A monitoring method and system for power grid terminal equipment

Abstract

The present invention discloses a monitoring method and system for power grid terminal equipment, which obtains current monitoring data of the power grid terminal equipment through a monitoring terminal, pre-processes the current monitoring data to obtain current processing data, stores the current processing data and sends it to a data middle station; sends an analysis model to a target edge server through the data middle station, and the analysis model is obtained by model training of the data middle station based on historical processing data; performs a first data analysis on the current processing data based on the analysis model through the target edge server to determine a first analysis result, determines a control instruction based on the first analysis result, and sends the control instruction to the monitoring terminal, so that the monitoring terminal controls the power grid terminal equipment. The data middle station and the edge server can collaborate to combine the characteristics of short edge computing task scheduling delay with the characteristics of strong data processing capability of the data middle station, so as to solve the problems of high broadband occupancy and high latency when cloud computing and edge computing run independently.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid monitoring, and in particular to a monitoring method and system for power grid terminal equipment. Background Art

[0002] Cloud computing is an expansion and interaction model based on Internet technology. The data middle platform based on cloud computing has powerful computing capabilities and can realize cross-domain data scheduling.

[0003] However, with the gradual deepening of the Internet of Things in the industry, the number of terminal devices has increased dramatically. If the data calculations of the huge number of user-side terminal devices rely entirely on the data middle platform, it will occupy a large amount of network bandwidth, resulting in bottlenecks in cloud computing performance and high latency, and will be unable to meet the real-time monitoring requirements of power grid big data. Summary of the invention

[0004] The present invention provides a monitoring method and system for power grid terminal equipment to solve the problems of high bandwidth occupancy and high latency in power grid monitoring that completely relies on a data middle station. The data middle station and edge computing collaborate to alleviate the computing pressure of the data middle station in multi-dimensional data processing.

[0005] According to one aspect of the present invention, a method for monitoring a power grid terminal device is provided, comprising:

[0006] Acquire current monitoring data of the power grid terminal device through the monitoring terminal, pre-process the current monitoring data to obtain current processed data, store the current processed data and send it to the data middle station;

[0007] The analysis model is sent to the target edge server through the data center, and the analysis model is obtained by the data center through model training based on historical processing data;

[0008] The target edge server obtains the current processing data from the monitoring terminal, performs a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determines a control instruction based on the first analysis result, and sends the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device.

[0009] Furthermore, after sending the control instruction to the monitoring terminal, the method further includes:

[0010] Determining incremental data based on the current processing data and the historical processing data through the target edge server, and sending the incremental data to the data middle station;

[0011] A second data analysis is performed on the incremental data through the data center to determine a second analysis result, and the analysis model is updated based on the second analysis result.

[0012] Furthermore, after the currently processed data is stored and sent to the data center, and before the analysis model is sent to the target edge server through the data center, it also includes:

[0013] Obtaining the current load of the edge server to which the monitoring terminal is connected through the data center;

[0014] If the current load of the edge server is greater than the underload threshold and less than the overload threshold, determining the edge server as a target edge server;

[0015] If the current load of the edge server is greater than or equal to the overload threshold, the data center controls the migration of the currently processed data from the edge server to any target edge server;

[0016] If the current load of the edge server is less than or equal to the underload threshold, the data center controls the migration of the currently processed data stored in the edge server to any target edge server, and controls the edge server to shut down;

[0017] The target edge server is an edge server whose current load is greater than the underload threshold and less than the overload threshold.

[0018] Furthermore, obtaining the current monitoring data of the power grid terminal device through the monitoring terminal includes:

[0019] Collect monitoring data of power grid terminal equipment in real time through a data collector, and store the monitoring data in a heterogeneous database;

[0020] A data query instruction is obtained through a monitoring terminal, and current monitoring data is obtained from a heterogeneous database in a global search manner based on the data query instruction.

[0021] Further, after preprocessing the current monitoring data to obtain current processed data, the method further includes:

[0022] storing the currently processed data in a heterogeneous database;

[0023] Correspondingly, obtaining the current processing data from the monitoring terminal through the target edge server includes:

[0024] The target edge server uses a local search method to obtain the current processing data stored in the heterogeneous database.

[0025] Further, the preprocessing of the current monitoring data to obtain current processed data includes:

[0026] The current monitoring data is sequentially subjected to data filtering, data cleaning and symmetrical encryption to obtain current processing data.

[0027] Furthermore, it also includes:

[0028] The first analysis result is synchronized to the data center through the target edge server.

[0029] According to another aspect of the present invention, there is provided a monitoring system for a power grid terminal device, comprising: a data middle station, a target edge server, a monitoring terminal and a power grid terminal device;

[0030] The monitoring terminal is used to obtain current monitoring data of the power grid terminal device, pre-process the current monitoring data to obtain current processed data, store the current processed data and send it to the data center;

[0031] The data middle station is used to send the analysis model to the target edge server, and the analysis model is obtained by the data middle station through model training based on historical processing data;

[0032] The target edge server is used to obtain the current processing data from the monitoring terminal, perform a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determine a control instruction based on the first analysis result, and send the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device.

[0033] The technical solution of the embodiment of the present invention obtains the current monitoring data of the power grid terminal equipment through the monitoring terminal, pre-processes the current monitoring data to obtain the current processing data, stores the current processing data and sends it to the data middle station; sends the analysis model to the target edge server through the data middle station, and the analysis model is obtained by model training of the data middle station based on historical processing data; the target edge server performs a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determines a control instruction based on the first analysis result, and sends the control instruction to the monitoring terminal, so that the monitoring terminal controls the power grid terminal equipment, which solves the problem of high broadband occupancy and high latency when cloud computing and edge computing run independently, and achieves the beneficial effect of combining the characteristics of short edge computing task scheduling delay and strong data processing capability of the data middle station through the collaboration of the data middle station and the edge server.

[0034] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 is a flow chart of a monitoring method for a power grid terminal device provided according to Embodiment 1 of the present invention;

[0037] Figure 2 This is an architecture diagram of a monitoring system for a power grid terminal device provided according to Embodiment 1 of the present invention;

[0038] Figure 3 is a flow chart of a monitoring method for a power grid terminal device provided according to a second embodiment of the present invention;

[0039] Figure 4 It is a structural diagram of a monitoring system for power grid terminal equipment provided according to Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Embodiment 1

[0043] Figure 1 A flowchart of a monitoring method for a power grid terminal device is provided for embodiment 1 of the present invention. This embodiment is applicable to the situation of real-time monitoring of terminal devices in a power grid. The method can be executed by a monitoring system of the power grid terminal device, and the monitoring system can be configured in the power grid system.

[0044] Figure 2 is a monitoring system for a power grid terminal device in the first embodiment of the present invention, such as Figure 2 As shown, the monitoring system of the power grid terminal equipment includes: a monitoring terminal, an edge server and a data middle station, and the monitoring terminal is connected to the data middle station and the edge server respectively, wherein the monitoring terminal is connected to an edge server, and an edge server is connected to multiple monitoring terminals. The connection method can be wired and / or wireless. The wireless method can use industrial Ethernet, 5G, Internet of Things, WiFi, Bluetooth wireless network to build a safe, reliable, and high-speed network system, and use protocols such as Profinet / Modbus / Ethercat, OPC UA, TCP / IP, MQTT, etc. for data transmission.

[0045] Among them, the monitoring terminal is a terminal device in the power grid terminal equipment that integrates data collection, data processing, display, control and human-computer interaction. The edge server refers to an open platform that integrates network, computing, storage, and application core capabilities on the side close to the terminal device. It can provide nearby services through local devices. Because it is closer to the power grid terminal device side, it can provide faster response. The data center can be a data center platform such as a cloud computing center, which has powerful computing capabilities and can realize cross-domain data scheduling.

[0046] like Figure 1 As shown, the method includes:

[0047] S110. Acquire current monitoring data of the power grid terminal device through the monitoring terminal, pre-process the current monitoring data to obtain current processed data, store the current processed data and send it to the data middle station.

[0048] Among them, the power grid terminal equipment refers to the monitored object in the monitoring system of the power grid terminal equipment, which can be the distribution equipment in the power grid or the user-side power equipment, for example, it can include distribution network transformers, distribution network lines, motor equipment, industrial equipment, etc.; the current monitoring data of the power grid terminal equipment can be the data of the power grid terminal equipment monitored at the current moment, for example, including voltage, current, active power, reactive power, harmonics, temperature, humidity and flow, etc.

[0049] The target edge server may be the edge server to which the monitoring terminal is connected, or may be another edge server that meets the preset conditions when the edge server to which the monitoring terminal is connected does not meet the preset conditions. The preset conditions may be conditions set according to actual needs, such as computing resources.

[0050] Specifically, the current monitoring data of the power grid terminal equipment is obtained through the monitoring terminal, and the current monitoring data is preprocessed to obtain the current processing data. The preprocessing may include: data filtering, data cleaning, data encryption and other operations. The current processing data obtained after preprocessing is stored and sent to the data middle station. The storage location of the current processing data can be a database or a local storage space of the monitoring terminal.

[0051] S120. Send the analysis model to the target edge server through the data center. The analysis model is obtained by model training based on historical processing data by the data center.

[0052] The data center obtains an analysis model by training the model through a machine learning method based on the processed data sent by the monitoring terminal in the historical time period. The model training method can adopt any existing model training method, and the analysis model is determined based on the type of monitoring data, which is not limited in the embodiment of the present invention.

[0053] The data center is positioned to provide data sharing and analysis application services for various professions and units. Based on the data analysis domain and management domain, it accumulates common data service capabilities and meets the needs of data sharing, value mining, analysis application and integration across professions horizontally and across different levels vertically through data services.

[0054] Specifically, the data center performs model training based on historical processing data to obtain an analysis model, and sends the analysis model to the target edge server, so that the target edge server performs data analysis on the current processing data based on the analysis model. Performing model training work with large computational workload in the data center can save server computing resources in the power grid.

[0055] S130. Performing a first data analysis on the currently processed data based on the analysis model through the target edge server to determine a first analysis result, determining a control instruction based on the first analysis result, and sending the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device.

[0056] Among them, the first data analysis is the operation of performing data analysis on the current processed data through the analysis model; the result output after the current processed data is input into the analysis model is the first analysis result, and the first analysis result can be used to reflect the power consumption information such as the power load and power consumption status of the power grid terminal equipment, for example, it can be voltage fluctuations, harmonic content and load forecast.

[0057] Specifically, the target edge server performs a first data analysis on the currently processed data based on the analysis model sent by the data middle station to obtain a first analysis result. Based on the first analysis result, the control instruction can be determined by querying a pre-stored control instruction mapping table, and the control instruction is sent to the monitoring terminal. The monitoring terminal can control the power grid terminal device according to the control instruction, thereby realizing automatic monitoring and control of the power grid terminal device.

[0058] The technical solution of this embodiment obtains the current monitoring data of the power grid terminal device through the monitoring terminal, pre-processes the current monitoring data to obtain the current processing data, stores the current processing data and sends it to the data middle station; sends the analysis model to the target edge server through the data middle station, and the analysis model is obtained by model training based on historical processing data by the data middle station; the target edge server performs a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determines a control instruction based on the first analysis result, and sends the control instruction to the monitoring terminal, so that the monitoring terminal controls the power grid terminal device. The data middle station and the edge server can collaborate to combine the short delay of edge computing task scheduling with the strong data processing capability of the data middle station. The target edge server processes data computing work within the local range, and the processed data is aggregated from the edge node to the data middle station. The data middle station performs big data mining and analysis, data application and model training, so as to solve the problems of high broadband occupancy and high latency when cloud computing and edge computing run independently.

[0059] Optionally, after sending the control instruction to the monitoring terminal, the method further includes:

[0060] Determine the incremental data based on the current processing data and historical processing data through the target edge server, and send the incremental data to the data center;

[0061] A second data analysis is performed on the incremental data through the data center to determine a second analysis result, and the analysis model is updated based on the second analysis result.

[0062] Among them, incremental data is used to represent new data relative to historical processing data. Therefore, by comparing the current processing data with the historical processing data, the data that does not contain the historical processing data in the current processing data is determined as incremental data. The second data analysis can use any existing data analysis method to analyze the incremental data to obtain the second analysis result, and the embodiment of the present invention does not limit the second data analysis. The second analysis result is used to reflect the impact of the state of the power grid terminal equipment on the parameters of the analysis model.

[0063] Specifically, the target edge server determines the incremental data and sends the incremental data to the data center. The data center performs a second data analysis on the incremental data to determine a second analysis result, and updates the analysis model according to the second analysis result. After each update of the analysis model, the data center needs to send the updated analysis model to the target edge server.

[0064] The advantage of doing this is that by updating the analysis model based on incremental data through the data middle platform, and pushing the updated analysis model to the target edge server, autonomous learning and updating can be achieved in a closed loop.

[0065] Embodiment 2

[0066] Figure 3 A flowchart of a monitoring method for a power grid terminal device provided in Embodiment 2 of the present invention. The relationship between this embodiment and the above embodiment is that after the currently processed data is stored and sent to the data middle station, and before the analysis model is sent to the target edge server through the data middle station, a technical feature for determining the target edge server is added, specifically including:

[0067] Obtain the current load of the edge server to which the monitoring terminal is connected through the data center;

[0068] If the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as the target edge server;

[0069] If the current load of the edge server is greater than or equal to the overload threshold, the data center controls the migration of the currently processed data from the edge server to any target edge server;

[0070] If the current load of the edge server is less than or equal to the underload threshold, the data center controls the migration of the currently processed data stored in the edge server to any target edge server, and controls the edge server to shut down;

[0071] The target edge server is an edge server whose current load is greater than an underload threshold and less than an overload threshold.

[0072] like Figure 3 As shown, the method includes:

[0073] S210, obtaining current monitoring data of the power grid terminal device through the monitoring terminal, preprocessing the current monitoring data to obtain current processed data, storing the current processed data and sending it to the data middle station;

[0074] S220. Obtain the current load of the edge server to which the monitoring terminal is connected through the data center.

[0075] Specifically, edge servers can provide users with faster responses through local computing. However, due to the space limitations of physical facilities, the computing capabilities of edge servers vary, which will lead to uneven loads and backlogs of "hot spots" on individual edge servers. Therefore, it is necessary to obtain the current load of the edge server to which the monitoring terminal is connected, and perform data scheduling based on the current load to solve the problem of load imbalance.

[0076] S230: If the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as a target edge server.

[0077] Among them, the underload threshold and the overload threshold can be flexibly set according to the computing capacity of each edge server. For example, the overload threshold can be set to 80% of the full load threshold, and the underload threshold can be set to 20% of the full load threshold. That is, the optimal load operation range of each edge server is between 20% and 80% of the full load.

[0078] Specifically, it is prioritized to determine whether the current load of the edge server to which the monitoring terminal is connected is within the optimal load operating range. If so, that is, the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as the target edge server without the need for data migration; if not, it is necessary to determine the target edge server based on the load conditions of each edge server in the monitoring system of the power grid terminal equipment.

[0079] S240: If the current load of the edge server is greater than or equal to the overload threshold, the data center controls the migration of the currently processed data from the edge server to any target edge server.

[0080] The target edge server is an edge server in the monitoring system of the power grid terminal device, whose current load is greater than the underload threshold and less than the overload threshold.

[0081] Specifically, if the current load of the edge server connected to the monitoring terminal obtained by the data middle station is greater than or equal to the underload threshold, then any target edge server that meets the load conditions in the monitoring system of the power grid terminal equipment is determined, and the data middle station controls the current processing data to migrate from the edge server to the determined target edge server, thereby reducing the load overhead of the edge server to which the monitoring terminal is connected.

[0082] Exemplarily, the method for the data middle station to control the migration of currently processed data from the edge server to the determined target edge server can be: generating a scheduling instruction based on the identifier of the determined target edge server, and sending the scheduling instruction to the edge server to which the monitoring terminal is connected, so that the edge server sends the currently processed data to the target edge server in the scheduling instruction.

[0083] S250: If the current load of the edge server is less than or equal to the underload threshold, the data center controls the migration of the currently processed data stored in the edge server to any target edge server, and controls the edge server to shut down.

[0084] Specifically, if the current load of the edge server connected to the monitoring terminal obtained by the data middle station is less than or equal to the overload threshold, then any target edge server in the monitoring system of the power grid terminal device that meets the load conditions is determined, and all currently processed data stored in the edge server is controlled by the data middle station to be migrated to the determined target edge server, and the edge server connected to the monitoring terminal is controlled to be shut down, thereby reducing the energy consumption overhead of edge computing.

[0085] Exemplarily, the method for the data middle station to control the migration of currently processed data from the edge server to the determined target edge server can be: generating a scheduling instruction based on the identifier of the determined target edge server, and sending the scheduling instruction to the edge server to which the monitoring terminal is connected, so that the edge server sends the currently processed data to the target edge server in the scheduling instruction.

[0086] S260: Send the analysis model to the edge server through the data center. The analysis model is obtained by model training based on historical processing data by the data center.

[0087] S270. Perform a first data analysis on the currently processed data based on the analysis model through the target edge server to determine a first analysis result, determine a control instruction based on the first analysis result, and send the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device.

[0088] The technical solution of this embodiment obtains the current monitoring data of the power grid terminal device through the monitoring terminal, pre-processes the current monitoring data to obtain the current processing data, stores the current processing data and sends it to the data middle station; obtains the current load of the edge server connected to the monitoring terminal through the data middle station; if the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as the target edge server; if the current load of the edge server is greater than or equal to the overload threshold, the current processing data is controlled to migrate from the edge server to any target edge server through the data middle station; if the current load of the edge server is less than or equal to the underload threshold, the current processing data is controlled to migrate from the edge server to any target edge server through the data middle station, and the edge server is controlled to be shut down; the analysis is performed through the data middle station The model is sent to the target edge server, and the analysis model is obtained by the data middle station through model training based on historical processing data; the target edge server performs a first data analysis on the current processing data based on the analysis model to determine the first analysis result, determines the control instruction based on the first analysis result, and sends the control instruction to the monitoring terminal so that the monitoring terminal can control the power grid terminal equipment. Through the collaboration of the data middle station and the edge server, the characteristics of short delay in scheduling edge computing tasks and the characteristics of strong data processing capabilities of the data middle station can be combined. The data computing work within the local range is processed by the target edge server, and the processed data is aggregated from the edge node to the data middle station. The data middle station performs big data mining and analysis, data application and model training, and solves the problems of high broadband occupancy and high latency when cloud computing and edge computing are independently running. At the same time, data migration is performed according to the load of the edge server to solve the problem of load imbalance and keep the load of each edge server in the optimal range.

[0089] Optionally, obtaining current monitoring data of the power grid terminal device through the monitoring terminal includes:

[0090] The monitoring data of power grid terminal equipment is collected in real time through the data collector, and the monitoring data is stored in a heterogeneous database;

[0091] The data query instruction is obtained through the monitoring terminal, and the current monitoring data is obtained from the heterogeneous database by a global search method based on the data query instruction.

[0092] Among them, since there are tens of thousands of power grid terminal devices in the power grid system, including large-scale heterogeneous networks composed of smart terminals and sensors, these heterogeneous networks often belong to different business systems and have different logics. A considerable amount of data is generated at the edge of the network at all times. For a large amount of heterogeneous data, a heterogeneous database is established to integrate heterogeneous data, so that data from different business systems can be directly accessed with a unified data view, unified tools, and unified data analysis and processing methods.

[0093] Specifically, the monitoring data of the power grid terminal device is collected in real time through the data collector, and the monitoring data is stored in the heterogeneous database. The data collector may be a sensor arranged on the power grid terminal device. When the monitoring data is used, the query instruction is obtained through the monitoring terminal, so that the monitoring terminal queries the heterogeneous database based on the data query instruction to obtain the current monitoring data. The query instruction may be generated based on the needs of the user.

[0094] When retrieving information collected by different collectors in different time periods, it is necessary to adopt overall control to achieve large-scale resource scheduling. Therefore, a global search can be used to query heterogeneous databases to obtain current monitoring data. The global search is performed using the artificial bee colony algorithm (CABC) combined with the decision tree C4.5 algorithm.

[0095] Among them, the artificial bee colony algorithm (ABC) is an optimization algorithm proposed by scholars by imitating the activities and behaviors of bee populations. It consists of two elements: nectar sources and artificial bees. Artificial bees are divided into three types of artificial bees: scout bees, follower bees and leader bees. The set of nectar sources is equivalent to the solution set of the problem. The leader bee is responsible for searching around the random nectar source to find a suitable nectar source, and feedback the quality, distance and other information of the nectar source to other bees. The follower bee will selectively choose the leader bee to follow according to probability for further mining. If the nectar source has not improved after several cycles, the scout bee will search for a new feasible solution as a new nectar source in the feasible solution to achieve the role conversion between the three types of bees.

[0096] The idea of ​​the decision tree C4.5 algorithm is introduced to classify individual bees in the population. According to the load classification standard, the leaf node with the most load classification attributes, that is, the lowest impurity (indicating the balanced distribution of the categories of the samples falling on the current node) is marked as the leading bee. The high classification accuracy of the C4.5 algorithm is used to improve the search ability of the artificial bee colony algorithm. When the search range is increased and the bee population increases, the algorithm accuracy can be effectively improved while the search rate can also be improved, resource occupation can be reduced, and cost overhead can be reduced.

[0097] When constructing a decision tree using the C4.5 algorithm, first select a root node to start training. Then select the corresponding features, and classify the samples based on the selected features, which mainly follows the IF-THEN classification rule. If the samples all belong to the same class, they are divided into the same node. If there are no other attributes to continue dividing, they are regarded as leaf nodes, otherwise continue to divide. If they are not in the same class, select the current node with the largest impurity, and then select new features, and classify according to the new features. Until the samples with the same feature attributes are divided into the same class, or there are no attributes to provide samples for further division. Finally, the decision tree pruning process is adopted to prune the obtained decision tree to remove overfitting and complex samples.

[0098] Optionally, after preprocessing the current monitoring data to obtain the current processed data, the following steps are further included:

[0099] Storing currently processed data in heterogeneous databases;

[0100] Accordingly, obtaining the current processing data from the monitoring terminal through the target edge server includes:

[0101] The current processing data stored in the heterogeneous database is obtained by local search through the target edge server.

[0102] For resource scheduling problems of latency-sensitive and bandwidth-intensive tasks, a local search is used to search the current processing data. The local search is performed using a differential artificial bee colony algorithm; based on the artificial bee colony algorithm, the search formula is modified to provide directionality to the search, and the step length is modified to speed up the search efficiency. At the same time, the coefficient factor is increased, and the search step length is adjusted accordingly as the number of iterations increases, which speeds up the search rate and improves the search efficiency in the later stages of the algorithm, while jumping out of the local optimal solution and ensuring the accuracy and stability of the solution.

[0103] For resource scheduling issues that are sensitive to latency and have high bandwidth usage, local range search is used to obtain corresponding data from heterogeneous databases. When retrieving information in different time periods and collected by different collectors, global range search is used to obtain corresponding data from heterogeneous databases. By retrieving data in a hierarchical manner, the search range is reduced, thereby reducing the delay problem of data retrieval.

[0104] Optionally, preprocessing the current monitoring data to obtain current processed data includes:

[0105] The current monitoring data is filtered, cleaned and symmetrically encrypted in turn to obtain the current processing data.

[0106] Among them, data filtering is used to filter redundant data in the current monitoring data, and data cleaning is used to process invalid values ​​and missing values ​​in the current monitoring data. Symmetric encryption is used to encrypt the current monitoring data. The embodiment of the present invention does not limit the specific implementation algorithms of data filtering, data cleaning and symmetric encryption.

[0107] For example, the characteristic of using symmetric encryption technology for data encryption is that the same key is used for encryption and decryption, that is, the encryption key can also be used as the decryption key. This method is called a symmetric encryption algorithm in cryptography. The symmetric encryption algorithm is simple and fast to use, the key is short, and it is difficult to decipher. For example, the DES data encryption algorithm or the IDEA encryption algorithm can be used.

[0108] Optionally, also include:

[0109] The first analysis result is synchronized to the data center through the target edge server.

[0110] Specifically, after determining the first analysis result, the target edge server not only needs to determine the control instruction based on the first analysis result and send the control instruction to the monitoring terminal so that the monitoring terminal can control the power grid terminal equipment; it also needs to synchronously send the first analysis result to the data middle station so that the data middle station can back up the first analysis result.

[0111] Embodiment 3

[0112] Figure 4 This is a schematic diagram of the structure of a monitoring system for a power grid terminal device provided by Embodiment 3 of the present invention. Figure 4 As shown, the system includes: a data center 310, a target edge server 320, a monitoring terminal 330, and a power grid terminal device 340:

[0113] The monitoring terminal 330 is used to obtain current monitoring data of the power grid terminal device, pre-process the current monitoring data to obtain current processing data, store the current processing data and send it to the data center;

[0114] The data center 310 is used to send the analysis model to the target edge server, and the analysis model is obtained by the data center through model training based on historical processing data;

[0115] The target edge server 320 is used to obtain the current processing data from the monitoring terminal, perform a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determine a control instruction based on the first analysis result, and send the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device.

[0116] Optionally, the target edge server 320 is further used for:

[0117] Determine incremental data based on the current processing data and the historical processing data, and send the incremental data to the data middle station;

[0118] The data center 310 is also used for:

[0119] A second data analysis is performed on the incremental data to determine a second analysis result, and the analysis model is updated based on the second analysis result.

[0120] Optionally, the data center 310 is further used for:

[0121] Obtaining a current load of an edge server to which the monitoring terminal is connected;

[0122] If the current load of the edge server is greater than the underload threshold and less than the overload threshold, determining the edge server as a target edge server;

[0123] If the current load of the edge server is greater than or equal to the overload threshold, the data center controls the migration of the currently processed data from the edge server to any target edge server;

[0124] If the current load of the edge server is less than or equal to the underload threshold, the data center controls the migration of the currently processed data stored in the edge server to any target edge server, and controls the edge server to shut down;

[0125] The target edge server is an edge server whose current load is greater than the underload threshold and less than the overload threshold.

[0126] Optionally, it also includes: data collector,

[0127] The data collector is used to collect monitoring data of power grid terminal equipment in real time and store the monitoring data in a heterogeneous database;

[0128] The monitoring terminal is used to obtain a data query instruction, and based on the data query instruction, a global search method is adopted to obtain current monitoring data from a heterogeneous database.

[0129] Optionally, the monitoring terminal 330 is further used for:

[0130] After preprocessing the current monitoring data to obtain current processed data, storing the current processed data in a heterogeneous database;

[0131] Accordingly, the target edge server 320 is specifically used for:

[0132] The target edge server uses a local search method to obtain the current processing data stored in the heterogeneous database.

[0133] Optionally, the monitoring terminal 330 is specifically used for:

[0134] The current monitoring data is sequentially subjected to data filtering, data cleaning and symmetrical encryption to obtain current processing data.

[0135] Optionally, the target edge server 320 is further used for:

[0136] Synchronize the first analysis result to the data center.

[0137] The monitoring system of the power grid terminal equipment provided in the embodiment of the present invention can execute the monitoring method of the power grid terminal equipment provided in any embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method.

[0138] It should be noted that the data center 310, the edge server 320, the monitoring terminal 330 and the power grid terminal device 340 all have a processor for controlling itself and the memory to implement the steps of the monitoring method embodiment of any of the above-mentioned power grid terminal devices. The processor may also be referred to as a CPU (Central Processing Unit). The processor may be an integrated circuit chip with signal processing capabilities. The processor may also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. In addition, the processor may be implemented by an integrated circuit chip.

[0139] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0140] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A monitoring method for a power grid terminal device, characterized in that: include: Acquire current monitoring data of the power grid terminal device through the monitoring terminal, pre-process the current monitoring data to obtain current processing data, store the current processing data in a heterogeneous database and send it to the data center; The current load of the edge server to which the monitoring terminal is connected is obtained through the data middle station; if the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as the target edge server; if the current load of the edge server is greater than or equal to the overload threshold, the current processing data is controlled to migrate from the edge server to any target edge server through the data middle station; if the current load of the edge server is less than or equal to the underload threshold, the current processing data stored in the edge server is controlled to migrate to any target edge server through the data middle station, and the edge server is controlled to be shut down; wherein, the target edge server is an edge server whose current load is greater than the underload threshold and less than the overload threshold; The analysis model is sent to the target edge server through the data center, and the analysis model is obtained by the data center through model training based on historical processing data; Acquire the current processing data from the monitoring terminal through the target edge server, perform a first data analysis on the current processing data based on the analysis model to determine a first analysis result, determine a control instruction based on the first analysis result, and send the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device; The controlling the migration of the currently processed data from the edge server to any target edge server through the data middle station includes: generating a scheduling instruction based on the determined identifier of the target edge server, and sending the scheduling instruction to the edge server to which the monitoring terminal is connected, so that the edge server sends the currently processed data to the target edge server in the scheduling instruction; The method of obtaining the current monitoring data of the power grid terminal device through the monitoring terminal includes: collecting the monitoring data of the power grid terminal device in real time through the data collector, and storing the monitoring data in the heterogeneous database; obtaining a data query instruction through the monitoring terminal, and obtaining the current monitoring data from the heterogeneous database by a global search method based on the data query instruction; the global search method adopts an artificial bee colony algorithm combined with a decision tree C4.5 algorithm; Acquiring the current processing data from the monitoring terminal through the target edge server includes: acquiring the current processing data stored in the heterogeneous database through the target edge server using a local search method; the local search method uses a differential artificial bee colony algorithm.

2. The method according to claim 1, characterized in that After sending the control instruction to the monitoring terminal, the method further includes: Determine incremental data based on the current processing data and the historical processing data through the target edge server, and send the incremental data to the data middle station; A second data analysis is performed on the incremental data through the data center to determine a second analysis result, and the analysis model is updated based on the second analysis result.

3. The method according to claim 1, characterized in that The preprocessing of the current monitoring data to obtain current processed data includes: The current monitoring data is sequentially subjected to data filtering, data cleaning and symmetrical encryption to obtain current processing data.

4. The method according to claim 1, characterized in that: Also includes: The first analysis result is synchronized to the data center through the target edge server.

5. A monitoring system for power grid terminal equipment, characterized in that: include: Data center, target edge server, monitoring terminal and power grid terminal equipment; The monitoring terminal is used to obtain the current monitoring data of the power grid terminal equipment, pre-process the current monitoring data to obtain the current processing data, store the current processing data in a heterogeneous database, and send it to the data center; The data middle station is used to obtain the current load of the edge server to which the monitoring terminal is connected; if the current load of the edge server is greater than the underload threshold and less than the overload threshold, the edge server is determined as the target edge server; if the current load of the edge server is greater than or equal to the overload threshold, the current processing data is controlled to migrate from the edge server to any target edge server; if the current load of the edge server is less than or equal to the underload threshold, the current processing data stored in the edge server is controlled to migrate to any target edge server, and the edge server is controlled to be shut down; wherein the target edge server is an edge server whose current load is greater than the underload threshold and less than the overload threshold; and the analysis model is sent to the target edge server, and the analysis model is obtained by the data middle station through model training based on historical processing data; The target edge server is configured to obtain the currently processed data from the monitoring terminal, perform a first data analysis on the currently processed data based on the analysis model to determine a first analysis result, determine a control instruction based on the first analysis result, and send the control instruction to the monitoring terminal so that the monitoring terminal controls the power grid terminal device; The controlling the migration of the currently processed data from the edge server to any target edge server comprises: generating a scheduling instruction based on the determined identifier of the target edge server, and sending the scheduling instruction to the edge server to which the monitoring terminal is connected, so that the edge server sends the currently processed data to the target edge server in the scheduling instruction; The obtaining of current monitoring data of the power grid terminal device comprises: collecting the monitoring data of the power grid terminal device in real time through a data collector, and storing the monitoring data in a heterogeneous database; obtaining a data query instruction through a monitoring terminal, and obtaining the current monitoring data from the heterogeneous database by a global search method based on the data query instruction; the global search method adopts an artificial bee colony algorithm combined with a decision tree C4.5 algorithm; Acquiring the current processing data from the monitoring terminal includes: acquiring the current processing data stored in the heterogeneous database by using a local search method through the target edge server; the local search method uses a differential artificial bee colony algorithm.

6. The system according to claim 5, characterized in that The target edge server is further used to determine incremental data based on the current processing data and the historical processing data, and send the incremental data to the data middle station; The data middle platform is also used to perform a second data analysis on the incremental data to determine a second analysis result, and update the analysis model based on the second analysis result.

Images