Method and device for evaluating extreme natural weather emergency mechanism

Abstract

The invention discloses an evaluation method and device for an extreme natural weather emergency mechanism, and relates to the field of power grid management. In the method, first prediction accuracy of extreme natural weather is determined according to a meteorological log in a to-be-evaluated emergency mechanism, and a first score is determined according to the first prediction accuracy; determining second prediction accuracy of the power grid fault according to a power grid state log in the to-be-evaluated emergency mechanism, and determining a second score according to the second prediction accuracy; determining a processing capability value according to a fault response speed and a power supply recovery speed for processing the fault event, and determining a third score according to the processing capability value and an influence level of extreme natural weather corresponding to the fault event; determining a fourth score according to the public opinion attitude about the processing condition of the fault event on the social media; and evaluating the emergency mechanism to be evaluated according to a total score obtained by performing weighted summation on the first score, the second score, the third score and the fourth score. By implementing the technical scheme provided by the invention, the effect of comprehensively evaluating the emergency mechanism is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of power grid management, and in particular to an evaluation method and device for an extreme natural weather emergency mechanism. Background Art

[0002] In the field of power grid management, with the frequent occurrence of natural disasters and increasingly severe extreme weather conditions, it is crucial to ensure the stable operation of the power system. As an important infrastructure in modern society, the safety of the power grid is not only related to the daily lives of residents and enterprises, but also directly affects the national economic and social stability. Therefore, establishing an effective emergency mechanism to deal with the stability of the power grid under extreme natural weather conditions is an important topic of current research.

[0003] At present, emergency mechanism evaluation usually adopts various methods, such as historical data analysis, fault prediction accuracy and other methods. These methods collect and analyze historical data to predict power grid failures under extreme natural weather conditions, and then take preventive measures. However, the existing emergency mechanism evaluation methods usually lack a comprehensive evaluation system to comprehensively consider the influence of multiple factors, which often leads to inaccurate evaluation of emergency mechanisms.

[0004] Therefore, how to construct an emergency mechanism evaluation method that can comprehensively consider multiple factors has become an urgent problem to be solved. Summary of the invention

[0005] This application provides an evaluation method and device for extreme natural weather emergency response mechanism, which can comprehensively reflect the performance of emergency response mechanism in various links such as prediction, early warning, processing and feedback. This comprehensive evaluation method provides a scientific decision-making basis for the power grid management platform, helps to continuously optimize and improve the emergency response mechanism, and enhance the overall ability and level of the power grid to cope with extreme natural weather.

[0006] In a first aspect of the present application, a method for evaluating an extreme natural weather emergency mechanism is provided, which is applied to a power grid management platform. The method comprises: Obtaining a meteorological log in the emergency mechanism to be evaluated, determining a first prediction accuracy of extreme natural weather according to the meteorological log, and determining a first score according to the first prediction accuracy, the meteorological log including daily actual meteorological data and meteorological forecast data within a preset time period, the extreme natural weather including a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; Obtaining a power grid status log in the emergency mechanism to be evaluated, determining a second prediction accuracy of a power grid failure according to the power grid status log, and determining a second score according to the second prediction accuracy, wherein the power grid status log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; Determine a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determine an impact level of the extreme natural weather corresponding to the fault event, and determine a third score according to the processing capability value and the impact level; Obtaining public opinion on social media regarding the handling of the fault event, and determining a fourth score according to the public opinion, wherein the public opinion includes positive, neutral, and negative; The first score, the second score, the third score and the fourth score are weightedly summed to obtain a total score of the emergency mechanism to be evaluated, and the emergency mechanism to be evaluated is evaluated according to the total score.

[0007] By adopting the above technical solution, the prediction accuracy of the emergency mechanism for extreme natural weather (first score) can be quantitatively evaluated by comparing the actual meteorological data in the meteorological log with the meteorological forecast data within the preset time period. This helps to timely discover and improve the prediction model, thereby improving the prediction accuracy of future extreme weather events and providing strong support for the power grid to take preventive measures in advance. Combining the meteorological log with the power grid status log, evaluating the second prediction accuracy of the power grid failure (second score) can reveal the effectiveness of the emergency mechanism in predicting power grid failures. This helps to identify potential power grid failure risks in advance, take preventive measures, and reduce the incidence and impact of power grid failures caused by extreme weather. By evaluating the fault response speed and power supply restoration speed to determine the processing capacity value (third score), and combining the impact level of extreme natural weather for comprehensive scoring, the efficiency and effectiveness of the emergency mechanism in responding to actual fault events can be intuitively reflected. This helps to discover and improve the weak links in the processing process, improve the overall fault handling capability, ensure that the power grid quickly restores power supply, and reduce social impact and economic losses. Obtaining public opinion attitudes on the handling of fault events from social media (fourth score) reflects the importance of public concern and satisfaction. This helps to understand the public's true views on the power grid's emergency response work, provides a social reference for the continuous improvement of the emergency mechanism, and enhances the public's trust and support for power grid companies. The total score of the emergency mechanism is obtained by weighted summation, and a comprehensive evaluation is conducted to fully reflect the performance of the emergency mechanism in various links such as prediction, early warning, processing and feedback. This comprehensive evaluation method provides a scientific decision-making basis for the power grid management platform, helps to continuously optimize and improve the emergency mechanism, and enhance the overall ability and level of the power grid to cope with extreme natural weather.

[0008] Optionally, determining the first prediction accuracy of extreme natural weather according to the meteorological log includes: A first sub-score is determined according to a first difference between the predicted time of occurrence of extreme natural weather and the actual time of occurrence of extreme natural weather, a second sub-score is determined according to a second difference between the predicted intensity of extreme natural weather and the actual intensity of extreme natural weather, and a third sub-score is determined according to the predicted type of extreme natural weather and the actual type of occurrence of extreme natural weather; The sub-prediction accuracy of a first target prediction event is determined according to the first sub-score, the second sub-score and the third sub-score, and the first prediction accuracy is obtained according to the average value of the sub-prediction accuracies of multiple first target prediction events.

[0009] By adopting the above technical solution, the prediction accuracy is refined into the evaluation of three dimensions of time, intensity and type (i.e., the first sub-score, the second sub-score and the third sub-score), which can more comprehensively and accurately measure the performance of the prediction model in extreme natural weather prediction. This refined evaluation helps to discover the advantages and disadvantages of the prediction model in different aspects and provide a more targeted direction for the optimization of the model. By calculating the time difference and the intensity difference to determine the sub-prediction accuracy, the deviation between the prediction and the actual can be quantified, thereby more intuitively reflecting the accuracy of the prediction. This helps to motivate the relevant team to continuously optimize the prediction model, reduce the prediction error, improve the prediction accuracy, and provide a more reliable basis for the early warning and prevention work of the power grid. By comparing the prediction type with the actual occurrence type to determine the third sub-score, the accuracy of the prediction model in identifying the extreme weather type can be ensured. This is crucial for the targeted prevention and response measures of the power grid, because different types of extreme weather may have different impacts on the power grid and different response strategies need to be adopted. By calculating the average value of the sub-prediction accuracy of multiple first target prediction events to obtain the first prediction accuracy, it is possible to achieve induction from individual to general, making the evaluation results more representative and credible. This comprehensive evaluation method helps to reduce the impact of the randomness of individual events on the overall evaluation results and improve the scientificity and fairness of the evaluation results.

[0010] Optionally, determining a second prediction accuracy of a power grid fault according to the power grid status log includes: Extracting a first feature related to a power grid failure from the meteorological log, the first feature including the occurrence time, intensity and duration of extreme natural weather, and extracting a second feature reflecting the power grid status from the power grid status log, the second feature including a load change trend, equipment health status and historical failure mode; Inputting the first feature and the second feature into a preset fault prediction model to obtain power grid fault prediction data, wherein the power grid fault prediction data includes a predicted occurrence time, a predicted occurrence location, and a predicted occurrence type; The second prediction accuracy is determined according to the power grid fault prediction data and the actual fault data.

[0011] By adopting the above technical solution, the first feature related to power grid failure (such as the occurrence time, intensity and duration of extreme natural weather) is extracted from the meteorological log, and the second feature reflecting the power grid status (such as load change trend, equipment health status and historical failure mode) is extracted from the power grid status log, so as to achieve multi-dimensional and comprehensive consideration of power grid failure prediction. This comprehensive feature analysis method can more comprehensively capture various factors that may cause power grid failure and improve the accuracy of prediction. The extracted first feature and second feature are input into the preset fault prediction model to obtain power grid failure prediction data. This prediction model based on big data and machine learning algorithm can automatically learn the complex relationship between historical fault data and current power grid status, so as to achieve accurate prediction of future power grid failures. This intelligent prediction method not only improves the prediction efficiency, but also reduces the influence of human intervention and subjective judgment. The power grid fault prediction data includes detailed information such as predicted occurrence time, predicted occurrence location and predicted occurrence type. This accurate prediction result helps power grid managers to understand the specific circumstances of possible failures in advance, formulate targeted preventive measures and emergency plans, so as to effectively reduce the impact of failures on power grid operation. By comparing the power grid failure prediction data with the actual failure data, the accuracy of the prediction model (i.e., the second prediction accuracy) can be quantitatively evaluated. This evaluation method helps to timely discover deficiencies and deviations in the prediction model and provide data support for the continuous optimization of the model. With the continuous optimization and iteration of the model, the prediction accuracy will continue to improve, providing more reliable protection for the safe and stable operation of the power grid.

[0012] Optionally, determining the second prediction accuracy according to the power grid fault prediction data and actual fault data includes: a fourth sub-score is determined according to a third difference between the predicted grid fault occurrence time and the actual grid fault occurrence time, a fifth sub-score is determined according to a fourth difference between the predicted grid fault location and the actual grid fault location, and a sixth sub-score is determined according to the predicted grid fault type and the actual grid fault type; The sub-prediction accuracy of the second target prediction event is determined according to the fourth sub-score, the fifth sub-score and the sixth sub-score, and the sub-prediction accuracy of multiple second target prediction events is weightedly summed according to the power grid fault prediction type to obtain the second prediction accuracy.

[0013] By adopting the above technical solution, the difference between the predicted occurrence time and the actual occurrence time of the power grid fault (the third difference), the difference between the predicted location and the actual location (the fourth difference), and the matching degree between the predicted type and the actual type are calculated respectively, so as to realize a multi-dimensional evaluation of the prediction accuracy. This evaluation method can more comprehensively reflect the performance of the prediction model in various aspects and provide a more detailed reference for the optimization of the model. By comparing key indicators such as the predicted time, location and type with the actual data, the accuracy of the prediction model in these aspects can be accurately quantified. This precise evaluation method helps to reduce the evaluation error and improve the reliability of the evaluation results. According to the power grid fault prediction type, the sub-prediction accuracies of multiple second target prediction events are weighted and summed to obtain the second prediction accuracy. This weighted summation method takes into account the differences in the impact of different fault types on the operation of the power grid, so that the evaluation results are more in line with the actual situation and enhance the pertinence and practicality of the evaluation. Through the multi-dimensional and precise evaluation of the prediction accuracy, the deficiencies and deviations of the prediction model in different aspects can be discovered in time. This helps relevant personnel to optimize the prediction model in a targeted manner, improve the prediction ability and accuracy of the model, and provide more reliable guarantees for the safe and stable operation of the power grid.

[0014] Optionally, determining the processing capability value according to a fault response speed and a power supply recovery speed for processing a fault event includes: Obtain the first time when the fault occurs, the second time when the maintenance personnel arrive at the scene to handle the fault, and the third time when the power supply is restored; A fault response speed is determined according to the first time and the second time, a power recovery speed is determined according to the second time and the third time, and a processing capacity value is obtained by weighted summing the fault response speed and the power recovery speed.

[0015] By adopting the above technical solution, key nodes such as the time of fault occurrence, the time when maintenance personnel arrive at the scene and the time of power restoration are quantified, which can accurately evaluate the response speed and recovery capability of the power grid management platform when handling fault events. This quantitative evaluation method makes the processing capacity value objective and comparable, and provides a scientific basis for evaluating the emergency response capability of the power grid management platform. Fault response speed is one of the important indicators for evaluating emergency response capability. By calculating the time difference from the occurrence of the fault to the arrival of the maintenance personnel at the scene, the response speed of the power grid management platform after receiving the fault report can be intuitively reflected. This evaluation method helps to motivate relevant personnel to improve emergency response efficiency, reduce fault handling time, and reduce the impact of faults on power grid operation. Power restoration speed is an important indicator for evaluating the recovery capability of the power grid management platform. By calculating the time difference from the arrival of maintenance personnel at the scene to the restoration of power supply, the recovery speed and efficiency of the power grid management platform after fault handling can be evaluated. This evaluation method helps the power grid management platform optimize the fault handling process, improve the speed of power restoration, and ensure that the power grid resumes normal operation as soon as possible. By weighted summing the fault response speed and power restoration speed, a comprehensive processing capacity value can be obtained. This value can fully reflect the overall capability and level of the power grid management platform when handling fault events. This comprehensive assessment method helps the power grid management platform to fully understand its own emergency response capabilities, identify existing problems and deficiencies, and provide direction for subsequent improvements and enhancements.

[0016] Optionally, the determining the impact level of the extreme natural weather corresponding to the fault event, and determining a third score according to the processing capability value and the impact level includes: Determine the type and intensity of the extreme natural weather corresponding to the fault event, match a preset impact level according to the type and intensity, and multiply the preset impact level by the processing capacity value to obtain a third score.

[0017] By adopting the above technical solution, determining the type and intensity of extreme natural weather corresponding to the fault event, and matching the preset impact level, an accurate assessment of the impact of weather factors on power grid faults can be achieved. This evaluation method takes into account the different impacts of different weather types and intensities on power grid operation, making the evaluation results more in line with the actual situation. The third score is obtained by multiplying the processing capacity value by the impact level of extreme natural weather. This step comprehensively considers the actual ability of the power grid management platform in handling fault events and the impact of external weather factors. This comprehensive evaluation method helps to more comprehensively understand the performance of the power grid management platform in dealing with extreme weather conditions and provide strong support for subsequent improvements and enhancements. By evaluating the third scores corresponding to different fault events, the power grid management platform can understand the extreme weather types that have a greater impact on the operation of the power grid, as well as its own ability to handle these faults. This helps the power grid management platform optimize resource allocation, invest more resources and energy in the prevention and response of extreme weather types that have a greater impact on the operation of the power grid, and improve the overall response capability.

[0018] Optionally, obtaining public opinion on social media regarding the handling of the fault event and determining a fourth score according to the public opinion includes: Using crawler technology or social media interfaces, crawling social media data containing keywords of the fault event, the keywords include fault name, location, time, power supply company name, emergency repair, and recovery, and the social media data includes posts, comments, and forwarding; Performing sentiment analysis on the social media data to identify sentiment tendency and intensity, wherein the sentiment tendency includes positive, negative and neutral, and the intensity includes praise, satisfaction, recognition, criticism, dissatisfaction and complaint; The public opinion score of each content in the social media data is determined according to the emotional tendency and intensity, and the fourth score is obtained by weighted averaging all the content.

[0019] By adopting the above technical solution, using crawler technology or social media interface, social media data related to fault events, including posts, comments, forwarding, etc., can be quickly captured. This method can reflect the public's views and attitudes on fault events and their handling in real time, providing valuable feedback information for the power grid management platform. By performing sentiment analysis on social media data, the public's sentiment tendency and intensity can be accurately identified. This analysis is not limited to simple positive, negative and neutral classification, but is further refined into specific emotions such as praise, satisfaction, recognition, criticism, dissatisfaction, and complaints. This detailed sentiment analysis helps the power grid management platform to have a deeper understanding of the public's real feelings and needs, and improve the insight and accuracy of decision-making. The public opinion score of each social media content is determined according to the sentiment tendency and intensity, and the fourth score is obtained by weighted average of all content. This quantitative evaluation method makes public opinion attitudes measurable and comparable, providing an objective evaluation standard for the power grid management platform. By comparing the fourth scores of different fault events, the public's satisfaction and recognition of the handling of different events can be intuitively understood. The process of determining the fourth score provides a clear direction for the power grid management platform to improve service quality. By analyzing the positive and negative factors in public opinion, the power grid management platform can identify the problems and deficiencies in its service process and formulate targeted improvement measures accordingly. This will help improve the service level of the power grid management platform and enhance the public's trust and satisfaction with it.

[0020] In a second aspect of the present application, an evaluation system for an extreme natural weather emergency mechanism is provided, comprising a weather module, a power grid module, a processing module, a public opinion module and an evaluation module, wherein: a weather module configured to obtain a meteorological log in the emergency mechanism to be evaluated, determine a first prediction accuracy of extreme natural weather according to the meteorological log, and determine a first score according to the first prediction accuracy, wherein the meteorological log includes daily actual meteorological data and meteorological forecast data within a preset time period, and the extreme natural weather includes a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; a power grid module configured to obtain a power grid state log in the emergency mechanism to be evaluated, determine a second prediction accuracy of a power grid failure according to the power grid state log, and determine a second score according to the second prediction accuracy, wherein the power grid state log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; a processing module configured to determine a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determine an impact level of extreme natural weather corresponding to the fault event, and determine a third score according to the processing capability value and the impact level; A public opinion module configured to obtain public opinion attitudes on social media regarding the handling of the fault event, and determine a fourth score according to the public opinion attitudes, wherein the public opinion attitudes include positive, neutral, and negative; An evaluation module is configured to perform a weighted summation of the first score, the second score, the third score and the fourth score to obtain a total score of the emergency mechanism to be evaluated, and evaluate the emergency mechanism to be evaluated according to the total score.

[0021] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0023] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By acquiring and analyzing the actual meteorological data and forecast data in the meteorological log, the accuracy of the first forecast of extreme natural weather can be quantified and a first score can be given accordingly. This method helps to understand the performance of the emergency mechanism in predicting extreme weather and provide a scientific basis for making preparations in advance, thereby reducing the losses caused by inaccurate forecasts; 2. Analyzing the power grid status logs, especially the power grid fault prediction data, in combination with the meteorological logs can evaluate the accuracy of power grid fault prediction under extreme weather conditions, which helps to timely discover potential power grid fault risks, determine the processing capacity value by quantifying the fault response speed and power supply restoration speed, and evaluate the efficiency of the emergency mechanism in combination with the impact level of extreme natural weather. This can not only intuitively reflect the effectiveness of the emergency mechanism in handling specific fault events, but also provide data support for further optimizing the emergency response process; 3. Collect public opinions on the handling of fault events through social media, and determine the fourth score based on this, making the evaluation process more comprehensive and objective. This method helps to understand the public's satisfaction and trust in power grid emergency management, and also encourages power grid management departments to respond more actively to public concerns and improve service quality; 4. By weighting and summing the scores of the four aspects, the total score of the emergency mechanism is obtained, thus achieving a comprehensive and systematic evaluation of the emergency mechanism. This comprehensive evaluation method takes into account both technical factors (such as prediction accuracy and processing capabilities) and social factors (such as public opinion attitudes), which helps to discover the strengths and weaknesses of the emergency mechanism and provide a clear direction for subsequent improvement and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the evaluation method of the extreme natural weather emergency mechanism disclosed in the embodiment of the present application; Figure 2 It is a module schematic diagram of the evaluation system of the extreme natural weather emergency mechanism disclosed in the embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0025] Explanation of the reference numerals: 201, weather module; 202, power grid module; 203, processing module; 204, public opinion module; 205, evaluation module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0026] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0027] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0028] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0029] This embodiment discloses an evaluation method for an extreme natural weather emergency mechanism, which is applied to a power grid management platform. Figure 1 is a flow chart of the evaluation method of the extreme natural weather emergency mechanism disclosed in the embodiment of the present application, such as Figure 1 As shown, the evaluation method includes the following steps: S110, obtaining a meteorological log in the emergency mechanism to be evaluated, determining a first prediction accuracy of extreme natural weather according to the meteorological log, and determining a first score according to the first prediction accuracy, wherein the meteorological log includes daily actual meteorological data and meteorological forecast data within a preset time period, and the extreme natural weather includes a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; Obtain the meteorological logs of the area where the emergency mechanism to be evaluated is implemented. Daily actual meteorological data refers to the daily meteorological conditions recorded in actual observations, including but not limited to specific values ​​such as temperature, precipitation, and wind speed. These data are the basis for evaluating the accuracy of the forecast; the meteorological forecast data within the preset time period refers to the data for predicting the meteorological conditions within the future preset time period based on various meteorological models and technical means. These data are used to compare with the actual data to evaluate the accuracy of the forecast. In order to evaluate the accuracy of the forecast, it is necessary to clearly define the meteorological conditions that are considered extreme natural weather. According to the description, these meteorological conditions include: the temperature is higher than the first temperature preset value (such as 37°C) or lower than the second temperature preset value (such as 0°C), the precipitation is greater than the first precipitation preset value (such as 20 mm / hour) or less than the second precipitation preset value (such as 20 days without precipitation), the wind speed is greater than the wind speed preset value (such as level 12), and a specific type of precipitation (such as freezing rain, hail). Using the collected meteorological logs, especially comparing the actual meteorological data with the forecast data, the accuracy of the forecast of extreme natural weather can be evaluated. Specifically, the forecast data is compared with the actual daily meteorological data item by item, with special attention paid to whether the occurrence of extreme weather events is accurately predicted. Based on the comparison results, the ratio of the number of accurate predictions to the total number of predictions is calculated as a quantitative indicator of the first prediction accuracy. Based on the first prediction accuracy, a score (first score) can be further determined to quantitatively evaluate the performance of the emergency mechanism in extreme natural weather prediction. The determination of this score may be based on a preset scoring standard, for example: if the prediction accuracy is higher than a certain threshold, a higher score is given; if the prediction accuracy is lower, the score is reduced accordingly. The specific calculation method and threshold of the score can be set according to the actual situation and evaluation needs.

[0030] Optionally, determining the first prediction accuracy of extreme natural weather according to the meteorological log includes: A first sub-score is determined according to a first difference between the predicted time of occurrence of extreme natural weather and the actual time of occurrence of extreme natural weather, a second sub-score is determined according to a second difference between the predicted intensity of extreme natural weather and the actual intensity of extreme natural weather, and a third sub-score is determined according to the predicted type of extreme natural weather and the actual type of occurrence of extreme natural weather; The sub-prediction accuracy of a first target prediction event is determined according to the first sub-score, the second sub-score and the third sub-score, and the first prediction accuracy is obtained according to the average value of the sub-prediction accuracies of multiple first target prediction events.

[0031] Suppose you are evaluating the accuracy of forecasts for extreme weather events in a region over the next week. The region's weather log records actual weather events as well as forecasted weather events. You are interested in three main types of extreme weather: heat, heavy rain, and high winds.

[0032] Forecast 1 (high temperature): The temperature was predicted to reach 38°C on the afternoon of the third day, but it actually occurred on the morning of the fourth day, with a temperature of 39°C. Forecast 2 (heavy rain): Heavy rain was predicted to occur on the evening of the fifth day, and it actually occurred on the evening of the fifth day, but the precipitation was predicted to be 100mm, and it actually occurred on the afternoon of the sixth day, but the wind actually reached 9. For the high temperature forecast, the time difference was one and a half days (from the afternoon of the third day to the morning of the fourth day), which may result in a lower sub-score, such as 60 points (out of 100 points) according to the scoring criteria. The time difference between the heavy rain and strong wind forecasts is 0, so the sub-scores for both forecasts in terms of time are full marks, i.e. 100 points. In the high temperature forecast, the actual temperature is 1°C higher than the forecast, which may result in a sub-score close to the full score, such as 95 points. In the heavy rain forecast, the actual precipitation is 20mm more than the forecast, which may result in a slightly lower sub-score, such as 85 points. In the high wind forecast, the actual wind force is one level higher than the forecast, which may also result in a lower sub-score, such as 75 points. All forecasts correctly predicted the type of extreme weather (high temperature, heavy rain, high winds), so the sub-scores for type matching for all three forecasts are full marks, that is, 100 points. For each predicted event, a comprehensive sub-forecast accuracy score can be calculated. For example, for the high temperature forecast: Sub-forecast accuracy = (time difference sub-score + intensity difference sub-score + type matching sub-score) / 3 = (60 + 95 + 100) / 3 = 85. Similarly, the sub-forecast accuracy of heavy rain and strong wind forecasts can be calculated. Finally, the average of the sub-forecast accuracy of all first target forecast events is taken to get the first forecast accuracy: First forecast accuracy = (high temperature sub-forecast accuracy + heavy rain sub-forecast accuracy + strong wind sub-forecast accuracy) / 3. Assuming that the sub-forecast accuracy of heavy rain and strong wind is 88 and 80 respectively (the assumed values ​​calculated based on the above logic), the first forecast accuracy = (85 + 88 + 80) / 3 = 84.33.

[0033] The prediction accuracy of extreme natural weather is comprehensively evaluated through three different dimensions (time, intensity and type). This multi-dimensional evaluation can more comprehensively reflect the performance of the prediction, because considering only one dimension (such as time or intensity) may ignore other key information, resulting in biased evaluation results. By converting the difference between the prediction and the actual (time difference, intensity difference) and the type matching situation into specific sub-scores, the embodiment of the present invention realizes the refined quantification of the prediction accuracy. This quantification method not only makes the evaluation results more intuitive and comparable, but also provides a specific improvement direction for the subsequent optimization of the prediction model. By calculating the average value of the sub-prediction accuracy of multiple first target prediction events to obtain the first prediction accuracy, the embodiment of the present invention can effectively reduce the impact of individual extreme or abnormal events on the overall evaluation results, making the evaluation results more stable and reliable. By quantitatively evaluating the prediction accuracy, the performance of the prediction model in different aspects can be clearly seen. This provides valuable feedback to model developers, helps them discover the shortcomings of the model and make targeted optimizations, and further improves the accuracy and reliability of the prediction model.

[0034] S120, obtaining a power grid status log in the emergency mechanism to be evaluated, determining a second prediction accuracy of a power grid failure according to the power grid status log, and determining a second score according to the second prediction accuracy, wherein the power grid status log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; The power grid status log is a detailed record generated during the daily operation of the power grid, including but not limited to key parameters such as voltage, current, power factor, load conditions, and equipment operating status. In addition, it also includes power grid fault prediction data generated in combination with meteorological logs (such as temperature, humidity, wind speed, rainfall, etc.). These data are usually automatically generated by the monitoring system and prediction model of the power grid and stored in the database. For evaluation, relevant log information needs to be extracted from these data sources. The daily working data of the power grid is analyzed in detail to understand the operating status, load changes, equipment health status, etc. of the power grid in different time periods. These data reflect the actual operation of the power grid and potential risk points. The power grid fault prediction data generated based on the meteorological log is compared with the actual working data. The prediction data usually includes the prediction of the time, location, type and severity of possible power grid failures in the future.

[0035] Evaluate the difference between the predicted time of the fault and the actual time of the fault, and calculate the time accuracy index. Compare the type and location of the predicted fault with the type and location of the actual fault, and evaluate the accuracy of the prediction in terms of type and location. If the prediction data includes a prediction of the severity of the fault, it is also necessary to evaluate the consistency between the predicted severity and the actual severity. Combine the above evaluation results to form a comprehensive evaluation of the accuracy of power grid fault prediction. Based on the evaluation results, formulate a scoring standard to convert the prediction accuracy into a specific score. The scoring standard can be weighted based on multiple dimensions such as time, type, location, severity, etc. According to the scoring standard, calculate the prediction accuracy score of the power grid fault (i.e., the second score). The second score reflects the effectiveness of the emergency mechanism in predicting power grid failures.

[0036] Optionally, determining a second prediction accuracy of a power grid fault according to the power grid status log includes: Extracting a first feature related to a power grid failure from the meteorological log, the first feature including the occurrence time, intensity and duration of extreme natural weather, and extracting a second feature reflecting the power grid status from the power grid status log, the second feature including a load change trend, equipment health status and historical failure mode; Inputting the first feature and the second feature into a preset fault prediction model to obtain power grid fault prediction data, wherein the power grid fault prediction data includes a predicted occurrence time, a predicted occurrence location, and a predicted occurrence type; The second prediction accuracy is determined according to the power grid fault prediction data and the actual fault data.

[0037] Identify and record the occurrence time of extreme natural weather (such as storms, lightning, high temperatures, etc.) that may affect the power grid from the meteorological log. These time points are key references for predicting the possible occurrence of power grid failures. In addition to the time of occurrence, the intensity and duration of these extreme natural weather events also need to be recorded. This information is critical for assessing their potential impact on the power grid. Analyze the load data of the power grid and identify the daily, weekly, monthly or seasonal change trends of the load. A sharp change in load may be a precursor to power grid failure. Extract health information of key power grid equipment through monitoring systems and maintenance records. Aging, damage or abnormal operation of equipment may increase the risk of power grid failure. Review past power grid failure records and analyze the type, location, cause and correlation between the failures. These historical data help identify the vulnerable links and common failure modes of the power grid. The extracted first and second features are used as inputs and input into the preset power grid failure prediction model. This model may be a complex machine learning or deep learning model that has been trained and optimized with a large amount of historical data. The model outputs prediction data of power grid failures based on the input features. These data include the predicted time of occurrence, the predicted location of occurrence and the predicted type of occurrence. The predicted time can be a specific date and time point, or a time period; the predicted location can be a specific device or area in the power grid; the predicted type refers to the specific nature of the fault (such as short circuit, open circuit, overload, etc.). Comparing predicted data with actual data: Compare the power grid fault prediction data with the actual fault data. This includes comparing the predicted occurrence time with the actual occurrence time, the predicted occurrence location with the actual occurrence location, and the predicted occurrence type with the actual occurrence type. Based on the comparison results, a series of accuracy indicators are calculated to evaluate the accuracy of the prediction. These indicators may include the mean or median of the time difference, the location matching rate, the type matching rate, etc. In addition, more complex evaluation methods, such as confusion matrix, ROC curve, etc., can be used to comprehensively evaluate the performance of the prediction model. Combine various accuracy indicators to form a comprehensive evaluation of the accuracy of power grid fault prediction. This evaluation result reflects the effectiveness and reliability of the preset fault prediction model in power grid fault prediction.

[0038] By combining extreme natural weather information in the meteorological log (such as the time of occurrence, intensity, and duration) with specific status information in the power grid status log (such as load change trends, equipment health, and historical failure modes), various factors that affect power grid failures can be considered more comprehensively. This multi-source data fusion method helps to improve the accuracy and reliability of power grid failure prediction. Based on the predicted fault information, power grid operators can more effectively allocate human, material, and financial resources to ensure that they can respond quickly and restore power supply when a fault occurs. This helps to reduce fault handling costs and improve resource utilization efficiency. Accurate fault prediction data provides strong support for power grid planning, construction, and operation. Based on this data, power grid managers can make decisions more scientifically, optimize the power grid structure, and improve the overall performance and disaster resistance of the power grid.

[0039] Optionally, determining the second prediction accuracy according to the power grid fault prediction data and actual fault data includes: a fourth sub-score is determined according to a third difference between the predicted grid fault occurrence time and the actual grid fault occurrence time, a fifth sub-score is determined according to a fourth difference between the predicted grid fault location and the actual grid fault location, and a sixth sub-score is determined according to the predicted grid fault type and the actual grid fault type; The sub-prediction accuracy of the second target prediction event is determined according to the fourth sub-score, the fifth sub-score and the sixth sub-score, and the sub-prediction accuracy of multiple second target prediction events is weightedly summed according to the power grid fault prediction type to obtain the second prediction accuracy.

[0040] The difference between the predicted time of occurrence of the power grid fault and the actual time of occurrence of the power grid fault is calculated, and this difference is called the third difference. It reflects the accuracy of the prediction in time. According to the size of the third difference, a scoring standard can be set to determine the fourth sub-score. Generally speaking, the smaller the difference, the more accurate the predicted time and the higher the fourth sub-score. The difference or distance between the predicted location of the power grid fault and the actual location of the power grid fault is calculated, and this difference is called the fourth difference. It reflects the accuracy of the prediction in spatial position. Similarly, according to the size of the fourth difference, a scoring standard can be set to determine the fifth sub-score. The smaller the difference, the more accurate the predicted location and the higher the fifth sub-score. For the comparison between the predicted type of the power grid fault and the actual type of the power grid fault, if the two are completely consistent, the highest sixth sub-score can be given; if they are partially consistent or inconsistent, the corresponding score is given according to the specific situation. This score reflects the accuracy of the prediction in type identification. Based on the sub-scores of the above three dimensions (the fourth sub-score, the fifth sub-score and the sixth sub-score), the sub-prediction accuracy of each second target prediction event can be calculated. This is usually done by weighted averaging the three sub-scores or according to other predetermined algorithms. Since different types of power grid faults may have different degrees of impact and importance on power grid operation, when calculating the second prediction accuracy, the sub-prediction accuracies of multiple second target prediction events can be weighted and summed according to the power grid fault prediction type. This ensures that those fault types that have a greater impact on power grid operation receive more attention in the prediction accuracy assessment. The result obtained by weighted summation is the second prediction accuracy, which fully reflects the comprehensive performance of the power grid fault prediction model in the three dimensions of prediction time, location and type.

[0041] By calculating the difference between the three dimensions of time, location and type, the sub-scores (the fourth sub-score, the fifth sub-score and the sixth sub-score) are determined. This method can more accurately reflect the deviation between the prediction and the actual situation. The evaluation of each dimension is carried out independently, which helps to identify the performance of the prediction model in different aspects, so as to carry out more targeted optimization. Considering that different fault types may have different degrees of impact on the operation of the power grid, the weighted summation method is used to summarize the sub-prediction accuracy of multiple second target prediction events, which can more reasonably reflect the overall prediction accuracy. This weighting method can be adjusted according to the actual situation to better meet the needs of power grid operation. By comparing the predicted data with the actual data and calculating the second prediction accuracy, the shortcomings of the prediction model can be discovered in time. This feedback mechanism helps model developers to continuously optimize and improve the model and improve the reliability and accuracy of the prediction model.

[0042] S130, determining a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determining an impact level of the extreme natural weather corresponding to the fault event, and determining a third score according to the processing capability value and the impact level; Fault response speed refers to the time interval from the occurrence of a fault event to the time when the grid operator begins to take countermeasures. A faster response speed usually means higher emergency response capabilities and shorter power outage time. Therefore, fault response speed is an important indicator for measuring the grid's processing capacity. Power restoration speed refers to the time from the start of a fault to the restoration of normal power supply to the grid. The power restoration speed not only reflects the maintenance efficiency of the grid operator, but is also directly related to the user's power experience. A faster power restoration speed can reduce the impact of power outages on users and improve user satisfaction. Based on the fault response speed and power restoration speed, a processing capacity value can be calculated. This value can be a weighted sum of the two speeds or a combination of them converted into scores. The higher the processing capacity value, the better the grid performs in responding to fault events. Extreme natural weather includes but is not limited to extreme meteorological conditions such as strong winds, heavy rains, lightning, hail, and heavy snow. These weather conditions may cause damage to grid facilities and lead to fault events. According to the intensity, duration, and actual impact of extreme natural weather on grid facilities, it can be divided into different impact levels. The higher the impact level, the greater the destructive power of extreme natural weather on the power grid, and the higher the possibility and severity of the fault event. After determining the processing capacity value and the impact level of extreme natural weather, these two factors can be combined for comprehensive evaluation. One possible approach is to regard the processing capacity value as a positive factor of the power grid's coping ability and the impact level as a negative factor. Through some algorithm (such as weighted summation, product method, etc.), these two factors are converted into a comprehensive evaluation score, namely the third score. The third score reflects the overall performance of the power grid in coping with fault events under specific extreme natural weather conditions. It not only takes into account the processing capacity of the power grid itself, but also the impact of external weather conditions on the operation of the power grid. Therefore, the third score is a more comprehensive and objective evaluation indicator that can provide valuable reference information for power grid managers.

[0043] Optionally, determining the processing capability value according to a fault response speed and a power supply recovery speed for processing a fault event includes: Obtain the first time when the fault occurs, the second time when the maintenance personnel arrive at the scene to handle the fault, and the third time when the power supply is restored; A fault response speed is determined according to the first time and the second time, a power recovery speed is determined according to the second time and the third time, and a processing capacity value is obtained by weighted summing the fault response speed and the power recovery speed.

[0044] First time (fault occurrence time): This refers to the time point when the fault event actually occurs. This time point marks the beginning of the fault event and is the starting point for all subsequent processing activities. Second time (maintenance personnel arrival time): This refers to the time point when the power grid maintenance personnel arrive at the fault site and start to handle the fault. The interval between this time point and the fault occurrence time reflects the response speed of the power grid operator to the fault event. Third time (power supply restoration time): This refers to the time point when the power grid resumes normal power supply. The interval between this time point and the maintenance personnel arrival time reflects the efficiency of the maintenance work and the speed of power supply restoration. Fault response speed: It can usually be obtained by calculating the difference between the second time and the first time. The smaller this difference is, the faster the power grid operator responds to the fault event. Power supply restoration speed: It can be obtained by calculating the difference between the third time and the second time. The smaller this difference is, the higher the efficiency of the maintenance work is and the faster the power supply is restored.

[0045] Since the importance of fault response speed and power supply restoration speed in evaluating the processing capacity of the power grid may be different, they need to be weighted and summed. The distribution of weights can be determined according to the actual situation and the needs of the power grid operator. For example, if the fault response speed is considered more important, it can be given a higher weight. The result of the weighted sum is the processing capacity value. The processing capacity value comprehensively reflects the overall performance of the power grid in responding to fault events, including the response speed and the power supply restoration speed. The higher the processing capacity value, the stronger the processing capacity of the power grid. By comparing the processing capacity values ​​between different time periods, different regions or different power grids, the operating efficiency and emergency response capabilities of the power grid can be evaluated. The processing capacity value can be used as a reference for power grid managers to formulate improvement measures, optimize resource allocation and improve emergency response capabilities. At the same time, it can also be used as an important indicator for users to evaluate the quality of power grid services.

[0046] By quantifying the key time nodes in the fault handling process, such as the time of fault occurrence, the time of arrival of maintenance personnel, and the time of power restoration, the evaluation of the power grid's processing capacity is made more objective and specific. This quantitative evaluation method helps to reduce the influence of subjective judgment and improve the accuracy and reliability of the evaluation results. At the same time, the two dimensions of fault response speed and power restoration speed are considered, which reflect the emergency response capability and maintenance and recovery capability of the power grid after the fault occurs. By comprehensively considering these two aspects, the overall performance of the power grid in fault handling can be more comprehensively evaluated. The fault response speed and power restoration speed are weighted and summed to obtain the processing capacity value. This weighted summation method can adjust the weights according to actual conditions to reflect the importance of different dimensions in the evaluation. For example, in some cases, the power restoration speed may be more important, so it can be given a higher weight. This flexibility makes the evaluation results more in line with actual needs. By evaluating the processing capacity value of the power grid regularly or irregularly, the power grid manager can timely understand the performance of the power grid in fault handling and formulate improvement measures accordingly. This helps to improve the management efficiency of the power grid and reduce the impact of faults on the operation of the power grid.

[0047] Optionally, the determining the impact level of the extreme natural weather corresponding to the fault event, and determining a third score according to the processing capability value and the impact level includes: Determine the type and intensity of the extreme natural weather corresponding to the fault event, match a preset impact level according to the type and intensity, and multiply the preset impact level by the processing capacity value to obtain a third score.

[0048] According to the experience of the power grid operator or industry standards, there is a preset extreme weather impact level table. In this table, different types of extreme weather and their different intensities correspond to different impact levels. For example, a strong storm with high intensity may be assigned a higher impact level, such as level 5 (assuming that the impact level is represented by numbers 1-5, 5 is the highest). According to the method described above, the first time when the fault occurs, the second time when the maintenance personnel arrive at the scene, and the third time when the power supply is restored are obtained. By calculation, the fault response speed is X hours / minutes and the power supply restoration speed is Y hours / minutes. Then, the two speeds are weighted and summed (assuming that the weight of the fault response speed is 0.6 and the weight of the power supply restoration speed is 0.4) to obtain the processing capacity value Z. The impact level is converted into a coefficient (for example, the corresponding coefficient of level 5 is 1.5, which can be set according to actual conditions), and then this coefficient is multiplied by the processing capacity value to obtain the third score. The third score = processing capacity value Z*impact level coefficient (1.5). For example, the processing capacity value Z is 80 (this is an assumed value, which will be obtained based on specific data in actual calculations), and the impact level coefficient is 1.5, then: the third score = 80*1.5 = 120.

[0049] The embodiment of the present invention not only considers the processing capacity of the power grid itself (i.e., the fault response speed and the power supply restoration speed), but also considers the impact of external extreme natural weather. By combining the two, the overall performance of the power grid under specific conditions can be more comprehensively evaluated. By determining the type and intensity of the extreme natural weather corresponding to the fault event and matching the preset impact level, the impact of external factors can be accurately quantified. This quantification method helps to reduce the uncertainty of subjective judgment and improve the accuracy and objectivity of the evaluation results. The preset impact level is multiplied by the processing capacity value to obtain the third score. This weighted multiplication method can reflect the degree of influence of external factors (extreme natural weather) on the processing capacity of the power grid. The higher the impact level, the greater the negative impact of the external factors on the power grid, so the result after multiplication will also be reduced accordingly, thereby more accurately reflecting the actual performance of the power grid under specific conditions. By calculating the third score, power grid managers can more clearly understand the performance of the power grid in dealing with different extreme natural weather conditions. This helps them make more scientific and reasonable decisions, such as optimizing resource allocation, strengthening equipment maintenance, and improving emergency response capabilities, so as to better deal with possible fault events in the future. The calculation result of the third score provides a quantifiable evaluation indicator for power grid managers to evaluate the performance of the power grid in fault handling. By calculating the third score regularly or irregularly and comparing it with historical data, grid managers can promptly identify the grid's shortcomings in coping with extreme natural weather and take corresponding improvement measures to continuously improve the reliability and stability of the grid.

[0050] S140, obtaining public opinion on the handling of the fault event on social media, and determining a fourth score according to the public opinion, where the public opinion includes positive, neutral, and negative; Social media platforms, such as Weibo, Twitter, and Facebook, are important channels for the public to express their opinions and attitudes. By monitoring and collecting posts, comments, and shared content related to the handling of specific fault events on these platforms, a large amount of user feedback can be obtained. The collected social media data needs to be subjected to sentiment analysis to determine the public's public opinion attitude towards the handling of fault events. These attitudes can usually be divided into positive, negative, and neutral. Positive attitude: It means that the public is satisfied with the handling results of the fault event, believes that the handling is timely and effective, or appreciates the responsibility and professionalism shown in the handling process. Neutral attitude: It means that the public has no particular emotional tendency towards the handling of the fault event, which may be due to the lack of sufficient information to form a clear opinion, or believes that the handling is neither particularly good nor particularly bad. Negative attitude: It means that the public is dissatisfied with the handling results of the fault event, believes that the handling is not timely and ineffective, or criticizes and blames the problems that occurred during the handling process. Based on the classified public opinion attitudes, an evaluation score (the fourth score) can be calculated to quantify the public's overall evaluation of the handling of the fault event. The calculation method of this score can be designed according to actual needs, but it usually takes into account the number, proportion and weight of various attitude types. For example, positive attitudes can be given higher scores, negative attitudes can be given lower scores, and neutral attitudes can be appropriately adjusted according to their proportion in the whole. Through weighted summation or other statistical methods, a fourth score that comprehensively reflects the attitude of public opinion can be obtained. The calculation results of the fourth score can provide valuable feedback for power grid managers. By understanding the public's views and attitudes on the handling of fault events, managers can evaluate whether their response measures are appropriate and whether there are areas that need improvement. At the same time, this information can also serve as an important reference for formulating similar event handling strategies in the future.

[0051] Optionally, obtaining public opinion on social media regarding the handling of the fault event and determining a fourth score according to the public opinion includes: Using crawler technology or social media interfaces, crawling social media data containing keywords of the fault event, the keywords include fault name, location, time, power supply company name, emergency repair, and recovery, and the social media data includes posts, comments, and forwarding; Performing sentiment analysis on the social media data to identify sentiment tendency and intensity, wherein the sentiment tendency includes positive, negative and neutral, and the intensity includes praise, satisfaction, recognition, criticism, dissatisfaction and complaint; The public opinion score of each content in the social media data is determined according to the emotional tendency and intensity, and the fourth score is obtained by weighted averaging all the content.

[0052] Identify a series of keywords related to the fault event, such as "power failure in XX city", "power outage on XX road", "emergency repair by XX power supply company", "power supply restoration", etc. Use crawler technology or social media interface: select appropriate tools (such as Python's Scrapy framework, Twitter API, Weibo API, etc.) to search and crawl relevant posts, comments and forwarding data on social media platforms (such as Weibo, WeChat, Douyin, Twitter, etc.) according to the above keywords. Eliminate content that is not related to the fault event, such as advertisements, spam, etc. Perform deduplication, denoising, conversion to lowercase (if applicable), and unified encoding on the captured text. Select appropriate sentiment analysis tools: Use natural language processing (NLP) libraries (such as NLTK, jieba word segmentation + SnowNLP, Baidu AI Open Platform, etc.) to perform sentiment analysis on the pre-processed text. If positive sentiment words and phrases such as praise, satisfaction, and recognition are identified, such as "rapid emergency repair", "good service", and "timely power supply restoration", the content is classified as positive attitude. If negative sentiment words and phrases such as criticism, dissatisfaction, and complaints are identified, such as "slow processing", "poor service", and "power outage time is too long", the content is classified as negative. If it is not clear whether it is positive or negative, or the content itself is relatively objective and has no sentiment, the content is classified as neutral. At the same time, the sentiment intensity of each piece of content is evaluated. For example, some praise may be just a light touch, while some complaints may be very strong. According to the sentiment tendency and intensity, a public opinion score is assigned to each piece of social media content. For example, the highest score for positive sentiment can be set to 5 points, neutral to 3 points, and the lowest score for negative sentiment to 1 point (or adjusted according to the specific situation), and then fine-tuned according to the sentiment intensity. The public opinion scores of all content are weighted averaged to calculate the fourth score of the public opinion attitude on the handling of this fault incident on the entire social media. When weighted averaging, the exposure of each piece of content (such as the number of likes and reposts) can be considered as a weight factor to more accurately reflect the overall trend of public opinion. For example, the fourth score for the handling of the failure incident was calculated to be 3.5 points (out of 5 points), which shows that the public has a certain positive view on the handling of the failure incident, but there is still a certain degree of dissatisfaction and complaints.

[0053] By performing sentiment analysis on captured social media data, we can accurately identify the public's sentiment tendencies (positive, negative, neutral) and intensity (praise, satisfaction, recognition, criticism, dissatisfaction, complaints) towards the handling of fault events. This sentiment analysis technology can deeply explore the public's real attitudes and emotions, providing a more objective and accurate basis for evaluation. The public opinion score of each social media content is determined according to the sentiment tendency and intensity, and the fourth score is obtained by weighted average of all content. This process realizes the quantitative evaluation of public opinion attitudes. This quantitative evaluation method makes the evaluation results more scientific, objective and comparable, which helps power grid managers to more accurately understand the public's views and attitudes on the handling of fault events. By obtaining and analyzing the public opinion attitudes on social media, power grid managers can have a more comprehensive understanding of the public's feedback and opinions on the handling of fault events. This helps them to find problems in a timely manner, adjust strategies, and optimize services, so as to better meet the needs and expectations of the public. At the same time, as a comprehensive evaluation indicator, the fourth score can also provide decision support for power grid managers and help them make more scientific and reasonable decisions.

[0054] S150: Perform a weighted sum of the first score, the second score, the third score, and the fourth score to obtain a total score of the emergency mechanism to be evaluated, and evaluate the emergency mechanism to be evaluated according to the total score.

[0055] The first score, second score, third score and fourth score are assigned different weights during the evaluation process to reflect their respective importance in the overall evaluation. The weight distribution can be adjusted according to the actual situation and evaluation objectives. For example, if the evaluation focuses on emergency response speed and recovery capabilities, the weight of the third score may be relatively high. The quality of the emergency mechanism to be evaluated is determined based on the total score.

[0056] This embodiment also discloses an evaluation system for extreme natural weather emergency response mechanisms. Figure 2 is a schematic diagram of a module of an evaluation system for an extreme natural weather emergency response mechanism disclosed in an embodiment of the present application, such as Figure 2 As shown, the evaluation system includes a weather module 201, a power grid module 202, a processing module 203, a public opinion module 204 and an evaluation module 205, wherein: The weather module 201 is configured to obtain a meteorological log in the emergency mechanism to be evaluated, determine a first prediction accuracy of extreme natural weather according to the meteorological log, and determine a first score according to the first prediction accuracy, wherein the meteorological log includes daily actual meteorological data and meteorological forecast data within a preset time period, and the extreme natural weather includes a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; A power grid module 202 is configured to obtain a power grid state log in the emergency mechanism to be evaluated, determine a second prediction accuracy of a power grid failure according to the power grid state log, and determine a second score according to the second prediction accuracy, wherein the power grid state log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; The processing module 203 is configured to determine a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determine an impact level of the extreme natural weather corresponding to the fault event, and determine a third score according to the processing capability value and the impact level; The public opinion module 204 is configured to obtain public opinion attitudes on social media regarding the handling of the fault event, and determine a fourth score according to the public opinion attitudes, wherein the public opinion attitudes include positive, neutral and negative; The evaluation module 205 is configured to perform a weighted summation of the first score, the second score, the third score and the fourth score to obtain a total score of the emergency mechanism to be evaluated, and evaluate the emergency mechanism to be evaluated according to the total score.

[0057] Optionally, the weather module 201 is configured to: A first sub-score is determined according to a first difference between the predicted time of occurrence of extreme natural weather and the actual time of occurrence of extreme natural weather, a second sub-score is determined according to a second difference between the predicted intensity of extreme natural weather and the actual intensity of extreme natural weather, and a third sub-score is determined according to the predicted type of extreme natural weather and the actual type of occurrence of extreme natural weather; The sub-prediction accuracy of a first target prediction event is determined according to the first sub-score, the second sub-score and the third sub-score, and the first prediction accuracy is obtained according to the average value of the sub-prediction accuracies of multiple first target prediction events.

[0058] Optionally, the power grid module 202 is configured to: Extracting a first feature related to a power grid failure from the meteorological log, the first feature including the occurrence time, intensity and duration of extreme natural weather, and extracting a second feature reflecting the power grid status from the power grid status log, the second feature including a load change trend, equipment health status and historical failure mode; Inputting the first feature and the second feature into a preset fault prediction model to obtain power grid fault prediction data, wherein the power grid fault prediction data includes a predicted occurrence time, a predicted occurrence location, and a predicted occurrence type; The second prediction accuracy is determined according to the power grid fault prediction data and the actual fault data.

[0059] Optionally, the power grid module 202 is configured to: a fourth sub-score is determined according to a third difference between the predicted grid fault occurrence time and the actual grid fault occurrence time, a fifth sub-score is determined according to a fourth difference between the predicted grid fault location and the actual grid fault location, and a sixth sub-score is determined according to the predicted grid fault type and the actual grid fault type; The sub-prediction accuracy of the second target prediction event is determined according to the fourth sub-score, the fifth sub-score and the sixth sub-score, and the sub-prediction accuracy of multiple second target prediction events is weightedly summed according to the power grid fault prediction type to obtain the second prediction accuracy.

[0060] Optionally, the processing module 203 is configured to: Obtain the first time when the fault occurs, the second time when the maintenance personnel arrive at the scene to handle the fault, and the third time when the power supply is restored; A fault response speed is determined according to the first time and the second time, a power recovery speed is determined according to the second time and the third time, and a processing capacity value is obtained by weighted summing the fault response speed and the power recovery speed.

[0061] Optionally, the processing module 203 is configured to: Determine the type and intensity of the extreme natural weather corresponding to the fault event, match a preset impact level according to the type and intensity, and multiply the preset impact level by the processing capacity value to obtain a third score.

[0062] Optionally, the public opinion module 204 is configured to: Using crawler technology or social media interfaces, crawling social media data containing keywords of the fault event, the keywords include fault name, location, time, power supply company name, emergency repair, and recovery, and the social media data includes posts, comments, and forwarding; Performing sentiment analysis on the social media data to identify sentiment tendency and intensity, wherein the sentiment tendency includes positive, negative and neutral, and the intensity includes praise, satisfaction, recognition, criticism, dissatisfaction and complaint; The public opinion score of each content in the social media data is determined according to the emotional tendency and intensity, and the fourth score is obtained by weighted averaging all the content.

[0063] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0064] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0065] The communication bus 302 is used to realize the connection and communication between these components.

[0066] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0067] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0068] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0069] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be optionally at least one storage device located away from the aforementioned processor 301. As Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of an evaluation method for an extreme natural weather emergency response mechanism.

[0070] exist Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the application program for storing the extreme natural weather emergency mechanism assessment method in the memory 305. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0071] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0072] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0077] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for evaluating an extreme natural weather emergency response mechanism, characterized in that: Applied to the power grid management platform, the evaluation method includes: Obtaining a meteorological log in the emergency mechanism to be evaluated, determining a first prediction accuracy of extreme natural weather according to the meteorological log, and determining a first score according to the first prediction accuracy, the meteorological log including daily actual meteorological data and meteorological forecast data within a preset time period, the extreme natural weather including a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; Obtaining a power grid status log in the emergency mechanism to be evaluated, determining a second prediction accuracy of a power grid failure according to the power grid status log, and determining a second score according to the second prediction accuracy, wherein the power grid status log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; Determine a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determine an impact level of the extreme natural weather corresponding to the fault event, and determine a third score according to the processing capability value and the impact level; Obtaining public opinion on social media regarding the handling of the fault event, and determining a fourth score according to the public opinion, wherein the public opinion includes positive, neutral, and negative; The first score, the second score, the third score and the fourth score are weightedly summed to obtain a total score of the emergency mechanism to be evaluated, and the emergency mechanism to be evaluated is evaluated according to the total score.

2. The extreme natural weather emergency response mechanism evaluation method according to claim 1, characterized in that: Determining the first prediction accuracy of extreme natural weather according to the meteorological log comprises: A first sub-score is determined according to a first difference between the predicted time of occurrence of extreme natural weather and the actual time of occurrence of extreme natural weather, a second sub-score is determined according to a second difference between the predicted intensity of extreme natural weather and the actual intensity of extreme natural weather, and a third sub-score is determined according to the predicted type of extreme natural weather and the actual type of occurrence of extreme natural weather; The sub-prediction accuracy of a first target prediction event is determined according to the first sub-score, the second sub-score and the third sub-score, and the first prediction accuracy is obtained according to the average value of the sub-prediction accuracies of multiple first target prediction events.

3. The extreme natural weather emergency mechanism evaluation method according to claim 1, characterized in that: Determining a second prediction accuracy of a power grid fault according to the power grid status log comprises: Extracting a first feature related to a power grid failure from the meteorological log, the first feature including the occurrence time, intensity and duration of extreme natural weather, and extracting a second feature reflecting the power grid status from the power grid status log, the second feature including a load change trend, equipment health status and historical failure mode; Inputting the first feature and the second feature into a preset fault prediction model to obtain power grid fault prediction data, wherein the power grid fault prediction data includes a predicted occurrence time, a predicted occurrence location, and a predicted occurrence type; The second prediction accuracy is determined according to the power grid fault prediction data and the actual fault data.

4. The extreme natural weather emergency response mechanism evaluation method according to claim 3 is characterized in that: Determining the second prediction accuracy according to the power grid fault prediction data and the actual fault data includes: a fourth sub-score is determined according to a third difference between the predicted grid fault occurrence time and the actual grid fault occurrence time, a fifth sub-score is determined according to a fourth difference between the predicted grid fault location and the actual grid fault location, and a sixth sub-score is determined according to the predicted grid fault type and the actual grid fault type; The sub-prediction accuracy of the second target prediction event is determined according to the fourth sub-score, the fifth sub-score and the sixth sub-score, and the sub-prediction accuracy of multiple second target prediction events is weightedly summed according to the power grid fault prediction type to obtain the second prediction accuracy.

5. The extreme natural weather emergency mechanism evaluation method according to claim 1, characterized in that: Determining the processing capacity value according to the fault response speed and power supply recovery speed of processing the fault event includes: Obtain the first time when the fault occurs, the second time when the maintenance personnel arrive at the scene to handle the fault, and the third time when the power supply is restored; A fault response speed is determined according to the first time and the second time, a power recovery speed is determined according to the second time and the third time, and a processing capacity value is obtained by weighted summing the fault response speed and the power recovery speed.

6. The extreme natural weather emergency mechanism assessment method according to claim 5, characterized in that: The determining of the impact level of the extreme natural weather corresponding to the fault event, and determining the third score according to the processing capability value and the impact level comprises: Determine the type and intensity of the extreme natural weather corresponding to the fault event, match a preset impact level according to the type and intensity, and multiply the preset impact level by the processing capacity value to obtain a third score.

7. The extreme natural weather emergency mechanism evaluation method according to claim 1, characterized in that: The obtaining of public opinion on social media regarding the handling of the fault event and determining a fourth score according to the public opinion includes: Using crawler technology or social media interfaces, crawling social media data containing keywords of the fault event, the keywords include fault name, location, time, power supply company name, emergency repair, and recovery, and the social media data includes posts, comments, and forwarding; Performing sentiment analysis on the social media data to identify sentiment tendency and intensity, wherein the sentiment tendency includes positive, negative and neutral, and the intensity includes praise, satisfaction, recognition, criticism, dissatisfaction and complaint; The public opinion score of each content in the social media data is determined according to the emotional tendency and intensity, and the fourth score is obtained by weighted averaging all the content.

8. An evaluation system for extreme natural weather emergency response mechanism, characterized in that: It includes weather module, power grid module, processing module, public opinion module and evaluation module, among which: a weather module configured to obtain a meteorological log in the emergency mechanism to be evaluated, determine a first prediction accuracy of extreme natural weather according to the meteorological log, and determine a first score according to the first prediction accuracy, wherein the meteorological log includes daily actual meteorological data and meteorological forecast data within a preset time period, and the extreme natural weather includes a temperature higher than a first temperature preset value or lower than a second temperature preset value, a precipitation greater than a first precipitation preset value or less than a second precipitation preset value, a wind force greater than a wind force preset value, and a preset type of precipitation; a power grid module configured to obtain a power grid state log in the emergency mechanism to be evaluated, determine a second prediction accuracy of a power grid failure according to the power grid state log, and determine a second score according to the second prediction accuracy, wherein the power grid state log includes daily working data of the power grid and power grid failure prediction data generated in combination with the meteorological log; a processing module configured to determine a processing capability value according to a fault response speed and a power supply restoration speed for processing a fault event, determine an impact level of extreme natural weather corresponding to the fault event, and determine a third score according to the processing capability value and the impact level; A public opinion module configured to obtain public opinion attitudes on social media regarding the handling of the fault event, and determine a fourth score according to the public opinion attitudes, wherein the public opinion attitudes include positive, neutral, and negative; An evaluation module is configured to perform a weighted summation of the first score, the second score, the third score and the fourth score to obtain a total score of the emergency mechanism to be evaluated, and evaluate the emergency mechanism to be evaluated according to the total score.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images