A power outage information aggregation pushing method and system

Abstract

The application discloses a power failure information aggregation pushing method and system, and the method comprises the following steps: obtaining the operation data of a target power supply device based on a received power failure consultation request, to generate a fault work order of the target power supply device and obtain a plurality of target users matched with the target power supply device; matching the device maintenance work order corresponding to the fault work order to obtain the corresponding maintenance track and the current maintenance personnel; obtaining the real-time positioning information of the current maintenance personnel to predict the fault maintenance time of the target power supply device according to the real-time positioning information and the maintenance track; obtaining the historical maintenance information of the current maintenance personnel and the structure information of the target power supply device to predict the maintenance duration of the target power supply device; generating the power failure information corresponding to the power failure consultation request based on the fault maintenance time, the maintenance duration and the maintenance track, and pushing the power failure information to the target users to improve the accuracy of the power failure information content and the pushing efficiency.

Description

Technical Field

[0001] This invention relates to the field of power management technology, specifically to a method and system for aggregating and pushing power outage information. Background Technology

[0002] With the development of technology and the increasing demand for electricity in people's daily lives, responding promptly to users' inquiries and providing them with specific and accurate outage information when power outages occur has become an important development direction for power supply companies in order to soothe users' emotional fluctuations caused by power outages.

[0003] In the current technology for disseminating power outage information, power supply companies often need to collect and process a large amount of operational data from power equipment to comprehensively analyze the authenticity of the outage event, which is time-consuming and affects the speed of power outage information dissemination. In addition, when a power outage actually occurs, it often causes power outages for users within a certain range. When different users within that range report power outages, it results in a large number of duplicate fault work orders for the same faulty equipment. This causes maintenance personnel to continuously receive fault work orders for the same faulty equipment, which not only affects the maintenance efficiency of maintenance personnel but also wastes data resources.

[0004] In addition, since fault recovery often takes time, existing technologies often only provide the cause of the power outage and the estimated repair time when sending out power outage notifications, which makes it impossible for users to obtain the power restoration progress in a timely manner, causing users to feel anxious about power usage. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention discloses a method and system for aggregating and pushing power outage information, which improves the accuracy of power outage information content and the efficiency of its delivery.

[0006] To achieve the above objectives, the present invention discloses a method for aggregating and pushing power outage information, characterized in that it includes:

[0007] Based on the received power outage consultation request, obtain the operating data of the target power supply equipment, generate a fault work order for the target power supply equipment and obtain a number of target users matching the target power supply equipment based on the operating data;

[0008] Match the fault work order with the corresponding equipment maintenance work order, and obtain the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order;

[0009] Obtain the real-time location information of the current maintenance personnel, and predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory;

[0010] Obtain the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment to predict the maintenance time of the target power supply equipment;

[0011] Based on the fault repair time, the repair duration, and the repair trajectory, power outage information corresponding to the power outage consultation request is generated, and the power outage information is pushed to the target user.

[0012] This invention discloses a method for aggregating and pushing power outage information. Upon receiving a power outage inquiry request from a user, the method first locates the corresponding target power supply equipment to generate a corresponding fault work order. This precise location of the fault source reduces data processing volume and improves message push efficiency. Secondly, it matches the fault work order with the corresponding equipment maintenance work order to obtain specific equipment maintenance information, such as maintenance trajectory and personnel. Based on this information, it predicts the power restoration progress, such as maintenance time, so that users can understand the specific power restoration progress based on the power outage information, improving user satisfaction. Finally, when pushing the power outage information, it pushes the information to several target users matched with the target power supply equipment. This avoids receiving duplicate power outage inquiry requests, improving information push efficiency, and ensures accurate delivery of the information to users affected by the power outage, guaranteeing message push accuracy.

[0013] As a preferred example, the step of obtaining operational data of the target power supply equipment based on the received power outage consultation request, and generating a fault work order for the target power supply equipment and obtaining a number of target users matching the target power supply equipment based on the operational data, includes:

[0014] Based on the power outage inquiry request, obtain the power outage location information fed back by the user, and determine the target power supply equipment to supply power to the power outage location information based on the power outage location information and the pre-saved power grid lines;

[0015] Collect real-time operating data of the target power supply equipment;

[0016] When it is determined that the target power supply equipment has a power outage fault based on the real-time operating data and preset operating parameter thresholds, the equipment data of the target power supply equipment is obtained, and a fault work order for the target power supply equipment is generated based on the equipment data; wherein, the equipment data includes the maintenance location information, environmental data, topology connection information and historical operating data of the target power supply equipment.

[0017] In the above scheme, upon receiving a power outage inquiry request from a user, the system first collects the operational information of the target power supply equipment supplying power to the user. This information is used to accurately determine whether the power outage is caused by the target power supply equipment or a fault in the user's own equipment, thus avoiding unnecessary subsequent data processing. Power outage messages are only pushed when a power outage is determined to be occurring in the target power supply equipment; that is, a fault work order is generated for the target power supply equipment to initiate repairs. The fault work order is created based on the equipment data of the target power supply equipment to improve the completeness of the work order content, thereby increasing the accuracy of subsequent work order matching.

[0018] As a preferred example, the step of obtaining operational data of the target power supply equipment based on the received power outage consultation request, and generating a fault work order for the target power supply equipment and obtaining a number of target users matching the target power supply equipment based on the operational data, includes:

[0019] When it is determined that the target power supply equipment has a power outage fault, the obtained topology connection information and the maintenance location information corresponding to the target power supply equipment are input into the pre-built deep learning model so as to predict the power outage range corresponding to the target power supply equipment through the deep learning model;

[0020] Multiple sub-regions pre-divided according to the power grid lines are obtained, and several target sub-regions for power outage are determined from the multiple sub-regions according to the power outage range;

[0021] Obtain the region identifier of each target sub-region, and determine a number of target users that match the target sub-region based on the region identifier.

[0022] In the above scheme, after determining that the target power supply equipment has failed, in order to avoid receiving and processing a large number of repetitive power outage inquiry requests, the scope of the power outage caused by the power outage of the target power supply equipment is first assessed, and multiple users within the power outage scope are obtained, so that the generated power outage information is pushed to each user in a timely manner, so that users affected by the power outage can know the corresponding power restoration progress in a timely manner, thereby improving user satisfaction with electricity use.

[0023] As a preferred example, the step of matching the fault work order with the corresponding equipment maintenance work order and obtaining the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order includes:

[0024] Retrieve several equipment maintenance work orders with execution status of pending execution or in progress from the pre-built database;

[0025] The content similarity between each equipment maintenance work order and the fault work order is obtained, so as to match the equipment maintenance work order corresponding to the fault work order based on the content similarity.

[0026] In the above scheme, based on the existing power maintenance process, corresponding equipment maintenance work orders are generated for pre-defined power outage plans. To improve the accuracy of power outage information generation, fault work orders and equipment maintenance work orders are first matched. Specifically, using the content similarity of work orders during the matching process can improve the accuracy of the matching.

[0027] As a preferred example, the step of obtaining the content similarity between each equipment maintenance work order and the fault work order, and matching the equipment maintenance work order corresponding to the fault work order based on the content similarity, includes:

[0028] For any of the aforementioned equipment maintenance work orders;

[0029] The equipment repair work order and the fault work order are input into a pre-built content similarity evaluation model, so that the title similarity and body text similarity of the equipment repair work order and the fault work order can be output through the content similarity evaluation model;

[0030] When the title similarity is greater than or equal to a preset similarity threshold and the body text similarity is greater than or equal to the similarity threshold, the equipment repair work order is determined to be the equipment repair work order corresponding to the fault work order.

[0031] In the above scheme, the constructed model is used to evaluate the title similarity and body similarity between the equipment maintenance work order and the fault work order, which improves both the accuracy and efficiency of work order matching, thereby improving the efficiency and accuracy of power outage information push.

[0032] As a preferred example, the step of matching the fault work order with the corresponding equipment maintenance work order and obtaining the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order includes:

[0033] Obtain the user terminal identifier and user identifier that respond to the equipment maintenance work order, so as to determine the current maintenance personnel based on the user identifier and locate the current maintenance personnel based on the user terminal identifier, thereby obtaining the current location information of the current maintenance personnel;

[0034] Obtain the equipment location information of each piece of equipment to be repaired in the equipment maintenance work order, and determine real-time traffic information based on the equipment location information and the current location information;

[0035] The maintenance trajectory corresponding to the equipment maintenance work order is generated based on the real-time traffic information, the equipment location information, and the current location information.

[0036] In the above scheme, after matching the equipment maintenance work order corresponding to the fault work order, the road condition information is determined based on the location information of the equipment to be repaired and the location information of the target power supply in the equipment maintenance work order, so as to avoid delaying the operation and maintenance progress due to real-time road conditions. Then, the maintenance trajectory corresponding to the equipment maintenance work order is formulated using the location information of the equipment to be repaired, the location information of the target power supply, and the road condition information, and the maintenance trajectory is sent to the operation and maintenance personnel, which can improve the efficiency of equipment operation and maintenance.

[0037] As a preferred example, generating the maintenance trajectory corresponding to the equipment maintenance work order based on the real-time traffic information, the equipment location information, and the current location information includes:

[0038] Based on the device location information and the current location information, the initial shortest trajectory between two adjacent devices to be repaired is determined by a preset shortest path search algorithm.

[0039] Based on the shortest trajectory, the fault area where two adjacent devices to be repaired are located is determined, and real-time traffic information of the fault area is obtained;

[0040] The initial shortest trajectory is updated based on the real-time traffic information to obtain the real-time shortest trajectory between two adjacent devices to be repaired.

[0041] Based on the device location information, the current location information, and the real-time shortest trajectory between two adjacent devices to be repaired, a repair trajectory corresponding to the device repair work order is generated.

[0042] In the above scheme, the location information of each pair of devices is used to determine the shortest path between them, thereby improving maintenance efficiency by reducing the length of the maintenance path. Specifically, when determining the shortest path, traffic information corresponding to the current shortest path is considered. The selection of the shortest path is updated based on this traffic information. For example, if the traffic information for the shortest path shows congestion, while the traffic information for a slightly longer second shortest path shows smooth traffic, then the second shortest path is selected as the shortest path. This method is used to plan the maintenance path, thereby improving maintenance efficiency.

[0043] As a preferred example, obtaining the real-time location information of the current maintenance personnel, and predicting the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory, includes:

[0044] The real-time location information of the current maintenance personnel is obtained, and a real-time operation and maintenance trajectory of the current maintenance personnel is generated based on the real-time location information; wherein, the real-time operation and maintenance trajectory includes the real-time location time and the real-time location information;

[0045] When it is determined from the real-time operation and maintenance trajectory that the current maintenance personnel are performing maintenance along the maintenance trajectory, the historical operation and maintenance information of the current maintenance personnel and the equipment information of each device to be maintained are obtained based on the user identifier.

[0046] The historical operation and maintenance information, the equipment information, the equipment location information, the maintenance location information of the target power supply equipment, and the real-time operation and maintenance trajectory are input into a pre-built deep learning model to predict the fault repair time of the target power supply equipment through the deep learning model.

[0047] In the above scheme, firstly, the real-time location information of maintenance personnel is monitored to enable them to perform maintenance according to the established maintenance path, ensuring the accuracy of power outage information. Secondly, when predicting maintenance time, a deep learning model is used to improve the accuracy and efficiency of the prediction. Historical maintenance information of maintenance personnel, equipment information and location information of the target power supply equipment, maintenance location information of the target power supply equipment, and the real-time maintenance trajectory are used as the basis for prediction, thereby improving the accuracy of the prediction and ultimately enhancing the accuracy of the power outage information.

[0048] As a preferred example, the step of obtaining the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment to predict the maintenance time of the target power supply equipment includes:

[0049] Based on the user identifier of the current maintenance personnel, retrieve the historical maintenance information corresponding to the current maintenance personnel; wherein, the historical maintenance information includes the historical maintenance equipment, the structural information of the historical maintenance equipment, and the maintenance time of the historical maintenance equipment;

[0050] Retrieve the structural information of the target power supply device based on its device identifier;

[0051] The structural information and the historical operation and maintenance information are input into a pre-built maintenance time prediction model, so as to output the maintenance time of the target power supply equipment through the maintenance time prediction model.

[0052] In the above solution, to provide users with more detailed power outage information, a model is used to predict the repair time for maintenance personnel to repair the target power supply equipment, thereby improving users' perception of the power restoration progress. Specifically, predicting the repair time based on the maintenance personnel's historical maintenance information and the structural information of the target power supply equipment improves the accuracy of the prediction, and consequently, the accuracy of the power outage information.

[0053] As a preferred example, the step of matching the fault work order with the corresponding equipment maintenance work order and obtaining the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order includes:

[0054] When no equipment maintenance work order corresponding to the fault work order can be matched based on the title similarity and the body text similarity, the fault work order is sent to several maintenance personnel who are in an idle state according to the maintenance location information of the target power supply equipment.

[0055] From among the various maintenance personnel, determine the target maintenance personnel who will respond to the fault work order, and designate the target maintenance personnel as the current maintenance personnel;

[0056] Obtain the current location information of the target maintenance personnel, and determine real-time traffic information based on the current location information and the maintenance location information;

[0057] The maintenance trajectory corresponding to the fault work order is generated based on the real-time traffic information, the maintenance location information, and the current location information.

[0058] In the above scheme, when no matching equipment maintenance work order can be found for the fault work order, it indicates that the fault of the target power supply equipment is temporary. In this case, in order to perform timely maintenance on the target power supply equipment, the fault work order is first sent to available maintenance personnel so that the maintenance personnel can accurately repair the target power supply equipment. Furthermore, a maintenance trajectory is established using the real-time location information of the maintenance personnel and the location information of the target power supply equipment, allowing users to understand the power restoration progress in a timely manner, thereby improving the user's power experience.

[0059] As a preferred example, obtaining the real-time location information of the current maintenance personnel, and predicting the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory, includes:

[0060] Obtain the real-time location information of the target maintenance personnel, and determine the movement speed of the target maintenance personnel based on the real-time location information;

[0061] The failure repair time of the target power supply equipment is predicted based on the moving speed and the repair trajectory.

[0062] In the above scheme, based on the current maintenance personnel accurately repairing the target power supply equipment, the repair time of the equipment can be accurately predicted by calculating the user's movement speed, thereby improving the efficiency and accuracy of power outage information formulation.

[0063] As a preferred example, the step of generating power outage information corresponding to the power outage consultation request based on the fault repair time, the repair duration, and the repair trajectory, and pushing the power outage information to the target user, includes:

[0064] Extract key information corresponding to the maintenance trajectory, maintenance duration, and fault maintenance time, and output structured power outage information based on the key information and pre-saved information templates;

[0065] Based on the area identifier and the pre-saved user information database, several target users within the power outage area and the message queue address of each target user are matched.

[0066] The power outage information is pushed to each of the target users according to the message queue address.

[0067] In the above scheme, after generating power outage information including the maintenance trajectory, maintenance duration, and fault repair time, in order to ensure that the power outage information can be accurately pushed to all users affected by the power outage of the target power supply equipment, the area identifier and the pre-saved user information database can be used to match the message queue address corresponding to each user, and then the power outage information can be accurately pushed according to the message queue address and the user information database.

[0068] As a preferred example, the step of pushing the power outage information to each target user according to the message queue address includes:

[0069] The power outage information is encapsulated using a message middleware;

[0070] Set message topics and routing rules for the encapsulated power outage information;

[0071] The encapsulated power outage information is pushed to each target user according to the message topic and routing rules.

[0072] In the above scheme, the power outage information is encapsulated through a message middleware to achieve rapid distribution of the power outage information. At the same time, message topics and routing rules are set for the encapsulated power outage information to ensure accurate delivery of the power outage information.

[0073] On the other hand, the present invention also discloses a power outage information aggregation and push system, including a power outage analysis module, a maintenance monitoring module, a maintenance prediction module, a duration prediction module and a message push module;

[0074] The power outage analysis module is used to obtain the operating data of the target power supply equipment based on the received power outage consultation request, so as to generate a fault work order for the target power supply equipment and obtain a number of target users matching the target power supply equipment based on the operating data;

[0075] The maintenance monitoring module is used to match the equipment maintenance work order corresponding to the fault work order, and to obtain the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order;

[0076] The maintenance prediction module is used to obtain the real-time location information of the current maintenance personnel, so as to predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory.

[0077] The duration prediction module is used to obtain the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment, so as to predict the maintenance duration of the target power supply equipment;

[0078] The message push module is used to generate power outage information corresponding to the power outage consultation request based on the fault repair time, the repair duration and the repair trajectory, and push the power outage information to the target user.

[0079] This invention discloses a power outage information aggregation and push system. Upon receiving a power outage inquiry request from a user, the system first locates the corresponding target power supply equipment to generate a corresponding fault work order. This precise location of the fault source reduces data processing volume and improves message push efficiency. Secondly, it matches the fault work order with the corresponding equipment maintenance work order to obtain specific equipment maintenance information, such as maintenance trajectory and personnel. Based on this information, it predicts the power restoration progress, such as maintenance time, so that users can understand the specific power restoration progress based on the power outage information, improving user satisfaction. Finally, when pushing the power outage information, it pushes the information to several target users matched with the target power supply equipment. This avoids receiving duplicate power outage inquiry requests, improving information push efficiency, and ensures accurate push of the information to users affected by the power outage, guaranteeing message push accuracy. Attached Figure Description

[0080] Figure 1 This is a flowchart illustrating a method for aggregating and pushing power outage information according to an embodiment of the present invention.

[0081] Figure 2 This invention discloses a schematic diagram of the structure of a power outage information aggregation and push system according to an embodiment of the present invention;

[0082] Figure 3 This is a flowchart illustrating a method for aggregating and pushing power outage information, as disclosed in another embodiment of the present invention. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0084] Example 1

[0085] This embodiment provides a method for aggregating and pushing power outage information to improve the efficiency and accuracy of the push notification. For details on the implementation of this push method, please refer to [link / reference needed]. Figure 1 It mainly includes steps 101 to 105, and the steps are mainly as follows:

[0086] Step 101: Based on the received power outage inquiry request, obtain the operating data of the target power supply equipment, and generate a fault work order for the target power supply equipment and obtain several target users matching the target power supply equipment according to the operating data;

[0087] Step 102: Match the equipment maintenance work order corresponding to the fault work order, and obtain the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order;

[0088] Step 103: Obtain the real-time location information of the current maintenance personnel, so as to predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory;

[0089] Step 104: Obtain the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment to predict the maintenance time of the target power supply equipment;

[0090] Step 105: Generate power outage information corresponding to the power outage consultation request based on the fault repair time, the repair duration, and the repair trajectory, and push the power outage information to the target user.

[0091] In some embodiments of this example, to improve the efficiency and accuracy of the push notification, step 101 preferably involves the following steps: generating work orders and identifying target users. Specifically, the steps are as follows:

[0092] Step 1011: Obtain the power outage location information fed back by the user based on the power outage consultation request, and determine the target power supply equipment to supply power to the power outage location information based on the power outage location information and the pre-saved power grid lines;

[0093] Step 1012: Collect real-time operating data of the target power supply equipment;

[0094] Step 1013: When it is determined that the target power supply equipment has a power outage fault based on the real-time operating data and the preset operating parameter threshold, the equipment data of the target power supply equipment is obtained, and a fault work order for the target power supply equipment is generated based on the equipment data; wherein, the equipment data includes the maintenance location information, environmental data, topology connection information and historical operating data of the target power supply equipment.

[0095] Step 1014: When it is determined that the target power supply equipment has a power outage fault, the obtained topology connection information and the maintenance location information corresponding to the target power supply equipment are input into the pre-built deep learning model so as to predict the power outage range corresponding to the target power supply equipment through the deep learning model;

[0096] Step 1015: Obtain multiple sub-regions pre-divided according to the power grid lines, so as to determine several target sub-regions for power outage from the multiple sub-regions according to the power outage range;

[0097] Step 1016: Obtain the region identifier of each target sub-region, so as to determine a number of target users matching the target sub-region based on the region identifier.

[0098] In this embodiment, after receiving a power outage inquiry request from a user, steps 1011 to 1013 first collect the operating information of the target power supply equipment supplying power to the user. This information is used to accurately determine whether the power outage is caused by the target power supply equipment or a fault in the user's own equipment, thus avoiding unnecessary data processing. Power outage messages are only pushed when the target power supply equipment is determined to have experienced a power outage, i.e., a fault work order is generated for the target power supply equipment to initiate repairs. The fault work order is created based on the equipment data of the target power supply equipment to improve the completeness of the work order content and thus increase the accuracy of subsequent work order matching. Steps 1014 to 1016, after determining that the target power supply equipment has malfunctioned, to avoid receiving and processing a large number of repetitive power outage inquiry requests, first assess the scope of the power outage caused by the target power supply equipment's outage and acquire multiple users within that scope. This allows the generated power outage information to be simultaneously pushed to each user, ensuring that affected users are promptly informed of the power restoration progress and improving user satisfaction with electricity usage.

[0099] In some implementations of this embodiment, after constructing the fault work order and determining the target users for information push, in order to improve the accuracy of the power outage information content, step 102 performs work order matching and current maintenance personnel matching through the following steps. The steps are as follows:

[0100] Step 1021: Retrieve several equipment maintenance work orders with execution status of pending execution or in progress from the pre-built database;

[0101] Step 1022: Obtain the content similarity between each of the equipment maintenance work orders and the fault work orders, so as to match the equipment maintenance work orders corresponding to the fault work orders based on the content similarity.

[0102] In this embodiment, steps 1021 to 1022 generate corresponding equipment maintenance work orders based on existing power maintenance plans for pre-defined power outages. To improve the accuracy of power outage information generation, fault work orders and equipment maintenance work orders are first matched. Specifically, using the content similarity of work orders during the matching process can improve matching accuracy.

[0103] Furthermore, in step 1022, when matching work orders based on the content similarity, the following steps can be used to improve the accuracy of the matching. These steps are:

[0104] Step 10221: For any of the equipment maintenance work orders; input the equipment maintenance work order and the fault work order into a pre-built content similarity evaluation model, so as to output the title similarity and body text similarity of the equipment maintenance work order and the fault work order through the content similarity evaluation model;

[0105] Step 10222: When the title similarity is greater than or equal to a preset similarity threshold and the body text similarity is greater than or equal to the similarity threshold, the equipment repair work order is determined to be the equipment repair work order corresponding to the fault work order;

[0106] Step 10223: Obtain the user terminal identifier and user identifier that respond to the equipment maintenance work order, so as to determine the current maintenance personnel based on the user identifier and locate the current maintenance personnel based on the user terminal identifier, and obtain the current location information of the current maintenance personnel;

[0107] Step 10224: Obtain the equipment location information of each piece of equipment to be repaired in the equipment maintenance work order, so as to determine the real-time traffic information based on the equipment location information and the current location information;

[0108] Step 10225: Generate the maintenance trajectory corresponding to the equipment maintenance work order based on the real-time traffic information, the equipment location information, and the current location information;

[0109] Step 10226: When no equipment maintenance work order corresponding to the fault work order can be matched based on the title similarity and the body text similarity, the fault work order is sent to several maintenance personnel who are in an idle state according to the maintenance location information of the target power supply equipment.

[0110] Step 10227: Determine the target maintenance personnel who will respond to the fault work order from among the maintenance personnel, and designate the target maintenance personnel as the current maintenance personnel;

[0111] Step 10228: Obtain the current location information of the target maintenance personnel, and determine real-time traffic information based on the current location information and the maintenance location information;

[0112] Step 10229: Generate the maintenance trajectory corresponding to the fault work order based on the real-time traffic information, the maintenance location information, and the current location information.

[0113] In this embodiment, steps 10221 to 10229 utilize a constructed model to evaluate the title and text similarity between equipment maintenance work orders and fault work orders. This improves both the accuracy and efficiency of work order matching, thereby enhancing the efficiency and accuracy of power outage information delivery. Specifically, after matching a fault work order with a corresponding equipment maintenance work order, road conditions are determined based on the location information of the equipment to be repaired and the target power supply in the equipment maintenance work order. This avoids delays in maintenance due to real-time road conditions. The maintenance trajectory corresponding to the equipment maintenance work order is then calculated using the location information of the equipment to be repaired, the target power supply, and the road conditions, and the maintenance trajectory is sent to maintenance personnel, improving equipment maintenance efficiency. Conversely, if no matching equipment maintenance work order is found for the fault work order, it indicates that the fault in the target power supply equipment is temporary. In this case, to ensure timely maintenance of the target power supply equipment, the fault work order is first sent to available maintenance personnel, enabling them to accurately repair the target power supply equipment. The maintenance trajectory is determined by using the real-time location information of the maintenance personnel and the location information of the target power supply equipment, so that users can keep track of the power restoration progress in a timely manner, thereby improving the user's power experience.

[0114] In some embodiments of this example, when generating the maintenance trajectory corresponding to the equipment maintenance work order in step 10225, the accuracy and efficiency of the maintenance trajectory can be improved through the following steps. These steps are:

[0115] Step 102251: Based on the device location information and the current location information, determine the initial shortest trajectory between two adjacent devices to be repaired using a preset shortest path search algorithm;

[0116] Step 102252: Based on the shortest trajectory, determine the fault area where two adjacent devices to be repaired are located, and obtain the real-time traffic information of the fault area;

[0117] Step 102253: Update the initial shortest trajectory based on the real-time traffic information to obtain the real-time shortest trajectory between two adjacent devices to be repaired;

[0118] Step 102254: Generate the maintenance trajectory corresponding to the equipment maintenance work order based on the equipment location information, the current location information, and the real-time shortest trajectory between two adjacent devices to be maintained.

[0119] In this embodiment, steps 102251 to 102254 utilize the location information of each pair of devices to determine the shortest path between them, thereby improving maintenance efficiency by reducing the length of the maintenance path. Specifically, when determining the shortest path, traffic information corresponding to the current shortest path is considered. The selection of the shortest path is updated based on this traffic information. For example, if the traffic information for the shortest path indicates congestion, while the traffic information for a slightly longer second shortest path indicates smooth traffic, the second shortest path is selected as the shortest path. This is how the maintenance path is planned, thereby improving maintenance efficiency.

[0120] In some implementations of this embodiment, when generating the maintenance trajectory corresponding to the equipment maintenance work order in step 10229, the accuracy and efficiency of the maintenance trajectory can be improved through the following steps. These steps are:

[0121] Step 102291: Obtain the real-time location information of the target maintenance personnel, so as to obtain the movement speed of the target maintenance personnel based on the real-time location information;

[0122] Step 102292: Predict the fault repair time of the target power supply equipment based on the moving speed and the repair trajectory.

[0123] In this embodiment, steps 102291 to 102292 are based on the current maintenance personnel accurately repairing the target power supply equipment. At this time, the repair time of the equipment can be accurately predicted by calculating the user's movement speed, thereby improving the efficiency and accuracy of power outage information formulation.

[0124] In some embodiments of this example, step 103 is preferred for improving the accuracy of fault repair time prediction, and the following steps can be used to predict the repair time. These steps are:

[0125] Step 1031: Obtain the real-time location information of the current maintenance personnel, and generate the real-time operation and maintenance trajectory of the current maintenance personnel based on the real-time location information; wherein, the real-time operation and maintenance trajectory includes the real-time location time and the real-time location information;

[0126] Step 1032: When it is determined from the real-time operation and maintenance trajectory that the current maintenance personnel are performing maintenance along the maintenance trajectory, the historical operation and maintenance information of the current maintenance personnel and the equipment information of each device to be maintained are obtained based on the user identifier;

[0127] Step 1033: Input the historical operation and maintenance information, the equipment information, the equipment location information, the maintenance location information of the target power supply equipment, and the real-time operation and maintenance trajectory into the pre-built deep learning model, so as to predict the fault repair time of the target power supply equipment through the deep learning model.

[0128] In this embodiment, the above steps first monitor the real-time location information of maintenance personnel to enable them to perform maintenance according to the maintenance path, ensuring the accuracy of power outage information. Secondly, when predicting maintenance time, a deep learning model is used to improve the accuracy and efficiency of the prediction. Historical maintenance information of maintenance personnel, equipment information and location information of the target power supply equipment, maintenance location information of the target power supply equipment, and the real-time maintenance trajectory are used as the basis for prediction, thereby improving the accuracy of the prediction and ultimately the accuracy of the power outage information.

[0129] In some embodiments of this example, step 104 predicts the maintenance time through the following steps to improve the accuracy of the prediction. These steps are:

[0130] Step 1041: Retrieve the historical maintenance information corresponding to the current maintenance personnel based on the user identifier of the current maintenance personnel; wherein, the historical maintenance information includes historically repaired equipment, structural information of the historically repaired equipment, and the repair time of the historically repaired equipment;

[0131] Step 1042: Retrieve the structural information of the target power supply device based on its device identifier;

[0132] Step 1043: Input the structural information and the historical operation and maintenance information into the pre-built maintenance time prediction model, so as to output the maintenance time of the target power supply equipment through the maintenance time prediction model.

[0133] In this embodiment, steps 1041 to 1043, in order to provide users with more detailed power outage information, utilize a model to predict the repair time for maintenance personnel to repair the target power supply equipment, thereby improving the user's perception of the power restoration progress. Specifically, when predicting the repair time, basing the prediction on the maintenance personnel's historical maintenance information and the structural information of the target power supply equipment can improve the accuracy of the prediction, and thus improve the accuracy of the power outage information.

[0134] In some embodiments of this example, step 105 pushes outage information through the following steps to improve the efficiency and accuracy of the push. The steps are as follows:

[0135] Step 1051: Extract the key information corresponding to the maintenance trajectory, maintenance duration, and fault maintenance time, and output structured power outage information based on the key information and pre-saved information template;

[0136] Step 1052: Based on the area identifier and the pre-saved user information database, match several target users within the power outage area and the message queue address of each target user;

[0137] Step 1053: Push the power outage information to each of the target users according to the message queue address.

[0138] In this embodiment, after generating power outage information including the maintenance trajectory, maintenance duration, and fault repair time, in order to ensure that the power outage information can be accurately pushed to all users affected by the power outage of the target power supply equipment, the area identifier and the pre-saved user information database can be used to match the message queue address corresponding to each user, and then the power outage information can be accurately pushed according to the message queue address and the user information database.

[0139] In some implementations of this embodiment, when pushing power outage information according to the message queue address in step 1053, the information can be pushed through the following steps to improve the efficiency of the push. The steps are as follows:

[0140] Step 10531: Encapsulate the power outage information using a message middleware;

[0141] Step 10532: Set the message topic and routing rules for the encapsulated power outage information;

[0142] Step 10533: Push the encapsulated power outage information to each target user according to the message topic and routing rules.

[0143] In this embodiment, the above steps encapsulate the power outage information through a message middleware to achieve rapid distribution of the power outage information. At the same time, message topics and routing rules are set for the encapsulated power outage information to ensure accurate delivery of the power outage information.

[0144] On the other hand, this embodiment also provides a power outage information aggregation and push system. For the specific structural composition of the push system, please refer to... Figure 2 It mainly includes a power outage analysis module 201, a maintenance monitoring module 202, a maintenance prediction module 203, a duration prediction module 204, and a message push module 205.

[0145] The power outage analysis module 201 is used to obtain the operating data of the target power supply equipment based on the received power outage consultation request, so as to generate a fault work order for the target power supply equipment and obtain a number of target users matching the target power supply equipment according to the operating data.

[0146] The maintenance monitoring module 202 is used to match the equipment maintenance work order corresponding to the fault work order, and to obtain the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order.

[0147] The maintenance prediction module 203 is used to obtain the real-time location information of the current maintenance personnel, so as to predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory.

[0148] The duration prediction module 204 is used to obtain the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment in order to predict the maintenance duration of the target power supply equipment.

[0149] The message push module 205 is used to generate power outage information corresponding to the power outage consultation request based on the fault repair time, the repair duration and the repair trajectory, and push the power outage information to the target user.

[0150] This embodiment provides a method and system for aggregating and pushing power outage information. Upon receiving a power outage inquiry request from a user, the system first locates the corresponding target power supply equipment to generate a corresponding fault work order. This precise location of the fault source reduces data processing volume and improves message push efficiency. Secondly, it matches the fault work order with the corresponding equipment maintenance work order to obtain specific equipment maintenance information, such as maintenance trajectory and personnel. Based on this information, it predicts the power restoration progress, such as maintenance time, so that users can understand the specific power restoration progress based on the power outage information, improving user satisfaction. Finally, when pushing the power outage information, it is sent to several target users matched with the target power supply equipment. This avoids receiving duplicate power outage inquiry requests, improving information push efficiency, and ensures accurate delivery of the power outage information to users affected by the power outage, guaranteeing message push accuracy.

[0151] Example 2

[0152] To improve the accuracy and efficiency of power outage information delivery, thereby enhancing user satisfaction with electricity usage, this embodiment provides a method for aggregating and pushing power outage information. The specific implementation process of this method can be found in [reference needed]. Figure 3 It mainly includes steps 301 to 307, the steps being:

[0153] Step 301: Collect the operating data of the power supply equipment corresponding to the power outage consultation request to determine whether the power supply equipment has a power outage fault. If a power outage fault is detected, obtain the topology information, location information and other equipment information of the power supply equipment to generate a fault work order for the power supply equipment.

[0154] In this embodiment, the step specifically involves: locating the power supply equipment that supplies power to the user based on the user's inquiry request, and collecting the equipment operation data of the power supply equipment online in real time; transmitting the collected data to a preset fault work order system, so that the fault work order system can analyze and process the transmitted equipment operation data according to preset power supply equipment operation parameter thresholds, and determine whether the power supply equipment has a power outage fault.

[0155] If the fault work order system detects a power outage fault in the power supply equipment, it automatically triggers the generation of a fault work order for that equipment. The system automatically acquires data from the power supply equipment to generate the fault work order. This data includes historical operating data, environmental data, topology connection information, and specific location information. The data from the faulty power supply equipment is added to the generated fault work order as preliminary fault information.

[0156] Specifically, upon receiving a user's inquiry request, the system obtains the user's address information. Based on this address information, it retrieves the corresponding power supply equipment from a pre-defined power grid topology. The operating data of this power supply equipment is collected every 5 minutes via an online real-time acquisition system. The collected data includes key parameters such as voltage, current, and power. The collected data is transmitted to the fault work order system via the MQTT protocol. Upon receiving the data, the system analyzes and processes it using pre-defined power supply equipment operating parameter thresholds. For example, if the voltage is below 220V, the current is greater than 100A, or the power factor is less than 8, the system determines that the equipment has a power outage fault. When a power outage fault is detected in a power supply device, the system automatically generates a fault work order for that device. Simultaneously, it queries the topology connection information of the device through the Neo4j graph database and calls the Gaode Map API to obtain the device's specific location coordinates and historical operating data. This historical operating data includes parameters such as voltage, current, and power, as well as environmental data such as temperature, humidity, and wind speed, and estimated outage range data. The topology and location information are then added to the fault work order.

[0157] Step 302: Calculate the content similarity between the generated fault work order and the pre-saved equipment maintenance work order, so as to determine whether the power supply equipment is in the maintenance plan based on the content similarity.

[0158] In this embodiment, the step specifically involves: retrieving several equipment maintenance work orders with execution statuses of pending execution or in progress from a pre-built database; inputting the fault work order and each of the equipment maintenance work orders into a pre-built vector space model; calculating the content similarity between the fault work order and each of the equipment maintenance work orders based on preset sub-position weights in the vector space model; and determining that the equipment maintenance work order is the maintenance work order corresponding to the fault work order when the content similarity is greater than a preset similarity threshold. The similarity is determined based on the word vectors of the title words and the word vectors of the text words in the work order to be dispatched, the word vectors of the title words and the word vectors of the text words in the historical work orders, and the word position weights; the word vectors of the text words are determined based on the word frequency of the text words; the word vectors of the title words are determined based on the word frequency of the title words; and the word position weights are determined based on the word's position in the title or text, with different word position weights corresponding to different positions.

[0159] Specifically, a dataset of equipment maintenance work orders is collected. This dataset includes multiple equipment maintenance work orders, each comprising a title and a body text. Each work order is segmented into words to obtain multiple title terms and body text terms. Frequency statistics are performed on each title term and each body text term to obtain their respective frequencies. For each title term, a title word vector is mapped to its frequency. For each body text term, a body text word vector is mapped to its frequency. A vector space model is constructed based on the vectors corresponding to the title terms and the word vectors corresponding to the body text terms in each equipment maintenance work order.

[0160] Furthermore, based on the word vectors of each title word contained in each equipment maintenance work order stored in the vector space model, and the preset word position weight of each title word, the title similarity between the fault work order and each equipment maintenance work order is calculated; based on the word vectors of each body text word contained in each equipment maintenance work order stored in the vector space model, and the preset word position weight of each body text word, the body text similarity between the fault work order and each equipment maintenance work order is calculated; based on the title similarity between the fault work order and each equipment maintenance work order, and the body text similarity between the fault work order and each equipment maintenance work order, the similarity between the fault work order and each equipment maintenance work order is calculated.

[0161] Step 303: When it is determined that the power supply equipment is under maintenance plan, obtain the maintenance trajectory corresponding to the equipment maintenance work order and track the current maintenance personnel in real time, so as to estimate the fault repair time based on the maintenance trajectory and the tracking information.

[0162] In this embodiment, the step specifically involves: obtaining the location information of each device to be repaired in the equipment maintenance work order, and locating the user terminal responding to the equipment maintenance work order to obtain the current geographical location of the maintenance personnel; determining the corresponding real-time traffic information based on the current geographical location and the location information, and generating a corresponding maintenance route based on the real-time traffic information; and inputting the maintenance route, the location information of the device to be repaired, the location information of the faulty device, and the personnel information of the maintenance personnel into a pre-built deep learning algorithm to output a pre-estimated fault repair time.

[0163] Specifically, starting with the currently faulty device, from the locations of the multiple devices to be repaired, excluding the current fault point, the fault location with the shortest distance to the current fault point is searched as the next fault point, generating a second path from the current fault point to the next fault point; this process is repeated, executing the second generation step repeatedly until all fault locations within the fault range are traversed, thereby determining the repair route; furthermore, real-time traffic information of the fault area where the current fault point and the next fault point are located is obtained; the second path is generated based on the traffic information, wherein the fault range includes the fault area; wherein the fault information also includes the fault type; this process repeats, executing the second generation step repeatedly until all fault locations within the fault range are traversed, including: sorting the fault locations according to the fault type to generate a fault sequence; and taking a preset number of fault locations as a group, sequentially performing the operation of determining the repair route on each group of fault locations in the fault sequence.

[0164] Furthermore, the trajectory signal of the device receiving the maintenance work order is acquired in real time to locate the real-time geographical location of the maintenance personnel based on the trajectory signal. The geographical location, the maintenance route, and the personnel information of the maintenance personnel are input into a pre-built deep learning algorithm to output a pre-estimated fault repair time.

[0165] Step 304: When it is determined that the equipment is not under maintenance, for the initially generated fault work order, match the corresponding target maintenance personnel from the maintenance personnel in the idle state according to the location information of the power supply equipment, track the target maintenance personnel, generate the movement trajectory of the target maintenance personnel, and estimate the fault repair time based on the movement trajectory and the fault location.

[0166] In this embodiment, the step specifically involves: acquiring the trajectory signal of the device receiving the maintenance work order in real time, locating the real-time geographical location of the maintenance personnel based on the trajectory signal, and generating the movement trajectory of the target maintenance personnel based on the real-time geographical location. The movement speed of the target maintenance personnel is estimated based on the movement trajectory, and then the fault repair time for the target maintenance personnel to reach the fault location is estimated based on the movement speed and the fault location.

[0167] Step 305: Obtain the personnel information of the target maintenance personnel or the current maintenance personnel and call the topology information of the power supply equipment, and use a deep learning algorithm to predict the maintenance time of the power supply equipment.

[0168] In this embodiment, the specific steps are as follows: A maintenance time calculation model is established. The maintenance time calculation model includes optimal statistical estimation and expert experience estimation. The expert experience estimation includes the following steps: establishing a basic on-site maintenance work time model; establishing an overall operation step time model based on various collected anomalies; determining the operation steps and sequence according to the anomaly type and on-site conditions; calculating the on-site maintenance time based on the scheduled processing steps. The optimal statistical estimation includes the following steps: a) using maximum likelihood estimation in the estimation to calculate the maximum likelihood estimate and confidence interval for each type of anomaly in each region; b) checking whether the data involved in the calculation exceeds the confidence interval. If not, output the maximum likelihood value; if so, remove the historical data exceeding the limit and execute step a; c) obtaining the on-site maintenance time based on the historical on-site processing time.

[0169] Step 306: Generate a structured power outage information report based on the maintenance trajectory, maintenance duration, and fault maintenance time, and obtain the power outage range corresponding to the power supply equipment through a preset deep learning algorithm, so as to push the power outage information report through a message middleware.

[0170] In this embodiment, the main steps are as follows: Based on the initially generated fault work order, a deep learning model is constructed using the acquired power supply equipment topology information and fault location coordinates. This model is then used to quickly estimate the power outage range. Based on the maintenance trajectory, maintenance duration, and estimated maintenance time, natural language processing technology is used to extract key information and generate a structured power outage information report. The generated structured power outage information report is encapsulated through a message middleware, with message topics and routing rules set to achieve rapid report distribution. Based on the power outage impact range, relevant affected departments are matched from a pre-configured department information database, and their message queue addresses are obtained. The power outage information report is pushed to the matched relevant department message queues, triggering internal collaborative processing within the departments. An equipment status monitoring algorithm is used to continuously acquire the operating status data of the power supply equipment. By comparing this data with the power outage status, it is determined whether power outage restoration or a change in the power outage range has occurred. If power outage restoration or a change in the power outage range is detected, the power outage information report is updated promptly, an information change notification is generated, and the notification content is encapsulated as a message. By using a message push platform, information change notifications are sent in real time to the message queues of the relevant matched departments to ensure that each department receives the latest power outage information.

[0171] Specifically, after identifying the cause and scope of the power outage, natural language processing (NLP) is used to analyze the information, extracting keywords such as "transformer failure" and "high-voltage line breakage," and generating a structured power outage information report based on a pre-defined template. The report includes outage time, estimated recovery time, affected area, and number of affected users, all formatted into a standardized data structure using a predefined JSON format. The generated report is then encapsulated using a Kafka messaging service with a "power outage event" message topic and routing rules set according to the affected area for rapid report distribution. Simultaneously, a matching process is performed from the departmental database to identify affected power supply stations, business halls, and other departments, obtaining their corresponding ActiveMQ queue addresses. The encapsulated power outage information report is then pushed to the relevant departmental queues via an API interface, triggering internal power outage handling procedures, including emergency repairs and customer notifications. During processing, the system employs an equipment status monitoring algorithm, collecting current and voltage data from the power supply equipment every 5 minutes. This data is compared with the outage status to determine if power restoration has occurred or if the outage area has changed. If the power load of users in a certain area is monitored to return from 0 to normal levels, it can be determined that the power outage in that area has been restored. Once a change occurs, the system will automatically update the power outage information report, generate an information change notification, and then encapsulate and push the message again through Kafka to ensure that all departments receive the latest power outage information in real time and adjust their work arrangements accordingly.

[0172] Step 307: Collect and analyze data throughout the entire process of power outage information identification and push. By establishing a power outage event knowledge base, summarize the characteristic patterns of different types of power outages, optimize the intelligence level of the fault work order system, and dynamically control the speed of information generation and push while improving identification accuracy, so as to achieve a balance between comprehensiveness and timeliness.

[0173] In this embodiment, the main steps are as follows: Based on relevant data of the power outage event, including the outage area, outage time, and cause, the relevant data is stored in a preset database. The power outage event data is classified according to preset power outage event classification rules to obtain different types of power outage event data. For each type of power outage event data, a clustering algorithm is used to extract features, obtaining feature patterns for each type of power outage event. The extracted power outage event feature patterns are stored in a preset knowledge base, forming a complete power outage event knowledge base. When new power outage event data is received, it is matched with the feature patterns in the knowledge base to determine which type of power outage event it belongs to. Based on the matching results, the corresponding type of power outage event processing solution is retrieved from the knowledge base, and a fault work order is automatically generated, improving the intelligence level of the fault work order system. By dynamically evaluating the accuracy of power outage event identification and the speed of information push, the data collection, analysis, and information push strategies are adjusted to achieve a balance between comprehensiveness and timeliness.

[0174] Specifically, by connecting with the power grid dispatch system, information such as outage area, outage time, and outage cause can be obtained in real time. For example, if a power outage occurs in a certain area of ​​a city, the system interface can obtain information on substations and lines within that area, showing an outage time of 10:00-12:00 on March 15, 2023, due to planned maintenance. The obtained data is cleaned and transformed using ETL tools and stored in the outage event table of a MySQL database. Based on preset outage event classification rules, a decision tree algorithm is used to classify the outage event data, resulting in different types of outage events such as planned maintenance, fault outages, and natural disasters. For each type of outage event data, K-means clustering is used for feature extraction. By calculating the Euclidean distance between different data points, data with high similarity are grouped into a cluster, resulting in feature patterns for fault outages such as: repair time exceeding 2 hours, affecting more than 1000 users, etc. The extracted outage event feature patterns are stored in a MongoDB knowledge base, forming a complete outage event knowledge base. When new power outage event data is received, it is matched against feature patterns in the knowledge base. By calculating the similarity between the new event data and each feature pattern, the type of the feature pattern with the highest similarity is determined as the event type. Based on the matching results, the corresponding power outage event handling plan is retrieved from the MongoDB knowledge base. For example, the handling plan for a fault power outage includes activating the emergency plan, dispatching repair personnel, etc., automatically generating a fault work order, and pushing it to the power GIS system and work order management system. By comparing the power outage event identification results with the manual classification results, the identification accuracy is calculated. At the same time, the time from receiving the power outage event data to pushing the fault work order is statistically analyzed to evaluate the information push speed. When the identification accuracy is below 90% or the information push time exceeds 10 minutes, the data collection, analysis, and information push strategies are adjusted and optimized, such as increasing the data collection frequency, optimizing algorithm parameters, and compressing data transmission, to achieve a balance between comprehensiveness and timeliness.

[0175] This embodiment provides a method for aggregating and pushing power outage information. By collecting real-time operating data of power supply equipment and combining it with preset thresholds, the authenticity of power outage events reported by users is automatically judged, and corresponding fault work orders are generated based on the judgment results. Next, to avoid duplicate reporting of the same power outage event for the same faulty equipment, this invention compares the fault work order with maintenance work orders sent by maintenance personnel to determine whether the faulty equipment is already in the maintenance plan, thereby avoiding duplicate work order reporting and thus avoiding impacting maintenance efficiency. After identifying the category of the fault work order, the maintenance personnel are tracked in real-time, and the maintenance trajectory in the fault work order is obtained to visualize the maintenance trajectory, allowing users to understand the maintenance progress in real time and reducing users' power anxiety. Furthermore, a structured report is generated, and the power outage report is quickly distributed by predicting the power outage range, so that users within the current power outage range can understand the corresponding power restoration progress in real time, avoiding the same inquiry request from different users and effectively improving the user's power experience. Full-process data analysis is used to establish a knowledge base, summarize feature patterns, and continuously optimize the system's intelligence level. This invention enables intelligent identification, dynamic updating, and accurate delivery of power outage information, effectively improving the efficiency of power system fault handling.

[0176] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for aggregating and pushing power outage information, characterized in that, include: Based on the received power outage inquiry request, the system obtains the operating data of the target power supply equipment to generate a fault work order for the target power supply equipment and obtains several target users matching the target power supply equipment. When it is determined that the target power supply equipment has a power outage fault, the obtained topology connection information and the maintenance location information corresponding to the target power supply equipment are input into a pre-built deep learning model to predict the power outage range corresponding to the target power supply equipment. Multiple sub-regions pre-divided according to the power grid lines are obtained to determine several target sub-regions of power outage based on the power outage range. A region identifier for each target sub-region is obtained to determine several target users matching the target sub-region based on the region identifier. Match the fault work order with the corresponding equipment repair work order, and obtain the repair trajectory and current repair personnel corresponding to the equipment repair work order; wherein, obtain the user terminal identifier and user identifier responding to the equipment repair work order, so as to determine the current repair personnel based on the user identifier and locate the current repair personnel based on the user terminal identifier, and obtain the current location information of the current repair personnel; obtain the equipment location information of each equipment to be repaired in the equipment repair work order, so as to determine the real-time traffic information based on the equipment location information and the current location information; generate the repair trajectory corresponding to the equipment repair work order based on the real-time traffic information, the equipment location information, and the current location information; wherein, based on the equipment location information and the current location information, determine the initial shortest trajectory between two adjacent equipment to be repaired through a preset shortest path search algorithm; determine the fault area where the two adjacent equipment to be repaired are located based on the shortest trajectory, and obtain the real-time traffic information of the fault area; and determine the fault area based on the real-time traffic information. The initial shortest trajectory is updated to obtain the real-time shortest trajectory between two adjacent devices to be repaired. Based on the device location information, the current location information, and the real-time shortest trajectory between the two adjacent devices, a repair trajectory corresponding to the device repair work order is generated. Several device repair work orders with execution status of pending execution or currently executing are retrieved from a pre-built database. The content similarity between each device repair work order and the fault work order is obtained to match the device repair work order corresponding to the fault work order based on the content similarity. For any device repair work order, the device repair work order and the fault work order are input into a pre-built content similarity evaluation model to output the title similarity and body text similarity between the device repair work order and the fault work order. When the title similarity is greater than or equal to a preset similarity threshold and the body text similarity is greater than or equal to the similarity threshold, the device repair work order is determined to be the device repair work order corresponding to the fault work order. The system acquires the real-time location information of the current maintenance personnel to predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory. Specifically, it acquires the real-time location information of the current maintenance personnel to generate a real-time operation and maintenance trajectory for the current maintenance personnel. The real-time operation and maintenance trajectory includes the real-time location time and the real-time location information. When it is determined that the current maintenance personnel are performing maintenance along the maintenance trajectory based on the real-time operation and maintenance trajectory, the system acquires the historical operation and maintenance information of the current maintenance personnel and the equipment information of each device to be repaired based on the user identifier. The system inputs the historical operation and maintenance information, the equipment information, the equipment location information, the repair location information of the target power supply equipment, and the real-time operation and maintenance trajectory into a pre-built deep learning model to predict the fault repair time of the target power supply equipment using the deep learning model. The system acquires the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment to predict the maintenance time of the target power supply equipment. Specifically, it retrieves the historical operation and maintenance information corresponding to the current maintenance personnel based on their user identifier. This historical operation and maintenance information includes historically repaired equipment, its structural information, and the maintenance time. It also retrieves the structural information of the target power supply equipment based on its equipment identifier. Finally, it inputs the structural information and the historical operation and maintenance information into a pre-built maintenance time prediction model to output the maintenance time of the target power supply equipment. Based on the fault repair time, the repair duration, and the repair trajectory, power outage information corresponding to the power outage consultation request is generated, and the power outage information is pushed to the target user.

2. The method for aggregating and pushing power outage information according to claim 1, characterized in that, The process of obtaining operational data of the target power supply equipment based on the received power outage consultation request, and generating a fault work order for the target power supply equipment and obtaining a number of target users matching the target power supply equipment based on the operational data, includes: Based on the power outage inquiry request, obtain the power outage location information fed back by the user, and determine the target power supply equipment to supply power to the power outage location information based on the power outage location information and the pre-saved power grid lines; Collect real-time operating data of the target power supply equipment; When it is determined that the target power supply equipment has a power outage fault based on the real-time operating data and preset operating parameter thresholds, the equipment data of the target power supply equipment is obtained, and a fault work order for the target power supply equipment is generated based on the equipment data; wherein, the equipment data includes the maintenance location information, environmental data, topology connection information and historical operating data of the target power supply equipment.

3. The method for aggregating and pushing power outage information according to claim 1, characterized in that, The step of generating the maintenance trajectory corresponding to the equipment maintenance work order based on the real-time traffic information, the equipment location information, and the current location information includes: Based on the device location information and the current location information, the initial shortest trajectory between two adjacent devices to be repaired is determined by a preset shortest path search algorithm. Based on the shortest trajectory, the fault area where two adjacent devices to be repaired are located is determined, and real-time traffic information of the fault area is obtained; The initial shortest trajectory is updated based on the real-time traffic information to obtain the real-time shortest trajectory between two adjacent devices to be repaired. Based on the device location information, the current location information, and the real-time shortest trajectory between two adjacent devices to be repaired, a repair trajectory corresponding to the device repair work order is generated.

4. The method for aggregating and pushing power outage information according to claim 1, characterized in that, The process of matching the fault work order with the corresponding equipment maintenance work order and obtaining the maintenance trajectory and current maintenance personnel for the corresponding equipment maintenance work order includes: When no equipment maintenance work order corresponding to the fault work order can be matched based on the title similarity and the body text similarity, the fault work order is sent to several maintenance personnel who are in an idle state according to the maintenance location information of the target power supply equipment. From among the various maintenance personnel, determine the target maintenance personnel who will respond to the fault work order, and designate the target maintenance personnel as the current maintenance personnel; Obtain the current location information of the target maintenance personnel, and determine real-time traffic information based on the current location information and the maintenance location information; The maintenance trajectory corresponding to the fault work order is generated based on the real-time traffic information, the maintenance location information, and the current location information.

5. The method for aggregating and pushing power outage information according to claim 4, characterized in that, The step of obtaining the real-time location information of the current maintenance personnel, and predicting the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory, includes: Obtain the real-time location information of the target maintenance personnel, and determine the movement speed of the target maintenance personnel based on the real-time location information; The failure repair time of the target power supply equipment is predicted based on the moving speed and the repair trajectory.

6. The method for aggregating and pushing power outage information according to claim 1, characterized in that, The process of generating power outage information corresponding to the power outage inquiry request based on the fault repair time, the repair duration, and the repair trajectory, and pushing the power outage information to the target user, includes: Extract key information corresponding to the maintenance trajectory, maintenance duration, and fault maintenance time, and output structured power outage information based on the key information and pre-saved information templates; Based on the area identifier and the pre-saved user information database, several target users within the power outage area and the message queue address of each target user are matched. The power outage information is pushed to each of the target users according to the message queue address.

7. The method for aggregating and pushing power outage information according to claim 6, characterized in that, The step of pushing the power outage information to each target user according to the message queue address includes: The power outage information is encapsulated using a message middleware; Set message topics and routing rules for the encapsulated power outage information; The encapsulated power outage information is pushed to each target user according to the message topic and routing rules.

8. A power outage information aggregation and push system, characterized in that, It includes a power outage analysis module, a maintenance monitoring module, a maintenance prediction module, a duration prediction module, and a message push module; The power outage analysis module is used to obtain the operating data of the target power supply equipment based on the received power outage consultation request, and to generate a fault work order for the target power supply equipment and obtain a number of target users matching the target power supply equipment based on the operating data; wherein, when it is determined that the target power supply equipment has a power outage fault, the obtained topology connection information and the maintenance location information corresponding to the target power supply equipment are input into a pre-built deep learning model, so as to predict the power outage range corresponding to the target power supply equipment through the deep learning model; obtain multiple sub-regions pre-divided according to the power grid lines, so as to determine a number of target sub-regions with power outage from the multiple sub-regions according to the power outage range; obtain the area identifier of each target sub-region, so as to determine a number of target users matching the target sub-region based on the area identifier; The maintenance monitoring module is used to match the equipment maintenance work order corresponding to the fault work order, and obtain the maintenance trajectory and current maintenance personnel corresponding to the equipment maintenance work order; wherein, it obtains the user terminal identifier and user identifier responding to the equipment maintenance work order, so as to determine the current maintenance personnel based on the user identifier and locate the current maintenance personnel based on the user terminal identifier, thereby obtaining the current location information of the current maintenance personnel; it obtains the equipment location information of each piece of equipment to be repaired in the equipment maintenance work order, so as to determine real-time traffic information based on the equipment location information and the current location information; it generates the maintenance trajectory corresponding to the equipment maintenance work order based on the real-time traffic information, the equipment location information, and the current location information; wherein, based on the equipment location information and the current location information, it determines the initial shortest trajectory between two adjacent pieces of equipment to be repaired through a preset shortest path search algorithm; it determines the fault area where the two adjacent pieces of equipment to be repaired are located based on the shortest trajectory, and obtains the real-time traffic information of the fault area; and it obtains the real-time traffic information of the fault area based on the real-time traffic information. The initial shortest trajectory is updated based on real-time traffic information to obtain the real-time shortest trajectory between two adjacent devices to be repaired. Based on the device location information, the current location information, and the real-time shortest trajectory between the two adjacent devices, a repair trajectory corresponding to the device repair work order is generated. Several device repair work orders with execution status of pending or ongoing are retrieved from a pre-built database. The content similarity between each device repair work order and the fault work order is obtained to match the device repair work order corresponding to the fault work order based on the content similarity. For any device repair work order, the device repair work order and the fault work order are input into a pre-built content similarity evaluation model to output the title similarity and body text similarity between the device repair work order and the fault work order. When the title similarity is greater than or equal to a preset similarity threshold and the body text similarity is greater than or equal to the similarity threshold, the device repair work order is determined to be the device repair work order corresponding to the fault work order. The maintenance prediction module is used to acquire the real-time location information of the current maintenance personnel, and to predict the fault repair time of the target power supply equipment based on the real-time location information and the maintenance trajectory. Specifically, the module acquires the real-time location information of the current maintenance personnel to generate a real-time operation and maintenance trajectory for the current maintenance personnel. The real-time operation and maintenance trajectory includes the real-time location time and the real-time location information. When it is determined that the current maintenance personnel are performing maintenance along the maintenance trajectory based on the real-time operation and maintenance trajectory, the module acquires the historical operation and maintenance information of the current maintenance personnel and the equipment information of each device to be repaired based on the user identifier. The historical operation and maintenance information, the equipment information, the equipment location information, the repair location information of the target power supply equipment, and the real-time operation and maintenance trajectory are input into a pre-built deep learning model to predict the fault repair time of the target power supply equipment through the deep learning model. The duration prediction module is used to obtain the historical maintenance information of the current maintenance personnel and the structural information of the target power supply equipment to predict the maintenance duration of the target power supply equipment. Specifically, it retrieves the historical operation and maintenance information corresponding to the current maintenance personnel based on their user identifier; the historical operation and maintenance information includes historically repaired equipment, the structural information of the historically repaired equipment, and the maintenance duration of the historically repaired equipment; it retrieves the structural information of the target power supply equipment based on its equipment identifier; and it inputs the structural information and the historical operation and maintenance information into a pre-built maintenance duration prediction model to output the maintenance duration of the target power supply equipment through the maintenance duration prediction model. The message push module is used to generate power outage information corresponding to the power outage consultation request based on the fault repair time, the repair duration and the repair trajectory, and push the power outage information to the target user.

Images