Cloud edge-based power internet of things anomaly point identification method and system

Abstract

The application provides a cloud edge-end-based power Internet of Things abnormal point identification method and system, identifies an interval in which an illegal event occurs in a terminal layer according to operation data of the terminal layer, and adjusts a monitoring layout of an edge layer to the terminal layer; determines a power operation abnormal path in the terminal layer according to a sensing terminal and a power terminal log of the terminal layer, and adjusts resource input of the edge layer to the terminal layer, so as to realize accurate configuration of edge computing resources; analyzes power terminal monitoring data, predicts a potential overload event of the terminal layer, and determines an abnormal point of the terminal layer; analyzes task execution data of a related area of the abnormal point in a cloud layer, determines a task execution conflict event, and sends a task change instruction to the terminal layer through the cloud layer, so that efficient and accurate abnormal identification and troubleshooting are realized through clustering and differentiation detection in the terminal layer, and the accuracy and real-time performance of power Internet of Things abnormal point detection are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power Internet of Things, and particularly relates to an electric power Internet of Things abnormal point identification method and system based on cloud edge and end. BACKGROUND

[0002] The electric power Internet of Things is an Internet of Things applied in the electric power field, which can improve the detection and regulation capabilities of the power grid and ensure the safe, stable and efficient operation of the power grid. As an important technology used in the electric power Internet of Things, edge computing can realize the local distributed deployment of the electric power Internet of Things. The electric power Internet of Things architecture mainly includes a terminal layer, an edge layer and a cloud layer. The terminal layer is deployed with electric power terminals for maintaining the normal operation of the power grid and sensing terminals for monitoring the electric power terminals. The edge layer is deployed with a plurality of edge computers for implementing edge monitoring on different terminal clusters in the terminal layer. The cloud layer is deployed with a cloud server and a data storage for analyzing data from the edge layer and realizing the operation scheduling of the electric power terminals in the terminal layer. The existing electric power Internet of Things mostly realizes the operation monitoring of the power grid in the order of the terminal layer, the edge layer and the cloud layer, which requires a large amount of resources to complete the one-by-one checking of all electric power terminals. The operation abnormality identification cannot be realized in the terminal layer through clustering and differentiation, which reduces the efficiency and accuracy of the fault and abnormality detection of the electric power Internet of Things. SUMMARY

[0003] In view of the defects in the prior art, the present application provides an electric power Internet of Things abnormal point identification method and system based on cloud edge and end. According to the operation data of the terminal layer, the interval of the terminal layer where an irregular event occurs is identified, so as to adjust the monitoring layout of the edge layer to the terminal layer. According to the sensing terminal and electric power terminal logs of the terminal layer, the electric power operation abnormal path in the terminal layer is determined, so as to adjust the resource input of the edge layer to the terminal layer and realize the accurate configuration of the edge computing resources. The electric power terminal monitoring data is analyzed, the potential overload event of the terminal layer is predicted, and the abnormal point of the terminal layer is determined. The task execution data of the associated area of the abnormal point is analyzed in the cloud layer, the task execution conflict event is determined, and the task change instruction is sent from the cloud layer to the terminal layer. Through the clustering and differentiation detection in the terminal layer, efficient and accurate abnormality identification and troubleshooting are realized, and the accuracy and real-time performance of the electric power Internet of Things abnormal point detection are improved.

[0004] The present application provides an electric power Internet of Things abnormal point identification method based on cloud edge and end, which comprises the following steps:

[0005] Step S1, obtaining the operation data of the terminal layer of the electric power Internet of Things, identifying the electric power terminal change of the terminal layer, identifying the interval of the terminal layer where an irregular event occurs; according to the distribution of the interval, adjusting the monitoring layout of the edge layer to the terminal layer;

[0006] Step S2, according to the monitoring layout, acquiring the sensing terminal and power terminal log of the terminal layer, determining the power operation abnormal path in the terminal layer; according to the power terminal layout in the power operation abnormal path, adjusting the resource input of the edge layer to the terminal layer;

[0007] Step S3, analyzing the collected power terminal monitoring data, predicting the potential overload event of the terminal layer; according to the attribute of the potential overload event, determining the abnormal point of the terminal layer;

[0008] Step S4, analyzing the task execution data of the associated area of the abnormal point in the cloud layer, determining the task execution conflict event, and sending a task change instruction to the terminal layer through the cloud layer.

[0009] In an embodiment of the present application, in the step S1, the operation data of the terminal layer of the power Internet of Things is acquired, the power terminal change of the terminal layer is identified, and the interval in which the terminal layer occurs an illegal event is identified; according to the distribution of the interval, the monitoring layout of the edge layer to the terminal layer is adjusted, including:

[0010] The power terminal operation data of the terminal layer of the power Internet of Things is acquired; wherein the power terminal operation data includes the access and disconnection state change data of the power terminal in the terminal layer; the power terminal operation data is identified in terms of power grid location and time of occurrence, and the interval in which the terminal layer occurs an illegal event is identified; wherein the illegal event interval refers to the interval in which the terminal layer corresponding power grid occurs a power transmission interruption time too long event;

[0011] According to the distribution position of all intervals in which the power transmission interruption time too long event occurs in the power grid, the power grid is divided into several sub-grid areas; according to the number and type of power terminals under each sub-grid area, the monitoring layout of the edge layer of the power Internet of Things to each sub-grid area is adjusted; wherein the monitoring layout refers to the number and type of edge computers accessed by the edge layer to each sub-grid area.

[0012] In an embodiment of the present application, in the step S2, according to the monitoring layout, acquiring the sensing terminal and power terminal log of the terminal layer, determining the power operation abnormal path in the terminal layer; according to the power terminal layout in the power operation abnormal path, adjusting the resource input of the edge layer to the terminal layer, including:

[0013] According to the monitoring layout, the working log of the sensing terminal and the working log of the power terminal of each sub-network area are acquired, so that the working data stream of the sensing terminal and the working data stream of the power terminal are extracted; the flow of the working data stream of the sensing terminal and the working data stream of the power terminal are compared to determine whether the sub-network area belongs to an abnormal sub-network area; and the power transmission relationship of all abnormal sub-network areas in the power grid is determined to determine the power operation abnormal path in the terminal layer.

[0014] According to the power grid node layout of the power terminal in the power operation abnormal path, the amount of computing resource and bandwidth resource of the corresponding power grid node in the terminal layer input by the edge layer is adjusted.

[0015] In an embodiment of the present application, in the step S3, the collected power terminal monitoring data is analyzed to predict a potential overload event of the terminal layer; and according to the attribute of the potential overload event, an abnormal point of the terminal layer is determined, including:

[0016] The power terminal monitoring data collected by the edge layer is analyzed to obtain the power operation load transfer trend between the power terminals in the power operation abnormal path, so as to predict a potential overload event of the terminal layer;

[0017] According to the occurrence time and spatial attribute of the potential overload event, the power terminal in the power operation abnormal path that fails is identified, so as to determine the abnormal point of the terminal layer; wherein the abnormal point refers to the power grid node where the power terminal fails and the nodes directly connected thereto.

[0018] In an embodiment of the present application, in the step S4, the task execution data of the associated area of the abnormal point is analyzed in the cloud layer to determine a task execution conflict event, so as to send a task change instruction to the terminal layer through the cloud layer, including:

[0019] According to the distribution position of all abnormal points in the power grid, the associated area of the abnormal point in the power grid is demarcated; the task execution data of all power terminals in the associated area of the abnormal point is analyzed in the cloud layer to determine a task execution conflict event occurred by all power terminals in the associated area; wherein the task execution conflict event refers to a power transmission conflict event occurred during the task execution of any two power terminals in the associated area;

[0020] According to the occurrence position and occurrence time of the task execution conflict event, a task suspension execution instruction corresponding to one of the two power terminals is sent to the terminal layer through the cloud layer.

[0021] The application also provides a cloud edge-end-based power Internet of Things abnormal point identification system, including:

[0022] a terminal layer identification module, configured to acquire running data of a terminal layer of the power Internet of Things, perform power terminal change identification on the terminal layer, and identify an interval in which an out-of-compliance event of the terminal layer occurs;

[0023] an edge layer adjustment module, configured to adjust a monitoring layout of the edge layer on the terminal layer according to a distribution of the interval;

[0024] an abnormal path determination module, configured to acquire sensor terminal and power terminal logs of the terminal layer according to the monitoring layout, and determine a power operation abnormal path in the terminal layer;

[0025] a resource input adjustment module, configured to adjust resource input of the edge layer on the terminal layer according to a power terminal layout in the power operation abnormal path;

[0026] an overload event prediction module, configured to analyze collected power terminal monitoring data, and predict a potential overload event of the terminal layer;

[0027] an abnormal point determination module, configured to determine an abnormal point of the terminal layer according to an attribute of the potential overload event;

[0028] a task change module, configured to analyze task execution data of an associated area of the abnormal point in the cloud layer, determine a task execution conflict event, and send a task change instruction to the terminal layer through the cloud layer.

[0029] In an embodiment of the present disclosure, the terminal layer identification module, configured to acquire running data of a terminal layer of the power Internet of Things, perform power terminal change identification on the terminal layer, and identify an interval in which an out-of-compliance event of the terminal layer occurs, includes:

[0030] acquire power terminal running data of the terminal layer of the power Internet of Things; wherein the power terminal running data includes access and disconnection state change data of a power terminal in the terminal layer; perform power grid location and time of occurrence identification on the power terminal running data, and identify an interval in which an out-of-compliance event of the terminal layer occurs; wherein the out-of-compliance event interval refers to an interval in which a power transmission interruption time is too long in a power grid corresponding to the terminal layer;

[0031] The edge layer adjustment module, configured to adjust a monitoring layout of the edge layer on the terminal layer according to a distribution of the interval, includes:

[0032] According to the distribution position of the interval of all power transmission interruption time too long events in the power grid, the power grid is divided into several sub-grid areas; according to the number and type of power terminals under each sub-grid area, the monitoring layout of the edge layer of the power Internet of Things to each sub-grid area is adjusted; wherein the monitoring layout refers to the number and type of edge computers accessed by the edge layer to each sub-grid area.

[0033] In an embodiment of the present application, the abnormal path determination module is configured to obtain the sensing terminal and power terminal logs of the terminal layer according to the monitoring layout, and determine the power operation abnormal path in the terminal layer, including:

[0034] According to the monitoring layout, the sensing terminal working log and the power terminal working log of each sub-grid area are obtained, from which the sensing terminal working data stream and the power terminal working data stream are extracted; by comparing the flow of the sensing terminal working data stream and the power terminal working data stream, it is determined whether the sub-grid area belongs to an abnormal sub-grid area; according to the power transmission relationship of all abnormal sub-grid areas in the power grid, the power operation abnormal path in the terminal layer is determined;

[0035] The resource input adjustment module is configured to adjust the resource input of the edge layer to the terminal layer according to the power terminal layout in the power operation abnormal path, including:

[0036] According to the power grid node layout of the power terminal in the power operation abnormal path, the amount of computing power resources and bandwidth resources of the edge layer to the corresponding power grid node in the terminal layer is adjusted.

[0037] In an embodiment of the present application, the overload event prediction module is configured to analyze the collected power terminal monitoring data and predict potential overload events of the terminal layer, including:

[0038] The power terminal monitoring data collected by the edge layer is analyzed to obtain the power operation load transfer trend between the power terminals in the power operation abnormal path, so as to predict potential overload events of the terminal layer;

[0039] The abnormal point determination module is configured to determine the abnormal point of the terminal layer according to the attributes of the potential overload event, including:

[0040] According to the occurrence time and spatial attributes of the potential overload event, the power terminal that fails in the power operation abnormal path is identified, so as to determine the abnormal point of the terminal layer; wherein the abnormal point refers to the power grid node where the power terminal that fails is located and the nodes directly connected thereto.

[0041] In one embodiment of the present application, the task change module is configured to analyze the task execution data of the associated region of the abnormal point in the cloud layer, determine a task execution conflict event, and send a task change instruction to the terminal layer through the cloud layer, including:

[0042] According to the distribution position of all abnormal points in the power grid, the associated region of the abnormal point in the power grid is demarcated; the task execution data of all power terminals in the associated region of the abnormal point is analyzed in the cloud layer to determine a task execution conflict event occurring in all power terminals in the associated region; wherein the task execution conflict event refers to a power transmission conflict event occurring during the execution of tasks by any two power terminals in the associated region;

[0043] According to the occurrence position and occurrence time of the task execution conflict event, a task suspension execution instruction corresponding to one of the two power terminals is sent to the terminal layer through the cloud layer.

[0044] Compared with the prior art, the power Internet of Things abnormal point identification method and system based on cloud edge and end identifies the interval in which the terminal layer occurs an illegal event according to the operation data of the terminal layer, so as to adjust the monitoring layout of the edge layer to the terminal layer; determines the power operation abnormal path in the terminal layer according to the sensing terminal and power terminal log of the terminal layer, so as to adjust the resource investment of the edge layer to the terminal layer, and realizes the accurate configuration of the edge computing resource; analyzes the power terminal monitoring data, predicts the potential overload event of the terminal layer, so as to determine the abnormal point of the terminal layer; analyzes the task execution data of the associated region of the abnormal point in the cloud layer, determines a task execution conflict event, so as to send a task change instruction to the terminal layer through the cloud layer, realizes efficient and accurate abnormal identification and investigation by clustering and distinguishing detection in the terminal layer, and improves the accuracy and real-time performance of the power Internet of Things abnormal point detection.

[0045] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0046] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description can also be obtained by those skilled in the art without creating any inventive labor.

[0048] Figure 1 is a flowchart of a cloud-edge-end based power internet of things anomaly point identification method provided by the present application.

[0049] Figure 2 is a cloud-edge-end architecture of the power internet of things.

[0050] Figure 3 is a framework diagram of a cloud-edge-end based power internet of things anomaly point identification system provided by the present application. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0052] Reference Figure 1 is a flowchart of a cloud-edge-end based power internet of things anomaly point identification method provided by the present application. The cloud-edge-end based power internet of things anomaly point identification method comprises:

[0053] Step S1, obtaining running data of a terminal layer of the power internet of things, performing power terminal change identification on the terminal layer, identifying an interval in which an irregular event occurs in the terminal layer, and adjusting a monitoring layout of an edge layer to the terminal layer according to a distribution of the interval;

[0054] Step S2, obtaining a sensing terminal and a power terminal log of the terminal layer according to the monitoring layout, determining a power running abnormal path in the terminal layer, and adjusting resource input of the edge layer to the terminal layer according to a power terminal layout in the power running abnormal path;

[0055] Step S3, analyzing collected power terminal monitoring data, predicting a potential overload event of the terminal layer, and determining an anomaly point of the terminal layer according to an attribute of the potential overload event;

[0056] Step S4, analyzing task execution data of an associated area of the anomaly point in a cloud layer, determining a task execution conflict event, and sending a task change instruction to the terminal layer through the cloud layer.

[0057] The cloud edge-based power Internet of Things anomaly point identification method identifies the interval of the terminal layer where the non-compliance event occurs according to the operation data of the terminal layer, so as to adjust the monitoring layout of the edge layer to the terminal layer; determines the power operation abnormal path in the terminal layer according to the sensing terminal and power terminal log of the terminal layer, so as to adjust the resource input of the edge layer to the terminal layer, and realizes the accurate configuration of the edge computing resource; analyzes the power terminal monitoring data, predicts the potential overload event of the terminal layer, so as to determine the anomaly point of the terminal layer; analyzes the task execution data of the associated area of the anomaly point in the cloud layer, determines the task execution conflict event, so as to send the task change instruction to the terminal layer through the cloud layer, and realizes efficient and accurate anomaly identification and troubleshooting through clustering and differentiation detection in the terminal layer, and improves the accuracy and real-time performance of the power Internet of Things anomaly point detection.

[0058] Preferably, in step S1, the operation data of the terminal layer of the power Internet of Things is obtained, the power terminal change identification of the terminal layer is performed, and the interval of the terminal layer where the non-compliance event occurs is identified; according to the distribution of the interval, the monitoring layout of the edge layer to the terminal layer is adjusted, including:

[0059] The power terminal operation data of the terminal layer of the power Internet of Things is obtained; wherein the power terminal operation data includes the access and disconnection state change data of the power terminal in the terminal layer; the power terminal operation data is subjected to power grid location and time of occurrence identification, and the interval of the terminal layer where the non-compliance event occurs is identified; wherein the non-compliance event interval refers to the interval of the terminal layer corresponding to the power grid where the power transmission interruption time is too long event;

[0060] According to the distribution position of all intervals where the power transmission interruption time is too long event in the power grid, the power grid is divided into several sub-grid areas; according to the number and type of power terminals under each sub-grid area, the monitoring layout of the edge layer of the power Internet of Things to each sub-grid area is adjusted; wherein the monitoring layout refers to the number and type of edge computers accessed by the edge layer to each sub-grid area.

[0061] In the above technical solution, please refer to Figure 2The power Internet of Things includes a cloud layer, an edge layer, and a terminal layer. The cloud layer includes a cloud server and cloud storage. The edge layer includes a plurality of edge computers, each of which can perform partitioned monitoring on the terminal layer. The terminal layer includes a plurality of power terminals and a plurality of sensing terminals. The power terminals can be, but are not limited to, power generation stations, power transmission lines, power transformation stations, power distribution stations, intelligent devices, and other power generation, power transmission, and power consumption terminals. The sensing terminals can be, but are not limited to, current sensors, voltage sensors, temperature sensors, and other terminals for monitoring the operating state of the power terminals. The terminal layer includes a large number of power interfaces, and the power terminals are connected to the power grid by connecting the power interfaces. The access of the power terminals in the power grid directly affects the operation of the power grid. The layout of the power terminals in the power grid is pre-planned, and the operation of the power terminals in a certain region of the power grid will affect the local or global operation of the power grid. For example, if a certain power terminal connected to the power grid is interrupted for a long time, it will affect the normal operation of the power grid region where the power terminal is located. At this time, the power grid region needs to be monitored and investigated. Therefore, the access and disconnection state change data of all power terminals in the terminal layer of the power Internet of Things are obtained, and the access and disconnection state change data of each power terminal are identified in terms of the power grid location and the occurrence time. The interval in which the terminal layer has a power transmission interruption duration greater than a preset time threshold is determined, which is used as the interval in which the terminal layer has an irregular event. This facilitates the subsequent division of the power grid based on the above interval. According to the distribution of all intervals in which the power transmission interruption time is too long in the power grid, the electrical connection relationship between all intervals in the power grid is determined, and intervals with direct electrical connection relationship are divided into the same sub-grid region, so as to divide the global range of the power grid into a plurality of sub-grid regions. According to the number and type of power terminals under each sub-grid region (such as the operating mode type of the power terminal), the number and type of edge computers connected to each sub-grid region in the edge layer of the power Internet of Things are adjusted. For example, the more the number of power terminals under a sub-grid region, the more the number of edge computers connected to the sub-grid region in the edge layer; the more complex the operating mode of the power terminal under the sub-grid region, the higher the computing power performance of the edge computer connected to the sub-grid region in the edge layer, so as to ensure that the edge layer can fully and timely monitor all sub-grid regions.

[0062] Preferably, in step S2, according to the monitoring layout, the sensing terminals and the power terminals of the terminal layer are obtained, and the power operation abnormal path in the terminal layer is determined; according to the power terminal layout in the power operation abnormal path, the resource input of the edge layer to the terminal layer is adjusted, including:

[0063] According to the monitoring layout, the working log of the sensing terminal and the working log of the power terminal in each sub-network area are obtained, and the working data stream of the sensing terminal and the working data stream of the power terminal are extracted therefrom; the flow of the working data stream of the sensing terminal and the working data stream of the power terminal are compared to determine whether the sub-network area belongs to an abnormal sub-network area; and the power transmission relationship of all abnormal sub-network areas in the power grid is determined to determine the power operation abnormal path in the terminal layer.

[0064] According to the power grid node layout of the power terminal in the power operation abnormal path, the amount of computing resources and bandwidth resources of the edge layer to the corresponding power grid nodes in the terminal layer is adjusted.

[0065] In the above technical solution, the edge layer of the power Internet of Things configures a corresponding number and type of edge computers for each sub-network area, and obtains the working log of the sensing terminal and the working log of the power terminal in each sub-network area. As known from the foregoing description, the sensing terminal in each sub-network area is used to monitor the internal power terminal, and in normal circumstances, the working data stream generated by the sensing terminal and the working data stream generated by the power terminal in each sub-network area should be matched in terms of data volume. When the sub-network area is abnormal, the working data stream generated by the sensing terminal and the working data stream generated by the power terminal in the same sub-network area will not match, and this mismatch mainly reflects that the flow of the working data stream of the sensing terminal and the working data stream of the power terminal changes inconsistently over time. Specifically, the flow of the working data stream of the sensing terminal and the working data stream of the power terminal is extracted from the working log of the sensing terminal and the working log of the power terminal in each sub-network area, respectively, and compared. If they do not match, it is determined that the sub-network area belongs to an abnormal sub-network area; otherwise, it is determined that the sub-network area belongs to a normal sub-network area. The power transmission relationship (such as the power transmission path topological relationship) of all abnormal sub-network areas in the power grid is determined to determine the power operation abnormal path in the terminal layer, wherein the power operation abnormal path includes all abnormal sub-network areas and other sub-network areas having a direct power connection relationship, which can comprehensively check the terminal layer. The layout of all power grid nodes where the power terminals in the power operation abnormal path are located is obtained, and the amount of computing resources and bandwidth resources of the edge computers in the edge layer to the corresponding power grid nodes in the terminal layer is adjusted to ensure that each power grid node is accurately and efficiently monitored by the edge layer.

[0066] Preferably, in step S3, the collected power terminal monitoring data is analyzed to predict potential overload events of the terminal layer; and according to the attributes of the potential overload events, the abnormal points of the terminal layer are determined, including:

[0067] The power terminal monitoring data collected by the edge layer is analyzed to obtain the power operation load transfer trend between the power terminals in the power operation abnormal path, so as to predict potential overload events of the terminal layer;

[0068] According to the time and space attributes of the potential overload event, a fault power terminal in the power operation abnormal path is identified to determine the abnormal point of the terminal layer; wherein the abnormal point refers to the power grid node where the fault power terminal is located and the nodes directly connected thereto.

[0069] In the above technical solution, the edge layer allocates the corresponding computing resource and bandwidth resource to each power grid node to collect the corresponding power terminal monitoring data from the power grid node; the monitoring data can include the current data and voltage data of the power terminal. The monitoring data is analyzed to obtain the power operation load transfer trend between the power terminals in the power operation abnormal path; the power operation load transfer trend refers to the change trend that the power terminal in the power operation abnormal path needs to transfer its current working load to other power terminals due to its overload work. The power terminal to which the working load is transferred is identified according to the power operation load transfer trend between the power terminals in the power operation abnormal path, and a potential overload event of the terminal layer is predicted, i.e., an event corresponding to the power terminal whose received working load exceeds its maximum bearable working load. Then, according to the time and space attributes of the potential overload event, a fault power terminal in the power operation abnormal path is identified to determine the power grid node where the fault power terminal is located and the nodes directly connected thereto, thereby providing a reliable basis for adjusting the task execution state of the power terminal in the terminal layer.

[0070] Preferably, in step S4, the task execution data of the associated area of the abnormal point is analyzed in the cloud layer to determine the task execution conflict event, and a task change instruction is sent from the cloud layer to the terminal layer, including:

[0071] According to the distribution position of all abnormal points in the power grid, the associated area of the abnormal point in the power grid is demarcated; the task execution data of all power terminals in the associated area of the abnormal point is analyzed in the cloud layer to determine the task execution conflict event occurred by all power terminals in the associated area; wherein the task execution conflict event refers to the power transmission conflict event occurred during the task execution of any two power terminals in the associated area.

[0072] According to the occurrence position and occurrence time of the task execution conflict event, a task suspension execution instruction for one of the two power terminals is sent from the cloud layer to the terminal layer.

[0073] In the technical solution, first, the associated region of the abnormal point in the power grid is demarcated according to the distribution position of all abnormal points in the power grid, wherein the associated region refers to the region inside the power grid that has a direct electrical connection relationship with the abnormal point. Then, the task execution data of all power terminals in the associated region of the abnormal point is analyzed in the cloud layer to identify power transmission conflict events (such as power transmission scheduling disorder events) occurring during the task execution of any two power terminals in the associated region. Then, according to the occurrence position and occurrence time of the task execution conflict event, a task suspension execution instruction for one of the two power terminals is sent from the cloud layer to the terminal layer, and when the terminal layer receives the task suspension execution instruction from the cloud layer, the corresponding power terminal switches to a state of suspending the execution of the current task, effectively suppressing the work disorder of different power terminals in the power grid, improving the accuracy and real-time performance of the abnormal point detection of the power internet of things, and ensuring the normal and stable operation of the power grid.

[0074] Referring to Figure 3 A framework schematic diagram of the cloud-edge-end-based power internet of things abnormal point identification system provided by the embodiment is provided. The cloud-edge-end-based power internet of things abnormal point identification system comprises:

[0075] The terminal layer identification module is configured to acquire the running data of the terminal layer of the power internet of things, perform power terminal change identification on the terminal layer, and identify the interval in which the terminal layer has an irregular event;

[0076] The edge layer adjustment module is configured to adjust the monitoring layout of the edge layer to the terminal layer according to the distribution of the interval;

[0077] The abnormal path determination module is configured to acquire the sensing terminal and power terminal log of the terminal layer according to the monitoring layout, and determine the power operation abnormal path in the terminal layer;

[0078] The resource input adjustment module is configured to adjust the resource input of the edge layer to the terminal layer according to the power terminal layout in the power operation abnormal path;

[0079] The overload event prediction module is configured to analyze the collected power terminal monitoring data and predict potential overload events of the terminal layer;

[0080] The abnormal point determination module is configured to determine the abnormal point of the terminal layer according to the attribute of the potential overload event;

[0081] The task change module is configured to analyze the task execution data of the associated region of the abnormal point in the cloud layer, determine the task execution conflict event, and send a task change instruction from the cloud layer to the terminal layer.

[0082] The cloud edge-based power Internet of Things anomaly point identification system identifies the interval of the terminal layer where the non-compliance event occurs according to the operation data of the terminal layer, so as to adjust the monitoring layout of the edge layer to the terminal layer; determines the power operation abnormal path in the terminal layer according to the sensing terminal and power terminal log of the terminal layer, so as to adjust the resource input of the edge layer to the terminal layer, and realizes the precise configuration of the edge computing resource; analyzes the power terminal monitoring data, predicts the potential overload event of the terminal layer, so as to determine the anomaly point of the terminal layer; analyzes the task execution data of the associated area of the anomaly point in the cloud layer, determines the task execution conflict event, so as to send the task change instruction to the terminal layer through the cloud layer, and realizes efficient and accurate anomaly identification and troubleshooting through clustering and differentiation detection in the terminal layer, and improves the accuracy and real-time performance of the power Internet of Things anomaly point detection.

[0083] Preferably, the terminal layer identification module is configured to obtain the operation data of the terminal layer of the power Internet of Things, perform power terminal change identification on the terminal layer, and identify the interval of the terminal layer where the non-compliance event occurs, including:

[0084] Obtain the power terminal operation data of the terminal layer of the power Internet of Things; wherein the power terminal operation data includes access and disconnection state change data of the power terminal in the terminal layer; perform power grid location and time of occurrence identification on the power terminal operation data, and identify the interval of the terminal layer where the non-compliance event occurs; wherein the non-compliance event interval refers to the interval of the terminal layer corresponding to the power grid where the power transmission interruption time is too long event;

[0085] The edge layer adjustment module is configured to adjust the monitoring layout of the edge layer to the terminal layer according to the distribution of the interval, including:

[0086] According to the distribution position of all intervals where the power transmission interruption time is too long event occurs in the power grid, the power grid is divided into several sub-network areas; according to the number and type of power terminals under each sub-network area, the monitoring layout of the edge layer of the power Internet of Things to each sub-network area is adjusted; wherein the monitoring layout refers to the number and type of edge computers accessed by the edge layer to each sub-network area.

[0087] In the above technical solution, please refer to Figure 2The power Internet of Things includes a cloud layer, an edge layer, and a terminal layer. The cloud layer includes a cloud server and cloud storage. The edge layer includes a plurality of edge computers, each of which can perform partitioned monitoring on the terminal layer. The terminal layer includes a plurality of power terminals and a plurality of sensing terminals. The power terminals can be, but are not limited to, power generation stations, power transmission lines, power transformation stations, power distribution stations, intelligent devices, and other power generation, power transmission, and power consumption terminals. The sensing terminals can be, but are not limited to, current sensors, voltage sensors, temperature sensors, and other terminals for monitoring the operating state of the power terminals. The terminal layer includes a large number of power interfaces, and the power terminals are connected to the power grid by connecting the power interfaces. The access of the power terminals in the power grid directly affects the operation of the power grid. The layout of the power terminals in the power grid is pre-planned, and the operation of the power terminals in a certain region of the power grid will affect the local or global operation of the power grid. For example, if a certain power terminal connected to the power grid is interrupted for a long time, it will affect the normal operation of the power grid region where the power terminal is located. At this time, the power grid region needs to be monitored and investigated. Therefore, the access and disconnection state change data of all power terminals in the terminal layer of the power Internet of Things are obtained, and the access and disconnection state change data of each power terminal are identified in terms of the power grid location and the occurrence time. The interval where the power terminal corresponding to the terminal layer has a power transmission interruption duration greater than a preset time threshold is determined, which is used as the interval where the terminal layer has an irregular event. This facilitates the subsequent division of the power grid based on the above interval. According to the distribution of all intervals where the power transmission interruption time is too long in the power grid, the electrical connection relationship between all intervals in the power grid is determined, and intervals with direct electrical connection relationship are divided into the same sub-grid region, so as to divide the global range of the power grid into a plurality of sub-grid regions. According to the number and type of power terminals (such as the operating mode type of the power terminal) under each sub-grid region, the number and type of edge computers connected to each sub-grid region in the edge layer of the power Internet of Things are adjusted. For example, the more the number of power terminals under the sub-grid region, the more the number of edge computers connected to the sub-grid region in the edge layer; the more complex the operating mode of the power terminal under the sub-grid region, the higher the computing power performance of the edge computer connected to the sub-grid region in the edge layer, so as to ensure that the edge layer can fully and timely monitor all sub-grid regions.

[0088] Preferably, the abnormal path determination module is configured to obtain the logs of the sensing terminals and the power terminals in the terminal layer according to the monitoring layout, and determine the power operation abnormal path in the terminal layer, including:

[0089] According to the monitoring layout, the working log of the sensing terminal and the working log of the power terminal in each sub-network area are obtained, and the working data stream of the sensing terminal and the working data stream of the power terminal are extracted therefrom; the flow of the working data stream of the sensing terminal and the working data stream of the power terminal is compared to determine whether the sub-network area belongs to an abnormal sub-network area; and the power transmission relationship of all abnormal sub-network areas in the power grid is determined to determine the power operation abnormal path in the terminal layer.

[0090] The resource investment adjustment module is configured to adjust the resource investment of the edge layer to the terminal layer according to the power terminal layout in the power operation abnormal path, including:

[0091] According to the power terminal layout in the power operation abnormal path, the amount of computing resource and bandwidth resource of the edge layer to the corresponding power grid node in the terminal layer is adjusted.

[0092] In the above technical solution, the edge layer of the power Internet of Things configures a corresponding number and type of edge computers for each sub-network area, and obtains the working log of the sensing terminal and the working log of the power terminal in each sub-network area. As known from the foregoing description, the sensing terminal in each sub-network area is used to monitor the internal power terminal, and in a normal case, the working data stream generated by the sensing terminal and the working data stream generated by the power terminal in each sub-network area should be matched in terms of data volume. When the sub-network area is abnormal, the working data stream generated by the sensing terminal and the working data stream generated by the power terminal in the same sub-network area will not match, and this mismatch mainly reflects that the flow of the working data stream of the sensing terminal and the working data stream of the power terminal changes inconsistently over time. Specifically, the flow of the working data stream of the sensing terminal and the working data stream of the power terminal is extracted from the working log of the sensing terminal and the working log of the power terminal in each sub-network area, respectively, and compared. If the two do not match, it is determined that the sub-network area belongs to an abnormal sub-network area; otherwise, it is determined that the sub-network area belongs to a normal sub-network area. The power transmission relationship (such as the power transmission path topological relationship) of all abnormal sub-network areas in the power grid is determined to determine the power operation abnormal path in the terminal layer, wherein the power operation abnormal path includes all abnormal sub-network areas and other sub-network areas having a direct power connection relationship, so that the terminal layer can be fully investigated. Furthermore, the layout of all power grid nodes where the power terminals in the power operation abnormal path are located is obtained, and the amount of computing resource and bandwidth resource of the edge computer in the edge layer to the corresponding power grid node in the terminal layer is adjusted to ensure that each power grid node is accurately and efficiently monitored by the edge layer.

[0093] Preferably, the overload event prediction module is configured to analyze the collected power terminal monitoring data to predict potential overload events of the terminal layer, including:

[0094] The terminal layer is analyzed according to the power terminal monitoring data collected by the edge layer, and the power operation load transfer trend between the power terminals in the power operation abnormal path is obtained, so as to predict the potential overload event of the terminal layer;

[0095] The abnormal point determination module is configured to determine the abnormal point of the terminal layer according to the attribute of the potential overload event, including:

[0096] According to the occurrence time and spatial attribute of the potential overload event, the power terminal that fails in the power operation abnormal path is identified, so as to determine the abnormal point of the terminal layer; wherein the abnormal point refers to the power grid node where the power terminal fails and the node directly connected thereto.

[0097] In the above technical solution, the edge layer allocates corresponding computing resources and bandwidth resources to each power grid node, so as to collect corresponding power terminal monitoring data from the power grid node. The monitoring data can include current data and voltage data of the power terminal. The monitoring data is analyzed to obtain the power operation load transfer trend between the power terminals in the power operation abnormal path. The power operation load transfer trend refers to the change trend that the power terminal in the power operation abnormal path needs to transfer its current working load to other power terminals due to its overload work. The power terminal to which the working load is transferred is identified according to the power operation load transfer trend between the power terminals in the power operation abnormal path, and the potential overload event of the terminal layer is predicted, i.e. the event corresponding to the power terminal whose received working load exceeds its maximum tolerable working load. According to the occurrence time and spatial attribute of the potential overload event, the power terminal that fails in the power operation abnormal path is identified, so as to determine the power grid node where the power terminal that fails in the terminal layer is located and the node directly connected thereto, thereby providing a reliable basis for subsequent adjustment of the task execution state of the power terminal in the terminal layer.

[0098] Preferably, the task change module is configured to analyze the task execution data of the associated region of the abnormal point in the cloud layer, determine the task execution conflict event, and send a task change instruction to the terminal layer through the cloud layer, including:

[0099] According to the distribution position of all abnormal points in the power grid, the associated region of the abnormal point in the power grid is demarcated; the task execution data of all power terminals in the associated region of the abnormal point is analyzed in the cloud layer, and the task execution conflict event occurring in all power terminals in the associated region is determined; wherein the task execution conflict event refers to the power transmission conflict event occurring during the task execution of any two power terminals in the associated region;

[0100] According to the occurrence position and occurrence time of the task execution conflict event, a task suspension execution instruction is sent to one of the two power terminals through the cloud layer.

[0101] In the technical solution, first, the correlation area of the abnormal points in the power grid is determined according to the distribution positions of all the abnormal points in the power grid, wherein the correlation area refers to an area in the power grid that has a direct electrical connection relationship with the abnormal points. Then, the task execution data of all the power terminals in the correlation area of the abnormal points is analyzed in the cloud layer to identify power transmission conflict events (such as power transmission scheduling disorder events) that occur during the task execution of any two power terminals in the correlation area. Then, according to the occurrence position and occurrence time of the task execution conflict events, a task suspension execution instruction for one of the two power terminals is sent from the cloud layer to the terminal layer, and when the terminal layer receives the task suspension execution instruction from the cloud layer, the corresponding power terminal switches to a state of suspending the execution of the current task, thereby effectively suppressing the work disorder of different power terminals in the power grid, improving the accuracy and real-time performance of the abnormal point detection of the power Internet of Things, and ensuring the normal and stable operation of the power grid.

[0102] From the content of the above embodiment, it can be known that the cloud edge terminal-based abnormal point identification method and system of the power Internet of Things identifies the interval in which the terminal layer has an irregular event according to the operation data of the terminal layer, thereby adjusting the monitoring layout of the edge layer to the terminal layer; determines the power operation abnormal path in the terminal layer according to the sensing terminal and power terminal logs of the terminal layer, thereby adjusting the resource input of the edge layer to the terminal layer, and realizing the accurate configuration of the edge computing resources; analyzes the power terminal monitoring data, predicts the potential overload event of the terminal layer, and thereby determines the abnormal points of the terminal layer; analyzes the task execution data of the correlation area of the abnormal points in the cloud layer, determines the task execution conflict event, and thereby sends a task change instruction from the cloud layer to the terminal layer, realizes efficient and accurate abnormal identification and investigation through clustering and differentiation detection in the terminal layer, and improves the accuracy and real-time performance of the abnormal point detection of the power Internet of Things.

[0103] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for identifying anomalies in the power Internet of Things based on cloud-edge-device architecture, characterized in that, It includes the following steps: Step S1: Obtain the operation data of the terminal layer of the power Internet of Things, identify power terminal changes in the terminal layer, and identify the intervals in which non-compliant events occur in the terminal layer; Based on the distribution of the intervals, adjust the monitoring layout of the edge layer on the terminal layer, including: Acquire power terminal operation data of the terminal layer of the power Internet of Things; wherein, the power terminal operation data includes access and disconnection status change data of power terminals within the terminal layer; identify the location and time of occurrence of the power grid in the power terminal operation data, and identify the interval in which non-compliant events occur in the terminal layer; wherein, the non-compliant event interval refers to the interval in which the power grid corresponding to the terminal layer experiences an event of excessively long power transmission interruption time; Based on the distribution of all intervals where power transmission interruption events with excessively long durations occur within the power grid, the power grid is divided into several sub-grid areas. Based on the number and type of power terminals under each sub-grid area, the monitoring layout of the edge layer of the power Internet of Things for each sub-grid area is adjusted. Herein, the monitoring layout refers to the number and type of edge computers accessed by the edge layer for each sub-grid area. Step S2: Based on the monitoring layout, obtain the sensor terminal and power terminal logs of the terminal layer, and determine the abnormal power operation path within the terminal layer; based on the power terminal layout within the abnormal power operation path, adjust the resource allocation of the edge layer to the terminal layer, including: Based on the monitoring layout, the sensor terminal operation log and power terminal operation log of each subnet area are obtained, and the sensor terminal operation data stream and power terminal operation data stream are extracted from them; the flow rates of the sensor terminal operation data stream and the power terminal operation data stream are compared to determine whether the subnet area belongs to an abnormal subnet area; based on the power transmission relationship of all abnormal subnet areas in the power grid, the abnormal power operation path in the terminal layer is determined; wherein, the abnormal power operation path includes all abnormal subnet areas and other subnet areas with direct power connection relationships. Based on the power grid node layout of the power terminal within the abnormal power operation path, the edge layer adjusts the amount of computing power and bandwidth resources invested in the corresponding power grid nodes within the terminal layer. Step S3: Analyze the collected power terminal monitoring data to predict potential overload events at the terminal layer; determine the anomalies at the terminal layer based on the attributes of the potential overload events. Step S4: Analyze the task execution data of the associated area of ​​the anomaly point in the cloud layer to determine the task execution conflict event, and then send a task change instruction to the terminal layer through the cloud layer.

2. The power IoT anomaly identification method based on cloud-edge-device as described in claim 1, characterized in that: In step S3, the collected power terminal monitoring data is analyzed to predict potential overload events at the terminal layer. Based on the attributes of the potential overload events, the anomalies in the terminal layer are determined, including: By analyzing the power terminal monitoring data collected by the edge layer, the power load transfer trend between power terminals within the abnormal power operation path is obtained, thereby predicting potential overload events of the terminal layer. Based on the occurrence time and spatial attributes of the potential overload event, identify the power terminal that has failed within the abnormal power operation path, thereby determining the abnormal point of the terminal layer; wherein, the abnormal point refers to the power grid node where the power terminal that failed is located and the nodes directly connected to it.

3. The power IoT anomaly identification method based on cloud-edge-device as described in claim 2, characterized in that: In step S4, the task execution data of the associated area of ​​the anomaly point is analyzed in the cloud to determine the task execution conflict event, and then a task change instruction is sent to the terminal layer through the cloud, including: Based on the distribution of all anomalies in the power grid, the associated regions of the anomalies within the power grid are delineated; the task execution data of all power terminals within the associated regions of the anomalies are analyzed in the cloud to determine the task execution conflict events that occur in all power terminals within the associated regions; wherein, the task execution conflict event refers to the power transmission conflict event that occurs during the execution of tasks by any two power terminals within the associated regions. Based on the location and time of the task execution conflict event, a task suspension command corresponding to one of the two power terminals is sent from the cloud layer to the terminal layer.

4. A power IoT anomaly identification system based on cloud-edge-device architecture, characterized in that: include: The terminal layer identification module is used to acquire operational data of the terminal layer of the power Internet of Things, identify power terminal changes in the terminal layer, and identify the intervals in which non-compliant events occur in the terminal layer, including: Acquire power terminal operation data of the terminal layer of the power Internet of Things; wherein, the power terminal operation data includes access and disconnection status change data of power terminals within the terminal layer; identify the location and time of occurrence of the power grid in the power terminal operation data, and identify the interval in which non-compliant events occur in the terminal layer; wherein, the non-compliant event interval refers to the interval in which the power grid corresponding to the terminal layer experiences an event of excessively long power transmission interruption time; An edge layer adjustment module is used to adjust the monitoring layout of the edge layer relative to the terminal layer according to the distribution of the intervals, including: Based on the distribution of all intervals where power transmission interruption events with excessively long durations occur within the power grid, the power grid is divided into several sub-grid areas. Based on the number and type of power terminals under each sub-grid area, the monitoring layout of the edge layer of the power Internet of Things for each sub-grid area is adjusted. Herein, the monitoring layout refers to the number and type of edge computers accessed by the edge layer for each sub-grid area. An abnormal path determination module is used to obtain the sensor terminal and power terminal logs of the terminal layer according to the monitoring layout, and determine the abnormal power operation path within the terminal layer, including: Based on the monitoring layout, the sensor terminal operation log and power terminal operation log of each subnet area are obtained, and the sensor terminal operation data stream and power terminal operation data stream are extracted from them; the flow rates of the sensor terminal operation data stream and the power terminal operation data stream are compared to determine whether the subnet area belongs to an abnormal subnet area; based on the power transmission relationship of all abnormal subnet areas in the power grid, the abnormal power operation path in the terminal layer is determined; wherein, the abnormal power operation path includes all abnormal subnet areas and other subnet areas with direct power connection relationships. The resource allocation adjustment module is used to adjust the resource allocation of the edge layer to the terminal layer based on the power terminal layout within the abnormal power operation path, including: Based on the power grid node layout of the power terminal within the abnormal power operation path, the edge layer adjusts the amount of computing power and bandwidth resources invested in the corresponding power grid nodes within the terminal layer. The overload event prediction module is used to analyze the collected power terminal monitoring data and predict potential overload events at the terminal layer. An anomaly detection module is used to determine anomalies in the terminal layer based on the attributes of the potential overload event. The task change module is used to analyze the task execution data of the associated area of ​​the anomaly point in the cloud layer, determine the task execution conflict event, and then send the task change instruction to the terminal layer through the cloud layer.

5. The power IoT anomaly identification system based on cloud-edge-device as described in claim 4, characterized in that: The overload event prediction module is used to analyze collected power terminal monitoring data and predict potential overload events at the terminal layer, including: By analyzing the power terminal monitoring data collected by the edge layer, the power load transfer trend between power terminals within the abnormal power operation path is obtained, thereby predicting potential overload events of the terminal layer. The anomaly detection module is used to determine the anomaly points of the terminal layer based on the attributes of the potential overload event, including: Based on the occurrence time and spatial attributes of the potential overload event, identify the power terminal that has failed within the abnormal power operation path, thereby determining the abnormal point of the terminal layer; wherein, the abnormal point refers to the power grid node where the power terminal that failed is located and the nodes directly connected to it.

6. The power IoT anomaly identification system based on cloud-edge-device as described in claim 5, characterized in that: The task change module is used to analyze the task execution data of the associated area of ​​the anomaly point in the cloud layer, determine the task execution conflict event, and then send a task change instruction to the terminal layer through the cloud layer, including: Based on the distribution of all anomalies in the power grid, the associated regions of the anomalies within the power grid are delineated; the task execution data of all power terminals within the associated regions of the anomalies are analyzed in the cloud to determine the task execution conflict events that occur in all power terminals within the associated regions; wherein, the task execution conflict event refers to the power transmission conflict event that occurs during the execution of tasks by any two power terminals within the associated regions. Based on the location and time of the task execution conflict event, a task suspension command corresponding to one of the two power terminals is sent from the cloud layer to the terminal layer.

Images