A method, system and device for collaborative communication of power grid data cloud and edge
By building a power equipment relationship network and adjusting the cloud-edge collaborative communication rules according to the operating status of the power grid, the problem of low reliability of cloud-edge collaborative communication in the power grid data is solved, and data transmission and processing efficiency is improved.
Patent Information
- Application Number
- CN202411182091.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-27
AI Technical Summary
The reliability of power grid data cloud-edge collaborative communication is not high, resulting in low data transmission and processing efficiency.
Build a power equipment relationship network, and adjust communication rules according to the degree of abnormality to improve reliability by evaluating the power grid operation status.
It improves the reliability of cloud-edge collaborative communication between power grid data, ensures that the data communication frequency and sampling parameters are improved in abnormal situations to ensure monitoring reliability, reduce resource waste in low abnormal situations, and avoid unnecessary data communication and sampling.
Smart Images

Figure CN118694664B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power grid data collaborative communication, and specifically, to a power grid data cloud-edge collaborative communication method, system and device. Background Art
[0002] Power grid data cloud-edge collaborative communication is a technology that combines cloud computing and edge computing. Its main purpose is to optimize the processing, transmission and storage of power grid data and improve the operation efficiency and response speed of the power grid. This method usually involves the following key steps:
[0003] 1. Equipment and network layout
[0004] Edge device deployment: Deploy edge computing devices at key nodes of the power grid (such as substations, distribution rooms, etc.);
[0005] Network connectivity: Ensure that all edge computing devices have a stable network connection to the cloud server, usually using a dedicated communication line or through a secure Internet connection;
[0006] 2. Data Collection and Preprocessing
[0007] Data collection: Edge computing devices are responsible for collecting data from grid equipment (such as smart meters, sensors, etc.) in real time;
[0008] Preprocessing: Edge computing devices perform preliminary processing on the collected data, such as data cleaning, format conversion, simple analysis, etc.
[0009] After data collection and preprocessing, the edge computing device needs to transmit the collected and processed data to the cloud server for corresponding analysis to achieve power grid data cloud-edge collaborative communication. However, the inventors have found that in the above-mentioned data collection, preprocessing and data transmission process, there is a problem of low reliability of power grid data cloud-edge collaborative communication. Summary of the invention
[0010] In view of this, the purpose of the present application is to provide a power grid data cloud-edge collaborative communication method, system and device to improve the problem of low reliability of power grid data cloud-edge collaborative communication existing in the prior art.
[0011] To achieve the above objectives, this application adopts the following technical solutions:
[0012] A power grid data cloud-edge collaborative communication method is applied to a cloud server, wherein the cloud server is communicatively connected to a plurality of edge computing devices, each of the edge computing devices is communicatively connected to at least one power data acquisition device, and each of the power data acquisition devices is respectively deployed for a power device in a target power grid operation area, the method comprising:
[0013] Obtaining a power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rule, wherein each power data sequence is used to reflect the power grid operation status of the power device at multiple time nodes included in the current time period, and the power data sequence is formed based on the edge computing device processing the power data collected by the power data collection device;
[0014] Based on each of the power data sequences, a corresponding power equipment relationship network is constructed;
[0015] Based on the power equipment relationship network, the operating status of the target power grid operating area is evaluated, the corresponding power grid operation abnormality data is output, and based on the power grid operation abnormality data, the corresponding target cloud-edge collaborative communication rules are determined;
[0016] The target cloud-edge collaborative communication rules are sent to each of the edge computing devices, wherein the edge computing devices are used to replace the current cloud-edge collaborative communication rules with the target cloud-edge collaborative communication rules, and the target cloud-edge collaborative communication rules include at least the frequency and compression method of data communication to the cloud server and the sampling parameters for collecting power data.
[0017] In a preferred embodiment of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of evaluating the operating status of the target power grid operating area based on the power equipment relationship network, outputting the corresponding power grid operation abnormality data, and determining the corresponding target cloud-edge collaborative communication rule based on the power grid operation abnormality data includes:
[0018] In the power equipment relationship network, feature association mining is performed on the power data sequence of each power equipment to obtain the target power grid state feature corresponding to each power equipment;
[0019] Based on the target power grid state characteristics corresponding to each of the electric power equipment, the operation state of the target power grid operation area is evaluated, and the corresponding power grid operation abnormality data is output;
[0020] Based on the abnormal power grid operation data, the corresponding target cloud-edge collaborative communication rules are determined, wherein the frequency and sampling parameters of data communication included in the target cloud-edge collaborative communication rules have a positive correlation with the degree of abnormality represented by the abnormal power grid operation data, and the compression ratio corresponding to the compression method of data communication included in the target cloud-edge collaborative communication rules has a negative correlation with the degree of abnormality represented by the abnormal power grid operation data.
[0021] In a preferred option of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of performing feature association mining on the power data sequence of each of the power equipment in the power equipment relationship network to obtain the target power grid state feature corresponding to each of the power equipment includes:
[0022] Determine the semantic features of the power grid state corresponding to each power device in the power device relationship network respectively, wherein each power device includes at least one target power device, and the semantic features of the power grid state corresponding to any one of the at least one target power device are obtained by performing feature mining on the power data sequence corresponding to the any one target power device;
[0023] According to the semantic features of the power grid state corresponding to each of the power devices, feature transfer and update are performed on each of the power devices, and the target power grid state features corresponding to each of the power devices are output.
[0024] In a preferred embodiment of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of respectively determining the power grid state semantic features corresponding to each power device in the power device relationship network includes:
[0025] For each electric power device in the electric power device relationship network, respectively perform vector processing on the electric power data corresponding to each time node in the electric power data sequence corresponding to the electric power device, output the electric power data semantic vector sequence corresponding to the electric power device, and respectively perform vector processing on each time node, output the time node semantic vector sequence corresponding to the electric power device;
[0026] The power data semantic vectors in the power data semantic vector sequence and the time node semantic vectors in the time node semantic vector sequence are fused according to corresponding time nodes to form a corresponding fused power data vector sequence;
[0027] Based on the fused power data vector sequence, corresponding power grid state semantic features are obtained.
[0028] In a preferred embodiment of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of obtaining the corresponding power grid state semantic features based on the fused power data vector sequence includes:
[0029] For the first fused power data vector in the fused power data vector sequence, performing full connection processing and linear mapping processing on the fused power data vector, and outputting a local power grid state semantic feature corresponding to the fused power data vector;
[0030] For each fused power data vector other than the first fused power data vector in the fused power data vector sequence, the fused power data vector is associated with the local power grid state semantic feature corresponding to the previous fused power data vector and a fusion process is performed, and the local power grid state semantic feature corresponding to the fused power data vector is output;
[0031] The local power grid state semantic features corresponding to each fused power data vector in the fused power data vector sequence are concatenated to form corresponding power grid state semantic features.
[0032] In a preferred choice of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of updating the characteristics of each power device according to the semantic characteristics of the power grid state corresponding to each power device and outputting the target power grid state characteristics corresponding to each power device includes:
[0033] In the power equipment relationship network, any power equipment is determined and the power equipment is used as the power equipment to be processed;
[0034] Taking the power device to be processed as the starting point, performing non-repetitive walking in the power device relationship network according to a predetermined minimum connectivity, and forming at least one power device walking path corresponding to the power device to be processed, wherein the connectivity between each other power device included in the power device walking path and the power device to be processed is not less than the minimum connectivity, and the connectivity is negatively correlated with the length of the shortest walking path between the corresponding two power devices to be processed;
[0035] For each of the power equipment wandering paths, according to the semantic features of the power grid status corresponding to each of the power equipment in the power equipment wandering path, the features of each of the power equipment are updated through feature transfer, and the updated power grid status features corresponding to the power equipment wandering path are output;
[0036] The updated power grid state feature corresponding to each power equipment wandering path in at least one power equipment wandering path corresponding to the power equipment to be processed is subjected to feature fusion processing, and the target power grid state feature corresponding to the power equipment to be processed is output.
[0037] In a preferred embodiment of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, for each of the power equipment wandering paths, according to the power grid state semantic features corresponding to each of the power equipment in the power equipment wandering path, the steps of transmitting and updating the features of each of the power equipment, and outputting the updated power grid state features corresponding to the power equipment wandering path include:
[0038] Performing a first mapping process in a feature space on the semantic feature of the power grid state corresponding to the last power device in the power device wandering path, outputting a first power grid state mapping feature, and performing a second mapping process and a third mapping process in a feature space on the semantic feature of the power grid state corresponding to the penultimate power device, outputting a second power grid state mapping feature and a third power grid state mapping feature;
[0039] Based on the first grid state mapping feature corresponding to the last power device and the second grid state mapping feature corresponding to the second to last power device, determining a correlation parameter between the grid state semantic features corresponding to the two power devices;
[0040] The correlation parameter between the semantic features of the power grid states corresponding to the two power devices and the third power grid state mapping feature corresponding to the penultimate power device are fused, so that the semantic feature of the power grid state corresponding to the last power device is fused into the semantic feature of the power grid state corresponding to the penultimate power device, so as to output the associated feature of the power grid state corresponding to the penultimate power device;
[0041] For each power device other than the last power device in the power device wandering path, the power grid state association feature corresponding to the power device is subjected to a first mapping process in the feature space, and the corresponding first power grid state mapping feature is output, and the power grid state semantic feature corresponding to the previous power device of the power device is subjected to a second mapping process and a third mapping process in the feature space, and the corresponding second power grid state mapping feature and the third power grid state mapping feature are output respectively;
[0042] Determining a correlation parameter between the semantic features of the power grid states corresponding to the two power devices based on the first power grid state mapping feature and the second power grid state mapping feature corresponding to the two power devices;
[0043] The correlation parameter between the semantic features of the power grid states corresponding to the two power devices and the third power grid state mapping feature corresponding to the latter power device are fused so that the semantic features of the power grid state corresponding to the latter power device are fused into the semantic features of the power grid state corresponding to the former power device, so as to output the associated features of the power grid state corresponding to the former power device;
[0044] Based on the semantic features of the power grid state corresponding to the first power device and the corresponding power grid state association features, the updated power grid state features corresponding to the wandering path of the power device are output.
[0045] In a preferred embodiment of the present application, in the above-mentioned power grid data cloud-edge collaborative communication method, the step of constructing a corresponding power equipment relationship network based on each of the power data sequences includes:
[0046] Acquire a power grid electrical topology map corresponding to the target power grid operation area, wherein the power grid electrical topology map has electrical connection relationships between the various power devices;
[0047] For each of the power data sequences, the power equipment corresponding to the power data sequence is used as a network node, and the power data sequence is used as the node attribute data corresponding to the network node, wherein one of the power equipment corresponds to at least one network node, and one of the network nodes corresponds to one of the power equipment, and each of the power data sequences includes a current data sequence corresponding to the power data acquisition device belonging to the current sensor, a voltage data sequence corresponding to the power data acquisition device belonging to the voltage sensor, a temperature data sequence corresponding to the power data acquisition device belonging to the temperature sensor, a geomagnetic storm data sequence corresponding to the power data acquisition device belonging to the geomagnetic field sensor, and a frequency data sequence corresponding to the power data acquisition device belonging to the frequency sensor;
[0048] The same network nodes of the corresponding power equipment are associated and set, and the network nodes of the corresponding power equipment that have direct connection in the power grid electrical topology diagram are associated to form a corresponding power equipment network.
[0049] The present application also provides a power grid data cloud-edge collaborative communication system, which is applied to a cloud server, wherein the cloud server is communicatively connected to a plurality of edge computing devices, each of which is communicatively connected to at least one power data acquisition device, and each of which is deployed for a power device in a target power grid operation area. The system includes:
[0050] A data sequence acquisition module, used to acquire a power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rule, wherein each power data sequence is used to reflect the power grid operation status of the power device at multiple time nodes included in the current time period, and the power data sequence is formed based on the edge computing device processing the power data collected by the power data collection device;
[0051] An equipment relationship network construction module, used to construct a corresponding power equipment relationship network based on each of the power data sequences;
[0052] An operation abnormality assessment module is used to evaluate the operation status of the target power grid operation area based on the power equipment relationship network, output corresponding power grid operation abnormality data, and determine the corresponding target cloud-edge collaborative communication rules based on the power grid operation abnormality data;
[0053] A collaborative communication rule sending module is used to send the target cloud-edge collaborative communication rules to each of the edge computing devices, wherein the edge computing device is used to replace the current cloud-edge collaborative communication rules with the target cloud-edge collaborative communication rules, and the target cloud-edge collaborative communication rules at least include the frequency and compression method of data communication to the cloud server and the sampling parameters for collecting power data.
[0054] Based on the above, the present application also provides an electronic device, including:
[0055] Memory for storing computer programs;
[0056] A processor connected to the memory is used to execute the computer program stored in the memory to implement the above-mentioned power grid data cloud-edge collaborative communication method.
[0057] The present application provides a power grid data cloud-edge collaborative communication method, system and device. First, the power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rules is obtained; secondly, based on each power data sequence, a corresponding power equipment relationship network is constructed; then, based on the power equipment relationship network, the operating status of the target power grid operating area is evaluated, and the corresponding power grid operation abnormal data is output, and based on the power grid operation abnormal data, the corresponding target cloud-edge collaborative communication rules are determined; finally, the target cloud-edge collaborative communication rules are sent to each edge computing device. Based on the above content, since an electric power equipment relationship network will be constructed, reliable power grid operation abnormality data can be obtained when the operating status is evaluated based on the electric power equipment relationship network with high information richness. Therefore, the reliability of the target cloud-edge collaborative communication rules determined based on the power grid operation abnormality data can be guaranteed, thereby ensuring the reliability of the power grid data cloud-edge collaborative communication, thereby improving the problem of low reliability of power grid data cloud-edge collaborative communication in the prior art; for example, when the abnormality is high, the frequency and sampling parameters of data communication can be increased, and the compression ratio of the corresponding compression method can be reduced to ensure the reliability of cloud-based monitoring. When the abnormality is low, the frequency and sampling parameters of data communication can be reduced, and the compression ratio of the corresponding compression method can be increased to avoid the problem of resource waste caused by data communication and sampling. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.
[0059] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.
[0060] Figure 2A flowchart of the power grid data cloud-edge collaborative communication method provided in an embodiment of the present application.
[0061] Figure 3 A schematic diagram of a power equipment relationship network provided in an embodiment of the present application.
[0062] Figure 4 A schematic diagram of feature transfer update provided in an embodiment of the present application.
[0063] Figure 5 A block diagram of the power grid data cloud-edge collaborative communication system provided in an embodiment of the present application. DETAILED DESCRIPTION
[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.
[0065] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0066] like Figure 1 As shown, an embodiment of the present application provides an electronic device. The electronic device may include a memory, a processor, and a power grid data cloud-edge collaborative communication system.
[0067] In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The power grid data cloud-edge collaborative communication system includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function modules and computer programs included in the power grid data cloud-edge collaborative communication system, so as to realize the power grid data cloud-edge collaborative communication method provided in an embodiment of the present application.
[0068] Optionally, the memory may be, but not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0069] Understandably, Figure 1 The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown, for example, may also include a communication unit for information exchange with other devices. For example, when the electronic device is used as a cloud server, it may also include a communication unit for communicating with an edge computing device. In addition, the edge computing device may be a gateway device, etc. In addition, the cloud server may be communicatively connected to multiple edge computing devices, each of which may be communicatively connected to at least one power data acquisition device (such as a current sensor, a voltage sensor, a geomagnetic field sensor, a frequency sensor, a temperature sensor, etc.), and each of the power data acquisition devices may be deployed for a power device in the target power grid operation area. In this way, the power data acquisition device may perform power data acquisition operations on the corresponding power device to obtain corresponding power data, such as current data, voltage data, geomagnetic storm data, frequency data, and temperature data, etc.
[0070] Combination Figure 2 The embodiment of the present application also provides a power grid data cloud-edge collaborative communication method applicable to the above electronic device. Among them, the method steps defined in the process related to the power grid data cloud-edge collaborative communication method can be implemented by the electronic device.
[0071] The following will Figure 2The specific process shown is explained in detail.
[0072] Step S110, obtaining the power data sequence sent by each of the edge computing devices based on the current cloud-edge collaborative communication rules.
[0073] In an embodiment of the present application, the electronic device can obtain the power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rules. In this way, multiple power data sequences can be obtained. One edge computing device can send at least one power data sequence, and the specific number is determined based on the number of power data acquisition devices connected to the edge computing device, that is, one power data sequence corresponds to one power data acquisition device. Each of the power data sequences is used to reflect the power grid operation status of multiple time nodes included in the current time period of the power equipment, and the power data sequence is formed based on the edge computing device processing the power data collected by the power data acquisition device, such as sampling. In addition, the current cloud-edge collaborative communication rules can be based on user configuration or determined based on historical power data.
[0074] Step S120: constructing a corresponding power equipment relationship network based on each of the power data sequences.
[0075] In the embodiment of the present application, the electronic device can construct a corresponding power device relationship network based on each of the power data sequences, that is, the power device relationship network can include each of the power data sequences.
[0076] Step S130, based on the power equipment relationship network, evaluate the operating status of the target power grid operating area, output the corresponding power grid operation abnormality data, and determine the corresponding target cloud-edge collaborative communication rules based on the power grid operation abnormality data.
[0077] In an embodiment of the present application, the electronic device can evaluate the operating status of the target power grid operating area based on the power equipment relationship network, output the corresponding power grid operation abnormality data, and determine the corresponding target cloud-edge collaborative communication rules based on the power grid operation abnormality data. Exemplarily, the power grid operation abnormality data can refer to the type of power grid operation abnormality (such as voltage abnormality, harmonic distortion, etc.), or the degree of operation abnormality (such as extremely high abnormality, high abnormality, general abnormality, low abnormality, extremely low abnormality, no abnormality, or 0-100 points, etc.).
[0078] Step S140: Send the target cloud-edge collaborative communication rules to each of the edge computing devices.
[0079] In an embodiment of the present application, the electronic device can send the target cloud-edge collaborative communication rule to each of the edge computing devices. The edge computing device is used to replace the current cloud-edge collaborative communication rule with the target cloud-edge collaborative communication rule, and the target cloud-edge collaborative communication rule includes at least the frequency of data communication with the cloud server (i.e., sending a power data sequence once every preset time interval) and compression method (such as compressing the data according to a preset ratio, and then sending the compressed data), and sampling parameters for collecting power data (which can be either the sampling rate of the power data collected by the edge computing device to the power data collection device, or the collection frequency of the power data collection device to collect power data).
[0080] Based on the above content, since an electric power equipment relationship network will be constructed, reliable power grid operation abnormality data can be obtained when the operating status is evaluated based on the electric power equipment relationship network with high information richness. Therefore, the reliability of the target cloud-edge collaborative communication rules determined based on the power grid operation abnormality data can be guaranteed, thereby ensuring the reliability of the power grid data cloud-edge collaborative communication, thereby improving the problem of low reliability of power grid data cloud-edge collaborative communication in the prior art; for example, when the abnormality is high, the frequency and sampling parameters of data communication can be increased, and the compression ratio of the corresponding compression method can be reduced to ensure the reliability of cloud-based monitoring. When the abnormality is low, the frequency and sampling parameters of data communication can be reduced, and the compression ratio of the corresponding compression method can be increased to avoid the problem of resource waste caused by data communication and sampling.
[0081] It should be noted that for step S120 , the specific method of constructing the corresponding power equipment relationship network is not limited and can be selected according to actual needs.
[0082] For example, in an alternative implementation, in order to improve the information representation capability of the power equipment relationship network and ensure that subsequent operations based on the power equipment relationship network can be more reliable, the above-mentioned step S120 can further include step S121, step S122 and step S123, and the specific content of each step is described as follows.
[0083] Step S121, obtaining a power grid electrical topology map corresponding to the target power grid operation area.
[0084] In an embodiment of the present application, a power grid electrical topology map corresponding to the target power grid operation area may be first obtained. The power grid electrical topology map includes electrical connection relationships between the various power devices, so that corresponding association settings can be performed based on the electrical connection relationships, and the performed association settings can be guaranteed to have high reliability for subsequent abnormality analysis, for example, the power devices with electrical connection relationships have associations in power data.
[0085] Step S122: for each of the power data sequences, the power equipment corresponding to the power data sequence is used as a relationship network node, and the power data sequence is used as node attribute data corresponding to the relationship network node.
[0086] In an embodiment of the present application, for each of the power data sequences, the power device corresponding to the power data sequence can be used as a network node, and the power data sequence can be used as the node attribute data corresponding to the network node. Among them, one power device corresponds to at least one network node, and one network node corresponds to one power device, that is, there is a one-to-one correspondence or a one-to-many relationship between the power device and the network node. For example, for power device A, when it is necessary to collect current and temperature of power device A, two corresponding power data sequences can be formed, and correspondingly, two network nodes will also be formed. In addition, each of the power data sequences includes a current data sequence corresponding to the power data acquisition device belonging to the current sensor, a voltage data sequence corresponding to the power data acquisition device belonging to the voltage sensor, a temperature data sequence corresponding to the power data acquisition device belonging to the temperature sensor, a geomagnetic storm data sequence corresponding to the power data acquisition device belonging to the geomagnetic field sensor, and a frequency data sequence corresponding to the power data acquisition device belonging to the frequency sensor. In other implementations, other data may also be included, which is not specifically limited or described here.
[0087] Step S123, the same network nodes of the corresponding power equipment are associated with each other, and the network nodes of the corresponding power equipment that are directly connected in the power grid electrical topology diagram are associated with each other to form a corresponding power equipment network.
[0088] In the embodiment of the present application, on the one hand, the same network nodes of the corresponding power equipment can be associated with each other (for example, the two network nodes corresponding to the power equipment A can be associated with each other), and the network nodes of the corresponding power equipment with direct connection in the power grid electrical topology diagram can be associated with each other to form a corresponding power equipment network, such as Figure 3 As shown, in Figure 3 The connecting lines indicate that associated settings have been made.
[0089] It should be noted that for the above step S130, the specific manner of performing operations such as evaluating the operating status of the target power grid operating area is not limited and can be selected according to actual needs.
[0090] For example, in an alternative implementation, in order to improve the reliability of operating status evaluation and determine reliable target cloud-edge collaborative communication rules, the above-mentioned step S130 may further include step S131, step S132 and step S133, and the specific content of each step is described as follows.
[0091] Step S131, in the power equipment relationship network, feature association mining is performed on the power data sequence of each of the power equipment to obtain the target power grid state feature corresponding to each of the power equipment.
[0092] In an embodiment of the present application, after constructing the power equipment relationship network, feature association mining can be performed on the power data sequence of each of the power equipment in the power equipment relationship network to obtain the target power grid state characteristics corresponding to each of the power equipment. In this way, the target power grid state characteristics not only include the semantic information of the power data of the power equipment itself, but also include the semantic information of the power data of the associated power equipment, so that the semantic representation capability of the target power grid state characteristics can be better.
[0093] Step S132: Based on the target power grid state characteristics corresponding to each of the electric power equipment, the operation state of the target power grid operation area is evaluated, and the corresponding power grid operation abnormality data is output.
[0094] In an embodiment of the present application, after mining the target grid state characteristics, the operating state of the target grid operating area can be evaluated based on the target grid state characteristics corresponding to each of the power equipment, and the corresponding grid operation abnormality data can be output. For example, the target grid state characteristics corresponding to each of the power equipment can be spliced, and then fully connected, and finally, the configured classification function or regression function can be output to obtain the corresponding grid operation abnormality data. It should be noted that step S131 and step S132 can be implemented by a neural network, such as an encoding neural network, a recurrent neural network, etc., and the neural network can be trained based on the corresponding sample power equipment relationship network and the actual grid operation abnormality data.
[0095] Step S133: Determine the corresponding target cloud-edge collaborative communication rules based on the abnormal power grid operation data.
[0096] In an embodiment of the present application, after obtaining the abnormal power grid operation data, the corresponding target cloud-edge collaborative communication rules can be determined based on the abnormal power grid operation data. Among them, the frequency and sampling parameters of data communication included in the target cloud-edge collaborative communication rules have a positive correlation with the degree of abnormality represented by the abnormal power grid operation data, and the compression ratio corresponding to the compression method of data communication included in the target cloud-edge collaborative communication rules has a negative correlation with the degree of abnormality represented by the abnormal power grid operation data. For example, the correspondence between various abnormal power grid operation data and various cloud-edge collaborative communication rules can be established in advance, so that the target cloud-edge collaborative communication rules can be determined based on the correspondence.
[0097] It can be understood that in the above-mentioned step S131, the specific method of performing feature association mining is not limited. For example, in an alternative implementation, in order to improve the reliability of feature association mining, the above-mentioned step S131 can further include step S131a and step S131b. The specific content of each step is described as follows.
[0098] Step S131a, respectively determining the semantic features of the power grid state corresponding to each power device in the power device relationship network.
[0099] In an embodiment of the present application, the semantic features of the power grid state corresponding to each power device in the power device relationship network can be determined respectively. Each power device includes at least one target power device, and the semantic features of the power grid state corresponding to any one of the at least one target power device are obtained by performing feature mining on the power data sequence corresponding to the any one of the target power devices. In other words, the power data sequence corresponding to power device B can be semantically mined to obtain the corresponding semantic features of the power grid state.
[0100] Step S131b, based on the semantic features of the power grid state corresponding to each of the power devices, feature transfer and update are performed on each of the power devices, and target power grid state features corresponding to each of the power devices are output.
[0101] In an embodiment of the present application, after obtaining the semantic characteristics of the power grid status, the characteristics of each power device can be transferred and updated based on the semantic characteristics of the power grid status corresponding to each power device, and the target power grid status characteristics corresponding to each power device can be output. In other words, the semantic characteristics of the power grid status corresponding to each power device are transferred and updated so that the updated characteristics can be obtained, that is, the corresponding target power grid status characteristics.
[0102] It can be understood that, in the above-mentioned step S131a, the specific method of determining the semantic features of the power grid status is not limited. For example, in an alternative implementation, in order to make the information expression capability of the semantic features of the power grid status more accurate and the semantics richer, the above-mentioned step S131a can further include step a1, step a2 and step a3, and the specific content of each step is described as follows.
[0103] Step a1, for each power equipment in the power equipment relationship network, respectively vectorize the power data corresponding to each time node in the power data sequence corresponding to the power equipment, output the power data semantic vector sequence corresponding to the power equipment, and respectively vectorize each time node, output the time node semantic vector sequence corresponding to the power equipment.
[0104] In an embodiment of the present application, for each power device in the power device relationship network, the power data corresponding to each time node in the power data sequence corresponding to the power device is vectorized, and the power data semantic vector sequence corresponding to the power device is output. In the power data semantic vector sequence, one power data semantic vector corresponds to the power data of one time node. For example, the power data can be mapped into a vector space to obtain the corresponding power data semantic vector, which can be specifically realized by a trained embedding model. In addition, each time node can be vectorized separately, and a time node semantic vector sequence corresponding to the power device is output, wherein the representation of the time node can be a serial number such as 1, 2, 3, or a specific time point, such as 11:42:03 on August 4, 2024. Similarly, the time node can also be embedded to obtain the corresponding time node semantic vector.
[0105] Step a2: fusing the power data semantic vectors in the power data semantic vector sequence and the time node semantic vectors in the time node semantic vector sequence according to corresponding time nodes to form a corresponding fused power data vector sequence.
[0106] In an embodiment of the present application, after obtaining the power data semantic vector sequence and the time node semantic vector sequence, the power data semantic vectors in the power data semantic vector sequence and the time node semantic vectors in the time node semantic vector sequence can be fused according to the corresponding time nodes to form a corresponding fused power data vector sequence. For example, the power data semantic vector corresponding to the first time node can be fused (such as splicing or superposition, etc.) with the time node semantic vector corresponding to the first time node to obtain a fused power data vector corresponding to the first time node, thereby forming a corresponding fused power data vector sequence.
[0107] Step a3: obtaining corresponding semantic features of the power grid state based on the fused power data vector sequence.
[0108] In an embodiment of the present application, after obtaining the fused power data vector sequence, the corresponding power grid state semantic feature can be obtained based on the fused power data vector sequence. For example, the fused power data vector sequence can be directly used as the corresponding power grid state semantic feature.
[0109] It can be understood that, in the above step a3, the specific manner of obtaining the semantic features of the power grid state is not limited. For example, in an alternative implementation, in order to obtain a better semantic representation capability of the semantic features of the power grid state, the above step a3 may further include:
[0110] First, for the first fused power data vector in the fused power data vector sequence, the fused power data vector can be fully connected and linearly mapped, and the local power grid state semantic features corresponding to the fused power data vector can be output. For example, the corresponding fully connected network layer can be used to process the vector to obtain the corresponding fully connected vector, and then the corresponding linear mapping network layer can be used to process the vector to obtain the corresponding local power grid state semantic features, wherein the linear mapping network layer can implement the corresponding linear mapping processing by including an activation function;
[0111] Secondly, for each fused power data vector other than the first fused power data vector in the fused power data vector sequence, the fused power data vector and the local power grid state semantic feature corresponding to the previous fused power data vector may be associated and fused, and the local power grid state semantic feature corresponding to the fused power data vector may be output; for example, the fused power data vector and the local power grid state semantic feature corresponding to the previous fused power data vector may be weighted summed, and then the result of the weighted summation may be fully connected and linearly mapped;
[0112] Then, the local power grid state semantic features corresponding to each fused power data vector in the fused power data vector sequence can be spliced to form corresponding power grid state semantic features, such as concat (local power grid state semantic feature 1, local power grid state semantic feature 2, local power grid state semantic feature 3, local power grid state semantic feature 4, ...).
[0113] It can be understood that in the above-mentioned step S131b, the specific method of performing feature transfer update is not limited. For example, in an alternative implementation, in order to take into account both the semantic representation capability of the formed target power grid state characteristics and the efficiency of performing feature transfer update, the above-mentioned step S131b may include step b1, step b2, step b3 and step b4, each step is described as follows.
[0114] Step b1, in the power equipment relationship network, determine any power equipment and use the power equipment as the power equipment to be processed.
[0115] In the embodiment of the present application, any power device can be determined in the power device relationship network and used as the power device to be processed. It should be noted that each power device can be used as the power device to be processed in sequence or in parallel, so as to obtain the target grid state characteristics corresponding to each power device.
[0116] Step b2: taking the power device to be processed as the starting point, performing non-repetitive walking in the power device relationship network according to a predetermined minimum connection degree, and forming at least one power device walking path corresponding to the power device to be processed.
[0117] In the embodiment of the present application, after the power equipment to be processed is determined, the power equipment to be processed can be used as a starting point, and a non-repeated walk is performed in the power equipment relationship network according to a predetermined minimum connection degree (i.e., in one walk, one power equipment appears only once, and in addition, in the case of Figure 3 In the power equipment relationship network shown, when one power equipment corresponds to multiple relationship network nodes, it may mean that one relationship network node appears only once), at least one power equipment wandering path corresponding to the power equipment to be processed is formed. When multiple power equipment wandering paths are formed, any two power equipment wandering paths are not the same, and there is no inclusion relationship. Moreover, the connection degree between each other power equipment included in the power equipment wandering path and the power equipment to be processed is not less than the minimum connection degree, and the connection degree is equal to the shortest wandering path between the corresponding two power equipment to be processed (for example, in Figure 3 In the figure, there can be two wandering paths between power equipment B and power equipment E, namely "BE" and "BCDE", with lengths of 2 and 4 respectively, and the shortest wandering path is "BE"), which are negatively correlated in length (such as the inverse of the shortest wandering path). That is to say, the smaller the shortest wandering path is, the closer the relationship between the two power equipment to be processed is, and the greater the corresponding connectivity is. In this way, by setting the minimum connectivity, it is possible to avoid integrating the features of power equipment with too small a connectivity degree, so that the reliability of feature association mining can be relatively higher.
[0118] Step b3, for each of the power equipment wandering paths, based on the semantic features of the power grid status corresponding to each of the power equipment in the power equipment wandering path, feature transfer and update of each of the power equipment, and output of the updated power grid status features corresponding to the power equipment wandering path.
[0119] In an embodiment of the present application, for each of the power equipment wandering paths, the features of each of the power equipment can be transferred and updated based on the semantic features of the power grid status corresponding to each of the power equipment in the power equipment wandering paths, and the updated power grid status features corresponding to the power equipment wandering paths can be output. The updates can be performed along the power equipment wandering paths, so that the correlation in the updates can match the correlation relationship between the power equipment, thereby ensuring its reliability.
[0120] Step b4, performing feature fusion processing on the updated grid state feature corresponding to each of the at least one power device wandering path corresponding to the power device to be processed, and outputting the target grid state feature corresponding to the power device to be processed.
[0121] In an embodiment of the present application, the updated grid state characteristics corresponding to each of the at least one power device wandering paths corresponding to the power device to be processed can be subjected to feature fusion processing to output the target grid state characteristics corresponding to the power device to be processed; exemplarily, for each of the power device wandering paths, the connectivity between each power device in the power device wandering path and the power device to be processed can be determined first, and then the minimum value or mean of each connectivity can be determined, and finally, based on the minimum value or mean, the updated grid state characteristics corresponding to each power device wandering path can be weightedly summed to obtain the target grid state characteristics corresponding to the power device to be processed.
[0122] It is understandable that in the above step b3, the specific method of performing feature transfer update is not limited. For example, in an alternative implementation, in order to enable the relevant information to be fully mined and integrated in the feature transfer update, thereby ensuring the reliability of the updated power grid state feature, combined with Figure 4 , the above step b3 may further include the following contents:
[0123] First, the semantic features of the power grid state corresponding to the last power device in the wandering path of the power device can be subjected to a first mapping process in the feature space, and a first power grid state mapping feature can be output, and the semantic features of the power grid state corresponding to the penultimate power device can be subjected to a second mapping process and a third mapping process in the feature space, and a second power grid state mapping feature and a third power grid state mapping feature can be output; illustratively, the first mapping process, the second mapping process and the third mapping process in the feature space can be implemented by three mapping matrices, such as multiplying by three mapping matrices respectively, and the three mapping matrices can be formed in the training process of the corresponding neural network;
[0124] Secondly, based on the first grid state mapping feature corresponding to the last power device and the second grid state mapping feature corresponding to the second to last power device, the correlation parameter between the grid state semantic features corresponding to the two power devices can be determined. For example, the first grid state mapping feature and the second grid state mapping feature can be dot-producted to obtain the correlation parameter.
[0125] Then, the correlation parameter between the grid state semantic features corresponding to the two power devices and the third grid state mapping feature corresponding to the penultimate power device may be fused (e.g., weighted based on the correlation parameter), so that the grid state semantic feature corresponding to the last power device is fused into the grid state semantic feature corresponding to the penultimate power device, so as to output the grid state association feature corresponding to the penultimate power device;
[0126] Afterwards, for each power device except the last power device in the power device wandering path, the power grid state association feature corresponding to the power device can be subjected to a first mapping process in the feature space, and the corresponding first power grid state mapping feature can be output, and the power grid state semantic feature corresponding to the previous power device of the power device can be subjected to a second mapping process and a third mapping process in the feature space, and the corresponding second power grid state mapping feature and the third power grid state mapping feature can be output respectively. As described above, the specific parameters of the corresponding mapping matrix can be different;
[0127] Further, the correlation parameter between the grid state semantic features corresponding to the two power devices may be determined based on the first grid state mapping feature and the second grid state mapping feature corresponding to the two power devices, as described above;
[0128] Furthermore, the correlation parameter between the grid state semantic features corresponding to the two power devices and the third grid state mapping feature corresponding to the latter power device can be fused, so that the grid state semantic feature corresponding to the latter power device is fused into the grid state semantic feature corresponding to the former power device, so as to output the grid state association feature corresponding to the former power device, as described above;
[0129] Finally, based on the grid state semantic features and the corresponding grid state association features corresponding to the first power device, the updated grid state features corresponding to the roaming path of the power device can be output. For example, the grid state semantic features and the corresponding grid state association features can be spliced or superimposed to obtain the corresponding updated grid state features.
[0130] Combination Figure 5 The embodiment of the present application also provides a power grid data cloud-edge collaborative communication system applicable to the above electronic device. The power grid data cloud-edge collaborative communication system may include a data sequence acquisition module, a device relationship network construction module, an operation anomaly assessment module, and a collaborative communication rule sending module.
[0131] The data sequence acquisition module can be used to acquire the power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rules, wherein each power data sequence is used to reflect the power grid operation status of multiple time nodes included in the current time period of the power device, and the power data sequence is formed based on the power data collected by the power data acquisition device by the edge computing device. In the embodiment of the present application, the data sequence acquisition module can be used to execute Figure 2 As shown in step S110, the relevant contents of the data sequence acquisition module can refer to the above description of step S110.
[0132] The device relationship network construction module can be used to construct a corresponding power device relationship network based on each of the power data sequences. In the embodiment of the present application, the device relationship network construction module can be used to execute Figure 2 As shown in step S120, the relevant contents of the device relationship network building module can refer to the above description of step S120.
[0133] The operation anomaly assessment module can be used to evaluate the operation status of the target power grid operation area based on the power equipment relationship network, output the corresponding power grid operation anomaly data, and determine the corresponding target cloud-edge collaborative communication rules based on the power grid operation anomaly data. Figure 2As shown in step S130, the relevant contents of the operation abnormality assessment module can refer to the above description of step S130.
[0134] The collaborative communication rule sending module can be used to send the target cloud-edge collaborative communication rule to each edge computing device, wherein the edge computing device is used to replace the current cloud-edge collaborative communication rule with the target cloud-edge collaborative communication rule, and the target cloud-edge collaborative communication rule at least includes the frequency and compression method of data communication to the cloud server and the sampling parameters of power data collection. Figure 2 As shown in step S140, the relevant contents of the collaborative communication rule sending module can refer to the above description of step S140.
[0135] In an embodiment of the present application, corresponding to the above-mentioned power grid data cloud-edge collaborative communication method applied to the electronic device, a computer-readable storage medium is also provided, in which a computer program is stored, and when the computer program is run, each step of the power grid data cloud-edge collaborative communication method is executed. Among them, each step executed when the aforementioned computer program is run will not be repeated here one by one, and reference can be made to the above explanation of the power grid data cloud-edge collaborative communication method.
[0136] To summarize, the present application provides a power grid data cloud-edge collaborative communication method, system and device. First, the power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rules is obtained; secondly, based on each power data sequence, a corresponding power equipment relationship network is constructed; then, based on the power equipment relationship network, the operating status of the target power grid operating area is evaluated, and the corresponding power grid operation abnormal data is output, and based on the power grid operation abnormal data, the corresponding target cloud-edge collaborative communication rules are determined; finally, the target cloud-edge collaborative communication rules are sent to each edge computing device. Based on the above content, since an electric power equipment relationship network will be constructed, reliable power grid operation abnormality data can be obtained when the operating status is evaluated based on the electric power equipment relationship network with high information richness. Therefore, the reliability of the target cloud-edge collaborative communication rules determined based on the power grid operation abnormality data can be guaranteed, thereby ensuring the reliability of the power grid data cloud-edge collaborative communication, thereby improving the problem of low reliability of power grid data cloud-edge collaborative communication in the prior art; for example, when the abnormality is high, the frequency and sampling parameters of data communication can be increased, and the compression ratio of the corresponding compression method can be reduced to ensure the reliability of cloud-based monitoring. When the abnormality is low, the frequency and sampling parameters of data communication can be reduced, and the compression ratio of the corresponding compression method can be increased to avoid the problem of resource waste caused by data communication and sampling.
[0137] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0138] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
[0139] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, electronic device, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code. It should be noted that in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.
[0140] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A power grid data cloud-edge collaborative communication method, characterized in that: Applied to a cloud server, the cloud server is communicatively connected to a plurality of edge computing devices, each of the edge computing devices is communicatively connected to at least one power data acquisition device, and each of the power data acquisition devices is respectively deployed for a power device in a target power grid operation area, the method comprising: Obtaining a power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rule, wherein each power data sequence is used to reflect the power grid operation status of the power device at multiple time nodes included in the current time period, and the power data sequence is formed based on the edge computing device processing the power data collected by the power data collection device; Based on each of the power data sequences, a corresponding power equipment relationship network is constructed; Based on the power equipment relationship network, the operating status of the target power grid operating area is evaluated, the corresponding power grid operation abnormality data is output, and based on the power grid operation abnormality data, the corresponding target cloud-edge collaborative communication rules are determined; Sending the target cloud-edge collaborative communication rule to each of the edge computing devices, wherein the edge computing device is used to replace the current cloud-edge collaborative communication rule with the target cloud-edge collaborative communication rule, and the target cloud-edge collaborative communication rule at least includes the frequency and compression method of data communication to the cloud server and the sampling parameters for collecting power data; Among them, the step of evaluating the operating status of the target power grid operating area based on the power equipment relationship network, outputting corresponding power grid operation abnormality data, and determining the corresponding target cloud-edge collaborative communication rules based on the power grid operation abnormality data includes: In the power equipment relationship network, feature association mining is performed on the power data sequence of each power equipment to obtain the target power grid state feature corresponding to each power equipment; Based on the target power grid state characteristics corresponding to each of the electric power equipment, the operation state of the target power grid operation area is evaluated, and the corresponding power grid operation abnormality data is output; Based on the abnormal power grid operation data, a corresponding target cloud-edge collaborative communication rule is determined, wherein the frequency and sampling parameters of data communication included in the target cloud-edge collaborative communication rule have a positive correlation with the degree of abnormality represented by the abnormal power grid operation data, and the compression ratio corresponding to the compression method of data communication included in the target cloud-edge collaborative communication rule has a negative correlation with the degree of abnormality represented by the abnormal power grid operation data; The step of constructing a corresponding power equipment relationship network based on each of the power data sequences includes: Acquire a power grid electrical topology map corresponding to the target power grid operation area, wherein the power grid electrical topology map has electrical connection relationships between the various power devices; For each of the power data sequences, the power equipment corresponding to the power data sequence is used as a network node, and the power data sequence is used as the node attribute data corresponding to the network node, wherein one of the power equipment corresponds to at least one network node, and one of the network nodes corresponds to one of the power equipment, and each of the power data sequences includes a current data sequence corresponding to the power data acquisition device belonging to the current sensor, a voltage data sequence corresponding to the power data acquisition device belonging to the voltage sensor, a temperature data sequence corresponding to the power data acquisition device belonging to the temperature sensor, a geomagnetic storm data sequence corresponding to the power data acquisition device belonging to the geomagnetic field sensor, and a frequency data sequence corresponding to the power data acquisition device belonging to the frequency sensor; The same network nodes of the corresponding power equipment are associated and set, and the network nodes of the corresponding power equipment that have direct connection in the power grid electrical topology diagram are associated to form a corresponding power equipment network.
2. The power grid data cloud-edge collaborative communication method according to claim 1, characterized in that: The step of performing feature association mining on the power data sequence of each power device in the power device relationship network to obtain the target power grid state feature corresponding to each power device includes: Determine the semantic features of the power grid state corresponding to each power device in the power device relationship network respectively, wherein each power device includes at least one target power device, and the semantic features of the power grid state corresponding to any one of the at least one target power device are obtained by performing feature mining on the power data sequence corresponding to the any one target power device; According to the semantic features of the power grid state corresponding to each of the power devices, feature transfer and update are performed on each of the power devices, and the target power grid state features corresponding to each of the power devices are output.
3. The power grid data cloud-edge collaborative communication method according to claim 2, characterized in that: The step of respectively determining the semantic features of the power grid state corresponding to each power device in the power device relationship network includes: For each electric power device in the electric power device relationship network, respectively perform vector processing on the electric power data corresponding to each time node in the electric power data sequence corresponding to the electric power device, output the electric power data semantic vector sequence corresponding to the electric power device, and respectively perform vector processing on each time node, output the time node semantic vector sequence corresponding to the electric power device; The power data semantic vectors in the power data semantic vector sequence and the time node semantic vectors in the time node semantic vector sequence are fused according to corresponding time nodes to form a corresponding fused power data vector sequence; Based on the fused power data vector sequence, corresponding power grid state semantic features are obtained.
4. The power grid data cloud-edge collaborative communication method according to claim 3 is characterized in that: The step of obtaining corresponding grid state semantic features based on the fused power data vector sequence includes: For the first fused power data vector in the fused power data vector sequence, performing full connection processing and linear mapping processing on the fused power data vector, and outputting a local power grid state semantic feature corresponding to the fused power data vector; For each fused power data vector other than the first fused power data vector in the fused power data vector sequence, the fused power data vector is associated with the local power grid state semantic feature corresponding to the previous fused power data vector and a fusion process is performed, and the local power grid state semantic feature corresponding to the fused power data vector is output; The local power grid state semantic features corresponding to each fused power data vector in the fused power data vector sequence are concatenated to form corresponding power grid state semantic features.
5. The power grid data cloud-edge collaborative communication method according to claim 2, characterized in that: The step of transmitting and updating the characteristics of each of the electric power devices according to the semantic characteristics of the electric power state corresponding to each of the electric power devices, and outputting the target electric power state characteristics corresponding to each of the electric power devices comprises: In the power equipment relationship network, any power equipment is determined and the power equipment is used as the power equipment to be processed; Taking the power device to be processed as the starting point, performing non-repetitive walking in the power device relationship network according to a predetermined minimum connectivity, and forming at least one power device walking path corresponding to the power device to be processed, wherein the connectivity between each other power device included in the power device walking path and the power device to be processed is not less than the minimum connectivity, and the connectivity is negatively correlated with the length of the shortest walking path between the corresponding two power devices to be processed; For each of the power equipment wandering paths, according to the semantic features of the power grid status corresponding to each of the power equipment in the power equipment wandering path, the features of each of the power equipment are updated through feature transfer, and the updated power grid status features corresponding to the power equipment wandering path are output; The updated power grid state feature corresponding to each power equipment wandering path in at least one power equipment wandering path corresponding to the power equipment to be processed is subjected to feature fusion processing, and the target power grid state feature corresponding to the power equipment to be processed is output.
6. The power grid data cloud-edge collaborative communication method according to claim 5, characterized in that: The step of updating the characteristics of each power device according to the semantic characteristics of the power grid state corresponding to each power device in the power device wandering path, and outputting the updated power grid state characteristics corresponding to the power device wandering path includes: Performing a first mapping process in a feature space on the semantic feature of the power grid state corresponding to the last power device in the power device wandering path, outputting a first power grid state mapping feature, and performing a second mapping process and a third mapping process in a feature space on the semantic feature of the power grid state corresponding to the penultimate power device, outputting a second power grid state mapping feature and a third power grid state mapping feature; Based on the first grid state mapping feature corresponding to the last power device and the second grid state mapping feature corresponding to the second to last power device, determining a correlation parameter between the grid state semantic features corresponding to the two power devices; The correlation parameter between the semantic features of the power grid states corresponding to the two power devices and the third power grid state mapping feature corresponding to the penultimate power device are fused, so that the semantic feature of the power grid state corresponding to the last power device is fused into the semantic feature of the power grid state corresponding to the penultimate power device, so as to output the associated feature of the power grid state corresponding to the penultimate power device; For each power device other than the last power device in the power device wandering path, the power grid state association feature corresponding to the power device is subjected to a first mapping process in the feature space, and the corresponding first power grid state mapping feature is output, and the power grid state semantic feature corresponding to the previous power device of the power device is subjected to a second mapping process and a third mapping process in the feature space, and the corresponding second power grid state mapping feature and the third power grid state mapping feature are output respectively; Determining a correlation parameter between the semantic features of the power grid states corresponding to the two power devices based on the first power grid state mapping feature and the second power grid state mapping feature corresponding to the two power devices; The correlation parameter between the semantic features of the power grid states corresponding to the two power devices and the third power grid state mapping feature corresponding to the latter power device are fused so that the semantic features of the power grid state corresponding to the latter power device are fused into the semantic features of the power grid state corresponding to the former power device, so as to output the associated features of the power grid state corresponding to the former power device; Based on the semantic features of the power grid state corresponding to the first power device and the corresponding power grid state association features, the updated power grid state features corresponding to the wandering path of the power device are output.
7. A power grid data cloud-edge collaborative communication system, characterized in that: Applied to a cloud server, the cloud server is communicatively connected to a plurality of edge computing devices, each of the edge computing devices is communicatively connected to at least one power data acquisition device, and each of the power data acquisition devices is deployed for a power device in a target power grid operation area, the system includes: A data sequence acquisition module, used to acquire a power data sequence sent by each edge computing device based on the current cloud-edge collaborative communication rule, wherein each power data sequence is used to reflect the power grid operation status of the power device at multiple time nodes included in the current time period, and the power data sequence is formed based on the edge computing device processing the power data collected by the power data collection device; An equipment relationship network construction module, used to construct a corresponding power equipment relationship network based on each of the power data sequences; An operation abnormality assessment module is used to evaluate the operation status of the target power grid operation area based on the power equipment relationship network, output corresponding power grid operation abnormality data, and determine the corresponding target cloud-edge collaborative communication rules based on the power grid operation abnormality data; A collaborative communication rule sending module, used to send the target cloud-edge collaborative communication rule to each of the edge computing devices, wherein the edge computing device is used to replace the current cloud-edge collaborative communication rule with the target cloud-edge collaborative communication rule, and the target cloud-edge collaborative communication rule at least includes the frequency and compression method of data communication to the cloud server and the sampling parameters for collecting power data; Among them, based on the power equipment relationship network, the operating status of the target power grid operating area is evaluated, the corresponding power grid operation abnormality data is output, and based on the power grid operation abnormality data, the corresponding target cloud-edge collaborative communication rules are determined, including: In the power equipment relationship network, feature association mining is performed on the power data sequence of each power equipment to obtain the target power grid state feature corresponding to each power equipment; Based on the target power grid state characteristics corresponding to each of the electric power equipment, the operation state of the target power grid operation area is evaluated, and the corresponding power grid operation abnormality data is output; Based on the abnormal power grid operation data, a corresponding target cloud-edge collaborative communication rule is determined, wherein the frequency and sampling parameters of data communication included in the target cloud-edge collaborative communication rule have a positive correlation with the degree of abnormality represented by the abnormal power grid operation data, and the compression ratio corresponding to the compression method of data communication included in the target cloud-edge collaborative communication rule has a negative correlation with the degree of abnormality represented by the abnormal power grid operation data; Wherein, based on each of the power data sequences, a corresponding power equipment relationship network is constructed, including: Acquire a power grid electrical topology map corresponding to the target power grid operation area, wherein the power grid electrical topology map has electrical connection relationships between the various power devices; For each of the power data sequences, the power equipment corresponding to the power data sequence is used as a network node, and the power data sequence is used as the node attribute data corresponding to the network node, wherein one of the power equipment corresponds to at least one network node, and one of the network nodes corresponds to one of the power equipment, and each of the power data sequences includes a current data sequence corresponding to the power data acquisition device belonging to the current sensor, a voltage data sequence corresponding to the power data acquisition device belonging to the voltage sensor, a temperature data sequence corresponding to the power data acquisition device belonging to the temperature sensor, a geomagnetic storm data sequence corresponding to the power data acquisition device belonging to the geomagnetic field sensor, and a frequency data sequence corresponding to the power data acquisition device belonging to the frequency sensor; The same network nodes of the corresponding power equipment are associated and set, and the network nodes of the corresponding power equipment that have direct connection in the power grid electrical topology diagram are associated to form a corresponding power equipment network.
8. An electronic device, characterized in that: include: Memory for storing computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the power grid data cloud-edge collaborative communication method described in any one of claims 1-6.
Citation Information
Patent Citations
Cloud edge collaboration system evaluation method and device, storage medium and electronic equipment
CN116896512A
Cloud edge collaboration method and system, computer equipment and storage medium
CN117240734A
