Power distribution IoT intelligent gateway cloud-edge collaborative multi-node control method and system under weak network

Abstract

The present invention discloses a method and system for cloud-edge collaborative multi-node management and control of a distribution Internet of Things intelligent gateway under a weak network, which relates to the field of cloud-edge collaborative technology, including: collecting operating data of power equipment, dividing it into real-time data and non-real-time data, generating real-time data tasks and non-real-time data tasks, evaluating the computing power performance requirements of real-time data tasks and non-real-time data tasks, obtaining multiple edge nodes connected to the distribution Internet of Things intelligent gateway, evaluating the performance redundancy value and network performance index of multiple edge nodes connected to the distribution Internet of Things intelligent gateway, establishing multiple transmission protocols under different network connection states, dividing the network connection state level according to the network performance index of the edge node, and adapting the corresponding transmission protocol, establishing a cloud-edge collaborative multi-point management and control model, and generating a weak network cloud-edge collaborative multi-node management and control scheme; the advantages of the present invention are: reducing the risk of data loss and improving the stability of the distribution network.

Description

Technical Field

[0001] The present invention relates to the field of cloud-edge collaborative technology, and specifically to a cloud-edge collaborative multi-node management and control method and system for a power distribution IoT intelligent gateway under a weak network. Background Art

[0002] A weak network refers to a network environment with problems such as insufficient bandwidth, high latency, high packet loss rate, unstable connection, network congestion, weak signal or improper configuration. These problems may lead to degraded network performance or unstable communications, affecting data transmission and application service quality.

[0003] Due to bandwidth limitations, high latency, high packet loss rate and unstable connection factors in a weak network environment, the cloud-edge collaborative multi-nodes of the distribution IoT intelligent gateway have low resource allocation efficiency and slow data processing speed, resulting in poor reliability and stability of cloud-edge collaborative management and control. Summary of the invention

[0004] In order to solve the above-mentioned technical problems, a method and system for cloud-edge collaborative multi-node management and control of a distribution Internet of Things intelligent gateway under a weak network is provided. This technical solution solves the above-mentioned problem that due to bandwidth limitations, high latency, high packet loss rate and unstable connection in a weak network environment, the cloud-edge collaborative multi-nodes of the distribution Internet of Things intelligent gateway have low resource allocation efficiency and slow data processing speed, resulting in poor reliability and stability of cloud-edge collaborative management and control.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] The cloud-edge collaborative multi-node control method of the power distribution IoT intelligent gateway under weak network includes:

[0007] Based on the power distribution IoT smart gateway, collect the operation data of power equipment;

[0008] The operation data of the power equipment is divided into real-time data and non-real-time data, and real-time data tasks and non-real-time data tasks are generated;

[0009] Evaluate the computing power performance requirements of real-time data tasks and non-real-time data tasks;

[0010] Get multiple edge nodes connected to the power distribution IoT smart gateway;

[0011] Evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the performance redundancy value of each edge node;

[0012] Evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the network performance index of each edge node;

[0013] Establish multiple transmission protocols under different network connection states;

[0014] The network connection status is classified according to the network performance index of the edge node, and the corresponding transmission protocol is adapted;

[0015] Establish a cloud-edge collaborative multi-point control model;

[0016] Based on the cloud-edge collaborative multi-point control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, the computing power performance requirements of the real-time data task are taken as input, and the edge nodes whose performance redundancy values ​​meet the computing power performance requirements of the real-time data task are taken as output. The computing power performance requirements of non-real-time data tasks are taken as input, and the edge nodes whose network performance index meets the computing power performance requirements of non-real-time data tasks are taken as output to generate a weak network cloud-edge collaborative multi-node control solution.

[0017] Among them, the cloud-edge collaborative multi-point control model expression is specifically:

[0018]

[0019] Where P is a weak network cloud-edge collaborative multi-node control solution, NPI' e is the standardized network performance index of the e-th edge node, f() is the transmission protocol mapping function, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, R e is the comprehensive performance redundancy value of the e-th edge node, and L is the total number of edge nodes.

[0020] Preferably, the computing power performance requirements for evaluating real-time data tasks and non-real-time data tasks specifically include:

[0021] Determine the central core parameters of edge nodes;

[0022] The real-time data tasks and non-real-time data tasks are quantitatively decomposed to obtain the total data volume of the real-time data tasks and non-real-time data tasks; the total data volume refers to the smallest data unit bit.

[0023] Determine the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks;

[0024] Calculate the central core processing time of processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task;

[0025] Calculate the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node;

[0026] Based on the total data volume of real-time data tasks and non-real-time data tasks, the running memory requirements of real-time data tasks and non-real-time data tasks, the central processing time for processing real-time data tasks and non-real-time data tasks, and the total number of central cores required, the computing power performance requirements of real-time data tasks and non-real-time data tasks are calculated through the computing power formula;

[0027] The specific calculation power formula is:

[0028]

[0029] In the formula, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, D 1 is the total data volume of the real-time data task, D 2 is the total data volume of non-real-time data tasks, H 1 is the memory requirement for real-time data tasks, H 2 is the running memory requirement of the non-real-time data task, Ct is the response time of executing the task, Central processing time for real-time data processing tasks, The central processing time for processing non-real-time data tasks, C core is the desired total number of central cores.

[0030] Preferably, evaluating the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway and obtaining the performance redundancy value of each edge node specifically includes:

[0031] Determine the ideal hardware performance index of the edge node based on the edge node construction parameters; the ideal hardware performance index of the edge node includes at least: CPU utilization, memory utilization and storage I / O utilization;

[0032] According to the real-time operation data of the edge nodes, collect the hardware utilization rate of the edge nodes and determine the real-time performance indicators of the hardware of the edge nodes;

[0033] Based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node, the comprehensive performance redundancy value of the edge node is calculated through the performance analysis algorithm;

[0034] The performance analysis algorithm is as follows:

[0035]

[0036] In the formula, R is the comprehensive performance redundancy value of the edge node, U j ' is the ideal performance index of the jth hardware of the edge node, U j is the jth hardware real-time performance indicator of the edge node.

[0037] Preferably, evaluating the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway and obtaining the network performance index of each edge node specifically includes:

[0038] Actively upload data packets for all edge nodes to test, and collect several network connection parameters of edge nodes under the upload data packet test; the network connection parameters include but are not limited to: bandwidth, delay (ping), packet loss rate and jitter;

[0039] Based on the measured values ​​of several network connection parameters of the edge node, score each network connection parameter to obtain several network connection parameter scores of the edge node;

[0040] Based on the importance of the network connection status corresponding to the measured values ​​of several network connection parameters of the edge node, a weight is assigned to each network connection parameter to obtain several network connection parameter weights of the edge node;

[0041] According to the scores of several network connection parameters of the edge node and the weights of several network connection parameters of the edge node, the network performance index of the edge node is calculated through a weighted formula;

[0042] Using a standardized formula, the network performance index of the edge node is standardized to obtain a standardized network performance index of the edge node;

[0043] The specific weighted formula is:

[0044]

[0045] Where NPI is the network performance index of the edge node, S k Score the kth network connection parameter of the edge node, W k is the kth network connection parameter weight of the edge node;

[0046] The specific standardized formula is:

[0047]

[0048] Where NPI' is the network performance index of the standardized edge node, NPI max is the maximum value of the network performance index of the edge node, NPI min is the minimum network performance index of the edge node.

[0049] Furthermore, a power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under a weak network is proposed, which is used to implement the power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control method under a weak network as described above, including:

[0050] Data segmentation module: The data segmentation module is used to collect the operation data of power equipment based on the power distribution IoT intelligent gateway, and segment it into real-time data and non-real-time data, and generate real-time data tasks and non-real-time data tasks;

[0051] A computing power demand assessment module, the computing power demand assessment module is electrically connected to the data segmentation module, and the computing power demand assessment module is used to assess the computing power performance requirements of real-time data tasks and non-real-time data tasks;

[0052] An edge node acquisition module, which is used to acquire multiple edge nodes connected to the power distribution IoT smart gateway;

[0053] A performance status evaluation module, which is electrically connected to the edge node acquisition module, and is used to evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the performance redundancy value of each edge node;

[0054] A network status evaluation module, the network status evaluation module is electrically connected to the edge node acquisition module, and the network status evaluation module is used to evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the network performance index of each edge node;

[0055] A multi-transmission protocol module is used to establish transmission protocols under multiple different network connection states;

[0056] A transmission protocol adaptation module, the transmission protocol adaptation module is electrically connected to the multi-transmission protocol module, and the transmission protocol adaptation module is used to classify the network connection status according to the network performance index of the edge node and adapt the corresponding transmission protocol;

[0057] Model building module: The model building module is used to establish a cloud-edge collaborative multi-point management and control model;

[0058] The model management and control module is electrically connected to the model construction module, the computing power requirement evaluation module, the performance status evaluation module and the network status evaluation module. Based on the cloud-edge collaborative multi-point management and control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, and the computing power performance requirement of the real-time data task is used as the input, and the edge node whose performance redundancy value meets the computing power performance requirement of the real-time data task is used as the output; the computing power performance requirement of the non-real-time data task is used as the input, and the edge node whose network performance index meets the computing power performance requirement of the non-real-time data task is used as the output, to generate a weak network cloud-edge collaborative multi-node management and control solution.

[0059] Optionally, the computing power demand assessment module includes:

[0060] A core parameter unit, which determines the central core parameters of edge nodes;

[0061] The task quantification unit quantifies and decomposes the real-time data tasks and the non-real-time data tasks to obtain the total data volume of the real-time data tasks and the non-real-time data tasks;

[0062] A memory requirement unit, which determines the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks;

[0063] A processing time unit, which calculates the central core processing time for processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task;

[0064] The core quantity unit calculates the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node;

[0065] The first computing unit calculates the computing power performance requirements of the real-time data tasks and the non-real-time data tasks through a computing power formula based on the total data volume of the real-time data tasks and the non-real-time data tasks, the running memory requirements of the real-time data tasks and the non-real-time data tasks, the central processing time for processing the real-time data tasks and the non-real-time data tasks, and the total central core quantity requirements.

[0066] Optionally, the performance status assessment module includes:

[0067] The ideal performance unit determines the ideal performance indicators of the edge node hardware based on the edge node construction parameters;

[0068] The real-time performance unit collects the hardware utilization rate of the edge node according to the real-time operation data of the edge node and determines the hardware real-time performance index of the edge node;

[0069] The second calculation unit calculates the comprehensive performance redundancy value of the edge node through a performance analysis algorithm based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node.

[0070] Optionally, the performance status assessment module includes:

[0071] The network connection parameter collection unit performs active data packet upload test on all edge nodes and collects several network connection parameters of the edge nodes under the upload data packet test;

[0072] A parameter evaluation unit, based on the measured values ​​of the network connection parameters of the edge node, scores each network connection parameter to obtain the network connection parameter scores of the edge node;

[0073] A parameter weight unit assigns a weight to each network connection parameter based on the importance of the network connection status corresponding to the measured values ​​of the network connection parameters of the edge node, thereby obtaining the weights of the network connection parameters of the edge node;

[0074] A third calculation unit calculates a network performance index of the edge node through a weighted formula according to the scores of the network connection parameters of the edge node and the weights of the network connection parameters of the edge node;

[0075] The standardization unit uses a standardization formula to standardize the network performance index of the edge node to obtain the network performance index of the standardized edge node.

[0076] Compared with the prior art, the present invention has the following beneficial effects:

[0077] The present invention proposes a cloud-edge collaborative multi-node management and control scheme for a distribution Internet of Things intelligent gateway under a weak network. By constructing a cloud-edge collaborative multi-point management and control model, the computing power performance requirements of real-time data tasks and non-real-time data tasks are determined. The computing power performance requirements of real-time data tasks and non-real-time data tasks are used as restriction conditions. The performance redundancy value of the edge node that meets the computing power performance requirements of the real-time data task is uploaded as the proximal node for data processing, and the network performance index of the edge node that meets the computing power performance requirements of the non-real-time data task is uploaded as the remote node for data processing, thereby reducing the risk of data loss and improving the stability of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is a flow chart of the cloud-edge collaborative multi-node control method of the power distribution IoT intelligent gateway under weak network;

[0079] Figure 2 A flow chart of the method for evaluating the computing power performance requirements of real-time data tasks and non-real-time data tasks;

[0080] Figure 3 A flow chart of a method for obtaining the performance redundancy value of each edge node;

[0081] Figure 4 A flow chart of a method for obtaining a network performance index of each edge node;

[0082] Figure 5 This is the framework diagram of the cloud-edge collaborative multi-node management and control system of the power distribution IoT intelligent gateway under weak network. DETAILED DESCRIPTION

[0083] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0084] Reference Figure 1 As shown in the figure, the cloud-edge collaborative multi-node control method of the power distribution IoT intelligent gateway under weak network includes:

[0085] Based on the power distribution IoT smart gateway, collect the operation data of power equipment;

[0086] The operation data of the power equipment is divided into real-time data and non-real-time data, and real-time data tasks and non-real-time data tasks are generated;

[0087] Evaluate the computing power performance requirements of real-time data tasks and non-real-time data tasks;

[0088] Get multiple edge nodes connected to the power distribution IoT smart gateway;

[0089] Evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the performance redundancy value of each edge node;

[0090] Evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the network performance index of each edge node;

[0091] Establish multiple transmission protocols under different network connection states;

[0092] The network connection status is graded according to the network performance index of the edge node, and the corresponding transmission protocol is adapted;

[0093] Establish a cloud-edge collaborative multi-point control model;

[0094] Based on the cloud-edge collaborative multi-point control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, the computing power performance requirements of the real-time data task are taken as input, and the edge nodes whose performance redundancy values ​​meet the computing power performance requirements of the real-time data task are taken as output. The computing power performance requirements of non-real-time data tasks are taken as input, and the edge nodes whose network performance index meets the computing power performance requirements of non-real-time data tasks are taken as output to generate a weak network cloud-edge collaborative multi-node control solution.

[0095] Among them, the cloud-edge collaborative multi-point control model expression is specifically:

[0096]

[0097] Where P is a weak network cloud-edge collaborative multi-node control solution, NPI' e is the standardized network performance index of the e-th edge node, f() is the transmission protocol mapping function, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, R e is the comprehensive performance redundancy value of the e-th edge node, and L is the total number of edge nodes.

[0098] This solution constructs a cloud-edge collaborative multi-point management and control model, and allocates the computing power performance requirements of real-time data to edge nodes whose proximal performance redundancy values ​​meet the computing power performance requirements to ensure the normal operation of the distribution IoT intelligent gateway under weak networks. Secondly, the edge nodes that meet the computing power performance requirements of non-real-time data are uploaded to the cloud according to the network performance index, thereby improving the transmission efficiency of non-real-time data under weak networks, realizing cloud-edge collaborative multi-node management of distribution IoT intelligent gateways under weak networks, reducing the risk of data loss, and improving the stability of the distribution network.

[0099] It should be noted that the network connection status is divided into levels according to the network performance index of the edge node, and the corresponding transmission protocol is adapted. For example, the network connection status is divided into three levels according to the network performance index, namely excellent, good and fair. TCP, UDP and RTP transmission protocols are created in advance based on the three levels. When the network performance index of the edge node rises or falls to the corresponding level, the transmission protocol is adaptively changed to ensure the quality and reliability of the transmitted data and reduce the loss of data packets.

[0100] Reference Figure 2 As shown in the figure, the computing power performance requirements for evaluating real-time data tasks and non-real-time data tasks specifically include:

[0101] Determine the central core parameters of edge nodes;

[0102] The real-time data tasks and non-real-time data tasks are quantitatively decomposed to obtain the total data volume of the real-time data tasks and non-real-time data tasks; the total data volume refers to the smallest data unit bit.

[0103] Determine the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks;

[0104] Calculate the central core processing time of processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task;

[0105] Calculate the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node;

[0106] Based on the total data volume of real-time data tasks and non-real-time data tasks, the running memory requirements of real-time data tasks and non-real-time data tasks, the central processing time for processing real-time data tasks and non-real-time data tasks, and the total number of central cores required, the computing power performance requirements of real-time data tasks and non-real-time data tasks are calculated through the computing power formula;

[0107] The specific calculation power formula is:

[0108]

[0109] In the formula, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, D 1 is the total data volume of the real-time data task, D 2 is the total data volume of non-real-time data tasks, H 1 is the memory requirement for real-time data tasks, H 2 is the running memory requirement of the non-real-time data task, Ct is the response time of executing the task, Central processing time for real-time data processing tasks, The central processing time for processing non-real-time data tasks, C core is the desired total number of central cores.

[0110] This solution quantitatively decomposes the tasks, determines the minimum unit of data volume (bit), and then evaluates the operating memory requirements. Based on the central core parameters of the edge node, it calculates the processing time and the number of central cores required. Finally, through the computing power formula, the amount of data, memory requirements, processing time and number of cores are comprehensively considered to calculate the computing power performance requirements for real-time and non-real-time data tasks. Through precise quantification and calculation, the computing power performance requirements of real-time and non-real-time data processing tasks in the edge computing environment provide a data basis for subsequent edge node decisions.

[0111] Reference Figure 3 As shown in the figure, the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway is evaluated, and the performance redundancy value of each edge node is obtained, which specifically includes:

[0112] Determine the ideal hardware performance index of the edge node based on the edge node construction parameters; the ideal hardware performance index of the edge node includes at least: CPU utilization, memory utilization and storage I / O utilization;

[0113] According to the real-time operation data of the edge nodes, collect the hardware utilization rate of the edge nodes and determine the real-time performance indicators of the hardware of the edge nodes;

[0114] Based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node, the comprehensive performance redundancy value of the edge node is calculated through the performance analysis algorithm;

[0115] The performance analysis algorithm is as follows:

[0116]

[0117] In the formula, R is the comprehensive performance redundancy value of the edge node, U j ' is the ideal performance index of the jth hardware of the edge node, U j is the jth hardware real-time performance indicator of the edge node.

[0118] It should be noted that the comprehensive performance redundancy value of the edge node refers to the remaining computing power performance under the unfinished tasks of the edge node, and the redundancy value is well known to those skilled in the art and will not be described in detail here.

[0119] Reference Figure 4 As shown in the figure, the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway is evaluated, and the network performance index of each edge node is obtained, which specifically includes:

[0120] Actively upload data packets for all edge nodes to test, and collect several network connection parameters of edge nodes under the upload data packet test; the network connection parameters include but are not limited to: bandwidth, delay (ping), packet loss rate and jitter;

[0121] Based on the measured values ​​of several network connection parameters of the edge node, score each network connection parameter to obtain several network connection parameter scores of the edge node;

[0122] Based on the importance of the network connection status corresponding to the measured values ​​of several network connection parameters of the edge node, a weight is assigned to each network connection parameter to obtain several network connection parameter weights of the edge node;

[0123] According to the scores of several network connection parameters of the edge node and the weights of several network connection parameters of the edge node, the network performance index of the edge node is calculated through a weighted formula;

[0124] Using a standardized formula, the network performance index of the edge node is standardized to obtain a standardized network performance index of the edge node;

[0125] The specific weighted formula is:

[0126]

[0127] Where NPI is the network performance index of the edge node, S k Score the kth network connection parameter of the edge node, W k is the kth network connection parameter weight of the edge node;

[0128] The specific standardized formula is:

[0129]

[0130] Where NPI' is the network performance index of the standardized edge node, NPI max is the maximum value of the network performance index of the edge node, NPI min is the minimum network performance index of the edge node.

[0131] It is understandable that due to the volatility of the network connection status, the corresponding network connection parameter measurement values ​​are also different. Therefore, the network performance index of each edge node is standardized so that it is on the same dimension to ensure the accuracy of subsequent data analysis.

[0132] Reference Figure 5 As shown, based on the same inventive concept of the power distribution IoT intelligent gateway cloud-edge collaborative multi-node control method under a weak network, a power distribution IoT intelligent gateway cloud-edge collaborative multi-node control system under a weak network is proposed, including:

[0133] Data segmentation module: The data segmentation module is used to collect the operation data of power equipment based on the power distribution IoT intelligent gateway, and segment it into real-time data and non-real-time data, and generate real-time data tasks and non-real-time data tasks;

[0134] A computing power demand assessment module, the computing power demand assessment module is electrically connected to the data segmentation module, and the computing power demand assessment module is used to assess the computing power performance requirements of real-time data tasks and non-real-time data tasks;

[0135] An edge node acquisition module, which is used to acquire multiple edge nodes connected to the power distribution IoT smart gateway;

[0136] A performance status evaluation module, which is electrically connected to the edge node acquisition module, and is used to evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the performance redundancy value of each edge node;

[0137] A network status evaluation module, the network status evaluation module is electrically connected to the edge node acquisition module, and the network status evaluation module is used to evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the network performance index of each edge node;

[0138] A multi-transmission protocol module is used to establish transmission protocols under multiple different network connection states;

[0139] A transmission protocol adaptation module, the transmission protocol adaptation module is electrically connected to the multi-transmission protocol module, and the transmission protocol adaptation module is used to classify the network connection status according to the network performance index of the edge node and adapt the corresponding transmission protocol;

[0140] Model building module: The model building module is used to establish a cloud-edge collaborative multi-point management and control model;

[0141] The model management and control module is electrically connected to the model construction module, the computing power requirement evaluation module, the performance status evaluation module and the network status evaluation module. Based on the cloud-edge collaborative multi-point management and control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, and the computing power performance requirement of the real-time data task is used as the input, and the edge node whose performance redundancy value meets the computing power performance requirement of the real-time data task is used as the output; the computing power performance requirement of the non-real-time data task is used as the input, and the edge node whose network performance index meets the computing power performance requirement of the non-real-time data task is used as the output, to generate a weak network cloud-edge collaborative multi-node management and control solution.

[0142] The computing power demand assessment module includes:

[0143] A core parameter unit, which determines the central core parameters of edge nodes;

[0144] The task quantification unit quantifies and decomposes the real-time data tasks and the non-real-time data tasks to obtain the total data volume of the real-time data tasks and the non-real-time data tasks;

[0145] A memory requirement unit, which determines the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks;

[0146] A processing time unit, which calculates the central core processing time for processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task;

[0147] The core quantity unit calculates the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node;

[0148] The first computing unit calculates the computing power performance requirements of the real-time data tasks and the non-real-time data tasks through a computing power formula based on the total data volume of the real-time data tasks and the non-real-time data tasks, the running memory requirements of the real-time data tasks and the non-real-time data tasks, the central processing time for processing the real-time data tasks and the non-real-time data tasks, and the total central core quantity requirements.

[0149] The performance status assessment module includes:

[0150] The ideal performance unit determines the ideal performance indicators of the edge node hardware based on the edge node construction parameters;

[0151] The real-time performance unit collects the hardware utilization rate of the edge node according to the real-time operation data of the edge node and determines the hardware real-time performance index of the edge node;

[0152] The second calculation unit calculates the comprehensive performance redundancy value of the edge node through a performance analysis algorithm based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node.

[0153] The performance status assessment module includes:

[0154] The network connection parameter collection unit performs active data packet upload test on all edge nodes and collects several network connection parameters of the edge nodes under the upload data packet test;

[0155] A parameter evaluation unit, based on the measured values ​​of the network connection parameters of the edge node, scores each network connection parameter to obtain the network connection parameter scores of the edge node;

[0156] A parameter weight unit assigns a weight to each network connection parameter based on the importance of the network connection status corresponding to the measured values ​​of the network connection parameters of the edge node, thereby obtaining the weights of the network connection parameters of the edge node;

[0157] A third calculation unit calculates a network performance index of the edge node through a weighted formula according to the scores of the network connection parameters of the edge node and the weights of the network connection parameters of the edge node;

[0158] The standardization unit uses a standardization formula to standardize the network performance index of the edge node to obtain the network performance index of the standardized edge node.

[0159] The use process of the power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under weak network is as follows:

[0160] Step 1: Based on the power distribution IoT smart gateway, collect the operation data of power equipment, divide it into real-time data and non-real-time data, and generate real-time data tasks and non-real-time data tasks;

[0161] Step 2: Determine the central core parameters of the edge nodes;

[0162] Step 3: Quantitatively decompose the real-time data tasks and non-real-time data tasks to obtain the total data volume of the real-time data tasks and non-real-time data tasks;

[0163] Step 4: Based on the total data volume of the real-time data tasks and the non-real-time data tasks, determine the running memory requirements of the real-time data tasks and the non-real-time data tasks;

[0164] Step 5: Calculate the central core processing time for processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task;

[0165] Step 6: Calculate the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node;

[0166] Step 7: Based on the total data volume of the real-time data tasks and the non-real-time data tasks, the running memory requirements of the real-time data tasks and the non-real-time data tasks, the central processing time for processing the real-time data tasks and the non-real-time data tasks, and the total number of central cores required, the computing power performance requirements of the real-time data tasks and the non-real-time data tasks are calculated through the computing power formula;

[0167] Step 8: Obtain multiple edge nodes connected to the power distribution IoT smart gateway;

[0168] Step 9: Determine the ideal performance indicators of the edge node hardware based on the edge node construction parameters;

[0169] Step 10: According to the real-time operation data of the edge node, collect the hardware utilization rate of the edge node and determine the real-time performance index of the hardware of the edge node;

[0170] Step 11: Based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node, the comprehensive performance redundancy value of the edge node is calculated through a performance analysis algorithm;

[0171] Step 12: Perform an active data packet upload test on all edge nodes, and collect several network connection parameters of the edge nodes under the upload data packet test;

[0172] Step 13: Based on the measured values ​​of the network connection parameters of the edge node, score each network connection parameter to obtain the scores of the network connection parameters of the edge node;

[0173] Step 14: Based on the importance of the network connection status corresponding to the measured values ​​of the network connection parameters of the edge node, a weight is assigned to each network connection parameter to obtain the weights of the network connection parameters of the edge node;

[0174] Step 15: Calculate the network performance index of the edge node through a weighted formula according to the scores of the network connection parameters of the edge node and the weights of the network connection parameters of the edge node;

[0175] Step 16: Using a standardized formula, the network performance index of the edge node is standardized to obtain a standardized network performance index of the edge node;

[0176] Step 17: Establishing multiple transmission protocols under different network connection states;

[0177] Step 18: Classify the network connection status according to the network performance index of the edge node and adapt the corresponding transmission protocol;

[0178] Step 19: Establish a cloud-edge collaborative multi-point control model;

[0179] Step 20: Based on the cloud-edge collaborative multi-point control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, the computing power performance requirement of the real-time data task is taken as the input, and the edge nodes whose performance redundancy value meets the computing power performance requirements of the real-time data task are taken as the output; the computing power performance requirement of the non-real-time data task is taken as the input, and the edge nodes whose network performance index meets the computing power performance requirements of the non-real-time data task are taken as the output, to generate a weak network cloud-edge collaborative multi-node control solution.

[0180] To sum up, the advantages of the present invention are: by constructing a cloud-edge collaborative multi-point management and control model, the computing power performance requirements of real-time data tasks and non-real-time data tasks are determined, and the computing power performance requirements of real-time data tasks and non-real-time data tasks are used as constraints. The performance redundancy value of the edge node that meets the computing power performance requirements of the real-time data task is uploaded as a near-end node for data processing, and the network performance index of the edge node that meets the computing power performance requirements of the non-real-time data task is uploaded as a remote node for data processing, thereby reducing the risk of data loss and improving the stability of the distribution network.

[0181] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A cloud-edge collaborative multi-node management and control method for power distribution IoT intelligent gateways under weak networks, characterized in that: include: Based on the power distribution IoT smart gateway, collect the operation data of power equipment; The operation data of the power equipment is divided into real-time data and non-real-time data, and real-time data tasks and non-real-time data tasks are generated; Evaluate the computing power performance requirements of real-time data tasks and non-real-time data tasks; Get multiple edge nodes connected to the power distribution IoT smart gateway; Evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the performance redundancy value of each edge node; Evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the network performance index of each edge node; Establish multiple transmission protocols under different network connection states; The network connection status is classified according to the network performance index of the edge node, and the corresponding transmission protocol is adapted; Establish a cloud-edge collaborative multi-point control model; Based on the cloud-edge collaborative multi-point control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, the computing power performance requirements of the real-time data task are taken as input, and the edge nodes whose performance redundancy values ​​meet the computing power performance requirements of the real-time data task are taken as output. The computing power performance requirements of non-real-time data tasks are taken as input, and the edge nodes whose network performance index meets the computing power performance requirements of non-real-time data tasks are taken as output to generate a weak network cloud-edge collaborative multi-node control solution. Among them, the cloud-edge collaborative multi-point control model expression is specifically: Where P is a weak network cloud-edge collaborative multi-node control solution, NPI' e is the standardized network performance index of the e-th edge node, f() is the transmission protocol mapping function, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, R e is the comprehensive performance redundancy value of the e-th edge node, and L is the total number of edge nodes.

2. According to the method for cloud-edge collaborative multi-node management and control of power distribution IoT intelligent gateway under weak network in claim 1, it is characterized in that: The computing power performance requirements for evaluating real-time data tasks and non-real-time data tasks include: Determine the central core parameters of edge nodes; The real-time data tasks and non-real-time data tasks are quantitatively decomposed to obtain the total data volume of the real-time data tasks and non-real-time data tasks; the total data volume refers to the smallest data unit bit. Determine the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks; Calculate the central core processing time of processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task; Calculate the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node; Based on the total data volume of real-time data tasks and non-real-time data tasks, the running memory requirements of real-time data tasks and non-real-time data tasks, the central processing time for processing real-time data tasks and non-real-time data tasks, and the total number of central cores required, the computing power performance requirements of real-time data tasks and non-real-time data tasks are calculated through the computing power formula; The specific calculation power formula is: In the formula, A 1 To meet the computing power performance requirements of real-time data tasks, A 2 For the computing power performance requirements of non-real-time data tasks, D 1 is the total data volume of the real-time data task, D 2 is the total data volume of non-real-time data tasks, H 1 is the memory requirement for real-time data tasks, H 2 is the running memory requirement of the non-real-time data task, Ct is the response time of executing the task, Central processing time for real-time data processing tasks, The central processing time for processing non-real-time data tasks, C core is the desired total number of central cores.

3. According to the method for cloud-edge collaborative multi-node management and control of power distribution IoT intelligent gateway under weak network in claim 2, it is characterized in that: Evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway and obtain the performance redundancy value of each edge node, including: Determine the ideal hardware performance index of the edge node based on the edge node construction parameters; the ideal hardware performance index of the edge node includes at least: CPU utilization, memory utilization and storage I / O utilization; According to the real-time operation data of the edge nodes, collect the hardware utilization rate of the edge nodes and determine the real-time performance indicators of the hardware of the edge nodes; Based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node, the comprehensive performance redundancy value of the edge node is calculated through the performance analysis algorithm; The performance analysis algorithm is as follows: In the formula, R is the comprehensive performance redundancy value of the edge node, U j ' is the ideal performance index of the jth hardware of the edge node, U j is the jth hardware real-time performance indicator of the edge node.

4. According to the method for cloud-edge collaborative multi-node management and control of power distribution IoT intelligent gateway under weak network according to claim 3, it is characterized in that: Evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the network performance index of each edge node, including: Actively upload data packets for all edge nodes to test and collect several network connection parameters of edge nodes under the upload data packet test; the network connection parameters include but are not limited to: bandwidth, delay (ping), packet loss rate and jitter; Based on the measured values ​​of several network connection parameters of the edge node, score each network connection parameter to obtain several network connection parameter scores of the edge node; Based on the importance of the network connection status corresponding to the measured values ​​of several network connection parameters of the edge node, a weight is assigned to each network connection parameter to obtain several network connection parameter weights of the edge node; According to the scores of several network connection parameters of the edge node and the weights of several network connection parameters of the edge node, the network performance index of the edge node is calculated through a weighted formula; Using a standardized formula, the network performance index of the edge node is standardized to obtain a standardized network performance index of the edge node; The specific weighted formula is: Where NPI is the network performance index of the edge node, S k Score the kth network connection parameter of the edge node, W k is the kth network connection parameter weight of the edge node; The specific standardized formula is: Where NPI' is the network performance index of the standardized edge node, NPI max is the maximum value of the network performance index of the edge node, NPI min is the minimum network performance index of the edge node.

5. The power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under weak network is characterized by: A method for implementing a cloud-edge collaborative multi-node management and control method for a power distribution IoT intelligent gateway under a weak network as described in any one of claims 1 to 4, comprising: Data segmentation module: The data segmentation module is used to collect the operation data of power equipment based on the power distribution IoT intelligent gateway, and segment it into real-time data and non-real-time data, and generate real-time data tasks and non-real-time data tasks; A computing power demand assessment module, the computing power demand assessment module is electrically connected to the data segmentation module, and the computing power demand assessment module is used to assess the computing power performance requirements of real-time data tasks and non-real-time data tasks; An edge node acquisition module, which is used to acquire multiple edge nodes connected to the power distribution IoT smart gateway; A performance status evaluation module, which is electrically connected to the edge node acquisition module, and is used to evaluate the real-time performance status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the performance redundancy value of each edge node; A network status evaluation module, the network status evaluation module is electrically connected to the edge node acquisition module, and the network status evaluation module is used to evaluate the real-time network connection status of multiple edge nodes connected to the power distribution IoT smart gateway, and obtain the network performance index of each edge node; A multi-transmission protocol module is used to establish transmission protocols under multiple different network connection states; A transmission protocol adaptation module, the transmission protocol adaptation module is electrically connected to the multi-transmission protocol module, and the transmission protocol adaptation module is used to classify the network connection status according to the network performance index of the edge node and adapt the corresponding transmission protocol; Model building module: The model building module is used to establish a cloud-edge collaborative multi-point management and control model; The model management and control module is electrically connected to the model construction module, the computing power requirement evaluation module, the performance status evaluation module and the network status evaluation module. Based on the cloud-edge collaborative multi-point management and control model, the network connection status level of the network performance index of the edge node is adapted to the corresponding transmission protocol as the basic condition, and the computing power performance requirement of the real-time data task is used as the input, and the edge node whose performance redundancy value meets the computing power performance requirement of the real-time data task is used as the output; the computing power performance requirement of the non-real-time data task is used as the input, and the edge node whose network performance index meets the computing power performance requirement of the non-real-time data task is used as the output, to generate a weak network cloud-edge collaborative multi-node management and control solution.

6. The power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under weak network according to claim 5 is characterized in that: The computing power demand assessment module includes: A core parameter unit, which determines the central core parameters of edge nodes; The task quantification unit quantifies and decomposes the real-time data tasks and the non-real-time data tasks to obtain the total data volume of the real-time data tasks and the non-real-time data tasks; A memory requirement unit, which determines the running memory requirements of the real-time data tasks and the non-real-time data tasks based on the total data volume of the real-time data tasks and the non-real-time data tasks; A processing time unit, which calculates the central core processing time for processing the real-time data task and the non-real-time data task according to the central core parameters of the edge node and the total data volume of the real-time data task and the non-real-time data task; The core quantity unit calculates the total number of central cores required based on the processing capacity of a single central core and the central processing time for processing real-time data tasks and non-real-time data tasks in the central core parameters of the edge node; The first computing unit calculates the computing power performance requirements of the real-time data tasks and the non-real-time data tasks through a computing power formula based on the total data volume of the real-time data tasks and the non-real-time data tasks, the running memory requirements of the real-time data tasks and the non-real-time data tasks, the central processing time for processing the real-time data tasks and the non-real-time data tasks, and the total central core quantity requirements.

7. The power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under weak network according to claim 5 is characterized in that: The performance status assessment module includes: The ideal performance unit determines the ideal performance indicators of the edge node hardware based on the edge node construction parameters; The real-time performance unit collects the hardware utilization rate of the edge node according to the real-time operation data of the edge node and determines the hardware real-time performance index of the edge node; The second calculation unit calculates the comprehensive performance redundancy value of the edge node through a performance analysis algorithm based on the ideal performance index of the edge node and the utilization rate of each performance index of the edge node.

8. The power distribution IoT intelligent gateway cloud-edge collaborative multi-node management and control system under weak network according to claim 5 is characterized in that: The performance status assessment module includes: The network connection parameter collection unit performs active data packet upload test on all edge nodes, and collects several network connection parameters of the edge nodes under the upload data packet test; A parameter evaluation unit, based on the measured values ​​of the network connection parameters of the edge node, scores each network connection parameter to obtain the network connection parameter scores of the edge node; A parameter weight unit assigns a weight to each network connection parameter based on the importance of the network connection status corresponding to the measured values ​​of the network connection parameters of the edge node, thereby obtaining the weights of the network connection parameters of the edge node; A third calculation unit calculates a network performance index of the edge node through a weighted formula according to the scores of the network connection parameters of the edge node and the weights of the network connection parameters of the edge node; The standardization unit uses a standardization formula to standardize the network performance index of the edge node to obtain the network performance index of the standardized edge node.

Images