A power-off and power-on visual management method, system, device and storage medium

Abstract

This invention discloses a power outage and restoration visualization management method, system, device, and storage medium, comprising: receiving power outage and restoration related data from multiple heterogeneous business systems; preprocessing the data and uniformly converting the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset; constructing a power grid topology diagram, using user-side alarm events as observation variables, and using a Bayesian probability model to infer the probability of fault occurrence from the end user nodes along the topology diagram, calculating the fault confidence of each medium-voltage branch line segment, and generating a list of suspected fault sections; overlaying the list of suspected fault sections onto a geographic information system map for visualization, and generating emergency repair work orders by combining the distribution of emergency repair resources and historical handling records; updating the equipment status in the topology diagram based on the on-site handling results, and verifying the accuracy of the fault inference results in a closed loop, enabling real-time monitoring of power outages and precise scheduling of power restoration and emergency repair work.

Description

Technical Field

[0001] This invention relates to the technical field of power outage and restoration management, and in particular to a visual management method, system, device, and storage medium for power outage and restoration. Background Technology

[0002] The current application of power outage and restoration management technology in China's power industry suffers from several shortcomings. Firstly, it fails to integrate the outage / restoration coverage, critical equipment operating status, and user spatial distribution into a geographically visualized display. Managers cannot quickly ascertain the extent of the outage's impact, the number of affected users, or the location of core loads, lacking intuitive data support for decision-making. Secondly, the level of intelligence in fault diagnosis is low. Alarm information suffers from transmission problems, frequent duplication and redundancy, false alarms, and missed alarms. Furthermore, there is a lack of intelligent filtering and prioritization mechanisms. Finding fault points relies primarily on the on-site experience of maintenance personnel, lacking a systematic intelligent diagnosis model, resulting in slow fault investigation and directly hindering power restoration. Moreover, it cannot accurately differentiate between important and ordinary users and push outage / restoration information in real time, nor can it provide differentiated services based on user level. Additionally, users struggle to quickly and conveniently obtain crucial information such as the cause of the outage, the progress of fault handling, and the estimated restoration time, lowering customer satisfaction.

[0003] Currently, power outage and restoration management technology lacks a system that can automatically and in real-time generate power outage and restoration data reports. It also lacks the ability to perform comprehensive statistical analysis from dimensions such as region, time period, fault type, and user type. This results in a lack of effective data support for post-outage and restoration management optimization, dynamic resource allocation, and long-term planning, hindering the formation of a scientific management loop. These problems directly reduce the efficiency of power outage and restoration management and also affect the improvement of service quality. Summary of the Invention

[0004] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a method, system, device, and storage medium for visualized power outage and restoration management, addressing the issues of insufficient fault handling and monitoring capabilities, inaccurate customer service, and slow response times in existing systems.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a visual management method for power outage and restoration, comprising: Receive power outage and restoration related data from multiple heterogeneous business systems, preprocess the power outage and restoration related data, and uniformly convert the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. Construct a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with the corresponding nodes in the topology graph; Using user-side alarm events as observed variables, and based on a Bayesian probability model, the probability of fault occurrence is deduced from the end user node along the topology graph, the fault confidence of each medium-voltage branch line segment is calculated, and a list of suspected fault sections is generated. The list of suspected faulty sections is overlaid onto a geographic information system map for visualization, and a repair task work order is generated by combining the distribution of emergency repair resources and historical handling records. Based on the on-site handling results, the equipment status in the topology diagram is updated, and the accuracy of the fault simulation results is verified in a closed loop, thus completing the full-process management of power outage and restoration events.

[0006] As a preferred embodiment of the power outage and restoration visualization management method described in this invention, wherein: The process of receiving power outage and restoration related data from multiple heterogeneous service systems and preprocessing the power outage and restoration related data includes: The power outage and restoration related data includes massive time-series data and structured business data; data cleaning, format standardization, time alignment and entity association processing are performed on the power outage and restoration related data to identify the spatial coordinate information contained therein, and coordinate data from different sources are uniformly converted to a preset geographic coordinate system; The structured data and spatial data are integrated and processed to form a fused multi-source power outage and restoration dataset.

[0007] As a preferred embodiment of the power outage and restoration visualization management method of the present invention, the method includes: constructing a power grid topology graph and associating user-side events in the multi-source power outage and restoration dataset with corresponding nodes in the topology graph. The topology diagram includes substations, switching stations, distribution lines, transformers, and user metering points as nodes, and establishes the connection relationship between nodes through power transmission paths; in the topology diagram, each node is assigned a unique identifier, which is used to uniquely determine its specific physical location in the power grid and its electrical connection attributes. Based on the spatial coordinate information in the multi-source power outage and restoration dataset and the location of the user-side event, the user-side event is associated with the corresponding node in the topology graph; The associated nodes are marked with status to obtain data on the impact of user-side events on the power grid operation status.

[0008] As a preferred embodiment of the power outage and restoration visualization management method described in this invention, the following steps are taken: using user-side alarm events as observation variables, and based on a Bayesian probability model, the probability of fault occurrence is deduced backward from the end user node along the topology diagram to calculate the fault confidence of each medium-voltage branch line segment and generate a list of suspected fault sections. This includes: using user-side alarm events as observation variables, combining the topology diagram, defining fault propagation constraint rules according to the physical characteristics of the power grid, and constructing a Bayesian probability model based on external multi-source data; the external multi-source data includes power grid topology, historical faults, real-time alarms, and user metering data. Branch line faults are treated as latent variables. Prior probabilities and conditional probabilities are statistically analyzed using a historical case database, and maximum likelihood estimation is used for parameter training. When an alarm message is received, starting from the affected users and metering points, the posterior probability is calculated using Bayes' theorem based on the topological connection relationship and historical failure probability. If multiple alarms occur simultaneously, it is assumed that the alarms are independent of each other. The joint probability of all alarms occurring at the same time is calculated, and the possibility of each branch line becoming the source of the fault is deduced in reverse. Along the direction from the end user node to the substation, the fault confidence of each medium-voltage branch line section is calculated step by step, and a list of suspected fault sections is generated based on the preset confidence threshold.

[0009] As a preferred embodiment of the power outage and restoration visualization management method described in this invention, the step of generating a list of suspected fault sections based on a preset confidence threshold includes: comparing the posterior probability of each medium-voltage branch line section with a preset confidence threshold, filtering out line sections with a posterior probability of fault not lower than the threshold, and arranging them in descending order of fault confidence to form a list of suspected fault sections.

[0010] As a preferred embodiment of the power outage and restoration visualization management method described in this invention, the following steps are taken: The list of suspected faulty sections is overlaid onto a geographic information system map for visualization, and a repair task work order is generated by combining the distribution of emergency repair resources and historical handling records: Each medium-voltage branch line segment in the suspected fault section list is overlaid onto a geographic information system map according to its geographical coordinate information, and the fault risk area is visualized. Synchronously access the emergency repair resource database to obtain the location, skill qualifications, and task load status of currently available emergency repair teams, and combine this with the average processing time, success rate, and path consumption time of similar faults in historical handling records; The optimal repair team is matched based on the fault confidence ranking of suspected fault sections, real-time distribution of repair resources, traffic network conditions, and historical handling efficiency. Automatically generate emergency repair work orders that include fault location, affected area, recommended repair personnel, estimated arrival time, and handling suggestions, and push them to the corresponding mobile application.

[0011] As a preferred embodiment of the power outage and restoration visualization management method described in this invention, the following steps are taken: based on the on-site handling results, the equipment status in the topology diagram is updated, and the accuracy of the fault simulation results is verified in a closed loop to complete the full-process management of power outage and restoration events, including: Receive on-site handling results from emergency repair personnel via mobile terminal, including the actual fault location, equipment status, and power restoration time; The operating status of the corresponding node in the topology graph is updated according to the handling results, and the alarm records and simulation results of this power outage and restoration event are compared with the actual fault location to calculate the judgment accuracy, misjudgment rate and missed judgment rate. If a medium-voltage branch line is identified as a suspected fault source by the system in multiple historical events but no fault is confirmed on-site, the basic prior probability of the fault for that line will be lowered. If the alarm coverage rate of downstream users is consistently low when a line experiences a real fault, the conditional probability of triggering an alarm when the line is faulty in the Bayesian model will be reduced. If a certain type of user metering equipment or area frequently generates abnormal alarms under fault-free conditions, the likelihood probability parameter of false alarms under normal conditions will be increased. Based on the adjusted prior and conditional probabilities, the Bayesian fault inference model is dynamically optimized, and the updated model is used for the visual management of power outage and restoration events.

[0012] Secondly, the present invention provides a power outage and restoration visualization management system, comprising: The data receiving module is used to receive power outage and restoration related data from multiple heterogeneous business systems, preprocess the power outage and restoration related data, and uniformly convert the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. The topology construction module is used to construct a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with corresponding nodes in the topology graph. The fault inference and calculation module is used to use user-side alarm events as observation variables, and based on the Bayesian probability model, to infer the probability of fault occurrence from the end user node along the topology graph, calculate the fault confidence of each medium-voltage branch line segment, and generate a list of suspected fault sections. The visualization module is used to overlay the list of suspected fault sections onto a geographic information system map for visualization, and generate emergency repair work orders by combining the distribution of emergency repair resources and historical handling records. The status update module is used to update the device status in the topology diagram based on the on-site handling results, and to verify the accuracy of the fault simulation results in a closed loop, thereby completing the full-process management of power outage and restoration events.

[0013] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the power outage and restoration visualization management method are implemented.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the power outage and restoration visualization management method.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By fusing multi-source data and based on Bayesian back inference and power grid topology models, this invention can improve the accuracy of transformer / feeder status monitoring and 10kV branch fault location, thereby increasing the efficiency of fault location; relying on geographic information system visualization and intelligent decision-making, it presents the power outage situation globally, shortens decision response time, and automatically generates various reports, reducing labor costs; intelligent alarm filtering and precise location shorten power restoration time, effectively reducing power outage losses for users; at the same time, it supports accurate power outage information push and self-service query, reducing call volume and complaint rate, and improving customer satisfaction and corporate image; the system also provides multi-dimensional retrospective analysis to support emergency repair optimization and rational resource allocation, comprehensively promoting lean power grid operation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the process flow of a power outage and restoration visualization management method according to an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0018] Example 1, referring to Figure 1 As one embodiment of the present invention, this embodiment provides a power outage and restoration visualization management method, including: S100: Receives power outage and restoration related data from multiple heterogeneous business systems, preprocesses the power outage and restoration related data, and uniformly converts the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. S200: Construct a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with the corresponding nodes in the topology graph; S300: Using user-side alarm events as observation variables, based on the Bayesian probability model, it reverse-engineers the probability of fault occurrence from the end user node along the topology graph, calculates the fault confidence of each medium-voltage branch line segment, and generates a list of suspected fault sections. S400: Overlay the list of suspected faulty sections onto the geographic information system map for visualization, and generate emergency repair work orders by combining the distribution of emergency repair resources and historical handling records. S500: Based on the on-site handling results, update the equipment status in the topology diagram, and verify the accuracy of the fault simulation results in a closed loop, completing the full-process management of power outage and restoration events.

[0019] It should be noted that this invention focuses on the development of visualization technology for power grid geographic information systems (GIS) to address the issue of fault handling and monitoring capabilities. Because the coordinate systems for power outage and restoration information differ across external systems, when integrated into this system, coordinate systems such as CGCS2000 and WGS84 are uniformly converted to the GCJ-02 coordinate system usable by the power grid GIS. This integrates the spatial information of the power grid with power outage and restoration data, forming a unified system that allows information such as power outage and restoration areas, equipment operating status, and user distribution to be displayed intuitively on a map. Furthermore, a multi-dimensional, precise query function is developed, making power outage and restoration monitoring more intuitive and information retrieval faster. This allows management personnel to quickly assess the impact of power outages and provides effective support for decision-making. To address the problem of slow and inaccurate fault diagnosis, the invention focuses on designing intelligent alarm filtering and fault analysis models. Develop an intelligent alarm filtering and prioritization function to accurately filter out duplicate and false alarms. Combined with line topology and actual operational data, establish a systematic intelligent fault diagnosis model to quickly pinpoint the fault area and location, significantly reducing troubleshooting time. To address inaccurate customer service and slow response times, establish a tiered customer service information push mechanism. Based on user level (e.g., important users, ordinary users), important users receive more detailed information such as power restoration time, equipment affected by the outage, and repair progress via SMS and the power supply service app. Ordinary users receive power restoration information via the power supply service app. This creates a distinct information service system, accurately matching the most important information to each user—such as the cause of the outage, repair progress, and power restoration time—and pushing it in real-time. This solves the problem of information gaps and improves customer satisfaction. Finally, to address insufficient statistical analysis capabilities, develop automated, multi-dimensional reporting and statistical analysis functions. A real-time data collection and processing system for power outages and restorations has been established, enabling automatic report generation and multi-dimensional statistical analysis. This system supports statistical analysis by region, time period, fault type, and user level, and allows for custom data retrieval, report export, and trend analysis. It provides data support for power outage and restoration management, including review and improvement, resource allocation, and long-term planning.

[0020] It should be noted that faults in 10kV medium-voltage branch lines are highly concealed and have a wide impact range, making rapid and accurate location crucial for minimizing power outage time. Traditional methods relying on manual experience to troubleshoot section by section are inefficient. The algorithm design of this invention comprises three parts: fusing multi-source data, locating the root cause of the fault, and implementing continuous optimization.

[0021] In this embodiment of the invention, step S100, which involves receiving power outage and restoration related data from multiple heterogeneous service systems and preprocessing the power outage and restoration related data, includes: The heterogeneous business systems include a WebGIS platform, a marketing system, an operations and distribution system, a massive data platform, a SCADA / OMS system, and a metrology automation system; The power outage and restoration related data includes massive time-series data and structured business data; the power outage and restoration related data are processed sequentially by data cleaning, format standardization, time alignment and entity association, the spatial coordinate information contained therein is identified, and coordinate data from different sources are uniformly converted to the preset geographic coordinate system; The structured data and spatial data are integrated and processed to form a fused multi-source power outage and restoration dataset.

[0022] It should be noted that fault assessment requires the integration of heterogeneous data from multiple sources, including power grid topology, real-time alarms, and user information. This invention employs a data association strategy of metadata mapping and pre-dynamic field matching to construct a unified data dictionary and field mapping rule base. Based on the interface documents and actual return values ​​provided by various external systems, metadata information returned by each data source is extracted. A standardized metadata model is established based on the business requirements of power grid fault assessment. This enables accurate matching of data fields from different systems such as WebGIS, SCADA, and metering automation.

[0023] In this embodiment of the invention, step S200, which involves constructing a power grid topology graph and associating user-side events from the multi-source power outage and restoration dataset with corresponding nodes in the topology graph, includes: The topology diagram includes substations, switching stations, distribution lines, transformers, and user metering points as nodes, and establishes the connection relationship between nodes through power transmission paths. In the topology diagram, each node is assigned a unique identifier, which is used to uniquely determine its specific physical location in the power grid and its electrical connection attributes. Based on the spatial coordinate information in the multi-source power outage and restoration dataset and the location of the user-side event, the user-side event is associated with the corresponding node in the topology graph; The associated nodes are marked with status to obtain data on the impact of user-side events on the power grid operation status.

[0024] In this embodiment of the invention, step S300 uses user-side alarm events as observation variables, and based on a Bayesian probability model, reverses the probability of fault occurrence from the end user node along the topology graph to calculate the fault confidence of each medium-voltage branch line segment and generate a list of suspected fault sections. This includes: using user-side alarm events as observation variables, combining the topology graph, defining fault propagation constraint rules according to the physical characteristics of the power grid, and constructing a Bayesian probability model based on external multi-source data; the external multi-source data includes power grid topology, historical faults, real-time alarms, and user metering data. Branch line faults are treated as latent variables. Prior probabilities and conditional probabilities are statistically analyzed using a historical case database, and maximum likelihood estimation is used for parameter training. When an alarm message is received, starting from the affected users and metering points, the posterior probability is calculated using Bayes' theorem based on the topological connection relationship and historical failure probability. If multiple alarms occur simultaneously, it is assumed that the alarms are independent of each other. The joint probability of all alarms occurring at the same time is calculated, and the possibility of each branch line becoming the source of the fault is deduced in reverse. Along the direction from the end user node to the substation, the fault confidence of each medium-voltage branch line section is calculated step by step, and a list of suspected fault sections is generated based on the preset confidence threshold.

[0025] In this embodiment of the invention, step S300, which generates a list of suspected fault sections based on a preset confidence threshold, includes: comparing the posterior probability of each medium-voltage branch line section with a preset confidence threshold, selecting line sections with a posterior probability of fault not lower than the threshold, and arranging them in descending order of fault confidence to form a list of suspected fault sections.

[0026] In this embodiment of the invention, the confidence threshold can be adjusted in real time according to the fault situation, and the invention does not impose any specific limitations.

[0027] It should be noted that this invention integrates power grid topology with a Bayesian probabilistic statistical model, using external multi-source data fusion to extract power grid topology, historical faults, real-time alarms, and user metering data. First, a standardized node-edge topology model and a hierarchical fault propagation graph are constructed, and fault propagation constraint rules are defined based on the physical characteristics of the power grid. Then, using branch line faults as latent variables and user alarms as observed variables, a Bayesian probabilistic model is established. Prior probabilities and conditional probabilities are statistically analyzed using a historical case database, and parameters are trained using maximum likelihood estimation. Finally, the model is validated and optimized using typical fault cases, forming a fault backpropagation system adapted to the target power grid. A fault propagation path model tracing back from the user to the substation is constructed.

[0028] Furthermore, upon receiving an alarm message, starting from the affected users and metering points, and combining topological connections and historical fault probabilities, the posterior probability is calculated using Bayes' theorem. This posterior probability represents the probability that the line actually failed given the current alarm. The formula is: (the probability of triggering the current alarm when the line fails × the baseline probability of the line itself failing) divided by (the total probability of the current alarm occurring under all possible conditions). If multiple alarms occur simultaneously, the joint probability of all alarms occurring simultaneously is calculated under the assumption that the alarms are independent. This reverse calculation of the probability that each branch line is the source of the fault significantly improves the accuracy and efficiency of fault location, transforming it from blind guessing based on experience to precise data-driven location.

[0029] In this embodiment of the invention, step S400 involves overlaying the list of suspected faulty sections onto a geographic information system map for visualization, and generating an emergency repair task work order by combining the distribution of emergency repair resources and historical handling records. Each medium-voltage branch line segment in the suspected fault section list is overlaid onto the geographic information system map according to its geographical coordinate information, and the fault risk area is visualized. Synchronously access the emergency repair resource database to obtain the location, skill qualifications, and task load status of currently available emergency repair teams, and combine this with the average processing time, success rate, and path consumption time of similar faults in historical handling records; The optimal repair team is matched based on the fault confidence ranking of suspected fault sections, real-time distribution of repair resources, traffic network conditions, and historical handling efficiency. Automatically generate emergency repair work orders that include fault location, affected area, recommended repair personnel, estimated arrival time, and handling suggestions, and push them to the corresponding mobile application.

[0030] In this embodiment of the invention, step S500 involves updating the equipment status in the topology diagram based on the on-site handling results and verifying the accuracy of the fault simulation results in a closed loop, thus completing the full-process management of power outage and restoration events, including: Receive on-site handling results from emergency repair personnel via mobile terminal. The handling results include the actual fault location, equipment status, and power restoration time. The operating status of the corresponding nodes in the topology diagram is updated based on the handling results. The alarm records and simulation results of this power outage and restoration event are compared with the actual fault location, and the accuracy rate, misjudgment rate and missed judgment rate are calculated. If a medium-voltage branch line is identified as a suspected fault source by the system in multiple historical events but no fault is confirmed on-site, the basic prior probability of the fault for that line will be lowered. If the alarm coverage rate of downstream users is consistently low when a line experiences a real fault, the conditional probability of triggering an alarm when the line is faulty in the Bayesian model will be reduced. If a certain type of user metering equipment or area frequently generates abnormal alarms under fault-free conditions, the likelihood probability parameter of false alarms under normal conditions will be increased. Based on the adjusted prior and conditional probabilities, the Bayesian fault inference model is dynamically optimized, and the updated model is used for the visual management of power outage and restoration events.

[0031] It should be noted that this embodiment of the invention, through high-quality fused data and reverse inference algorithms, continuously learns and optimizes to quickly output reliable fault location results on conventional hardware. By receiving real-time alarm streams from systems such as SCADA and metering automation, and based on the full-link topology of stations, lines, transformers, and users, it automatically associates affected equipment and user ranges, filtering out invalid information. Based on topological connectivity and probability models, it reverse-engineers from the power outage user end to calculate the fault probability of each upstream branch line and generates a list of suspected branches sorted by probability. Finally, the model, combined with a preset initial reliability threshold or a threshold manually set on the system, automatically outputs the most likely fault section and directly pushes the judgment conclusion to the emergency repair work order, achieving a minute-level closed loop from alarm to analysis and location.

[0032] Furthermore, by recording and analyzing in real time the accuracy, false positive rate, and false negative rate of the judgment results under different parameter configurations, regarding the false positive rate issue, if a certain line is frequently misjudged as a fault source, its historical fault statistics need to be re-verified, and the basic fault probability adjusted to better reflect actual operational risks. Simultaneously, the probability of downstream alarms occurring when a line fails also needs to be adjusted; if alarm coverage is incomplete when a line fails, this probability value needs to be lowered. Similarly, the probability of false alarms occurring when the line is normal needs to be adjusted; if a certain type of equipment has a high false alarm rate, its false alarm probability parameter should be specifically increased to reduce the interference caused by false alarms and continuously improve the accuracy of fault location.

[0033] Example 2: The above example is an illustrative scheme of a power outage and restoration visualization management method. It should be noted that the technical solution of this power outage and restoration visualization management system belongs to the same concept as the technical solution of the power outage and restoration visualization management method described above. Details not described in detail in this example of the power outage and restoration visualization management system can be found in the description of the technical solution of the power outage and restoration visualization management method described above.

[0034] This embodiment of a power outage and restoration visualization management system includes: The data receiving module is used to receive power outage and restoration related data from multiple heterogeneous business systems, preprocess the power outage and restoration related data, and uniformly convert the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. The topology construction module is used to build a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with the corresponding nodes in the topology graph. The fault inference and calculation module is used to use user-side alarm events as observation variables, and based on the Bayesian probability model, to infer the probability of fault occurrence from the end user node along the topology graph, calculate the fault confidence of each medium-voltage branch line segment, and generate a list of suspected fault sections. The visualization module is used to overlay the list of suspected fault sections onto the geographic information system map for visualization, and generate emergency repair work orders by combining the distribution of emergency repair resources and historical handling records. The status update module is used to update the equipment status in the topology diagram based on the on-site handling results, and to verify the accuracy of the fault simulation results in a closed loop, thus completing the full-process management of power outage and restoration events.

[0035] Furthermore, in this embodiment of the invention, a power outage and restoration visualization management system adopts a three-layer architecture design, based on reliable data and supported by an intelligent engine, and ultimately realizes a complete closed loop of monitoring, judgment, handling and analysis in business scenarios, thereby improving power outage response efficiency and customer service level.

[0036] Furthermore, the data layer serves as the core for multi-source data aggregation and highly reliable storage. Leveraging standardized RESTful / WebService interfaces and message queues such as Kafka, it integrates data from WebGIS, marketing systems, distribution systems, massive data platforms, SCADA / OMS, and metering automation systems. It enables real-time incremental collection and scheduled full synchronization of various data types, including equipment operation data, user profiles, alarm events, and work order workflows. During synchronization, rules for format validation, integrity checks, and outlier removal are embedded to ensure the compliance of the integrated data.

[0037] Data storage employs a hybrid distributed database cluster of HBase and MySQL: the distributed NoSQL database stores massive amounts of time-series data such as SCADA real-time monitoring and metering alarms; the relational database stores structured business data such as user information, work order data, and topology relationships. Simultaneously, data backup and disaster recovery mechanisms are established to ensure efficient read / write, elastic scaling, and secure and reliable storage for massive amounts of data.

[0038] Furthermore, the support layer serves as the core technology engine and capability hub, with three core components: a GIS engine, a data fusion engine, and an intelligent algorithm engine. These three work together to complete data processing and intelligent analysis. The GIS engine provides basic services such as high-precision map rendering, power grid topology analysis, spatial location query, and area delineation. It can support the visualization and interaction of the main and distribution network topology map and provide spatial support for fault area location. The data fusion engine is used to establish a complete data processing chain encompassing access, cleaning, transformation, association, and integration, and to handle standardized processing of heterogeneous data from multiple data sources. Through cleaning operations such as data deduplication, missing value completion, and outlier removal, it achieves accurate matching of data fields from multiple data sources, ultimately outputting standardized data in a unified format. The intelligent algorithm engine adopts an offline training and online inference deployment approach, integrating core algorithm models such as alarm filtering, fault reverse inference, and intelligent learning, and provides algorithm call interfaces to support intelligent business scenarios such as application-layer fault assessment.

[0039] Furthermore, the application layer includes four core functional modules: real-time power outage and restoration monitoring, customer service control, automated alarm management, and report statistics management. It supports multi-terminal access via Web (management backend) and mobile (emergency repair APP), forming a complete business loop of monitoring, analysis, handling, and statistics. Real-time power outage and restoration monitoring is used to visually display information such as the power grid operation status, the distribution of power outage areas, and the progress of emergency repairs, and supports multi-dimensional filtering and querying. Customer service control integrates functions such as power outage information push, customer inquiry response, and power outage impact range statistics to help carry out proactive customer service; Automated alarm management is used to classify, filter, and automatically push alarm information, thereby reducing the interference of invalid alarms on staff. The report statistics management is used to automatically generate multi-dimensional reports such as power outage fault analysis, judgment accuracy, and repair efficiency, providing data reference for business decisions.

[0040] This embodiment also provides an electronic device applicable to the power outage and restoration visualization management method, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the power outage and restoration visualization management method proposed in the above embodiments.

[0041] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the power outage and restoration visualization management method proposed in the above embodiments.

[0042] The storage medium proposed in this embodiment and the method for implementing power outage and restoration visualization management proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0043] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0044] Example 3, referring to Table 1, is an embodiment of the present invention. The present invention verifies the beneficial effects of the present invention through comparative experiments.

[0045] In this embodiment, a comparative test was conducted in a pilot area, and the test results are shown in Table 1: Table 1 Comparison Results

[0046] As shown in Table 1, compared to traditional power outage and restoration management and fault diagnosis technologies, this invention achieves multi-dimensional technological breakthroughs and value enhancements through core designs involving architectural innovation, algorithm optimization, and business integration. Based on the system's multi-source data fusion capabilities at the data layer and the support layer's data fusion engine, it achieves deep integration and unified management of data from multiple business systems such as WebGIS, marketing, operations and distribution, and SCADA / OMS, completely solving the problems of data fragmentation, inconsistent standards, and poor correlation in traditional technologies. Through a full-process processing of data cleaning, transformation, and correlation, the completeness, consistency, and availability of power outage and restoration data are significantly improved, providing a high-quality data foundation for subsequent analysis and decision-making. Based on the core algorithm model, key monitoring and judgment indicators are made more accurate: First, the accuracy rate of transformer and feeder power outage and restoration status monitoring reaches over 90%, with real-time response delay controlled within 15 minutes, ensuring the timeliness and reliability of power outage and restoration; second, the accuracy rate of reverse inference of 10kV medium-voltage branch line faults reaches over 60%, with fault location controlled within 30 minutes, significantly improving the efficiency of fault location compared to traditional manual troubleshooting methods.

[0047] Leveraging GIS engine visualization technology at the support layer and intelligent decision-making at the application layer, the system provides a comprehensive and intuitive view of power outages and restorations. Management personnel can quickly and easily grasp core information such as the outage scope, fault distribution, and repair progress, reducing decision-making response time by over 30%. Simultaneously, the system's automated reporting function automatically generates detailed outage reports for customers, key users, and various custom reports, significantly reducing the labor costs of manual statistics and reporting, and improving data reporting efficiency and accuracy. Intelligent alarm filtering algorithms reduce invalid alarm interference, avoiding wasted effort from maintenance personnel; combined with fault reverse inference algorithms and analysis models, it enables rapid and accurate fault location, providing scientific support for emergency resource allocation and repair route planning. Ultimately, this can reduce power restoration time by an average of 20%-40%, effectively reducing power outage losses for users and improving the reliability of the power grid. It also enables accurate transmission of power outage and restoration information to different types of users, providing convenient power outage information query services for the public and reducing call volume on the 95598 hotline. Furthermore, based on the visualization of power outage and restoration status, it ensures the restoration of power to key users, reduces customer complaints, improves customer satisfaction, and further enhances the social image of the power company. The system has multi-dimensional data statistics and analysis functions, which can comprehensively review power outage and restoration events, fault handling effects, and resource utilization efficiency, providing decision support for optimizing power outage and restoration management and rationally allocating emergency repair resources, and promoting further refined management of enterprises.

[0048] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A visual management method for power outage and restoration, characterized in that, include: Receive power outage and restoration related data from multiple heterogeneous business systems, preprocess the power outage and restoration related data, and uniformly convert the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. Construct a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with the corresponding nodes in the topology graph; Using user-side alarm events as observed variables, and based on a Bayesian probability model, the probability of fault occurrence is deduced from the end user node along the topology graph, the fault confidence of each medium-voltage branch line segment is calculated, and a list of suspected fault sections is generated. The list of suspected faulty sections is overlaid onto a geographic information system map for visualization, and a repair task work order is generated by combining the distribution of emergency repair resources and historical handling records. Based on the on-site handling results, the equipment status in the topology diagram is updated, and the accuracy of the fault simulation results is verified in a closed loop, thus completing the full-process management of power outage and restoration events.

2. The power outage and restoration visualization management method as described in claim 1, characterized in that, The process of receiving power outage and restoration related data from multiple heterogeneous service systems and preprocessing the power outage and restoration related data includes: The power outage and restoration related data includes massive time-series data and structured business data; data cleaning, format standardization, time alignment and entity association processing are performed on the power outage and restoration related data to identify the spatial coordinate information contained therein, and coordinate data from different sources are uniformly converted to a preset geographic coordinate system; The structured data and spatial data are integrated and processed to form a fused multi-source power outage and restoration dataset.

3. The power outage and restoration visualization management method as described in claim 2, characterized in that, Constructing a power grid topology graph and associating user-side events from the multi-source power outage and restoration dataset with corresponding nodes in the topology graph includes: The topology diagram includes substations, switching stations, distribution lines, transformers, and user metering points as nodes, and establishes the connection relationship between nodes through power transmission paths; in the topology diagram, each node is assigned a unique identifier, which is used to uniquely determine its specific physical location in the power grid and its electrical connection attributes. Based on the spatial coordinate information in the multi-source power outage and restoration dataset and the location of the user-side event, the user-side event is associated with the corresponding node in the topology graph; The associated nodes are marked with status to obtain data on the impact of user-side events on the power grid operation status.

4. The power outage and restoration visualization management method as described in claim 3, characterized in that, Using user-side alarm events as observed variables, and based on a Bayesian probability model, the probability of fault occurrence is inferred backward from the end user node along the topology diagram. The fault confidence of each medium-voltage branch line segment is calculated, and a list of suspected fault sections is generated. This includes: using user-side alarm events as observed variables, combining the topology diagram, defining fault propagation constraint rules according to the physical characteristics of the power grid, and constructing a Bayesian probability model based on external multi-source data; the external multi-source data includes power grid topology, historical faults, real-time alarms, and user metering data. Branch line faults are treated as latent variables. Prior probabilities and conditional probabilities are statistically analyzed using a historical case database, and maximum likelihood estimation is used for parameter training. When an alarm message is received, starting from the affected users and metering points, the posterior probability is calculated using Bayes' theorem based on the topological connection relationship and historical failure probability. If multiple alarms occur simultaneously, it is assumed that the alarms are independent of each other. The joint probability of all alarms occurring at the same time is calculated, and the possibility of each branch line becoming the source of the fault is deduced in reverse. Along the direction from the end user node to the substation, the fault confidence of each medium-voltage branch line section is calculated step by step, and a list of suspected fault sections is generated based on the preset confidence threshold.

5. The power outage and restoration visualization management method as described in claim 4, characterized in that, The step of generating a list of suspected fault sections based on a preset confidence threshold includes: comparing the posterior probability of each medium-voltage branch line section with a preset confidence threshold, filtering out line sections with a posterior probability of fault not lower than the threshold, and arranging them in descending order of fault confidence to form a list of suspected fault sections.

6. The power outage and restoration visualization management method as described in claim 5, characterized in that, The list of suspected faulty sections is overlaid onto a geographic information system map for visualization. Combined with the distribution of emergency repair resources and historical handling records, an emergency repair task work order is generated, including: Each medium-voltage branch line segment in the suspected fault section list is overlaid onto a geographic information system map according to its geographical coordinate information, and the fault risk area is visualized. Synchronously access the emergency repair resource database to obtain the location, skill qualifications, and task load status of currently available emergency repair teams, and combine this with the average processing time, success rate, and path consumption time of similar faults in historical handling records; The optimal repair team is matched based on the fault confidence ranking of suspected fault sections, real-time distribution of repair resources, traffic network conditions, and historical handling efficiency. Automatically generate emergency repair work orders that include fault location, affected area, recommended repair personnel, estimated arrival time, and handling suggestions, and push them to the corresponding mobile application.

7. The power outage and restoration visualization management method as described in claim 6, characterized in that, Based on the on-site handling results, the equipment status in the topology diagram is updated, and the accuracy of the fault simulation results is verified in a closed loop, completing the full-process management of power outage and restoration events, including: Receive on-site handling results from emergency repair personnel via mobile terminal, including the actual fault location, equipment status, and power restoration time; The operating status of the corresponding node in the topology graph is updated according to the handling results, and the alarm records and simulation results of this power outage and restoration event are compared with the actual fault location to calculate the judgment accuracy, misjudgment rate and missed judgment rate. If a medium-voltage branch line is identified as a suspected fault source by the system in multiple historical events but no fault is confirmed on-site, the basic prior probability of the fault for that line will be lowered. If the alarm coverage rate of downstream users is consistently low when a line experiences a real fault, the conditional probability of triggering an alarm when the line is faulty in the Bayesian model will be reduced. If a certain type of user metering equipment or area frequently generates abnormal alarms under fault-free conditions, the likelihood probability parameter of false alarms under normal conditions will be increased. Based on the adjusted prior and conditional probabilities, the Bayesian fault inference model is dynamically optimized, and the updated model is used for the visual management of power outage and restoration events.

8. A power outage and restoration visualization management system, applied to the method described in any one of claims 1-7, characterized in that, include: The data receiving module is used to receive power outage and restoration related data from multiple heterogeneous business systems, preprocess the power outage and restoration related data, and uniformly convert the spatial coordinate data therein to a preset geographic coordinate system to form a fused multi-source power outage and restoration dataset. The topology construction module is used to construct a power grid topology graph and associate user-side events in the multi-source power outage and restoration dataset with corresponding nodes in the topology graph. The fault inference and calculation module is used to use user-side alarm events as observation variables, and based on the Bayesian probability model, to infer the probability of fault occurrence from the end user node along the topology graph, calculate the fault confidence of each medium-voltage branch line segment, and generate a list of suspected fault sections. The visualization module is used to overlay the list of suspected fault sections onto a geographic information system map for visualization, and generate emergency repair work orders by combining the distribution of emergency repair resources and historical handling records. The status update module is used to update the device status in the topology diagram based on the on-site handling results, and to verify the accuracy of the fault simulation results in a closed loop, thereby completing the full-process management of power outage and restoration events.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the power outage and restoration visualization management method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the power outage and restoration visualization management method according to any one of claims 1 to 7.

Images