An edge-computing-based power internet-of-things panoramic monitoring method and device

Abstract

The application relates to an edge-computing-based power internet-of-things panoramic monitoring method and device, and belongs to the technical field of power systems. Through node importance evaluation based on ontology-scene historical information driving and data transmission optimization based on power internet-of-things panoramic monitoring gain, the accuracy and comprehensiveness of power internet-of-things data node importance evaluation are improved, and then the transmission node selection strategy is optimized through prediction of power internet-of-things panoramic monitoring gain, so that efficient transmission of edge-computing gateway data is ensured. Efficient edge-end interaction of the power internet-of-things is realized.

Description

Technical Field

[0001] This invention relates to a panoramic monitoring method and device for power IoT based on edge computing, belonging to the field of power system technology. Background Technology

[0002] Currently, with the rapid advancement of new power system construction, the high proportion of distributed renewable energy, primarily wind and solar power, is significantly impacting grid operation, making the distribution network's operating status more complex and variable, controlling more diverse objects, and increasing the number of terminals. This places higher demands on the grid's state perception and data processing capabilities. The emergence of the power Internet of Things (IoT) provides new opportunities for the intelligent and sustainable development of power systems. By connecting various parts of the electrical system, the power IoT can significantly enhance grid state perception, data processing, and information transmission capabilities, thereby improving grid reliability and efficiency and providing strong support for the coordinated development of renewable energy systems and the grid. However, with the access of massive numbers of data acquisition terminals, the data that edge computing gateways in the power IoT need to process is also constantly increasing, placing higher demands on edge computing gateways to achieve efficient data transmission. Traditional power IoT edge computing methods and devices still need to address the following challenges:

[0003] First, traditional methods for assessing the importance of IoT data nodes typically only consider the topological attributes of the nodes, lacking a synergistic consideration of the node's historical information and the historical information of the scene. They ignore the changing trends of multidimensional node data and the impact of historical information on node importance, resulting in an inability to accurately and comprehensively assess the importance of IoT data nodes.

[0004] Second, traditional methods often select data transmission nodes based on a single factor, such as data backlog. When the node's working status or channel status changes, they cannot guarantee efficient data transmission from the edge computing gateway, thus affecting the overall efficiency of the power Internet of Things.

[0005] In view of the above-mentioned shortcomings, the present invention aims to create a power IoT panoramic monitoring method and device based on edge computing, so as to make it more valuable for industrial applications. Summary of the Invention

[0006] To address the aforementioned technical problems, the present invention aims to provide a power IoT panoramic monitoring method and device based on edge computing. By using ontology-scene historical information-driven node importance assessment and power IoT panoramic monitoring gain-based data transmission optimization, the accuracy and comprehensiveness of power IoT data node importance assessment are improved. Furthermore, by predicting the power IoT panoramic monitoring gain and optimizing the transmission node selection strategy, efficient transmission of edge computing gateway data is ensured. This achieves efficient edge-end interaction in the power IoT.

[0007] The present invention provides a panoramic monitoring method for power IoT based on edge computing, the specific steps of which are as follows:

[0008] Based on the recent upward and downward fluctuation trends of the data uploaded by each data node, the importance of the data nodes driven by the ontology's historical information is obtained.

[0009] Based on the difference between the recent data uploaded by each data node and the historical data at the same time, as well as the similarity information with historical fault data, the importance of data nodes driven by scene historical information is obtained.

[0010] Based on the importance of data nodes driven by ontology historical information and the importance of data nodes driven by scene historical information, the importance of data nodes is evaluated using an ontology-scene historical information driven method, and the node importance is obtained.

[0011] By optimizing the data transmission based on the gain of power IoT panoramic monitoring, the edge computing gateway evaluates the gain of the power IoT panoramic monitoring brought by the data uploaded by the data node based on the node importance, the data backlog of the edge computing gateway, and the historical channel status between the data node and the edge server, and selects the data node with the largest gain to upload data, thereby realizing power IoT panoramic monitoring based on edge computing.

[0012] Furthermore, the data nodes have a total of There are , a total of The set of times is defined as: (a set of times) The data node set is defined as follows: , Indicates the first The data node, at the _ , Heavenly At any given moment, a data node collects multidimensional data on the operation of its corresponding power equipment; the data set is defined as follows: ,in For the dimensions of the data, Represents data nodes At the Heavenly The data collected at the [time]th moment Dimensional data.

[0013] Furthermore, the importance of data nodes driven by the ontology's historical information. , represented as:

[0014] (1)

[0015] in, The first data node uploaded Weights of dimensional data The first one stored for the gateway Heavenly The total number of moments prior to the given moment, i.e., the importance of data nodes driven by ontology historical information, is based on the number of moments prior to the given moment. Heavenly Data uploaded at that time and before The calculation is based on historical data at time point n. The formula means that the least squares method is used to obtain the nth... Heavenly a moment and before The slope of the straight line approximated by historical data at a given time point. If the absolute value of this slope is too large or too small, it indicates that the data node... Recent uploaded data has shown fluctuating upward or downward trends, data nodes The corresponding power equipment is highly likely to fail, and the data nodes The importance of this node is relatively high, and data from this node should be uploaded first.

[0016] Furthermore, the importance of data nodes driven by the historical information of the scenario is... It can be calculated as:

[0017] (2)

[0018] in, The total number of days of historical data stored in the gateway. and These are the weights for the historical data difference at the same moment and the similarity weights for historical fault data, respectively. In the formula, the first term represents the weight of the first term. Heavenly Data nodes at each time point The uploaded data compared to the past The first term indicates whether there are significant discrepancies between historical data at the same time of day, and the second term indicates whether the data uploaded by recent nodes is similar to historical fault data; the formula means that when data nodes... When recently uploaded data differs significantly from historical data at the same time, or when the data is similar to historical failure scenarios, data nodes... The corresponding power equipment is highly likely to fail, and the data nodes This node has a relatively high importance and its data needs to be uploaded first; the gateway stores historical fault datasets. ,in, This represents the total number of historical fault data sets. For the first Groups of multidimensional data, represented as .

[0019] Furthermore, the importance of the node is... , represented as:

[0020] (3)

[0021] in, and These are weights driven by ontology historical information and weights driven by scene historical information, respectively.

[0022] Furthermore, the specific steps of the data transmission optimization method based on the panoramic monitoring gain of the power IoT are as follows:

[0023] S2.1: The edge computing gateway first queries the data node and the edge server regarding the connection at the [missing information - likely a date or time]. The historical channel status of the day, including historical transmission bandwidth. Historical signal-to-interference-to-noise ratio For data transmission latency Make predictions;

[0024] S2.2: Based on the backlog of historical gateway data For computation delay To make predictions, among which Representing the Heavenly The gateway backlog is the time when node data is received.

[0025] S2.3: Based on node importance Data transmission latency and computational latency Gain of power IoT panoramic monitoring of data uploaded by nodes Make a prediction:

[0026] S2.4: The edge computing gateway arranges the node set according to the panoramic monitoring gain of the power IoT, and directs the node set towards the node with the highest gain. Data upload permissions are distributed to each node, in the... At any given time, data is uploaded to the gateway through a permitted node.

[0027] Furthermore, the data transmission delay prediction method described in S2.1 is as follows:

[0028] (4)

[0029] in For the first Heavenly Data nodes at each time point The amount of data transmitted For experience weight, To The historical transmission bandwidth before a certain time is averaged using empirical weights, and the distance is... The more recent the data, the greater its weight and the greater its impact on the prediction results; the same applies to the signal-to-interference-plus-noise ratio.

[0030] Furthermore, S2.2 calculates the latency. The prediction method is as follows: First, the historical calculation delay sequence is obtained based on the accumulation of historical data. :

[0031] (5)

[0032] in For the first Heavenly Data nodes at each time point The amount of data transmitted to the gateway Gateway as data node Allocated computing resources To process each bit of data node The computational resources required for the data; further empirical weighting of the historical latency series to obtain the predicted computational latency. :

[0033] (6)

[0034] This formula represents the first... The historical calculation delay up to time n is averaged using empirical weights, and the delay from time n is... The closer the calculation delay data is to a given moment, the greater its weight and the greater its impact on the prediction result.

[0035] Furthermore, the gain of panoramic monitoring of the power Internet of Things in S2.3 The prediction is represented as:

[0036] (7)

[0037] in The weight of data transmission latency relative to computation latency is used; the greater the importance of a node, the smaller its data transmission latency and computation latency, and the greater the gain in the panoramic monitoring of the power Internet of Things brought by uploading data from that node; therefore, it should be given priority in the data transmission process.

[0038] A power IoT panoramic monitoring device based on edge computing includes:

[0039] Ontology historical information processing module: used to obtain the importance of data nodes driven by ontology historical information based on information such as the upward and downward fluctuation trends of recent data uploaded by each data node;

[0040] Scene history information processing module: It is used to obtain the importance of data nodes driven by scene history information based on the difference between the recent data uploaded by each data node and the historical data at the same time, and the similarity with historical fault data.

[0041] Node Importance Assessment Module: Based on the results of the ontology history information processing module and the scene history information processing module, this module assesses the importance of data nodes using an ontology-scene history information-driven approach.

[0042] The monitoring gain evaluation module is used to evaluate the power IoT panoramic monitoring gain of the data uploaded by the data node based on the data backlog of the edge computing gateway, the historical channel status between the data node and the edge server, and the node importance evaluation module.

[0043] Data transmission optimization module: used to select nodes for uploading data based on the power IoT panoramic monitoring gain obtained from the monitoring gain evaluation module;

[0044] Storage module: Used to store historical data uploaded by each data node, historical fault data, historical fault records, historical channel status between data nodes and edge servers, and to provide the above information to each module;

[0045] 5G communication module: Used to communicate with power Internet of Things data nodes, receive data uploaded by nodes, transmit it to this device for processing, and distribute data upload permissions to nodes selected by the data transmission optimization module;

[0046] Power module: Responsible for supplying power to the various modules in the edge computing gateway device suitable for panoramic monitoring of power IoT.

[0047] By means of the above-described solution, the present invention has at least the following advantages:

[0048] (1) This invention proposes a node importance assessment method based on ontology-scene historical information, which considers both node ontology historical information and scene historical information. The edge computing gateway receives historical data, current time, and historical fault occurrence information uploaded by each data node. It calculates the data node importance driven by ontology historical information of the power Internet of Things data node through information such as the upward and downward fluctuation trends of recently uploaded data. It calculates the data node importance driven by scene historical information through information such as the difference between recently uploaded data and historical data at the same time, and the similarity with historical fault data. Furthermore, the importance of data nodes is calculated based on the data node importance driven by ontology historical information and the data node importance driven by scene historical information, thereby improving the accuracy and comprehensiveness of power Internet of Things data node importance assessment.

[0049] (2) This invention proposes a data transmission optimization method based on the gain of power IoT panoramic monitoring. The edge computing gateway evaluates the gain of the data uploaded by the data node for power IoT panoramic monitoring based on the node importance, the data backlog of the edge computing gateway, and the historical channel status between the data node and the edge server. The data transmission of edge nodes with high gain is preferentially selected to realize the data transmission optimization of power IoT panoramic monitoring gain, thereby improving the data processing efficiency of important nodes.

[0050] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show a certain embodiment of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the power IoT panoramic monitoring device based on edge computing according to the present invention;

[0053] Figure 2 This is a flowchart of the power IoT panoramic monitoring method based on edge computing according to the present invention. Detailed Implementation

[0054] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0055] like Figure 1 As shown, this invention proposes an edge computing gateway device suitable for panoramic monitoring of power IoT, including an entity historical information processing module, a scene historical information processing module, a node importance assessment module, a monitoring gain assessment module, a data transmission optimization module, a storage module, a 5G communication module, and a power supply module. The modules are described below:

[0056] Ontology historical information processing module: Based on the recent data fluctuation trends and fluctuation decline trends of each data node, the importance of the data node is obtained by driving the ontology historical information.

[0057] Scene history information processing module: Based on the difference between the recent data uploaded by each data node and the historical data at the same time, as well as the similarity with historical fault data, the importance of data nodes driven by scene history information is obtained.

[0058] Node Importance Assessment Module: Based on the results of the ontology historical information processing module and the scene historical information processing module, the module assesses the importance of data nodes using an ontology-scene historical information-driven approach.

[0059] Monitoring Gain Evaluation Module: Based on the data backlog of the edge computing gateway, the historical channel status between the data node and the edge server, and the node importance obtained by the node importance evaluation module, the module evaluates the power IoT panoramic monitoring gain of the data uploaded by the data node.

[0060] Data transmission optimization module: Based on the power IoT panoramic monitoring gain obtained from the monitoring gain evaluation module, select the node to upload data.

[0061] Storage module: The storage module stores information such as historical data uploaded by each data node, historical fault data, historical fault records, and historical channel status between data nodes and edge servers, and provides the above information to each module.

[0062] 5G communication module: Responsible for communicating with power IoT data nodes, receiving data uploaded by the nodes, transmitting it to this device for processing, and distributing data upload permissions to nodes selected by the data transmission optimization module.

[0063] Power module: Responsible for supplying power to the various modules in the edge computing gateway device suitable for panoramic monitoring of power IoT.

[0064] Based on the above-mentioned device, this invention proposes a power IoT panoramic monitoring method based on edge computing, comprising two parts: a node importance assessment method driven by ontology-scene historical information and a data transmission optimization method based on power IoT panoramic monitoring gain. The process is as follows: Figure 2 As shown, the specific description is as follows:

[0065] 1. A node importance assessment method driven by ontology-scene historical information

[0066] In the context of the development of the power Internet of Things (IoT), massive numbers of power IoT terminals act as data nodes, collecting data from power equipment and uploading it to edge computing gateways for processing to support the safe and stable operation of the distribution network. However, due to limited communication resources, it is difficult to fully meet the data processing requirements of all data nodes. Therefore, it is necessary to reasonably assess the importance of data nodes and prioritize uploading data from important nodes. This invention proposes a node importance assessment method driven by ontology-scene historical information. The edge computing gateway assesses the importance of data nodes based on the historical data uploaded by each data node, the current time, and historical fault occurrences, using an ontology-scene historical information-driven method. The specific steps are as follows:

[0067] S1.1: Data node importance driven by ontology historical information. This invention considers... The set of times is defined as: (a set of times) Considering There are 1 data node, and the data node set is defined as follows: , Indicates the first The data node. In the... Heavenly At any given moment, a data node collects multi-dimensional data such as temperature, current, and voltage from its corresponding power equipment. This data set is defined as follows: ,in For the dimensions of the data, Represents data nodes At the Heavenly The data collected at the [time]th moment Dimensional data. Subsequently, the data nodes upload the data to the edge computing gateway. Based on the data uploaded by each data node, combined with historical data, the current time, and historical fault occurrences, the edge computing gateway evaluates the importance of the data nodes using an ontology-scene historical information-driven method. The steps for calculating the importance of data nodes driven by ontology historical information are as follows.

[0068] The importance of data nodes driven by ontology historical information reflects directly observable anomalies in data uploaded from the same data node on the same day. For example, there might be an upward or downward trend in fluctuations in equipment operating information such as temperature and voltage uploaded in the past few uploads. Defined in the... Heavenly At that moment, the data node The importance of data nodes driven by ontological historical information is , represented as

[0069] (1)

[0070] in, The first data node uploaded Weights of dimensional data The first one stored for the gateway Heavenly The total number of moments prior to the given moment, i.e., the importance of data nodes driven by ontology historical information, is based on the number of moments prior to the given moment. Heavenly Data uploaded at that time and before The calculation is based on historical data from the nth moment. The formula means that the th moment is obtained using the least squares method. Heavenly a moment and before The slope of the straight line approximated by historical data at a given time point. If the absolute value of this slope is too large or too small, it indicates that the data node... Recent uploaded data has shown fluctuating upward or downward trends, data nodes The corresponding power equipment is highly likely to fail, and the data nodes The importance of this node is relatively high, and data from this node should be uploaded first.

[0071] S1.2: Data Node Importance Driven by Scene Historical Information. The importance of data nodes driven by scene historical information reflects anomalies that can be analyzed based on historical fault occurrences. Examples include temperatures within the normal range but higher than the same historical time, or similarities between previously uploaded data and historical fault scenarios. The historical fault dataset stored by the gateway is defined as follows: ,in, This represents the total number of historical fault data sets. For the first Groups of multidimensional data can be represented as Therefore, in the first Heavenly At that moment, the data node The importance of data nodes driven by historical information of the scenario is It can be calculated as

[0072] (2)

[0073] in, The total number of days of historical data stored in the gateway. and These are the weights for the historical data difference at the same moment and the similarity weights for historical fault data, respectively. In the formula, the first term represents the weight of the first term. Heavenly Data nodes at each time point The uploaded data compared to the past The first term indicates whether there are significant discrepancies between historical data at the same time each day, and the second term indicates whether the data uploaded by recent nodes is similar to historical fault data. The formula means that when data nodes... When recently uploaded data differs significantly from historical data at the same time, or when the data is similar to historical failure scenarios, data nodes... The corresponding power equipment is highly likely to fail, and the data nodes The importance of this node is relatively high, and data from this node should be uploaded first.

[0074] S1.3: Node Importance Evaluation Based on Ontology-Scene History Information. To comprehensively evaluate the importance of data nodes, based on the ontology history information-driven data node importance and scene history information-driven data node importance obtained in S1.2 and S1.3, a node importance based on ontology-scene history information is further calculated, defined in the first... Heavenly At that moment, the data node The importance of data nodes is , represented as

[0075] (3)

[0076] in, and These are weights driven by ontology historical information and weights driven by scene historical information, respectively.

[0077] 2. Data transmission optimization method based on power IoT panoramic monitoring gain

[0078] A node importance assessment method based on ontology-scene history information is used to evaluate node importance. Following the evaluation, this invention further proposes a data transmission optimization method based on the gain of power IoT panoramic monitoring. The edge computing gateway evaluates the gain of power IoT panoramic monitoring brought by the data uploaded by the data node based on node importance, data backlog status of the edge computing gateway, and the historical channel status between the data node and the edge server, and selects the data node with the largest gain to upload data. The specific steps are as follows:

[0079] S2.1: The edge computing gateway first queries the data node and the edge server regarding the connection at the [missing information - likely a date or time]. The historical channel status of the day, including historical transmission bandwidth. Historical signal-to-interference-to-noise ratio For data transmission latency Prediction is performed. The data transmission latency prediction method is as follows:

[0080] (4)

[0081] in For the first Heavenly Data nodes at each time point The amount of data transmitted For experience weight, Indicates to The historical transmission bandwidth before a certain time is averaged using empirical weights, and the distance is... The more recent the data, the greater its weight and the greater its impact on the prediction results; the same applies to the signal-to-interference-plus-noise ratio.

[0082] S2.2: Based on the backlog of historical gateway data For computation delay To make predictions, among which Representing the Heavenly The timeframe represents the gateway backlog when node data is received. First, the historical computation delay sequence is obtained based on the historical data backlog. :

[0083] (5)

[0084] in For the first Heavenly Data nodes at each time point The amount of data transmitted to the gateway Gateway as data node Allocated computing resources To process each bit of data node The computational resources required for the data. Further empirical weighting of the historical latency series yields the predicted computational latency. :

[0085] (6)

[0086] This formula represents the first... The historical calculation delay up to time n is averaged using empirical weights, and the delay from time n is... The closer the calculation delay data is to a given moment, the greater its weight and the greater its impact on the prediction result.

[0087] S2.3: Based on node importance Data transmission latency and computational latency Gain of power IoT panoramic monitoring of data uploaded by nodes Make a prediction:

[0088] (7)

[0089] in This represents the weight of data transmission latency relative to computation latency. The greater the node's importance and the smaller its data transmission and computation latency, the greater the gain in the overall power IoT monitoring brought by uploading data from that node. Therefore, it should be prioritized during data transmission.

[0090] S2.4: The edge computing gateway arranges the node set according to the panoramic monitoring gain of the power IoT, and directs the node set towards the node with the highest gain. Data upload permissions are distributed to each node, in the... At any given time, data is uploaded to the gateway through a permitted node.

[0091] (1) This invention proposes a node importance assessment method based on ontology-scene historical information. When assessing the importance of power Internet of Things (IoT) data nodes, it considers the impact of node ontology historical information, scene historical information, and data change trends on node importance assessment. The ontology historical information-driven data node importance is calculated based on information such as the upward and downward fluctuation trends of recently uploaded data. The scene historical information-driven data node importance is calculated based on information such as the difference between recently uploaded data and historical data at the same time, and the similarity with historical fault data. Finally, by combining ontology and scene historical information, the importance of power IoT data nodes is further obtained, thus achieving an accurate and comprehensive assessment of the importance of power IoT data nodes.

[0092] (2) This invention proposes a data transmission optimization method based on the gain of power IoT panoramic monitoring. The edge computing gateway evaluates the gain of power IoT panoramic monitoring obtained by uploading data through the data node based on node importance, data transmission latency, and computation latency. The computation latency is predicted by the data backlog situation of the edge computing gateway, and the data transmission latency can be predicted by the edge computing gateway based on the historical channel status (historical transmission bandwidth, historical signal-to-interference-plus-noise ratio) between the data node and the edge server. Then, data upload permissions are distributed to the data nodes with the largest gain to improve the transmission and processing efficiency of important data nodes and ensure timely response to important data nodes.

[0093] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A panoramic monitoring method for power IoT based on edge computing, characterized in that... The specific steps are as follows: Based on the recent upward and downward fluctuation trends of data uploaded by each data node, the importance of the data nodes is determined by the ontology's historical information. The data nodes have a total of... There are , a total of The set of times is defined as _ time moments. The data node set is defined as follows: , Indicates the first The data node, at the _ , Heavenly At any given moment, a data node collects multidimensional data on the operation of its corresponding power equipment; the data set is defined as follows: ,in For the dimensions of the data, Represents data nodes At the Heavenly The data collected at the [time]th moment Dimensional data; Based on the difference between the recent data uploaded by each data node and the historical data at the same time, as well as the similarity information with historical fault data, the importance of data nodes driven by scene historical information is obtained. Based on the importance of data nodes driven by ontology historical information and the importance of data nodes driven by scene historical information, the importance of data nodes is evaluated using an ontology-scene historical information driven method, and the node importance is obtained. By optimizing the data transmission based on the gain of power IoT panoramic monitoring, the edge computing gateway evaluates the gain of the power IoT panoramic monitoring brought by the data uploaded by the data node based on the node importance, the data backlog of the edge computing gateway, and the historical channel status between the data node and the edge server, and selects the data node with the largest gain to upload data, thereby realizing power IoT panoramic monitoring based on edge computing. The specific steps of the data transmission optimization method based on the panoramic monitoring gain of power IoT are as follows: S2.1: The edge computing gateway first queries the data node and the edge server regarding the connection at the [missing information - likely a date or time]. The historical channel status of the day, including historical transmission bandwidth. Historical signal-to-interference-to-noise ratio For data transmission latency Make predictions; S2.2: Based on the backlog of historical gateway data For computation delay To make predictions, among which Representing the Heavenly The gateway backlog is the time when node data is received. S2.3: Based on node importance Data transmission latency and computational latency Gain of power IoT panoramic monitoring of data uploaded by nodes Make a prediction: S2.4: The edge computing gateway arranges the node set according to the panoramic monitoring gain of the power IoT, and directs the node set towards the node with the highest gain. Data upload permissions are distributed to each node, in the... At any given time, data is uploaded to the gateway through a permitted node.

2. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: The importance of data nodes driven by the ontology's historical information , is represented as: (1) in, The first data node uploaded Weights of dimensional data The first one stored for the gateway Heavenly The total number of moments prior to the given moment, i.e., the importance of data nodes driven by ontology historical information, is based on the number of moments prior to the given moment. Heavenly Data uploaded at that time and before The calculation is based on historical data at time t. The formula means that the least squares method is used to obtain the t... Heavenly a moment and before The slope of the straight line approximated by historical data at a given time point. If the absolute value of this slope is too large or too small, it indicates that the data node... Recent uploaded data has shown fluctuating upward or downward trends, data nodes The corresponding power equipment is highly likely to fail, and the data nodes The importance of this node is relatively high, and data from this node should be uploaded first.

3. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: The importance of the data nodes driven by the historical information of the scenario is: It can be calculated as: (2) in, The total number of days of historical data stored in the gateway. and These are the weights for the historical data difference at the same moment and the similarity weights for historical fault data, respectively. In the formula, the first term represents the weight of the first term. Heavenly Data nodes at each time point The uploaded data compared to the past The first term indicates whether there are significant discrepancies between historical data at the same time of day, and the second term indicates whether the data uploaded by recent nodes is similar to historical fault data; the formula means that when data nodes... When recently uploaded data differs significantly from historical data at the same time, or when the data is similar to historical failure scenarios, data nodes... The corresponding power equipment is highly likely to fail, and the data nodes This node has a relatively high importance and its data needs to be uploaded first; the gateway stores historical fault datasets. ,in, This represents the total number of historical fault data sets. For the first Groups of multidimensional data, represented as .

4. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: The importance of the node is , is represented as: (3) in, and These are weights driven by ontology historical information and weights driven by scene historical information, respectively. The importance of data nodes is driven by ontology historical information. The importance of data nodes is driven by historical information of the scene.

5. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: The data transmission delay prediction method described in S2.1 is as follows: (4) in For the first Heavenly Data nodes at each time point The amount of data transmitted For experience weight, To The historical transmission bandwidth before a certain time is averaged using empirical weights, and the distance is... The more recent the data, the greater its weight and the greater its impact on the prediction results; the same applies to the signal-to-interference-plus-noise ratio.

6. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: S2.2 Calculation delay The prediction method is as follows: First, the historical calculation delay sequence is obtained based on the backlog of historical data. : (5) in For the first Heavenly Data nodes at each time point The amount of data transmitted to the gateway Gateway as data node Allocated computing resources To process each bit of data node The computational resources required for the data; further empirical weighting of the historical latency series to obtain the predicted computational latency. : (6) This formula represents the first... The historical calculation delay up to the nth time point is averaged using empirical weights, and the delay from the nth time point is... The closer the calculation delay data is to a given moment, the greater its weight and the greater its impact on the prediction result.

7. The power IoT panoramic monitoring method based on edge computing according to claim 1, characterized in that: S2.3 Power Internet of Things Panoramic Monitoring Gain The prediction is represented as: (7) in This represents the weight of data transmission latency relative to computation latency. The greater the importance of a node, and the smaller the data transmission latency and computation latency, the greater the gain in the panoramic monitoring of the power Internet of Things brought by uploading data from that node; It should be given priority during data transmission.

8. A power IoT panoramic monitoring device based on edge computing, used to execute the method described in any one of claims 1-7, characterized in that... include: Ontology historical information processing module: used to obtain the importance of data nodes driven by ontology historical information based on information such as the upward and downward fluctuation trends of recent data uploaded by each data node; Scene history information processing module: It is used to obtain the importance of data nodes driven by scene history information based on the difference between the recent data uploaded by each data node and the historical data at the same time, and the similarity with historical fault data. Node Importance Assessment Module: Based on the results of the ontology history information processing module and the scene history information processing module, this module assesses the importance of data nodes using an ontology-scene history information-driven approach. The monitoring gain evaluation module is used to evaluate the power IoT panoramic monitoring gain of the data uploaded by the data node based on the data backlog of the edge computing gateway, the historical channel status between the data node and the edge server, and the node importance evaluation module. Data transmission optimization module: used to select nodes for uploading data based on the power IoT panoramic monitoring gain obtained from the monitoring gain evaluation module; Storage module: Used to store historical data, historical fault data, historical fault records, and historical channel status information between data nodes and edge servers uploaded by each data node, and to provide the above information to each module; 5G communication module: Used to communicate with power Internet of Things data nodes, receive data uploaded by nodes, transmit it to this device for processing, and distribute data upload permissions to nodes selected by the data transmission optimization module; Power module: Responsible for supplying power to the various modules in the edge computing gateway device suitable for panoramic monitoring of power IoT.

Images