows: Periodic monitoring link l ij Quality indicators, including RSSI, bandwidth and load; Update link weight w ij And store it in the adjacency matrix; ; Among them, α, β, γ are weight coefficients; RSSI ij Link l ij Signal strength; BW ij Link l ij The remaining available bandwidth of C j is the current load of the receiving node; Defined in the network graph G=(N,A), find the optimal path P from the source node s to the target node t s→t , so that the total path weight is minimized; N is the set of intelligent edge nodes in the network; A is the set of communication links in the network; Use Dijkstra algorithm to generate the optimal path to the target node in real time based on the dynamic adjacency matrix. Initialize the shortest distance d[i]=∞ of the current node i, except d[s]=0 for i∈N; Update the distance of neighbor nodes: ; Continue iterating unvisited nodes until the target node t is visited; Total path weight:

[0032] When link l ij The RSSI or bandwidth changes, and the weight w is adjusted in real time ij , recalculate path P; if RSSI ij Too low or BW ij = 0, marking the weight as infinity (∞) and triggering replanning.

[0033] In this embodiment, a point-to-point direct communication link is established according to the optimal communication path between nodes without cloud intervention; the routing table is updated regularly and the communication path is dynamically adjusted, as follows: Each node independently maintains a dynamic routing table to record all target nodes and their communication paths. The routing table structure of node i is : ; Among them, NextHop is the next hop node to the target node; Each edge node searches its routing table according to the target node t and obtains the next hop node NextHop: NextHop=R i [t].NextHop; If NextHop is the target node t, then communicate directly; R i [t] is the routing table obtained according to the target node t; Otherwise, the data is forwarded to the next hop node until it reaches the target node; Data transmission is forwarded hop by hop through the nodes in the target path according to the following process: The source node generates path P according to the routing table s→t ; The data packet is forwarded hop by hop according to each node in the path; The target node t receives and processes the data; In the Hongmeng OS environment, its distributed soft bus is used to support direct communication between different edge devices, eliminating intermediate nodes and cloud intervention.

[0034] In this embodiment, the MQTT / CoAP protocol stack is optimized to improve the efficiency and reliability of message transmission in response to the real-time requirements of the power system, as follows: MQTT (Message Queue Telemetry Transport) uses a lightweight version of MQTT-SN to adapt to the situation where edge devices have limited resources: UDP-based transmission (not TCP) is used to avoid complex connection management and improve real-time performance; CoAP (Constrained Application Protocol) uses the Non-confirmable mode to process low-priority messages and reduce handshake delays; optimizes ACK messages and combines UDP's fast timeout retransmission mechanism to ensure reliable transmission of critical data; Use efficient compression to enable CBOR (Concise Binary Object Representation) encoding for MQTT / CoAP message content, replacing traditional JSON to reduce data volume; enable data merging and compression in scenarios with high link load, bundle multiple data for transmission, and reduce message-level overhead; The power system requires communications with different reliability levels, such as low-latency protection signals and high-reliability data collection, and improved QoS strategies: MQTT QoS mechanism (for real-time power scenarios): QoS 0: low priority monitoring signal, used for non-critical monitoring data, using UDP transmission to reduce the overhead of establishing connections; QoS 1: Medium priority power collection signal, optimized retransmission strategy, set dynamic retransmission timeout Tretry(i'), adjust the message retransmission interval according to network delay and packet loss rate:

[0035] in, For the Number of retries, is the basic retransmission time, is the maximum retransmission time; QoS 2: Key protection control signal: With 5G URLLC slicing, delays caused by retransmission are avoided. In CoAP, the timeout assumption model is improved: Dynamically update the estimated value of the timeout RTT (return time) based on the particularity of the collection environment: ; Where a is the weight; R(s') refers to the actual network return time measured at the current time s'; Use EWMA (exponentially weighted moving average) to smooth the measured RTT and reduce unnecessary retries, especially in unstable networks; Integrate a time synchronization module based on PTP (Precision Time Protocol) / NTP (Network Time Protocol) in the protocol stack to ensure strict timestamp accuracy of power protection signals and event messages; Add a priority identifier Tenqueue to the MQTT / CoAP message (such as an additional priority field in the message header) and process it through a dual-queue scheduling mechanism (high / normal priority queue).

[0036] High priority: abnormal information of power grid equipment (such as overload and short circuit signals); Normal priority: monitoring data, diagnostic status; Tenqueue=P(x)+Tarrival(x) Among them, P(x) represents the priority of message x, Tarrival(x) represents the message arrival time, and high priority ensures priority consumption.

[0037] In this embodiment, the network slicing feature of 5G is used to allocate dedicated network resources for different types of power applications to improve communication quality, as follows: According to the different business requirements of the power system, three types of slices are divided: URLLC (Ultra-Reliable Low-Latency Communication) slicing: used for protection and control communications, such as line protection, fault detection, and automation control signals, requiring a latency of less than 1ms; Configure high-priority resource scheduling to ensure bandwidth and reliability (such as preventing data packet loss through HARQ (Hybrid ARQ)); eMBB (enhanced mobile broadband) slice: used to transmit data that requires high bandwidth, such as power status monitoring and video monitoring; The configuration throughput is greater than the threshold and is adapted to the intelligent monitoring system to transmit big data waveforms, event records or analysis results; mMTC (massive machine type communication) slicing is used to collect data from wide-area power terminal devices (such as electricity meters and distribution network equipment) to optimize terminal energy consumption and signaling overhead.

[0038] Network slice resource allocation is based on the actual load demand of the power system and is adjusted dynamically: Slice resource allocation function: Slice resource Rs is allocated by the ratio Decisions are made and dynamically adjusted according to task priority and latency requirements: ; Among them, P s is the slice priority; Q s Real-time load for slicing; R total is the total network bandwidth; For all n' The weighted sum of the priority and real-time load factor of each slice. The subscript i'' is used to traverse all slices. Based on the real-time monitoring of the power system, the network slice orchestrator is used to dynamically adjust the weight of each slice, for example: When an accident occurs, the weight of the URLLC slice protecting the communication will increase significantly; In normal monitoring scenarios, eMBB and mMTC slices are allocated more resources.

[0039] The QoS flow mechanism of 5G is utilized, combined with the priority of MQTT / CoAP messages, to provide the most suitable slices for power communications of different priorities.

[0040] Configure 5QI (5G QoS Identifier) ​​to define latency and reliability targets for each flow: The high priority field matches the URLLC; The medium and low priority fields match eMBB or mMTC.

[0041] In this embodiment, S4 is specifically: The power system's identity authentication and permission management are designed based on the zero-trust principle. A unique identifier is configured for each edge node, and a PKI public key system is used to establish secure communication between devices. Real-time identity authentication is implemented when nodes are connected, accessed or communicated to continuously verify the reliability of the equipment.

[0042] Introduce blockchain in data exchange and exception handling: Use blockchain to record device behavior logs (such as operating status or instructions) to ensure that data cannot be tampered with; provide on-chain evidence storage capabilities to achieve data auditing and tracking.

[0043] Use lightweight AES encryption algorithm to encrypt all edge node communication data; UDP-based TLS protects CoAP transmission.

[0044] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0046] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0047] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0048] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. The power edge cloud collaborative management method based on Hongmeng OS is characterized by: The following steps are included S1: Deploy intelligent edge nodes, set up embedded AI acceleration units, and run Hongmeng OS on all intelligent edge nodes, using its distributed characteristics to achieve seamless collaboration of devices; S2: Use Mesh network technology to achieve self-organizing connection of edge devices and design direct communication paths between edge devices to achieve rapid processing of local problems; S3: Optimize the MQTT / CoAP protocol stack to meet the real-time requirements of the power system, and use the network slicing feature of 5G to allocate dedicated network resources for different types of power applications; S4: Build a security architecture based on the zero-trust principle, enhance device access control and data transmission security, use blockchain technology for identity authentication and data integrity protection, and prevent data tampering and forgery.

2. The power edge cloud collaborative management method based on Hongmeng OS according to claim 1 is characterized in that: The S1 is specifically as follows: S11: Each intelligent edge node runs Hongmeng OS and acts as an intelligent terminal, with edge storage, computing power, distributed service capabilities, and supports AI reasoning. Through the broadcast discovery mechanism of the soft bus in the Hongmeng OS distributed architecture, dynamic networking between edge nodes is completed, a collaborative working environment is established, and storage, computing and task management resource sharing of distributed devices is realized. Through the Hongmeng OS distributed communication framework, direct communication between edge nodes is prioritized, and task linkage management is carried out in conjunction with the cloud: S12: Based on the distributed soft bus of Hongmeng OS, node devices automatically discover and complete spatial perception and connection; edge node resources are shared, including storage and computing power, and multiple nodes collaborate to perform large tasks; S13: Through the Hongmeng OS distributed communication framework, point-to-point data exchange between edge devices is prioritized to reduce dependence on the cloud.

3. The power edge cloud collaborative management method based on Hongmeng OS according to claim 2 is characterized in that: Based on the distributed soft bus of Hongmeng OS, node devices automatically discover and complete spatial perception and connection, as follows: In the Hongmeng OS environment, the distributed soft bus supports the use of broadcast discovery protocol to complete spatial perception of multiple nodes through broadcast access or peripheral scanning; Based on the node identifier, each device automatically authenticates and securely joins the group; Perceive the relative distance and topological relationship between nodes, and generate collaborative communication network topology based on the network quality and hardware capabilities of the devices; Each node contributes its local storage space to the collaborative group: Use Hongmeng OS's distributed file system to achieve file storage distribution and shared access; multiple nodes collaborate to write front-end data collection records and dynamically synchronize them to other edge nodes; Cold data and hot data are extracted from different nodes through hierarchical management to improve access speed; In the collaborative group, tasks are dynamically divided based on the size of computing tasks and node load conditions: CPU / NPU-intensive tasks are preferentially distributed to high-performance nodes; storage-intensive tasks are distributed and managed in nodes with larger storage space; and the distributed soft bus is used to uniformly schedule requests to call the resources of other idle nodes to execute large-scale analysis tasks.

4. The power edge cloud collaborative management method based on Hongmeng OS according to claim 1 is characterized in that: The S2 is specifically as follows: Deploy Mesh network technology, and intelligent edge devices automatically join the Mesh network using the broadcast discovery mechanism after they are powered on; Establish network topology and generate adjacency matrix. Edge nodes periodically send node information based on the distributed soft bus of Hongmeng OS, including node ID and cost information. The cost information includes link delay, RSSI, load, and remaining bandwidth. The information of directly connected nodes is stored through the adjacency table: A dj (i) = {j∈Nodes | j is directly connected to i} The adjacency matrix is ​​expressed as: ； w ij represents the link weight; Dynamically plan the optimal communication path between nodes based on link quality evaluation; Data is transmitted between devices through multi-hop routing, while node status is detected in real time, and failed nodes are automatically bypassed. After each routing adjustment is completed, the network topology is synchronously updated to the entire network; Based on the optimal communication path between nodes, a point-to-point direct communication link is established without cloud intervention; the routing table is updated regularly and the communication path is adjusted dynamically; Through edge layer collaboration, problems can be quickly detected. If a node failure or temperature abnormality is found, the adjacent node will directly take over the corresponding task, and local problems can be quickly handled through data collaboration and control logic transmission. Through dynamic task distribution and linkage between edge nodes, fast scheduling of distributed logic can be achieved; With the distributed architecture, Hongmeng OS decomposes the computing, data sharing and control logic of the device to the edge layer.

5. The power edge cloud collaborative management method based on Hongmeng OS according to claim 4 is characterized in that: The optimal communication path between nodes is dynamically planned based on link quality evaluation, as follows: Periodic monitoring link l ij Quality indicators, including RSSI, bandwidth and load; Update link weight w ij And store it in the adjacency matrix; ； Among them, α, β, γ are weight coefficients; RSSI ij Link l ij Signal strength; BW ij Link l ij The remaining available bandwidth of C j is the current load of the receiving node; Defined in the network graph G=(N,A), find the optimal path P from the source node s to the target node t s→t , so that the total path weight is minimized; N is the set of intelligent edge nodes in the network; A is the set of communication links in the network; Use Dijkstra algorithm to generate the optimal path to the target node in real time based on the dynamic adjacency matrix. Initialize the shortest distance d[i]=∞ of the current node i, except d[s]=0 for i∈N; Update the distance of neighbor nodes: ； Continue iterating unvisited nodes until the target node t is visited; Total path weight: ； When link l ij The RSSI or bandwidth changes, and the weight w is adjusted in real time ij , recalculate path P; if RSSI ij Too low or BW ij = 0, marking the weight as infinity (∞) and triggering replanning.

6. The power edge cloud collaborative management method based on Hongmeng OS according to claim 4 is characterized in that: According to the optimal communication path between nodes, a point-to-point direct communication link is established without cloud intervention; the routing table is updated regularly and the communication path is dynamically adjusted, as follows: Each node independently maintains a dynamic routing table to record all target nodes and their communication paths. The routing table structure of node i is : ； Among them, NextHop is the next hop node to the target node; R i [t] is the routing table obtained according to the target node t; Each edge node searches its routing table according to the target node t and obtains the next hop node NextHop: NextHop=R i [t].NextHop; If NextHop is the target node t, then communication is direct; Otherwise, the data is forwarded to the next hop node until it reaches the target node; Data transmission is forwarded hop by hop through the nodes in the target path according to the following process: The source node generates path P according to the routing table s→t ; The data packet is forwarded hop by hop according to each node in the path; The target node t receives and processes the data; In the Hongmeng OS environment, its distributed soft bus is used to support direct communication between different edge devices, eliminating intermediate nodes and cloud intervention.

7. The power edge cloud collaborative management method based on Hongmeng OS according to claim 1 is characterized in that: The MQTT / CoAP protocol stack is optimized for the real-time requirements of the power system, as follows: MQTT uses a lightweight version of MQTT-SN to adapt to the situation where edge devices have limited resources: UDP-based transmission is used to avoid complex connection management and improve real-time performance; CoAP uses the Non-confirmable mode to process low-priority messages and reduce handshake delays. It optimizes ACK messages and combines UDP's fast timeout retransmission mechanism to ensure reliable transmission of critical data. Use efficient compression to enable CBOR encoding for MQTT / CoAP message content, replacing traditional JSON to reduce data volume; enable data merging and compression in scenarios with high link load, bundle multiple data for transmission, and reduce message-level overhead; Improved QoS policy: MQTT QoS mechanism: QoS 0: low priority monitoring signal, used for non-critical monitoring data, using UDP transmission to reduce the overhead of establishing connections; QoS 1: Medium priority power collection signal, optimized retransmission strategy, set dynamic retransmission timeout Tretry(i'), adjust the message retransmission interval according to network delay and packet loss rate: ； in, For the Number of retries, is the basic retransmission time, is the maximum retransmission time; QoS 2: Key protection control signal: With 5G URLLC slicing, delays caused by retransmission are avoided. In CoAP, the timeout assumption model is improved: Dynamically update the estimated timeout RTT value based on the particularity of the collection environment: ； Where a is the weight; R(s') refers to the actual network return time measured at the current time s'; Use EWMA to smoothly measure RTT and reduce unnecessary retries, especially in unstable networks; Integrate a PTP / NTP-based time synchronization module into the protocol stack to ensure strict timestamp accuracy for power protection signals and event messages; Add priority identifiers to MQTT / CoAP messages and process them through a dual-queue scheduling mechanism.

8. The power edge cloud collaborative management method based on Hongmeng OS according to claim 7 is characterized in that: The network slicing feature of 5G is used to allocate dedicated network resources for different types of power applications, as follows: According to the different business requirements of the power system, three types of slices are divided: URLLC slice: used for protection and control communications, including line protection, fault detection, and automation control signals; Configure high-priority resource scheduling and prevent data packet loss through HARQ; eMBB slice: used to transmit data that requires high bandwidth, including power status monitoring and video surveillance data; The configuration throughput is greater than the threshold and is adapted to the intelligent monitoring system to transmit big data waveforms, event records or analysis results; mMTC slicing collects data from wide-area power terminal devices, optimizing terminal energy consumption and signaling overhead; Network slice resource allocation is based on the actual load demand of the power system and is adjusted dynamically: Slice resource allocation function: Slice resource Rs is allocated by the ratio Decisions are made and dynamically adjusted according to task priority and latency requirements: ； Among them, P s is the slice priority; Q s Real-time load for slicing; R total is the total network bandwidth; is the weighted sum of the priorities and real-time load factors of all n slices; Based on the real-time monitoring of the power system, the network slice orchestrator is used to dynamically adjust the weight of each slice. The QoS flow mechanism of 5G is utilized, combined with the priority of MQTT / CoAP messages, to provide the most suitable slices for power communications of different priorities.

9. The power edge cloud collaborative management method based on Hongmeng OS according to claim 1 is characterized in that: The S4 is specifically: Design the identity authentication and permission management of the power system according to the zero-trust principle, configure a unique identifier for each edge node, and use the PKI public key system to establish secure communication between devices; implement real-time identity authentication when nodes are connected, accessed or communicated, and continuously verify the reliability of the equipment; Introducing blockchain in data exchange and exception handling: Using blockchain to record device behavior logs to ensure that data cannot be tampered with; providing on-chain evidence storage capabilities to achieve data auditing and tracking; Use lightweight AES encryption algorithm to encrypt all edge node communication data; UDP-based TLS protects CoAP transmission.

10. A power edge cloud collaborative management system based on Hongmeng OS, characterized in that: It includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically executes the steps in a power edge cloud collaborative management method based on Hongmeng OS as described in any one of claims 1 to 9.

Images