Power grid situational awareness methods, devices, equipment, and readable storage media

Abstract

This application relates to a power grid situation awareness method, device, computer equipment, computer-readable storage medium, and computer program product. The method includes: locating multiple target entities corresponding to a target object from a preset knowledge graph, and determining the associated entities corresponding to the multiple target entities; for each target entity, acquiring entity information of the target entity and entity relationship information between the target entity and associated entities; constructing power grid situation association features corresponding to the target object based on the entity information and entity relationship information of the multiple target entities; performing power grid situation detection at the distribution transformer level and at the user level based on the power grid situation association features, obtaining a first power grid situation detection result and a second power grid situation detection result; and merging the first power grid situation detection result and the second power grid situation detection result into a target power grid situation awareness result. This method can improve the accuracy of power grid situation awareness.

Description

Technical Field

[0001] This application relates to the field of smart grid technology, and in particular to a power grid situational awareness method, device, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] Distributed power sources are characterized by randomness and volatility. Their large-scale integration into the distribution network can alter the power flow distribution, leading to issues such as power reversal and voltage exceeding limits. This can affect the stability of the distribution network's voltage operation and introduce unsafe factors into the network's operation. Therefore, accurate perception of the power grid situation is essential.

[0003] In traditional technologies, the power grid status is usually perceived by measuring some circuit parameters on the user side, at the distribution transformer, and on the line side.

[0004] However, this method of power grid situation awareness has a relatively simple perception dimension and a shallow perception level, resulting in low accuracy of power grid situation awareness. Summary of the Invention

[0005] Therefore, it is necessary to provide a power grid situation awareness method, device, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of power grid situation awareness in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a power grid situational awareness method, including:

[0007] Locate multiple target entities corresponding to a target object from a preset knowledge graph, and determine the associated entities corresponding to the multiple target entities. The target entities are used to represent the target object or the power grid basic equipment, parameter measurement points and customer service seats corresponding to the target object.

[0008] For each target entity, obtain the entity information of the target entity and the entity relationship information between the target entity and the associated entity;

[0009] Based on the entity information and entity relationship information of the multiple target entities, construct the power grid situation association features corresponding to the target objects;

[0010] The target object is determined to be the target power grid area. Based on the power grid situation correlation characteristics, the power grid situation of the target power grid area is detected at the distribution transformer level to obtain the first power grid situation detection result.

[0011] The target object is identified as the target electricity user. Based on the power grid situation correlation characteristics, user-level power grid situation detection is performed on the target electricity user to obtain a second power grid situation detection result.

[0012] The first power grid situation detection result and the second power grid situation detection result are combined into the target power grid situation awareness result.

[0013] In one embodiment, the step of performing distribution transformer-level grid status detection on the target grid area based on the grid status correlation characteristics to obtain a first grid status detection result includes:

[0014] According to the first feature extraction model, the transformer maintenance information feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer maintenance information feature, maintenance record detection is performed on the target power grid area to obtain maintenance record detection results; according to the second feature extraction model, the transformer equipment status feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer equipment status feature, transformer fault type detection is performed on the target power grid area to obtain transformer fault type detection results; based on the transformer equipment status feature, transformer branch fault prediction is performed on the target power grid area to obtain branch fault prediction results; the maintenance record detection results, transformer fault type detection results, and branch fault prediction results are merged into the first power grid status detection result.

[0015] In one embodiment, the step of performing user-level grid situation detection on the target electricity user based on the grid situation correlation characteristics to obtain a second grid situation detection result includes:

[0016] According to the third feature extraction model, the power load characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the power load characteristics, the power status of the target power user is detected to obtain the power status detection result; according to the fourth feature extraction model, the customer service risk correlation characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the customer service risk correlation characteristics, the customer service risk of the target power user is detected to obtain the customer service risk detection result; the power status detection result and the customer service risk detection result are merged into the second power grid situation detection result.

[0017] In one embodiment, the method further includes:

[0018] The process involves acquiring power grid data text, which includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to the multiple target objects, and customer service information text of multiple customer service agents. Based on the customer service information text, pairwise associations between the multiple customer service agents and the multiple target objects are detected to obtain a first association detection result. Based on the equipment parameter text of the basic power grid equipment corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects, pairwise associations between the multiple basic power grid equipment and the multiple parameter measurement points are detected to obtain a second association detection result. Based on the first association detection result, the second association detection result, the equipment parameter text of the multiple basic power grid equipment, and the measurement parameter text of the multiple parameter measurement points, a pre-defined knowledge graph is constructed.

[0019] In one embodiment, detecting the pairwise correlation between the multiple power grid basic devices and the multiple parameter measurement points based on the device parameter text of the power grid basic devices corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects includes:

[0020] For each basic power grid device, the device parameter text is segmented to obtain a first segmentation result, wherein the first segmentation result includes device parameter words, first word type information corresponding to the device parameter words, and first word description information corresponding to the device parameter words; based on the device parameter words, the first word type information, and the first word description information, a first entity information feature corresponding to the basic power grid device is constructed; for each parameter measurement point, the measurement parameter text is segmented to obtain a second segmentation result, wherein the second segmentation result includes measurement parameter words, second word type information corresponding to the measurement parameter words, and second word description information corresponding to the measurement parameter words; based on the measurement parameter words, the second word type information, and the second word description information, a second entity information feature corresponding to the parameter measurement point is constructed; based on the first entity information features of the multiple basic power grid devices and the second entity information features of the multiple parameter measurement points, the pairwise correlation between the multiple basic power grid devices and the multiple parameter measurement points is detected.

[0021] In one embodiment, the step of constructing a preset knowledge graph based on the correlation detection results, the equipment parameter text of the multiple power grid basic equipment, and the measurement parameter text of the multiple parameter measurement points includes:

[0022] A reference entity is selected from the multiple basic power grid devices and the multiple parameter measurement points, and the associated entity corresponding to the reference entity is determined; the equipment parameter text of the multiple basic power grid devices and the measurement parameter text of the multiple parameter measurement points are obtained; based on the association detection results between the reference entity and the associated entity, the equipment parameter text and the measurement parameter text, an entity relationship extension subgraph is constructed; the entity relationship extension subgraphs of the multiple reference entities are merged to obtain a knowledge graph.

[0023] Secondly, this application also provides a power grid situational awareness device, comprising:

[0024] The determination module is used to locate multiple target entities corresponding to a target object from a preset knowledge graph, and determine the associated entities corresponding to the multiple target entities, wherein the target entities are used to represent the target object, or the power grid basic equipment, parameter measurement point or customer service seat corresponding to the target object;

[0025] The acquisition module is used to acquire, for each target entity, the entity information of the target entity and the entity relationship information between the target entity and the associated entity;

[0026] The feature construction module is used to construct the power grid status association features corresponding to the target object based on the entity information and entity relationship information of the multiple target entities;

[0027] The first power grid status detection module is used to determine that the target object is a target power grid distribution area, and to perform power grid status detection at the distribution transformer level on the target power grid distribution area based on the power grid status correlation characteristics, so as to obtain the first power grid status detection result;

[0028] The second power grid situation detection module is used to determine the target object as the target power user, and to perform user-level power grid situation detection on the target power user based on the power grid situation correlation characteristics, so as to obtain the second power grid situation detection result.

[0029] The merging module is used to merge the first power grid situation detection result and the second power grid situation detection result into the target power grid situation awareness result.

[0030] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0031] The process involves locating multiple target entities corresponding to a target object from a pre-defined knowledge graph, and determining associated entities corresponding to these target entities. The target entities represent the target object or its corresponding basic power grid equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and its associated entities are acquired. Based on the entity information and entity relationship information of the multiple target entities, a power grid situation association feature corresponding to the target object is constructed. The target object is determined to be a target power grid distribution area. Based on the power grid situation association feature, power grid situation detection at the distribution transformer level is performed on the target power grid distribution area to obtain a first power grid situation detection result. The target object is determined to be a target electricity user. Based on the power grid situation association feature, user-level power grid situation detection is performed on the target electricity user to obtain a second power grid situation detection result. The first power grid situation detection result and the second power grid situation detection result are merged into a target power grid situation awareness result.

[0032] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0033] The process involves locating multiple target entities corresponding to a target object from a pre-defined knowledge graph, and determining associated entities corresponding to these target entities. The target entities represent the target object or its corresponding basic power grid equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and its associated entities are acquired. Based on the entity information and entity relationship information of the multiple target entities, a power grid situation association feature corresponding to the target object is constructed. The target object is determined to be a target power grid distribution area. Based on the power grid situation association feature, power grid situation detection at the distribution transformer level is performed on the target power grid distribution area to obtain a first power grid situation detection result. The target object is determined to be a target electricity user. Based on the power grid situation association feature, user-level power grid situation detection is performed on the target electricity user to obtain a second power grid situation detection result. The first power grid situation detection result and the second power grid situation detection result are merged into a target power grid situation awareness result.

[0034] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0035] The process involves locating multiple target entities corresponding to a target object from a pre-defined knowledge graph, and determining associated entities corresponding to these target entities. The target entities represent the target object or its corresponding basic power grid equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and its associated entities are acquired. Based on the entity information and entity relationship information of the multiple target entities, a power grid situation association feature corresponding to the target object is constructed. The target object is determined to be a target power grid distribution area. Based on the power grid situation association feature, power grid situation detection at the distribution transformer level is performed on the target power grid distribution area to obtain a first power grid situation detection result. The target object is determined to be a target electricity user. Based on the power grid situation association feature, user-level power grid situation detection is performed on the target electricity user to obtain a second power grid situation detection result. The first power grid situation detection result and the second power grid situation detection result are merged into a target power grid situation awareness result.

[0036] The aforementioned power grid situation awareness method, device, computer equipment, computer-readable storage medium, and computer program product first locate multiple target entities corresponding to a target object from a preset knowledge graph, and determine the associated entities corresponding to these target entities. The target entities represent the target object or its corresponding basic power grid equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and the associated entities are obtained. Then, based on the entity information and entity relationship information of multiple target entities, power grid situation association features corresponding to the target object are constructed. This allows for the extraction of rich power grid situation associations from the preset knowledge graph. These features help improve the accuracy of power grid situation awareness. Furthermore, by identifying the target object as the target power grid distribution area and performing distribution transformer-level power grid situation detection on the target power grid distribution area based on the power grid situation correlation features, a first power grid situation detection result is obtained. Additionally, by identifying the target object as the target electricity user and performing user-level power grid situation detection on the target electricity user based on the power grid situation correlation features, a second power grid situation detection result is obtained. This allows for power grid situation detection from both the distribution transformer and user levels, resulting in richer perception dimensions and finer granularity at the user level, leading to a deeper perception layer and thus improving the accuracy of power grid situation awareness. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating a power grid situational awareness method in one embodiment of this application;

[0039] Figure 2 This is a schematic diagram of the process for performing distribution transformer-level power grid status detection on a target power grid area in one embodiment of this application;

[0040] Figure 3 This is a schematic diagram of a process for performing user-level power grid status detection on a target power user in one embodiment of this application;

[0041] Figure 4 This is a structural block diagram of a power grid situational awareness device in one embodiment of this application;

[0042] Figure 5 This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] In one embodiment, such as Figure 1 As shown, a power grid situational awareness method is provided. This embodiment illustrates the method applied to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0045] Step 202: Locate multiple target entities corresponding to the target object from the preset knowledge graph, and determine the associated entities corresponding to the multiple target entities. The target entities are used to represent the target object or the power grid basic equipment, parameter measurement points and customer service seats corresponding to the target object.

[0046] In this embodiment, a preset knowledge graph is set for the target power grid. The preset knowledge graph includes multiple entities, each entity corresponding to a target object, the basic equipment of the target object, the parameter measurement point of the target object, or the customer service seat of the target object in the target power grid. If an entity corresponds to basic equipment of the power grid, the entity information contained in the entity can be the basic equipment parameter information of that basic equipment. If an entity corresponds to a parameter measurement point, the entity information contained in the entity can be the real-time data of the distribution transformer measured based on that parameter measurement point. Each transformer substation in the power grid can correspond to one or more entities in the preset knowledge graph. If an entity corresponds to a target object, the entity information contained in the entity can be the basic equipment parameter information of all basic equipment of the power grid corresponding to the target object, and the real-time data of the distribution transformer of all parameter measurement points corresponding to the target object. If an entity corresponds to a customer service seat, the entity information contained in the entity can be the text communication information between the customer service seat and the electricity user.

[0047] It should be noted that as the basic equipment parameter information of the power grid and the real-time data of the distribution transformer based on the parameter measurement points are continuously collected in real time, the entity information in the preset knowledge graph will also be dynamically updated in real time.

[0048] As an example, basic power grid equipment parameter information can include transformer information, terminal equipment information, switch information, and power grid statistics. Among them, transformer information can include transformer model, rated power, connection method, and transformer type; terminal equipment information can include terminal equipment type and terminal equipment installation location; switch information can include switch equipment model, installation location, and switch status; and power grid statistics can include the total load and peak load of the power grid area.

[0049] As an example, real-time data for distribution transformers can include measurement point number, meter reading time information, measured active power, measured reactive power, and three-phase voltage data.

[0050] As an example, step 202 includes: searching for multiple target entities corresponding to the target object from a preset knowledge graph based on the object identifier of the target object, and searching for the associated entities corresponding to each target entity in the preset knowledge graph. The associated entities are entities in the preset knowledge graph that have connection edges with the target entity. The target entity corresponds to the target object, the basic power grid equipment corresponding to the target object, the parameter measurement point of the target object, or the customer service seat corresponding to the target object. The target object can be a power grid area or an electricity user.

[0051] Step 204: For each target entity, obtain the entity information of the target entity and the entity relationship information between the target entity and related entities.

[0052] Among them, entity information is used to characterize the basic power grid equipment parameter information of the basic power grid equipment corresponding to the target entity, or to characterize the real-time distribution transformer data measured by the parameter measurement point corresponding to the target entity, or to characterize the real-time distribution transformer data measured by all parameter measurement points in the target object corresponding to the target entity and the basic power grid equipment parameter information of all basic power grid equipment, or to characterize the call text information of the customer service agent corresponding to the target entity; entity relationship information is used to characterize the relationship between the target entity and related entities. This relationship can be a device connection relationship, a monitoring and monitored relationship, a parallel relationship, a parent-child relationship, etc.

[0053] Step 206: Based on the entity information and entity relationship information of multiple target entities, construct the power grid status association features corresponding to the target object.

[0054] As an example, step 206 includes: extracting features from the entity information of each target entity to obtain entity information features of multiple target entities; extracting features from the entity relationship information between each target entity and associated entities to obtain entity relationship information features of multiple target entities; and fusing the entity information features and entity relationship information features of multiple target entities to obtain the power grid status association features corresponding to the target object.

[0055] As an example, step 206 includes: extracting features from the entity information of each target entity to obtain entity information features of multiple target entities; extracting features from the entity information of the associated entities of each target entity to obtain associated entity information features of multiple associated entities; extracting features from the entity relationship information between each target entity and its associated entities to obtain entity relationship information features of multiple target entities; and fusing the entity information features, entity relationship information features, and associated entity information features of multiple target entities to obtain the power grid situation correlation features corresponding to the target object. This allows for the addition of entity information of associated entities to the entity information and entity relationship information of the target entities to construct power grid situation correlation features, enriching the information used to construct these features and improving their accuracy, thus laying the foundation for improving the accuracy of power grid situation awareness.

[0056] Step 208: Determine the target object as the target power grid area. Based on the power grid status correlation characteristics, perform power grid status detection at the distribution transformer level on the target power grid area to obtain the first power grid status detection result.

[0057] In this embodiment, a first time series model is set up to detect the changes in the power grid status of the target power grid area in the future time interval.

[0058] As an example, step 208 includes: determining the target object as the target power grid area, inputting the power grid situation correlation features into the first time series model, detecting the change in the power grid situation of the target power grid area over time in the future time interval, and obtaining the first power grid situation detection result.

[0059] Step 210: Determine the target object as the target electricity user, and perform user-level power grid status detection on the target electricity user based on the power grid status correlation characteristics to obtain the second power grid status detection result.

[0060] In this embodiment, a second time series model is set up, while the first time series model is used to detect the changes in the power grid status of the target electricity user in the future time interval.

[0061] As an example, step 210 includes: determining the target object as the target electricity user, inputting the grid situation correlation features into the second time series model, detecting the change in the grid situation of the target electricity user over time in the future time interval, and obtaining the second grid situation detection result.

[0062] Step 212: Combine the first power grid situation detection result and the second power grid situation detection result into the target power grid situation awareness result.

[0063] In the aforementioned power grid situation awareness method, firstly, multiple target entities corresponding to the target object are located from a preset knowledge graph, and associated entities corresponding to these target entities are determined. Target entities represent the target object or its corresponding power grid basic equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and its associated entities are obtained. Then, based on the entity information and entity relationship information of multiple target entities, power grid situation association features corresponding to the target object are constructed. This method can extract rich power grid situation association features from the preset knowledge graph, which helps improve power grid situation awareness. To further improve the accuracy of power grid situation awareness, the system identifies the target object as the target power grid distribution area and performs power grid situation detection at the distribution transformer level based on the power grid situation correlation characteristics, obtaining a first power grid situation detection result. It also identifies the target object as the target electricity user and performs user-level power grid situation detection based on the power grid situation correlation characteristics, obtaining a second power grid situation detection result. This allows for power grid situation detection from both the distribution transformer and user levels, resulting in richer perception dimensions and finer granularity at the user level, leading to a deeper level of perception and thus improving the accuracy of power grid situation awareness.

[0064] In one exemplary embodiment, such as Figure 2As shown, step 208 includes steps 302 to 312. Wherein:

[0065] Step 302: Based on the first feature extraction model, extract the transformer maintenance information features of the target power grid area from the power grid status correlation features.

[0066] Step 304: Based on the characteristics of the transformer area maintenance information, perform maintenance record detection on the target power grid transformer area to obtain the maintenance record detection results.

[0067] In this embodiment, a first feature extraction model is provided, which is used to extract information features related to the maintenance information of the target power grid area from the power grid status correlation features.

[0068] As an example, steps 302 to 304 include: inputting the power grid status correlation features into the first feature extraction model to extract the substation maintenance information features of the target power grid area from the power grid status correlation features; based on the substation maintenance information features, detecting the number of power grid maintenances, the type of power grid maintenance, and the location of power grid maintenance in the target power grid area, and using the detected number of power grid maintenances, the type of power grid maintenance, and the location of power grid maintenance as the maintenance record detection result.

[0069] Step 306: Extract the distribution equipment status features of the target power grid area from the power grid status correlation features according to the second feature extraction model.

[0070] Step 308: Based on the status characteristics of the transformer substations, perform fault type detection on the target power grid transformer substations to obtain the fault type detection results.

[0071] In this embodiment, a second feature extraction model is provided, which is used to extract information features related to the faults that have occurred in the target power grid area from the power grid status correlation features.

[0072] As an example, steps 306 to 308 include: extracting the distribution equipment status features of the target power grid area from the power grid status association features by inputting the power grid situation association features into the second feature extraction model; and performing fault type detection on the equipment faults that have occurred in the target power grid area based on the distribution equipment status features to obtain the distribution area fault type detection result. The fault types include, but are not limited to, front-line fault types, equipment defect fault types, user fault types, and planned power outage fault types.

[0073] Step 310: Based on the equipment status characteristics of the distribution area, perform branch fault prediction for the target power grid distribution area to obtain the branch fault prediction results.

[0074] Among them, a branch fault refers to a fault in the energized line of the target power grid, such as a tripping circuit breaker or a blown fuse.

[0075] As an example, step 310 includes: predicting the location of a potential branch fault in the target power grid distribution area within a first future time interval based on the equipment status characteristics of the distribution area, and obtaining a first branch fault prediction result; predicting the location of a potential branch fault in the target power grid distribution area within a second future time interval based on the equipment status characteristics of the distribution area, and obtaining a second branch fault prediction result; and merging the first branch fault prediction result and the second branch fault prediction result into a final branch fault prediction result.

[0076] The second future time interval is longer than the first future time interval. The first future time interval can be the next few hours or days, while the second future time interval can be the next few months or years.

[0077] Step 312: Combine the maintenance record detection results, the transformer area fault type detection results, and the branch fault prediction results into the first power grid status detection result.

[0078] In this embodiment, the power grid situation perception of the target power grid area is realized from three dimensions: maintenance record detection, fault type judgment, and branch fault prediction. This makes the perception dimensions of the power grid situation perception richer and thus improves the accuracy of the power grid situation perception.

[0079] In one exemplary embodiment, such as Figure 3 As shown, step 210 includes steps 402 to 410. Wherein:

[0080] Step 402: Based on the third feature extraction model, extract the power load features corresponding to the target power user from the power grid status correlation features.

[0081] Step 404: Based on the characteristics of the electrical load, perform electrical status detection on the target electrical user and obtain the electrical status detection results.

[0082] The target electricity user can be an electricity user of the target power grid. This electricity user can be an individual user or a regional user representing an electricity consumption area. For example, a district of a city can be considered as a regional user. In this embodiment, a third feature extraction model is set up to extract information features related to electricity load from the power grid status correlation features.

[0083] As an example, steps 402 to 404 include: inputting the power grid status correlation features into the third feature extraction model to extract the power load features corresponding to the target power user from the power grid status correlation features; predicting the power load changes of the target power user for each preset power load type according to the power load features, and obtaining the power load change prediction results corresponding to each preset power load type; predicting the power load changes of the target power user for each preset load phase according to the power load features, and obtaining the load phase change prediction results for each preset load phase; and merging the power load change prediction results corresponding to each preset power load type and the load phase change prediction results for each preset load phase into the power status detection result.

[0084] The preset power load type can be industrial load type, agricultural load type, transportation load type, and residential load type, etc.; the preset load phase can be phase A, phase B, or phase C; the power load change prediction result can be the power load change curve over time for each preset power load type; the load phase change prediction result can be the power load change curve over time for each preset load phase.

[0085] Step 406: Based on the fourth feature extraction model, extract the customer service risk association features corresponding to the target electricity user from the power grid status association features.

[0086] Step 408: Based on the customer service risk association characteristics, perform customer service risk detection on the target electricity user and obtain the customer service risk detection results.

[0087] In this embodiment, a fourth feature extraction model is provided, which is used to extract information features related to customer service call information from the power grid status correlation features.

[0088] As an example, steps 406 to 408 include: inputting the power grid situation correlation features into the fourth feature extraction model to extract the customer service risk correlation features corresponding to the target electricity user from the power grid situation correlation features; based on the customer service risk correlation features, detecting the customer service risk type and the probability of customer service risk occurring for the target electricity user, and using the detected customer service risk type and the probability of customer service risk occurring as the customer service risk detection result.

[0089] Step 410: Combine the power consumption status detection results and customer service risk detection results into a second power grid status detection result.

[0090] In the above embodiments, for the target electricity user, the power grid situation perception can be carried out from two perception dimensions: the electricity consumption status of the target electricity user and the customer service risk. This makes the perception dimensions of the power grid situation perception richer and thus improves the accuracy of the power grid situation perception.

[0091] In one embodiment, the power grid situation awareness method further includes:

[0092] Obtain power grid data text, which includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to multiple target objects, and customer service information text of multiple customer service agents; based on the customer service information text, detect the pairwise association relationships between multiple customer service agents and multiple target objects to obtain a first association relationship detection result; based on the equipment parameter text of basic power grid equipment corresponding to multiple target objects and the measurement parameter text of parameter measurement points corresponding to multiple target objects, detect the pairwise association relationships between multiple basic power grid equipment and multiple parameter measurement points to obtain a second association relationship detection result; based on the first association relationship detection result, the second association relationship detection result, the equipment parameter text of multiple basic power grid equipment, and the measurement parameter text of multiple parameter measurement points, construct a preset knowledge graph.

[0093] In this embodiment, the power grid data collected is usually in the form of power grid data text. Therefore, a pre-defined knowledge graph can be constructed by performing some text processing on the power grid data text.

[0094] Specifically, the process involves acquiring power grid data text, which includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to multiple target objects, and customer service information text of multiple customer service agents. For each equipment parameter text, text processing is performed to obtain a first processed text. For each measurement parameter text, text processing is performed to obtain a second processed text. For each customer service agent, the customer service information text is processed to obtain a third processed text. Text processing methods may include text segmentation, synonym replacement, and text classification. Text features are extracted from each first processed text to obtain multiple first text features. Text features are extracted from each second processed text to obtain multiple second text features. Text features are extracted from each third processed text to obtain multiple third text features. Based on multiple... A third text feature is used to detect the pairwise association between each target object and each customer service representative, resulting in a first association detection result. Based on multiple first and second text features, the association between each basic power grid device and each parameter measurement point is detected, resulting in a second association detection result. Based on multiple first, second, and third processed texts, multiple entities of a pre-defined knowledge graph are constructed, where each entity corresponds to a target object, basic power grid device, parameter measurement point, or customer service representative. Each entity contains entity information consisting of first, second, and third processed texts, or a text set composed of first and second processed texts. Based on the first and second association detection results, connection edges for each entity are constructed, resulting in a pre-defined knowledge graph, where connection edges between entities are used to represent the association between entities.

[0095] In one embodiment, based on the equipment parameter text of the power grid basic equipment corresponding to multiple target objects and the measurement parameter text of the parameter measurement points corresponding to multiple target objects, the association between each pair of multiple power grid basic equipment and multiple parameter measurement points is detected, including:

[0096] For each basic power grid device, the device parameter text is segmented to obtain a first segmentation result, which includes the device parameter word, the corresponding first word type information, and the corresponding first word description information. Based on the device parameter word, the first word type information, and the first word description information, a first entity information feature corresponding to the basic power grid device is constructed. For each parameter measurement point, the measurement parameter text is segmented to obtain a second segmentation result, which includes the measurement parameter word, the corresponding second word type information, and the corresponding second word description information. Based on the measurement parameter word, the second word type information, and the second word description information, a second entity information feature corresponding to the parameter measurement point is constructed. Based on the first entity information features of multiple basic power grid devices and the second entity information features of multiple parameter measurement points, the pairwise correlation between multiple basic power grid devices and multiple parameter measurement points is detected.

[0097] In this embodiment, a thesaurus and a word segmentation library are provided. The thesaurus is used as a reference library for synonym replacement of power grid data text, and the word segmentation library is used as a reference library for word segmentation of power grid data text. The word segmentation library includes word type information and word description information corresponding to each word segmentation. The word description information is used to describe the detailed semantic information of the word segmentation, and the word type information is used to characterize the word type information of the word segmentation in the power grid data.

[0098] Specifically, for each basic power grid device, the device parameter text is replaced with synonyms based on a thesaurus; the replaced device parameter text is segmented based on a word segmentation corpus to obtain the first segmentation result, which includes the device parameter word, the first word type information corresponding to the device parameter word, and the first word description information corresponding to the device parameter word. The first word type information can be the first word type identifier; word features are extracted from the device parameter word to obtain the device parameter word features, and text features are extracted from the first word description information to obtain the first word description text features; the device parameter word features, the first word type identifier, and the first word description text features are concatenated to obtain the first entity information features of the basic power grid device.

[0099] Furthermore, for each parameter measurement point, the measurement parameter text of the parameter measurement point is replaced with synonyms according to the thesaurus; the measurement parameter text after synonym replacement is segmented according to the word segmentation corpus to obtain a second word segmentation result, wherein the second word segmentation result includes the measurement parameter word, the second word type information corresponding to the measurement parameter word, and the second word description information corresponding to the measurement parameter word, and the second word type information can be the second word type identifier; word features are extracted from the measurement parameter word to obtain measurement parameter word features, and text features are extracted from the second word description information to obtain second word description text features; the measurement parameter word features, the second word type identifier, and the second word description text features are concatenated to obtain the second entity information features of the power grid basic equipment.

[0100] Furthermore, for any two target entity information features among multiple first entity information features and multiple second entity information features, the two target entity information features are concatenated to obtain concatenated entity information features. By inputting the concatenated entity information features into the relationship detection model, the correlation between the power grid basic equipment or parameter measurement points corresponding to the two target entity information features is detected, thereby obtaining the pairwise correlation between any two of the multiple power grid basic equipment and multiple parameter measurement points.

[0101] In one embodiment, a pre-defined knowledge graph is constructed based on the correlation detection results, the equipment parameter text of multiple basic power grid devices, and the measurement parameter text of multiple parameter measurement points, including:

[0102] A reference entity is selected from multiple basic power grid equipment and multiple parameter measurement points, and the associated entities corresponding to the reference entity are determined; the equipment parameter text of multiple basic power grid equipment and the measurement parameter text of multiple parameter measurement points are obtained; based on the association detection results between the reference entity and the associated entities, the equipment parameter text and the measurement parameter text, an entity relationship extension subgraph is constructed; the entity relationship extension subgraphs of multiple reference entities are merged to obtain a knowledge graph.

[0103] Specifically, a benchmark entity is selected from multiple basic power grid devices and multiple parameter measurement points. Based on the detection results of the pairwise correlations between the multiple basic power grid devices and multiple parameter measurement points, the associated entities corresponding to the benchmark entities are found. The first processed text corresponding to the equipment parameter text of the multiple basic power grid devices and the second processed text corresponding to the measurement parameter text of the multiple parameter measurement points are obtained. Based on all the processed texts and all the second processed texts, the benchmark entity and the associated entities are used as entities in the knowledge graph, and the correlation between the benchmark entity and the associated entities is used as the edges of the knowledge graph to construct an entity relationship extension subgraph. The entity relationship extension subgraphs corresponding to all selected benchmark entities are merged to obtain the preset knowledge graph.

[0104] As an example, all basic power grid equipment and parameter measurement points are selected once as the reference entity.

[0105] In the above embodiments, firstly, power grid data text is acquired. This power grid data text includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to multiple target objects, and customer service information text of multiple customer service agents. Based on the customer service information text, the pairwise relationships between multiple customer service agents and multiple target objects are detected, resulting in a first relationship detection result. Based on the equipment parameter text of the basic power grid equipment corresponding to multiple target objects and the measurement parameter text of parameter measurement points corresponding to multiple target objects, the pairwise relationships between multiple basic power grid equipment and multiple parameter measurement points are detected, resulting in a second relationship detection result. This allows for processing power grid data text in a text processing manner, detecting pairwise relationships between multiple customer service agents and multiple target objects, as well as pairwise relationships between multiple basic power grid equipment and multiple parameter measurement points. Furthermore, based on the first relationship detection result, the second relationship detection result, the equipment parameter text of multiple basic power grid equipment, and the measurement parameter text of multiple parameter measurement points, a pre-defined knowledge graph can be constructed, laying the foundation for power grid situational awareness.

[0106] In one embodiment, firstly, based on the object identifier of the target object, multiple target entities corresponding to the target object are searched from a preset knowledge graph, and then associated entities corresponding to each target entity are searched in the preset knowledge graph. The associated entities are entities in the preset knowledge graph that have connection edges with the target entity. The target entity corresponds to the target object, the basic power grid equipment corresponding to the target object, the parameter measurement point of the target object, or the customer service seat corresponding to the target object. The target object can be a power grid substation or a power user. For each target entity, the entity information of the target entity and the entity relationship information between the target entity and the associated entities are obtained. Feature extraction is performed on the entity information of each target entity to obtain entity information features of multiple target entities. Feature extraction is performed on the entity information of the associated entities of each target entity to obtain associated entity information features of multiple associated entities. Feature extraction is performed on the entity relationship information between each target entity and the associated entities to obtain entity relationship information features of multiple target entities. The entity information features, entity relationship information features, and associated entity information features of multiple target entities are fused to obtain the power grid status association features corresponding to the target object.

[0107] Furthermore, by inputting the power grid status correlation features into the first feature extraction model, the maintenance information features of the target power grid distribution area are extracted from the power grid status correlation features. Based on the maintenance information features, the number of power grid maintenances, the type of power grid maintenance, and the location of power grid maintenance in the target power grid distribution area are detected, and the detected number of power grid maintenances, the type of power grid maintenance, and the location of power grid maintenance are used together as the maintenance record detection result. By inputting the power grid status correlation features into the second feature extraction model, the equipment status features of the target power grid distribution area are extracted from the power grid status correlation features. Based on the equipment status features, the fault type detection is performed on the equipment faults that have occurred in the target power grid distribution area to obtain the fault type detection result of the distribution area. The fault types include, but are not limited to, front-line fault types, equipment defect fault types, user fault types, and planned power outage fault types. Based on the equipment status characteristics of the distribution area, the location of possible branch faults in the target power grid distribution area within the first future time interval is predicted, resulting in the first branch fault prediction result. Based on the equipment status characteristics of the distribution area, the location of possible branch faults in the target power grid distribution area within the second future time interval is predicted, resulting in the second branch fault prediction result. The first and second branch fault prediction results are combined into the final branch fault prediction result. The maintenance record detection results, distribution area fault type detection results, and branch fault prediction results are combined into the first power grid status detection result.

[0108] Furthermore, by inputting the power grid status correlation features into the third feature extraction model, the power load features corresponding to the target power user are extracted from the power grid status correlation features. Based on the power load features, the power load changes of the target power user for each preset power load type are predicted, resulting in power load change prediction results for each preset power load type. Based on the power load features, the power load changes of the target power user for each preset load phase are predicted, resulting in load phase change prediction results for each preset load phase. The power load change prediction results for each preset power load type and the load phase change prediction results for each preset load phase are combined into the power consumption status detection result. By inputting the power grid status correlation features into the fourth feature extraction model, the customer service risk correlation features corresponding to the target power user are extracted from the power grid status correlation features. Based on the customer service risk correlation features, the customer service risk type and the probability of customer service risk occurrence for the target power user are detected, and the detected customer service risk type and the probability of customer service risk occurrence are used as the customer service risk detection result. The power consumption status detection result and the customer service risk detection result are combined into the second power grid status detection result.

[0109] Finally, the first power grid situation detection result and the second power grid situation detection result are merged into the target power grid situation awareness result.

[0110] In the above embodiments, power grid situation detection can be performed from both the distribution transformer level and the user level. The perception dimensions of power grid situation awareness are richer, and the granularity of power grid situation detection at the user level is more refined, with a deeper perception level, thus improving the accuracy of power grid situation awareness.

[0111] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0112] Based on the same inventive concept, this application also provides a power grid situational awareness device for implementing the power grid situational awareness method described above. The solution provided by this device is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more power grid situational awareness device embodiments provided below can be found in the limitations of the power grid situational awareness method described above, and will not be repeated here.

[0113] In one exemplary embodiment, such as Figure 4 As shown, a power grid situation awareness device is provided, comprising: a determination module 502, an acquisition module 504, a feature construction module 506, a first power grid situation detection module 508, a second power grid situation detection module 510, and a merging module 512, wherein:

[0114] The determination module 502 is used to locate multiple target entities corresponding to the target object from a preset knowledge graph, and determine the associated entities corresponding to the multiple target entities, wherein the target entities are used to represent the target object, or the power grid basic equipment, parameter measurement point or customer service seat corresponding to the target object;

[0115] The acquisition module 504 is used to acquire, for each target entity, the entity information of the target entity and the entity relationship information between the target entity and the associated entity;

[0116] The feature construction module 506 is used to construct the power grid situation association features corresponding to the target object based on the entity information and entity relationship information of the multiple target entities;

[0117] The first power grid situation detection module 508 is used to determine that the target object is a target power grid distribution area, and to perform power grid situation detection at the distribution transformer level on the target power grid distribution area according to the power grid situation correlation characteristics, so as to obtain the first power grid situation detection result;

[0118] The second power grid situation detection module 510 is used to determine the target object as the target power user, and to perform user-level power grid situation detection on the target power user based on the power grid situation association characteristics, so as to obtain the second power grid situation detection result.

[0119] The merging module 512 is used to merge the first power grid situation detection result and the second power grid situation detection result into the target power grid situation awareness result.

[0120] In one embodiment, the first power grid status detection module is further configured to:

[0121] According to the first feature extraction model, the transformer maintenance information feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer maintenance information feature, maintenance record detection is performed on the target power grid area to obtain maintenance record detection results; according to the second feature extraction model, the transformer equipment status feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer equipment status feature, transformer fault type detection is performed on the target power grid area to obtain transformer fault type detection results; based on the transformer equipment status feature, transformer branch fault prediction is performed on the target power grid area to obtain branch fault prediction results; the maintenance record detection results, transformer fault type detection results, and branch fault prediction results are merged into the first power grid status detection result.

[0122] In one embodiment, the second power grid status detection module is further configured to:

[0123] According to the third feature extraction model, the power load characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the power load characteristics, the power status of the target power user is detected to obtain the power status detection result; according to the fourth feature extraction model, the customer service risk correlation characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the customer service risk correlation characteristics, the customer service risk of the target power user is detected to obtain the customer service risk detection result; the power status detection result and the customer service risk detection result are merged into the second power grid situation detection result.

[0124] In one embodiment, the power grid situation awareness device further includes:

[0125] A knowledge graph construction module is used to acquire power grid data text, wherein the power grid data text includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to the multiple target objects, and customer service information text of multiple customer service agents; based on the customer service information text, the module detects the pairwise association relationships between the multiple customer service agents and the multiple target objects to obtain a first association relationship detection result; based on the equipment parameter text of the basic power grid equipment corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects, the module detects the pairwise association relationships between the multiple basic power grid equipment and the multiple parameter measurement points to obtain a second association relationship detection result; based on the first association relationship detection result, the second association relationship detection result, the equipment parameter text of the multiple basic power grid equipment, and the measurement parameter text of the multiple parameter measurement points, a preset knowledge graph is constructed.

[0126] In one embodiment, the knowledge graph construction module is further configured to:

[0127] For each basic power grid device, the device parameter text is segmented to obtain a first segmentation result, wherein the first segmentation result includes device parameter words, first word type information corresponding to the device parameter words, and first word description information corresponding to the device parameter words; based on the device parameter words, the first word type information, and the first word description information, a first entity information feature corresponding to the basic power grid device is constructed; for each parameter measurement point, the measurement parameter text is segmented to obtain a second segmentation result, wherein the second segmentation result includes measurement parameter words, second word type information corresponding to the measurement parameter words, and second word description information corresponding to the measurement parameter words; based on the measurement parameter words, the second word type information, and the second word description information, a second entity information feature corresponding to the parameter measurement point is constructed; based on the first entity information features of the multiple basic power grid devices and the second entity information features of the multiple parameter measurement points, the pairwise correlation between the multiple basic power grid devices and the multiple parameter measurement points is detected.

[0128] In one embodiment, the knowledge graph construction module is further configured to:

[0129] A reference entity is selected from the multiple basic power grid devices and the multiple parameter measurement points, and the associated entity corresponding to the reference entity is determined; the equipment parameter text of the multiple basic power grid devices and the measurement parameter text of the multiple parameter measurement points are obtained; based on the association detection results between the reference entity and the associated entity, the equipment parameter text and the measurement parameter text, an entity relationship extension subgraph is constructed; the entity relationship extension subgraphs of the multiple reference entities are merged to obtain a knowledge graph.

[0130] Each module in the aforementioned power grid situational awareness device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0131] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores power grid text data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a power grid situational awareness method.

[0132] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0133] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0134] The process involves locating multiple target entities corresponding to a target object from a pre-defined knowledge graph, and determining associated entities corresponding to these target entities. The target entities represent the target object or its corresponding basic power grid equipment, parameter measurement points, and customer service seats. For each target entity, entity information and entity relationship information between the target entity and its associated entities are acquired. Based on the entity information and entity relationship information of the multiple target entities, a power grid situation association feature corresponding to the target object is constructed. The target object is determined to be a target power grid distribution area. Based on the power grid situation association feature, power grid situation detection at the distribution transformer level is performed on the target power grid distribution area to obtain a first power grid situation detection result. The target object is determined to be a target electricity user. Based on the power grid situation association feature, user-level power grid situation detection is performed on the target electricity user to obtain a second power grid situation detection result. The first power grid situation detection result and the second power grid situation detection result are merged into a target power grid situation awareness result.

[0135] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0136] According to the first feature extraction model, the transformer maintenance information feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer maintenance information feature, maintenance record detection is performed on the target power grid area to obtain maintenance record detection results; according to the second feature extraction model, the transformer equipment status feature of the target power grid area is extracted from the power grid status correlation feature; based on the transformer equipment status feature, transformer fault type detection is performed on the target power grid area to obtain transformer fault type detection results; based on the transformer equipment status feature, transformer branch fault prediction is performed on the target power grid area to obtain branch fault prediction results; the maintenance record detection results, transformer fault type detection results, and branch fault prediction results are merged into the first power grid status detection result.

[0137] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0138] According to the third feature extraction model, the power load characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the power load characteristics, the power status of the target power user is detected to obtain the power status detection result; according to the fourth feature extraction model, the customer service risk correlation characteristics corresponding to the target power user are extracted from the power grid situation correlation characteristics; based on the customer service risk correlation characteristics, the customer service risk of the target power user is detected to obtain the customer service risk detection result; the power status detection result and the customer service risk detection result are merged into the second power grid situation detection result.

[0139] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0140] The process involves acquiring power grid data text, which includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to the multiple target objects, and customer service information text of multiple customer service agents. Based on the customer service information text, pairwise associations between the multiple customer service agents and the multiple target objects are detected to obtain a first association detection result. Based on the equipment parameter text of the basic power grid equipment corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects, pairwise associations between the multiple basic power grid equipment and the multiple parameter measurement points are detected to obtain a second association detection result. Based on the first association detection result, the second association detection result, the equipment parameter text of the multiple basic power grid equipment, and the measurement parameter text of the multiple parameter measurement points, a pre-defined knowledge graph is constructed.

[0141] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0142] For each basic power grid device, the device parameter text is segmented to obtain a first segmentation result, wherein the first segmentation result includes device parameter words, first word type information corresponding to the device parameter words, and first word description information corresponding to the device parameter words; based on the device parameter words, the first word type information, and the first word description information, a first entity information feature corresponding to the basic power grid device is constructed; for each parameter measurement point, the measurement parameter text is segmented to obtain a second segmentation result, wherein the second segmentation result includes measurement parameter words, second word type information corresponding to the measurement parameter words, and second word description information corresponding to the measurement parameter words; based on the measurement parameter words, the second word type information, and the second word description information, a second entity information feature corresponding to the parameter measurement point is constructed; based on the first entity information features of the multiple basic power grid devices and the second entity information features of the multiple parameter measurement points, the pairwise correlation between the multiple basic power grid devices and the multiple parameter measurement points is detected.

[0143] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0144] A reference entity is selected from the multiple basic power grid devices and the multiple parameter measurement points, and the associated entity corresponding to the reference entity is determined; the equipment parameter text of the multiple basic power grid devices and the measurement parameter text of the multiple parameter measurement points are obtained; based on the association detection results between the reference entity and the associated entity, the equipment parameter text and the measurement parameter text, an entity relationship extension subgraph is constructed; the entity relationship extension subgraphs of the multiple reference entities are merged to obtain a knowledge graph.

[0145] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0146] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0147] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0148] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0149] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A power grid situational awareness method, characterized in that, The method includes: Locate multiple target entities corresponding to a target object from a preset knowledge graph, and determine the associated entities corresponding to the multiple target entities. The target entities are used to represent the target object or the power grid basic equipment, parameter measurement points and customer service seats corresponding to the target object. For each target entity, obtain the entity information of the target entity and the entity relationship information between the target entity and the associated entity; Based on the entity information and entity relationship information of the multiple target entities, construct the power grid situation association features corresponding to the target objects; The target object is identified as a target power grid distribution area. Based on a first feature extraction model, the distribution area maintenance information features of the target power grid distribution area are extracted from the power grid status correlation features. Based on the distribution area maintenance information features, maintenance record detection is performed on the target power grid distribution area to obtain maintenance record detection results. Based on a second feature extraction model, the distribution area equipment status features of the target power grid distribution area are extracted from the power grid status correlation features. Based on the distribution area equipment status features, distribution area fault type detection is performed on the target power grid distribution area to obtain distribution area fault type detection results. Based on the distribution area equipment status features, branch fault prediction is performed on the target power grid distribution area to obtain branch fault prediction results. The maintenance record detection results, the distribution area fault type detection results, and the branch fault prediction results are merged into a first power grid status detection result. The target object is identified as the target electricity user. Based on the third feature extraction model, the electricity load characteristics corresponding to the target electricity user are extracted from the power grid situation correlation features. Based on the electricity load characteristics, the electricity status of the target electricity user is detected to obtain an electricity status detection result. Based on the fourth feature extraction model, the customer service risk correlation characteristics corresponding to the target electricity user are extracted from the power grid situation correlation features. Based on the customer service risk correlation characteristics, the customer service risk of the target electricity user is detected to obtain a customer service risk detection result. The electricity status detection result and the customer service risk detection result are merged into a second power grid situation detection result. The first power grid situation detection result and the second power grid situation detection result are combined into the target power grid situation awareness result.

2. The method according to claim 1, characterized in that, The method further includes: Obtain power grid data text, wherein the power grid data text includes equipment parameter text of basic power grid equipment corresponding to multiple target objects, measurement parameter text of parameter measurement points corresponding to the multiple target objects, and customer service information text of multiple customer service agents; Based on the customer service information text, detect the pairwise association between the multiple customer service agents and the multiple target objects to obtain the first association detection result; Based on the equipment parameter text of the power grid basic equipment corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects, the pairwise correlation between the multiple power grid basic equipment and the multiple parameter measurement points is detected to obtain the second correlation detection result; Based on the first association detection result, the second association detection result, the equipment parameter text of the multiple basic power grid devices, and the measurement parameter text of the multiple parameter measurement points, a preset knowledge graph is constructed.

3. The method according to claim 2, characterized in that, The step of detecting the pairwise correlation between multiple power grid basic devices and multiple parameter measurement points based on the device parameter text of the power grid basic equipment corresponding to the multiple target objects and the measurement parameter text of the parameter measurement points corresponding to the multiple target objects includes: For each basic power grid device, the device parameter text of the basic power grid device is segmented into words to obtain a first segmentation result, wherein the first segmentation result includes device parameter words, first word type information corresponding to the device parameter words, and first word description information corresponding to the device parameter words; based on the device parameter words, the first word type information, and the first word description information, a first entity information feature corresponding to the basic power grid device is constructed. For each parameter measurement point, the measurement parameter text of the parameter measurement point is segmented into words to obtain a second word segmentation result. The second word segmentation result includes the measurement parameter word, the second word type information corresponding to the measurement parameter word, and the second word description information corresponding to the measurement parameter word. Based on the measurement parameter word, the second word type information, and the second word description information, a second entity information feature corresponding to the parameter measurement point is constructed. Based on the first entity information features of the plurality of basic power grid equipment and the second entity information features of the plurality of parameter measurement points, the pairwise correlation between the plurality of basic power grid equipment and the plurality of parameter measurement points is detected.

4. A power grid situational awareness device, characterized in that, The device includes: The determination module is used to locate multiple target entities corresponding to a target object from a preset knowledge graph, and determine the associated entities corresponding to the multiple target entities, wherein the target entities are used to represent the target object, or the power grid basic equipment, parameter measurement point or customer service seat corresponding to the target object; The acquisition module is used to acquire, for each target entity, the entity information of the target entity and the entity relationship information between the target entity and the associated entity; The feature construction module is used to construct the power grid status association features corresponding to the target object based on the entity information and entity relationship information of the multiple target entities; The first power grid situation detection module is used to determine that the target object is a target power grid distribution area; extract the distribution area maintenance information features of the target power grid distribution area from the power grid situation association features according to a first feature extraction model; perform maintenance record detection on the target power grid distribution area according to the distribution area maintenance information features to obtain maintenance record detection results; extract the distribution area equipment status features of the target power grid distribution area from the power grid situation association features according to a second feature extraction model; perform distribution area fault type detection on the target power grid distribution area according to the distribution area equipment status features to obtain distribution area fault type detection results; perform distribution area branch fault prediction on the target power grid distribution area according to the distribution area equipment status features to obtain branch fault prediction results; and merge the maintenance record detection results, the distribution area fault type detection results, and the branch fault prediction results into a first power grid situation detection result. The second power grid situation detection module is used to determine that the target object is a target electricity user, extract the electricity load characteristics corresponding to the target electricity user from the power grid situation correlation characteristics according to the third feature extraction model, perform electricity status detection on the target electricity user according to the electricity load characteristics, and obtain electricity status detection results; extract the customer service risk correlation characteristics corresponding to the target electricity user from the power grid situation correlation characteristics according to the fourth feature extraction model, perform customer service risk detection on the target electricity user according to the customer service risk correlation characteristics, and obtain customer service risk detection results; and merge the electricity status detection results and the customer service risk detection results into the second power grid situation detection result. The merging module is used to merge the first power grid situation detection result and the second power grid situation detection result into the target power grid situation awareness result.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images