Natural disaster power grid disaster damage assessment method and system based on big data analysis

Abstract

The invention provides a natural disaster power grid disaster damage assessment method and system based on big data analysis, and relates to the technical field of power grid disaster damage assessment, and the method comprises the steps: determining the situation evolution characteristics of power grid shutdown of a target city power grid under natural disasters and emergencies through all fault attribute nodes; according to the risk aggregation identifier, determining a power loss index set during disaster-caused recovery of the target city power grid; determining a loss interaction amount when the target urban power grid is subjected to recovery strategy updating according to the disaster damage correction information, and determining a disaster damage response decision when the target urban power grid is influenced by an emergency according to the loss interaction amount and the situation evolution characteristics; and performing synchronous matching on the disaster damage information of the power grid of the target city in the power supply database of the power distribution network of the target city through the disaster damage identification constraint and the disaster damage response decision, and generating a risk assessment report. And multi-dimensional accurate assessment is carried out on the disaster damage condition of the power grid, so that the accuracy of disaster damage assessment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid disaster loss assessment, and more specifically, the present application relates to a natural disaster power grid disaster loss assessment method and system based on big data analysis. Background Art

[0002] Power grid disaster loss assessment refers to the collection and analysis of power grid loss data in natural disasters or emergencies to assess the extent of damage to power grid facilities, downtime, load recovery, and the safety and stability of post-disaster power grid operations. It uses a large amount of historical power grid operation data, weather data, disaster occurrence data, and power grid equipment health status data, combined with advanced data processing technology and algorithm models, to accurately simulate and predict the response of the power grid after a disaster.

[0003] The existing natural disaster power grid loss assessment method based on big data analysis is to achieve an accurate assessment of post-disaster losses of the power grid by integrating multi-source data, advanced algorithms and model analysis. First, a big data platform is built by collecting multi-dimensional data such as power grid operation data (such as load, equipment status), meteorological data, geographic information, disaster occurrence data, etc. in real time. These data are collected and processed through IoT sensors, remote sensing technology and post-disaster assessment tools. Secondly, using machine learning and data mining technology, combined with data from historical disaster events, a disaster loss prediction model is constructed to assess the loss level of the power grid in different types of disasters. Model analysis includes factors such as the degree of damage to power grid equipment, the scope of power outages, and the speed of load recovery, helping to identify key nodes and weak links. Furthermore, based on artificial intelligence and optimization algorithms, the priority and resource allocation requirements of power grid recovery are evaluated, the effects of different recovery strategies are simulated, and the recovery plan is adjusted in real time. However, existing natural disaster power grid damage assessment methods based on big data analysis usually rely on static historical data and preset damage models, lacking real-time tracking and in-depth analysis of the power grid's operating status and post-disaster dynamic changes, resulting in lags and limitations in the damage prediction results during the assessment process, making it difficult to adjust the power grid restoration strategy in a timely manner and inaccurate resource scheduling, thus affecting the efficiency and quality of post-disaster recovery. Therefore, how to accurately assess the damage status of the power grid in multiple dimensions in combination with the dynamic evolution characteristics of the damaged power grid during the post-disaster power grid restoration process in order to improve the accuracy of damage assessment is a problem faced by the industry. Summary of the invention

[0004] The present application provides a natural disaster power grid damage assessment method and system based on big data analysis, which can conduct a multi-dimensional and accurate assessment of the power grid damage status during the post-disaster power grid recovery process, combined with the dynamic evolution characteristics of the damaged power grid, to improve the accuracy of the damage assessment.

[0005] In a first aspect, the present application provides a method for assessing power grid damage caused by natural disasters based on big data analysis, the method comprising the following steps:

[0006] Obtain the power grid operation data of the historical power outage areas of the target city power grid through the big data resource library;

[0007] Classifying the power grid operation data by fault scenario, obtaining the fault attribute node corresponding to each fault scenario, and determining the state evolution characteristics of the target city power grid outage under natural disasters and emergencies according to all the fault attribute nodes;

[0008] Determine the risk aggregation identifier of the target city power grid under different disaster-causing factors, determine the power loss index set of the target city power grid during disaster recovery according to the risk aggregation identifier, and then determine the disaster loss identification constraint of the target city power grid when it is affected by the emergency event through the power loss index set;

[0009] Obtaining the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid, determining the loss interaction amount when the target city power grid updates the restoration strategy according to the disaster correction information, and then determining the disaster response decision when the target city power grid is affected by the emergency event according to the loss interaction amount and the situation evolution characteristics;

[0010] The disaster loss information of the target city power grid is synchronously matched in the target city distribution network power supply database through the disaster loss identification constraint and the disaster loss response decision, and a risk assessment report is generated.

[0011] In this embodiment, the fault scenario classification is performed on the power grid operation data to obtain the fault attribute node corresponding to each fault scenario, specifically including:

[0012] Obtain the fault distribution characteristics when a fault occurs in the target city power grid;

[0013] Determining a fault scenario of a target city power grid according to the fault distribution characteristics;

[0014] The power grid operation data is mapped to the fault scenarios to obtain a fault attribute node corresponding to each fault scenario.

[0015] In this embodiment, the situation evolution characteristics of the target city power grid outage under natural disasters and emergencies are determined based on all fault attribute nodes, specifically including:

[0016] Determine the operation risk scale of the target city power grid when it is shut down according to the fault attribute node;

[0017] The operation risk scale is used to determine the evolution characteristics of the target city power grid outage situation under natural disasters and emergencies.

[0018] In this embodiment, determining the risk aggregation identifier of the target city power grid under different disaster factors specifically includes:

[0019] Obtain risk identification nodes of the target city power grid in the planned outage model;

[0020] The risk aggregation identifier of the target city power grid under different disaster-causing factors is determined according to the risk identification node.

[0021] In this embodiment, determining the damage identification constraints when the target city power grid is affected by an emergency event through the power loss index set specifically includes:

[0022] Determining the fault difference amount of the target city power grid operation according to the power failure index set;

[0023] Determine the dispatching discrete information during the emergency dispatching of the power grid through the fault difference amount;

[0024] The damage identification constraints when the target city power grid is affected by an emergency event are determined according to the fault difference amount and the dispatch discrete information.

[0025] In this embodiment, determining the disaster response decision when the target city power grid is affected by an emergency event based on the loss interaction amount and the situation evolution characteristics specifically includes:

[0026] Determine a fault response label when a sudden fault occurs in the target city power grid according to the loss interaction amount;

[0027] Determine the disaster loss fusion information when the target city power grid is affected by the fault according to the situation evolution characteristics;

[0028] The disaster response decision when the target city power grid is affected by an emergency event is determined by the fault response tag and the disaster fusion information.

[0029] In this embodiment, the power grid operation data refers to a data set generated during the operation of the power system and reflecting the state, performance and operation events of the power grid.

[0030] In this embodiment, the disaster damage correction information refers to key information for adjusting and correcting disaster damage assessment during the power grid restoration process after the power grid encounters a natural disaster or an emergency.

[0031] In this embodiment, the power failure indicator set represents a quantitative set of the degree of recovery of the power failure state during the power grid disaster recovery process.

[0032] In a second aspect, the present application provides a natural disaster power grid damage assessment system based on big data analysis, which is used to perform a natural disaster power grid damage assessment method based on big data analysis, and the assessment system includes:

[0033] A data acquisition module is used to obtain the power grid operation data of the historical power outage area of ​​the target city power grid through a big data resource library;

[0034] A situation analysis module is used to classify the fault scenarios of the power grid operation data, obtain the fault attribute nodes corresponding to each fault scenario, and determine the situation evolution characteristics of the target city power grid outage under natural disasters and emergencies based on all fault attribute nodes;

[0035] A disaster loss identification module is used to determine the risk aggregation identifier of the target city power grid under different disaster-causing factors, determine the power loss index set of the target city power grid during disaster recovery according to the risk aggregation identifier, and then determine the disaster loss identification constraints of the target city power grid when it is affected by the emergency event through the power loss index set;

[0036] A disaster response module is used to obtain the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid, determine the loss interaction amount when the target city power grid updates the restoration strategy according to the disaster correction information, and then determine the disaster response decision when the target city power grid is affected by the emergency event according to the loss interaction amount and the situation evolution characteristics;

[0037] The synchronous matching module is used to synchronously match the disaster information of the target city power grid in the target city distribution network power supply database through the disaster identification constraint and the disaster response decision, and generate a risk assessment report.

[0038] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0039] The power grid operation data of the historical power outage area of ​​the target city power grid is obtained through the big data resource library; the fault scenario of the power grid operation data is classified to obtain the fault attribute node corresponding to each fault scenario, and the situation evolution characteristics of the power grid outage of the target city power grid under natural disasters and emergencies are determined according to all fault attribute nodes; the risk aggregation identifier of the target city power grid under different disaster-causing factors is determined, and the power loss index set of the target city power grid during disaster recovery is determined according to the risk aggregation identifier, and then the disaster loss identification constraint of the target city power grid when it is affected by the emergency is determined through the power loss index set; the disaster loss correction information synchronized in the operation database of the power outage area of ​​the target city power grid is obtained, and the loss interaction amount when the target city power grid updates the recovery strategy is determined according to the disaster loss correction information, and then the disaster loss response decision of the target city power grid when it is affected by the emergency is determined according to the loss interaction amount and the situation evolution characteristics; the disaster loss information of the target city power grid is synchronously matched in the power supply database of the target city distribution network through the disaster loss identification constraint and the disaster loss response decision, and a risk assessment report is generated.

[0040] It can be seen that in this application, firstly, by analyzing the power grid operation data, the problem of incomplete power grid historical data and poor timeliness in the traditional method is effectively solved, and the comprehensiveness and accuracy of the power grid fault data are ensured; then, by determining the situation evolution characteristics, the hierarchical classification processing of the power grid fault data is realized, and the evolution characteristics of the power grid outage under different fault scenarios are identified, which solves the problem of insufficient grasp of the fault evolution law in the prior art; secondly, by determining the disaster loss identification constraints, the problems of inaccurate risk identification and incomplete recovery indicators in the disaster recovery process are effectively solved, so that the power grid recovery process can be dynamically adjusted according to the disaster-causing factors, and refined risk assessment and constraint setting are realized; then, by determining the disaster loss response decision, the problem of delayed power grid recovery strategy and insufficient manual adjustment is solved, and by dynamically obtaining the disaster loss correction information and loss interaction, the real-time update and optimization of the recovery strategy is realized, and the response speed and recovery efficiency of the power grid are improved; finally, by matching the data in the power supply database of the target city distribution network, the automatic synchronous matching of disaster loss information and the efficient generation of risk assessment reports are realized, which solves the problems of untimely post-disaster recovery decisions and delayed assessment results, and provides scientific risk assessment and decision support for power grid managers, and improves the accuracy of disaster loss assessment.

[0041] In summary, the technical solution adopted in this application can combine the dynamic evolution characteristics of the damaged power grid during the post-disaster power grid recovery process to conduct a multi-dimensional and accurate assessment of the power grid's damage status, thereby improving the accuracy of the damage assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0043] Figure 1 is a flow chart of a natural disaster power grid damage assessment method based on big data analysis provided by the present invention;

[0044] Figure 2 is a schematic diagram of a flow chart for determining a power failure indicator set according to the present invention;

[0045] Figure 3 is a schematic diagram of a process for determining the amount of loss interaction provided by the present invention;

[0046] Figure 4 It is a module structure diagram of a natural disaster power grid damage assessment system based on big data analysis provided by the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and drawings. The schematic implementation modes and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention. It should be noted that the present invention is already in the actual development and use stage.

[0048] Embodiment 1

[0049] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 As shown in FIG. 1 , this figure is an exemplary flow chart of a natural disaster power grid damage assessment method based on big data analysis according to this embodiment of the present application, and the assessment method includes the following steps:

[0050] In step S1, the power grid operation data of the historical power outage area of ​​the target city power grid is obtained through the big data resource library.

[0051] In the specific implementation, first, based on the data interface established by the power grid operation management platform, call the historical power outage data related to the target city power grid in the big data resource library, including core fields such as power outage time, fault area, fault type, recovery time, and number of affected users. Secondly, use data extraction, transformation, and loading (ETL) tools to collect data in batches or incrementally to ensure data integrity and timeliness. For example, the power outage data of a specified city in the past five years can be filtered out by timestamp and regional ID. Third, perform data preprocessing, including deduplication, missing value repair, and format standardization operations, such as unifying non-standard time formats to "YYYY-MM-DD HH:MM:SS" and eliminating invalid records. Finally, the data is stored in the power grid fault history database.

[0052] It should be noted that, in the present invention, power grid operation data refers to a data set generated during the operation of the power system that reflects the state, performance and operation events of the power grid.

[0053] In step S2, the power grid operation data is classified according to fault scenarios to obtain fault attribute nodes corresponding to each fault scenario, and the situation evolution characteristics of the target city power grid outage under natural disasters and emergencies are determined based on all fault attribute nodes.

[0054] In this embodiment, the fault scenario classification of the power grid operation data and obtaining the fault attribute node corresponding to each fault scenario can be implemented by the following steps:

[0055] Obtain the fault distribution characteristics when a fault occurs in the target city power grid;

[0056] Determining a fault scenario of a target city power grid according to the fault distribution characteristics;

[0057] The power grid operation data is mapped to the fault scenarios to obtain a fault attribute node corresponding to each fault scenario.

[0058] In the specific implementation, first, the fault distribution information is extracted through the historical power grid operation data, covering the dimensions of time, space, frequency and cause of the fault. For example: count the time periods when faults occur (such as morning peak and night); analyze the high-incidence and low-incidence areas of fault area distribution; identify the number and periodicity of faults in different regions; distinguish different disaster-causing factors such as natural disasters (typhoons, lightning strikes), equipment aging, and external force damage, and then use statistical analysis tools (such as heat maps and distribution curves) to visualize the fault distribution characteristics, reveal the regularity and key distribution areas of power grid faults, and obtain the fault distribution characteristics when the target city power grid fails. Then, according to the fault distribution characteristics, the fault scenarios are divided into several categories, such as: Mild scenario: small-scale, short-duration local power outages (such as equipment tripping). Moderate scenario: medium-sized area, multi-user power outages (such as substation failures). Severe scenario: large-scale, long-term power outages (such as multi-regional power outages caused by typhoons). Combined with the main causes of the faults (natural disasters, equipment failures, emergencies), the fault scenarios are classified and identified to form a scenario set. Finally, using data labeling and rule mapping technology, the historical power grid operation data is matched to the determined fault scenario. For example, the fault data caused by a typhoon (fault time, area, recovery time, number of affected users) will be mapped to the "natural disaster-severe scenario". After the mapping is completed, the fault attribute nodes are extracted from each fault scenario. These nodes are the key elements to describe the fault characteristics, such as: fault area, fault cause, fault duration and number of affected users.

[0059] It should be noted that, in the present invention, the fault distribution characteristics represent the distribution patterns of power grid faults in time, space, frequency and causes; the fault scenario is a specific context formed by classifying power grid faults based on the distribution characteristics and impact degree of the faults; the fault attribute node refers to the quantitative indicator of the key data points of the fault characteristics, including fault time, area, cause, duration and number of users affected.

[0060] In this embodiment, the following steps can be used to determine the state evolution characteristics of the target city power grid outage under natural disasters and emergencies based on all fault attribute nodes:

[0061] Determine the operation risk scale of the target city power grid when it is shut down according to the fault attribute node;

[0062] The operation risk scale is used to determine the evolution characteristics of the target city power grid outage situation under natural disasters and emergencies.

[0063] In the specific implementation, first, based on the fault attribute nodes (fault area, cause, duration, number of affected users, etc.), the data of these nodes are aggregated to form a feature set of power grid outage. For example: fault area distribution (such as high-incidence area, diffusion area), outage duration range (short, medium, long), the number of affected users (mild <1000, medium 1000-10000, severe >10000), and then based on the feature set of power grid outage, a quantitative scale of operation risk is constructed, for example, low risk: localized fault, fewer affected users, short outage time; medium risk: fault spread to multiple regions, moderate recovery time; high risk: large-scale outage, wide impact on users, long recovery time (such as multi-regional faults under natural disasters), and then mathematical models (such as entropy weight method, hierarchical analysis method) are used to assign weights to each fault attribute node, and it is comprehensively calculated as an operation risk index, and combined with the fault duration and the number of affected users, the risk index is evaluated as "high risk", and the corresponding risk scale is "high risk-severe scenario", that is, the operation risk scale is obtained. Then, based on the fault risk scale, the fault evolution process of the power grid under natural disasters and emergencies is modeled to form a situation evolution chain: cause → diffusion → peak → recovery, and key indicators are extracted from the evolution process, such as: fault diffusion speed (the rate at which the fault area expands); the number of users affected by the peak (the maximum number of affected users); the duration of the outage (the total time from the occurrence of the fault to the completion of recovery); fault recovery efficiency (the ratio of fault repair time to resource input), and the extracted evolution indicators are used as situation evolution characteristics.

[0064] It should be noted that, in the present invention, the operating risk scale represents a scale for quantitative risk assessment of the power grid outage status, which is used to describe the severity and risk level of power grid failures; the situation evolution characteristics describe the dynamic evolution law of power grid outages under natural disasters or emergencies, including key stages such as the cause of the failure, the diffusion process, the peak impact and the recovery process.

[0065] In step S3, the risk aggregation identifier of the target city power grid under different disaster factors is determined, and the power loss index set of the target city power grid during disaster recovery is determined according to the risk aggregation identifier, and then the disaster loss identification constraint of the target city power grid when it is affected by the emergency is determined through the power loss index set.

[0066] In this embodiment, the risk aggregation identification of the target city power grid under different disaster-causing factors can be determined by the following steps:

[0067] Obtain risk identification nodes of the target city power grid in the planned outage model;

[0068] The risk aggregation identifier of the target city power grid under different disaster-causing factors is determined according to the risk identification node.

[0069] In the specific implementation, first, a planned outage model is constructed, that is, the planned outage model is a virtual operation scenario established by simulating the outage risk of the power grid under different scenarios. The model generates the planned outage status of the power grid in different time periods and regions by inputting data such as the equipment operation status, historical fault data, maintenance plan, and load distribution of the power grid. For example, regular equipment maintenance and load peak diversion operations will form specific outage nodes. Then, the planned outage model is run to identify the risk nodes in the power grid. These nodes usually include high-risk equipment: such as substations that have been operating under long-term overload or seriously aged transmission lines; key outage areas: areas with a large impact on the power supply load or dense important users; potential disaster-causing factor trigger points: such as areas with high incidence of natural disasters (typhoons, floods), and areas with external force damage risks (construction areas). The risk nodes in the power grid are used as risk identification nodes. Then, the disaster-causing factors that cause power grid failures are classified, including: natural disasters; equipment failures; external force damage; human factors. Based on different disaster-causing factors, the risk identification nodes are aggregated and analyzed to form a risk aggregation mark. A weight is assigned to each risk identification node, and the weight value is determined according to factors such as the historical frequency of faults, the scope of impact, and the importance of equipment. The risk of each node is matched with the corresponding disaster-causing factor to form an association between the node and the disaster-causing factor. By aggregating the risk nodes of all disaster-causing factors, a risk aggregation mark for the power grid is generated.

[0070] It should be noted that, in the present invention, the planned outage model refers to a simulation of the outage state of the power grid due to factors such as equipment maintenance and load distribution, which is used to identify possible risk nodes in the power grid; risk identification nodes refer to key areas, equipment or lines where faults or risks may exist during power grid operation; risk aggregation identifiers represent the state characteristics of the overall risk state of the power grid and potential failure trends.

[0071] Preferably, in this embodiment, the power failure indicator set for disaster recovery of the target city power grid is determined according to the risk aggregation identifier, referring to Figure 2 As shown, this figure is a schematic diagram of a process for determining a power failure indicator set in some embodiments of the present invention. In this embodiment, determining the power failure indicator set can be implemented by the following steps:

[0072] In step S31, recovery feedback information of the target city power grid during disaster recovery is determined according to the risk aggregation identifier;

[0073] In step S32, a load classification sample in the target city power grid restoration process is determined through the restoration feedback information;

[0074] In step S33, the power failure attribute probability when the load is graded and restored in the target city power grid is determined;

[0075] In step S34, a power failure indicator set for disaster recovery of the target city power grid is determined according to the load classification samples and the power failure attribute probability.

[0076] In the specific implementation, first, locate the key nodes and areas with high risks in the power grid, such as high-fault equipment, large-scale power outage areas, etc., and then collect the following data in real time through the monitoring system of the power grid dispatching center during the disaster recovery process of the power grid: fault recovery time: the gradual recovery time of each area; recovery progress feedback: equipment repair status, proportion of users with power supply recovery, etc.; load fluctuation feedback: the change of power load during the recovery process, that is, to obtain recovery feedback information; then, based on the recovery feedback information, classify the load data in the recovery process to form load classification samples. Load classification can be carried out according to the following principles: important loads: such as hospitals, communication base stations, government agencies, etc., which have the highest priority for power grid recovery; secondary loads: such as commercial areas, industrial areas, etc., which have high requirements for power supply time, but are not emergency loads; general loads: such as ordinary residential power supply, rural areas, etc., which can be gradually restored after the important loads are restored, and then extract load classification samples from historical recovery data and real-time feedback to form a data set of recovery time, repair priority and power supply fluctuation under different load levels. Then, based on the load classification samples, the probability of power outage at different load levels during the disaster recovery process is statistically analyzed. This can be achieved using historical statistical analysis or probability distribution models. For example: the probability of power outage for important loads is low, the recovery time is short, and the power supply stability is high; the probability of power outage for secondary loads is medium, the recovery time is long, and the power supply fluctuation is obvious; the probability of power outage for general loads is high, the recovery time is longest, and the stability is poor. Finally, the load classification samples are associated with the power outage attribute probability to form a power outage index set, which includes the following key indicators: power outage recovery time: the average time required for various types of loads to restore power supply; unrestored load ratio: the proportion of unrestored loads at different stages; power outage probability distribution: the power outage probability and distribution law of each load level; power supply stability: stability indicators such as voltage and frequency after power supply is restored.

[0077] It should be noted that, in the present invention, the recovery feedback information represents the equipment repair status, power supply restoration progress, load fluctuation and other information collected during the power grid restoration process, which is used to evaluate the power grid restoration effect; the load classification sample refers to the classification of load data in the recovery process according to load priority, forming a classification data set; the power outage attribute probability refers to the probability of the load experiencing a power outage during the recovery process, including quantitative statistics of factors such as recovery time and the proportion of unrestored load; the power outage index set represents a quantitative set of the degree of recovery of the power outage state during the disaster recovery process of the power grid, which is used to evaluate the power grid restoration efficiency and stability.

[0078] In this embodiment, the following steps can be used to determine the damage identification constraints when the target city power grid is affected by an emergency event through the power loss index set:

[0079] Determining the fault difference amount of the target city power grid operation according to the power failure index set;

[0080] Determine the dispatching discrete information during the emergency dispatching of the power grid through the fault difference amount;

[0081] The damage identification constraints when the target city power grid is affected by an emergency event are determined according to the fault difference amount and the dispatch discrete information.

[0082] In the specific implementation, first, combined with the power failure index set, the recovery time, power failure probability, and unrecovered load ratio of each load level in the power grid are obtained, and the difference in the operating state before and after the power grid fault is compared. For example, under normal conditions, the voltage in a certain area is stable at 220kV, and after the fault occurs, it drops to 180kV, with a voltage deviation of 40kV. The normal load power supply rate is 100%, and the power supply rate in some areas drops to 50% after the fault, and the load recovery deviation is 50%, which will not be repeated here. Then, based on the fault difference, the load distribution changes and dispatch execution deviations in the power grid are analyzed, and the deviation between the emergency dispatch instruction and the actual dispatch execution result is extracted, and the deviation is used as the dispatch discrete information during the emergency dispatch of the power grid. Finally, according to the fault difference, the voltage, current, power flow and other parameters when the power grid is restored are analyzed to see whether they exceed the safe operation limit. For example, a large voltage drop may cause system instability. According to the dispatch discrete information, the execution capacity upper limit of the power grid emergency dispatch is determined, and the resource dispatch bottleneck in power supply restoration is identified. Then, combined with the power failure index set, the duration and priority constraints faced by different load levels during the recovery process are determined.

[0083] It should be noted that, in the present invention, the fault difference represents the actual operation deviation of the power grid under the influence of an emergency event; the dispatch discrete information refers to the power distribution deviation and the degree of dispatch imbalance caused by the fault difference in the emergency dispatch process of the power grid; the disaster loss identification constraint represents the fault recovery limit, resource allocation bottleneck and other restrictive conditions that exist in the disaster recovery process of the power grid.

[0084] In step S4, the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid is obtained, and the loss interaction amount when the target city power grid updates the recovery strategy is determined based on the disaster correction information, and then the disaster response decision when the target city power grid is affected by the emergency is determined based on the loss interaction amount and the situation evolution characteristics.

[0085] In specific implementation, the following methods can be used to obtain the synchronized disaster correction information in the operation database of the power grid blackout area in the target city: first, the power grid operation database stores the real-time monitoring data, historical operation records and disaster correction strategies of the power grid equipment. In order to ensure the accuracy of the data, the key operating parameters of each blackout area, such as voltage, current, power flow and load fluctuation, must be obtained in real time through the data acquisition system (such as the SCADA system). Secondly, by parameterizing the disaster correction model of the blackout area, combined with the fault recovery model and the load recovery model, the disaster correction information of the power grid under different disaster scenarios can be obtained. The disaster correction information mainly includes: the power supply correction amount during post-disaster recovery, the time required for equipment repair during the recovery process, the load fluctuation amount after the fault occurs, etc. By comparing with historical data, the correction information can effectively reflect the actual loss and repair progress of the power grid recovery under emergencies. In addition, the disaster correction information will be processed and optimized through the big data analysis platform, combined with the predictive maintenance algorithm and the post-disaster assessment model, to accurately estimate the disaster correction amount of the power grid during the recovery process, which will not be repeated here.

[0086] It should be noted that, in the present invention, the disaster damage correction information refers to the key information for adjusting and correcting the disaster damage assessment during the power grid restoration process after the power grid encounters a natural disaster or an emergency.

[0087] Preferably, in this embodiment, the loss interaction amount when the target city power grid performs restoration strategy update is determined according to the disaster damage correction information, referring to Figure 3 As shown, this figure is a schematic diagram of the process of determining the loss interaction amount in some embodiments of the present invention. In this embodiment, determining the loss interaction amount can be implemented by the following steps:

[0088] In step S41, the load coordination characteristics when the target city power grid is damaged are extracted from the disaster damage correction information;

[0089] In step S42, resource adaptation information for load restoration of the target city power grid is determined according to the disaster damage correction information;

[0090] In step S43, determining the power distribution confidence level when the target city power grid performs restoration strategy update;

[0091] In step S44, the loss interaction amount when the target city power grid performs a restoration strategy update is determined according to the load coordination characteristics, the resource adaptation information and the power distribution confidence level.

[0092] In the specific implementation, first, during the post-disaster recovery process, the demand fluctuations of different load areas are considered, and the recovery order and priority of each load area are clarified according to the actual situation of post-disaster recovery. The actual data of the load recovery in each area can be obtained from the disaster damage correction information, and the load change trend, recovery time and repair priority can be analyzed. Then, according to the disaster damage correction information, the recovery demand of each load area is analyzed, and the resource adaptation plan is formulated in combination with the actual resource status of the power grid (such as remaining power generation capacity, transmission line status, equipment repair progress, etc.). For example, if there is a shortage of resources in a certain area (such as insufficient power generation capacity) during the recovery process, the resource allocation needs to be adjusted through the dispatching system to avoid bottlenecks or conflicts in load recovery; then, according to the disaster damage correction information, combined with the recovery progress, load priority and resource allocation of each area, the confidence level of different recovery stages is established. High priority areas (such as hospitals, communication facilities, etc.) will be given higher confidence and priority, that is, the distribution confidence level is obtained. Finally, the load coordination characteristics, resource adaptation information and distribution confidence level are combined to analyze the interactive effects of different recovery strategies and resource allocation schemes. Calculate the comprehensive loss of the power grid during the restoration process under a given restoration strategy (such as losses caused by load restoration delays, insufficient resources, etc.), that is, the loss interaction.

[0093] It should be noted that, in the present invention, load coordination characteristics refer to the coordination method and adjustment strategy of loads in different regions during the power grid recovery process, reflecting characteristics such as load fluctuation, recovery progress and recovery order; resource adaptation information refers to the load recovery resource allocation plan based on the actual resource conditions during the power grid recovery process, ensuring that resources reasonably and efficiently support the recovery of various regions of the power grid; distribution confidence level refers to the priority and recovery control trust level dynamically set according to the recovery progress and resource requirements of each region during the recovery process; loss interaction refers to the comprehensive loss caused by the interaction between the recovery strategies and resource scheduling of each region during the power grid recovery process.

[0094] In this embodiment, the disaster response decision when the target city power grid is affected by an emergency event can be determined by the loss interaction amount and the situation evolution characteristics by using the following steps:

[0095] Determine a fault response label when a sudden fault occurs in the target city power grid according to the loss interaction amount;

[0096] Determine the disaster loss fusion information when the target city power grid is affected by the fault according to the situation evolution characteristics;

[0097] The disaster response decision when the target city power grid is affected by an emergency event is determined by the fault response tag and the disaster fusion information.

[0098] In the specific implementation, first, the recovery status of the power grid is divided into several levels according to the size of the loss interaction and the difficulty of recovery of different areas of the power grid. For example, a high loss interaction may correspond to a high-priority response label, such as the "emergency repair" label, indicating that the area must be restored first and more resources may need to be allocated; a low loss interaction may correspond to a lower priority response label, such as the "conventional repair" label, indicating that the recovery of the area can be moderately delayed and more urgent areas are given priority. Then, based on the situation evolution characteristics of the power grid (such as the evolution trend of the post-disaster power grid operation, the recovery progress, the load recovery, etc.), various types of post-disaster data are fused to identify the bottlenecks, key issues and possible future evolution trends of power grid recovery. For example, the disaster loss correction information and load coordination characteristics can be combined with the power grid recovery situation to further deduce the overall performance of post-disaster recovery and the resource bottlenecks that may be faced. According to the situation evolution characteristics, the disaster loss information of different stages (such as initial loss, equipment damage, long-term recovery, etc.) is integrated to optimize the disaster loss assessment process. Finally, the fault response label is combined with the disaster loss fusion information to form a decision basis, so as to determine the best recovery strategy. For example: for areas marked as "emergency response", more power resources are dispatched and equipment repair is accelerated based on the disaster and loss fusion information; for low-priority areas, a more flexible recovery plan is formulated based on the overall progress and resource status of post-disaster recovery; optimization decisions are made, that is, based on the disaster and loss fusion information, the resource allocation and scheduling strategy for power grid recovery can be optimized to ensure that the recovery work is carried out efficiently and minimize disaster losses.

[0099] It should be noted that, in the present invention, the fault response label represents the characteristics of classifying and marking the power grid recovery status according to data such as loss interaction volume when the power grid encounters a sudden fault or disaster, reflecting the priority of recovery and the urgency of resource scheduling; disaster loss fusion information refers to information that performs a global assessment of the disaster loss status of the post-disaster power grid in multiple dimensions during the power grid recovery process; disaster loss response decision refers to the resource allocation strategy for post-disaster recovery of the power grid under the influence of an emergency.

[0100] In step S5, the disaster information of the target city power grid is synchronously matched in the target city distribution network power supply database through the disaster identification constraint and the disaster response decision, and a risk assessment report is generated.

[0101] In the specific implementation, first, through the disaster loss identification constraint (i.e., the analysis and restriction conditions of various potential losses in the process of post-disaster recovery of the power grid), combined with the disaster loss response decision (i.e., the specific plan adjusted according to the power grid recovery strategy), the actual performance of post-disaster losses in different regions and types can be accurately identified. By sorting and analyzing this information, specific disaster loss information can be generated for each power supply area, including power outage area, recovery progress, equipment damage, etc. Next, the synchronization matching process begins. The power grid's disaster loss information needs to be connected with the target city distribution network power supply database. The database contains real-time data such as load, equipment operation status, and historical fault records of each distribution area. Through the matching mechanism, the disaster loss information will be automatically updated to each related module according to the distribution area and equipment information in the database. For example, if a serious fault occurs in a certain area, the disaster loss information will update the distribution status of the area, including data such as power outage time, load recovery progress, and equipment repair status. Finally, based on these synchronized data, the system will generate a risk assessment report. The risk assessment report will list in detail the recovery status of the power grid under emergencies, loss assessment, and possible future risks. The risk assessment report provides grid managers with a clear basis for decision making.

[0102] It can be seen that in this application, firstly, by analyzing the power grid operation data, the problem of incomplete power grid historical data and poor timeliness in the traditional method is effectively solved, and the comprehensiveness and accuracy of the power grid fault data are ensured; then, by determining the situation evolution characteristics, the hierarchical classification processing of the power grid fault data is realized, and the evolution characteristics of the power grid outage under different fault scenarios are identified, which solves the problem of insufficient grasp of the fault evolution law in the prior art; secondly, by determining the disaster loss identification constraints, the problems of inaccurate risk identification and incomplete recovery indicators in the disaster recovery process are effectively solved, so that the power grid recovery process can be dynamically adjusted according to the disaster-causing factors, and refined risk assessment and constraint setting are realized; then, by determining the disaster loss response decision, the problem of delayed power grid recovery strategy and insufficient manual adjustment is solved, and by dynamically obtaining the disaster loss correction information and loss interaction, the real-time update and optimization of the recovery strategy is realized, and the response speed and recovery efficiency of the power grid are improved; finally, by matching the data in the power supply database of the target city distribution network, the automatic synchronous matching of disaster loss information and the efficient generation of risk assessment reports are realized, which solves the problems of untimely post-disaster recovery decisions and delayed assessment results, and provides scientific risk assessment and decision support for power grid managers, and improves the accuracy of disaster loss assessment.

[0103] In summary, the technical solution adopted in this application can combine the dynamic evolution characteristics of the damaged power grid during the post-disaster power grid recovery process to conduct a multi-dimensional and accurate assessment of the power grid's damage status, thereby improving the accuracy of the damage assessment.

[0104] Embodiment 2

[0105] This application provides a natural disaster power grid damage assessment system based on big data analysis, referring to Figure 4 As shown, this figure is a schematic diagram of an evaluation system according to this embodiment of the present application, and the evaluation system includes:

[0106] The data acquisition module 100 is used to acquire the power grid operation data of the historical power outage area of ​​the target city power grid through the big data resource library;

[0107] The situation analysis module 200 is used to classify the fault scenarios of the power grid operation data, obtain the fault attribute nodes corresponding to each fault scenario, and determine the situation evolution characteristics of the target city power grid outage under natural disasters and emergencies based on all the fault attribute nodes;

[0108] The disaster loss identification module 300 is used to determine the risk aggregation identifier of the target city power grid under different disaster-causing factors, determine the power loss index set when the target city power grid is restored from the disaster according to the risk aggregation identifier, and then determine the disaster loss identification constraint when the target city power grid is affected by the emergency through the power loss index set;

[0109] The disaster response module 400 is used to obtain the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid, determine the loss interaction amount when the target city power grid updates the restoration strategy according to the disaster correction information, and then determine the disaster response decision when the target city power grid is affected by the emergency event according to the loss interaction amount and the situation evolution characteristics;

[0110] The synchronous matching module 500 is used to synchronously match the disaster information of the target city power grid in the target city distribution network power supply database through the disaster identification constraint and the disaster response decision, and generate a risk assessment report.

[0111] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A natural disaster power grid damage assessment method based on big data analysis, characterized in that: The evaluation method comprises the following steps: Obtain the power grid operation data of the historical power outage areas of the target city power grid through the big data resource library; Classifying the power grid operation data by fault scenario, obtaining the fault attribute node corresponding to each fault scenario, and determining the state evolution characteristics of the target city power grid outage under natural disasters and emergencies according to all the fault attribute nodes; Determine the risk aggregation identifier of the target city power grid under different disaster-causing factors, determine the power loss index set of the target city power grid during disaster recovery according to the risk aggregation identifier, and then determine the disaster loss identification constraint of the target city power grid when it is affected by the emergency event through the power loss index set; Obtaining the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid, determining the loss interaction amount when the target city power grid updates the restoration strategy according to the disaster correction information, and then determining the disaster response decision when the target city power grid is affected by the emergency event according to the loss interaction amount and the situation evolution characteristics; The disaster loss information of the target city power grid is synchronously matched in the target city distribution network power supply database through the disaster loss identification constraint and the disaster loss response decision, and a risk assessment report is generated.

2. A natural disaster power grid damage assessment method based on big data analysis according to claim 1, characterized in that: The power grid operation data is classified into fault scenarios, and the fault attribute nodes corresponding to each fault scenario are obtained, specifically including: Obtain the fault distribution characteristics when a fault occurs in the target city power grid; Determining a fault scenario of a target city power grid according to the fault distribution characteristics; The power grid operation data is mapped to the fault scenarios to obtain a fault attribute node corresponding to each fault scenario.

3. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: The situation evolution characteristics of the target city power grid outage under natural disasters and emergencies are determined based on all fault attribute nodes, including: Determine the operation risk scale of the target city power grid when it is shut down according to the fault attribute node; The operation risk scale is used to determine the evolution characteristics of the target city power grid outage situation under natural disasters and emergencies.

4. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: Determine the risk aggregation identification of the target city power grid under different disaster factors, including: Obtain risk identification nodes of the target city power grid in the planned outage model; The risk aggregation identifier of the target city power grid under different disaster-causing factors is determined according to the risk identification node.

5. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: The damage identification constraints of the target city power grid when affected by an emergency event are determined by the power loss index set specifically include: Determining the fault difference amount of the target city power grid operation according to the power failure index set; Determine the dispatching discrete information during the emergency dispatching of the power grid through the fault difference amount; The damage identification constraints when the target city power grid is affected by an emergency event are determined according to the fault difference amount and the dispatch discrete information.

6. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: Determining the disaster response decision when the target city power grid is affected by an emergency event based on the loss interaction amount and the situation evolution characteristics specifically includes: Determine a fault response label when a sudden fault occurs in the target city power grid according to the loss interaction amount; Determine the disaster loss fusion information when the target city power grid is affected by the fault according to the situation evolution characteristics; The disaster response decision when the target city power grid is affected by an emergency event is determined by the fault response tag and the disaster fusion information.

7. A natural disaster power grid damage assessment method based on big data analysis according to claim 1, characterized in that: The grid operation data refers to a set of data generated during the operation of the power system that reflects the status, performance and operation events of the grid.

8. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: The disaster damage correction information refers to key information for adjusting and correcting disaster damage assessment during the power grid restoration process after the power grid encounters a natural disaster or emergency.

9. The method for natural disaster power grid damage assessment based on big data analysis according to claim 1 is characterized in that: The power failure index set represents a quantitative set of the degree of recovery of the power failure state during the power grid disaster recovery process.

10. A natural disaster power grid damage assessment system based on big data analysis, used to execute a natural disaster power grid damage assessment method based on big data analysis as claimed in any one of claims 1 to 9, characterized in that: The evaluation system comprises: A data acquisition module is used to obtain the power grid operation data of the historical power outage area of ​​the target city power grid through a big data resource library; A situation analysis module is used to classify the fault scenarios of the power grid operation data, obtain the fault attribute nodes corresponding to each fault scenario, and determine the situation evolution characteristics of the target city power grid outage under natural disasters and emergencies based on all fault attribute nodes; A disaster loss identification module is used to determine the risk aggregation identifier of the target city power grid under different disaster-causing factors, determine the power loss index set of the target city power grid during disaster recovery according to the risk aggregation identifier, and then determine the disaster loss identification constraints of the target city power grid when it is affected by the emergency event through the power loss index set; A disaster response module is used to obtain the disaster correction information synchronized in the operation database of the power outage area of ​​the target city power grid, determine the loss interaction amount when the target city power grid updates the restoration strategy according to the disaster correction information, and then determine the disaster response decision when the target city power grid is affected by the emergency event according to the loss interaction amount and the situation evolution characteristics; The synchronous matching module is used to synchronously match the disaster information of the target city power grid in the target city distribution network power supply database through the disaster identification constraint and the disaster response decision, and generate a risk assessment report.

Images