Method and device for calculating similarity between power emergency events, computer equipment and readable storage medium

Abstract

The invention relates to the technical field of power system emergency management, and provides a method and device for calculating the similarity between power emergency events, computer equipment and a readable storage medium. The method comprises the steps of inputting disaster damage characteristics of a historical power emergency event and a target power emergency event into a similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines a similarity measurement method and weight of the disaster damage characteristics according to data types, importance and influence degrees of the disaster damage characteristics, and a similarity calculation formula is obtained; obtaining a similarity score between the target power emergency event and each historical power emergency event according to the disaster damage characteristics of the historical power emergency events and the target power emergency event and a similarity calculation formula; and determining a most similar historical power emergency event list of the target power emergency event according to the similarity score. By adopting the method, the accurate calculation of the similarity between the power emergency events can be realized, so that the power emergency events can be efficiently and accurately processed.

Description

Technical Field

[0001] The present application relates to the technical field of power system emergency management, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for calculating the similarity between power emergency events. Background Art

[0002] As an important infrastructure in modern society, the stable operation of the power system plays a vital role in economic and social development. However, factors such as natural disasters and man-made accidents often cause power system failures, seriously affecting power supply safety. Therefore, it is very important to handle complex and changing power emergency events efficiently and accurately.

[0003] Traditional methods of handling power emergencies rely on manual experience and limited historical data, and the efficiency and accuracy of power emergency handling are low. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for calculating the similarity between power emergency events in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for calculating the similarity between power emergency events, including:

[0006] Inputting the disaster loss characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains the similarity calculation formula;

[0007] According to the disaster loss characteristics of the historical power emergency events and the target power emergency events, and the similarity calculation formula, a similarity score between the target power emergency event and each historical power emergency event is obtained;

[0008] According to the similarity scores, a list of historical power emergency events most similar to the target power emergency event is determined.

[0009] In one embodiment, the disaster damage characteristics include at least any one of emergency treatment basic impact characteristics, emergency treatment combined impact characteristics, emergency treatment interactive impact characteristics, time series characteristics, geographical characteristics and domain knowledge characteristics.

[0010] In one embodiment, before inputting the basic impact characteristics of emergency treatment, the combined impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment, the time series characteristics, the geographical characteristics and the domain knowledge characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, the method further includes:

[0011] Obtaining initial impact characteristics of emergency response input by power emergency personnel that have an impact on the handling of historical power emergency events and target power emergency events;

[0012] Performing a feature importance assessment on the initial impact features of the emergency treatment to obtain basic impact features of the emergency treatment of historical power emergency events and target power emergency events;

[0013] According to the emergency treatment basic impact characteristics of the historical power emergency events and the target power emergency events, the emergency treatment combined impact characteristics and the emergency treatment interactive impact characteristics of the historical power emergency events and the target power emergency events are obtained;

[0014] According to the time series data and geographic information system data of historical power emergency events and target power emergency events, the time series characteristics and geographic characteristics of the historical power emergency events and target power emergency events are obtained.

[0015] In one embodiment, the emergency treatment combined impact characteristics and emergency treatment interactive impact characteristics of the historical power emergency events and the target power emergency events are obtained according to the emergency treatment basic impact characteristics of the historical power emergency events and the target power emergency events, including:

[0016] Combining two or more of the emergency handling basic impact features of the historical power emergency events and the target power emergency events to obtain the emergency handling combined impact features of the historical power emergency events and the target power emergency events;

[0017] According to the interactive relationship between the basic impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events are obtained.

[0018] In one embodiment, before inputting the damage characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, the method further includes:

[0019] According to the calculation characteristics of the similarity between power emergency events and the data characteristics of disaster damage characteristics, several candidate models to be trained are screened;

[0020] According to the damage characteristics of historical power emergency events, a training feature set is obtained;

[0021] According to the training feature set, each candidate model to be trained is trained to obtain a plurality of candidate models;

[0022] The candidate model with the best performance is selected as the recommendation model for the similarity calculation formula.

[0023] In one embodiment, after determining the most similar historical power emergency event list to the target power emergency event, the method further includes:

[0024] According to the most similar historical power emergency event list of the target power emergency event, corresponding response measures and resource allocation plans for each most similar historical power emergency event are obtained;

[0025] According to the characteristics and basic information of the target power emergency event, the response measures and resource allocation plans corresponding to the most similar historical power emergency events are adjusted to obtain the initial response measures and initial resource allocation plans corresponding to the target power emergency event.

[0026] In one embodiment, after obtaining the initial response measures and initial resource allocation plan corresponding to the target power emergency event, the method further includes:

[0027] Obtain real-time data on the progress of target power emergency events;

[0028] Processing and analyzing the real-time data of the progress of the target power emergency event to obtain the evaluation effect of the initial response measures and initial resource allocation plan;

[0029] According to the evaluation results of the initial response measures and the initial resource allocation plan, the initial response measures and the initial resource allocation plan are adjusted to obtain the adjusted response measures and resource allocation plan for the target power emergency event.

[0030] In a second aspect, the present application also provides a device for calculating the similarity between power emergency events, including:

[0031] A similarity calculation formula acquisition module is used to input the disaster loss characteristics of historical power emergency events and target power emergency events into a similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains a similarity calculation formula;

[0032] A similarity score acquisition module, used to obtain a similarity score between the target power emergency event and each historical power emergency event according to the disaster loss characteristics of the historical power emergency event and the target power emergency event and the similarity calculation formula;

[0033] The event list acquisition module is used to determine the most similar historical power emergency event list of the target power emergency event according to the similarity score.

[0034] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the above method.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to execute the above method.

[0036] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and the computer program is executed by a processor to execute the above method.

[0037] The above-mentioned method, device, computer equipment, computer-readable storage medium and computer program product for calculating the similarity between power emergency events input the disaster loss characteristics of historical power emergency events and target power emergency events into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains the similarity calculation formula; according to the disaster loss characteristics of historical power emergency events and target power emergency events, and the similarity calculation formula, obtains the similarity score between the target power emergency event and each historical power emergency event; according to the similarity score, determines the most similar historical power emergency event list of the target power emergency event. When calculating the similarity between power emergency events, the present application integrates multi-factor disaster loss characteristics and comprehensively considers the complexity and diversity of power emergency events, which can improve the accuracy of the assessment of the impact and severity of power emergency events, can realize the accurate calculation of the similarity between power emergency events, and can also provide a more comprehensive and scientific basis for the response measures and resource allocation plans for power emergency events, so as to efficiently and accurately handle power emergency events. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A diagram of an application environment of a method for calculating similarities between power emergency events in one embodiment;

[0040] Figure 2 is a flow chart of a method for calculating similarity between power emergency events in one embodiment;

[0041] Figure 3 A schematic diagram of a flow chart of a disaster damage feature acquisition step in one embodiment;

[0042] Figure 4 is a structural block diagram of a device for calculating similarities between power emergency events in one embodiment;

[0043] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] The present application embodiment provides a method for calculating the similarity between power emergency events. The present embodiment can be executed by a computer device, such as Figure 1 As shown, the computer device can obtain the damage characteristics of historical power emergency events and target power emergency events, and then determine the most similar historical power emergency event list of the target power emergency event. It is understandable that the computer device can be implemented through a server, a terminal, or an interactive system between a terminal and a server. In this embodiment, the method includes Figure 2 The steps shown are:

[0046] Step S201, input the disaster loss characteristics of historical power emergency events and target power emergency events into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains the similarity calculation formula.

[0047] The power emergency event whose similarity to historical power emergency events is to be calculated may be referred to as a target power emergency event.

[0048] The disaster damage characteristics may be multi-factor disaster damage characteristics, and the disaster damage characteristics may include at least any one of the basic impact characteristics of emergency treatment, the combined impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment, the time series characteristics, the geographical characteristics and the domain knowledge characteristics.

[0049] Similarity measurement methods can include Euclidean distance measurement method, cosine similarity measurement method, Manhattan distance measurement method and Jaccard similarity measurement method. Among them, the Euclidean distance measurement method is applicable to numerical feature data, and can calculate the straight-line distance between two feature vectors; the cosine similarity measurement method is applicable to high-dimensional sparse feature data, and can calculate the cosine value of the angle between two feature vectors; the Manhattan distance measurement method is applicable to numerical feature data, and can calculate the sum of the absolute differences of two feature vectors in each dimension; the Jaccard similarity measurement method is applicable to set feature data, and can calculate the ratio of the intersection and union of two feature sets.

[0050] The disaster loss characteristics of historical power emergency events and target power emergency events can be input into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact of the disaster loss characteristics, and obtains the similarity calculation formula shown in formula (1) according to the similarity measurement method and weight of the disaster loss characteristics. Among them, the higher the importance and impact of the disaster loss characteristics, the higher the weight of the disaster loss characteristics.

[0051] (1)

[0052] In the formula, Similarity(A, B) represents the similarity score between power emergency event A and power emergency event B; Represents the weight of the i-th disaster loss feature; similarity_method ( , ) represents the similarity measurement method of the i-th disaster loss characteristics of power emergency event A and power emergency event B.

[0053] For example, when the similarity measurement method for determining the disaster damage characteristics is the Euclidean distance measurement method, the similarity calculation formula shown in formula (2) can be obtained.

[0054] (2)

[0055] Step S202, according to the disaster loss characteristics of the historical power emergency events and the target power emergency event, and the similarity calculation formula, obtain the similarity score between the target power emergency event and each historical power emergency event.

[0056] The disaster loss characteristics of historical power emergency events and target power emergency events can be substituted into the similarity calculation formula to obtain the similarity score between the target power emergency event and each historical power emergency event.

[0057] Step S203: determining a list of historical power emergency events most similar to the target power emergency event according to the similarity scores.

[0058] According to the similarity scores between the target power emergency event and each historical power emergency event, the historical power emergency events can be arranged in descending order, and the top N historical power emergency events with the highest similarity scores can be selected to form a list of historical power emergency events most similar to the target power emergency event.

[0059] The similarity scores between the target power emergency event and each historical power emergency event, as well as the list of the most similar historical power emergency events to the target power emergency event, can be organized into a structured similarity report, which includes detailed information on each most similar historical power emergency event, such as the identifier (ID), occurrence time, location, disaster type, and impact scope of the most similar historical power emergency event.

[0060] In the above-mentioned method for calculating the similarity between power emergency events, when calculating the similarity between power emergency events, the multi-factor disaster loss characteristics are integrated, and the complexity and diversity of power emergency events are fully considered. This can improve the accuracy of the assessment of the impact and severity of power emergency events, and can achieve accurate calculation of the similarity between power emergency events. It can also provide a more comprehensive and scientific basis for the response measures and resource allocation plans for power emergency events, so as to handle power emergency events efficiently and accurately.

[0061] In one embodiment, the disaster damage characteristics include at least any one of emergency treatment basic impact characteristics, emergency treatment combined impact characteristics, emergency treatment interactive impact characteristics, time series characteristics, geographical characteristics and domain knowledge characteristics.

[0062] The basic impact characteristics of emergency response refer to the characteristics that have a basic impact on the emergency response of power emergencies, which may include disaster type, size of the affected area, duration, cost of restoring damaged facilities, population density of the affected area, power network structure of the affected area, meteorological conditions, and historical similar power emergencies. The power network structure includes information such as grid nodes, lines, and substations; meteorological conditions include meteorological parameters such as temperature, rainfall, and wind speed when the power emergency occurs.

[0063] The combined impact characteristics of emergency treatment represent new characteristics that have an impact on the emergency treatment of power emergencies, which are obtained by combining two or more basic impact characteristics of emergency treatment. For example, the combined impact characteristics of emergency treatment may be a disaster risk index and a recovery cost; wherein the disaster risk index may be calculated in combination with the disaster type and meteorological conditions.

[0064] The interactive impact feature of emergency treatment represents a new feature that has an impact on the emergency treatment of power emergencies based on the interactive relationship between the basic impact features of emergency treatment. For example, the interactive impact feature of emergency treatment can be population density and restoration cost; where population density and restoration cost are used to evaluate the impact of population density on restoration cost.

[0065] Time series features can be extracted from the time series data corresponding to power emergency events. For example, the frequency of similar power emergency events in a certain area over a period of time can be calculated as the historical event frequency feature, which is a time series feature.

[0066] Geographical features can be extracted from the geographic information system data corresponding to the power emergency event. For example, the distance from the affected area to the nearest substation can be calculated as the distance to the nearest substation feature, and the distance to the nearest substation feature is a geographical feature.

[0067] Domain knowledge features can be obtained based on the professional knowledge of the power industry. For example, the degree to which the power grid is vulnerable to damage under specific conditions can be evaluated as a power grid vulnerability index feature, which is a domain knowledge feature; the distribution of emergency resources (such as repair teams and material reserves) in the affected area can also be evaluated as an emergency resource distribution feature, which is a domain knowledge feature.

[0068] In this embodiment, the disaster loss characteristics include at least any one of the basic impact characteristics of emergency treatment, the combined impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment, the time series characteristics, the geographical characteristics and the domain knowledge characteristics. Therefore, when calculating the similarity between power emergency events, the multi-factor disaster loss characteristics are integrated, and the complexity and diversity of power emergency events are fully considered, so that the accurate calculation of the similarity between power emergency events can be achieved.

[0069] In one embodiment, before inputting the basic impact characteristics of emergency treatment, combined impact characteristics of emergency treatment, interactive impact characteristics of emergency treatment, time series characteristics, geographical characteristics and domain knowledge characteristics of historical power emergency events and target power emergency events into the similarity calculation formula recommendation model, the method provided by the present application also includes Figure 3The steps shown are: step S301, obtaining the initial impact characteristics of emergency treatment input by the power emergency personnel, which have an impact on the handling of historical power emergency events and target power emergency events; step S302, performing feature importance evaluation on the initial impact characteristics of emergency treatment, and obtaining the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events; step S303, obtaining the combined impact characteristics of emergency treatment and the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events based on the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events; step S304, obtaining the time series characteristics and geographic characteristics of historical power emergency events and target power emergency events based on the time series data and geographic information system data of historical power emergency events and target power emergency events.

[0070] Before obtaining the initial impact characteristics of emergency treatment that have an impact on the handling of historical power emergency events and target power emergency events input by power emergency personnel, that is, before extracting the characteristics of historical power emergency events and target power emergency events, it is necessary to collect various types of original data related to historical power emergency events and target power emergency events from multiple data sources, and clean and standardize the collected various types of original data to ensure the quality and consistency of the collected various types of original data, and meet the requirements of subsequent processing and analysis. The specific steps of data collection, data cleaning and data standardization are as follows:

[0071] Various raw data related to historical power emergency events and target power emergency events can be collected from multiple data sources. The data sources can include historical power emergency event records, meteorological data, Geographic Information System (GIS) data, and power network topology data.

[0072] Among them, historical power emergency event records can be obtained from the internal database of the power company, including the time, location, type, impact range, duration and recovery cost of historical power emergency events; meteorological data can be obtained from the Meteorological Bureau or a third-party meteorological service platform, including meteorological parameters such as temperature, rainfall, wind speed and humidity, as well as historical meteorological records; geographic information system data can be obtained from the Surveying and Mapping Bureau or a commercial geographic information system platform, including geographic information such as topography, population density and transportation network; power network topology data can be obtained from the power grid database maintained within the power company, including infrastructure information such as power grid nodes, lines and substations.

[0073] For data sources that provide an Application Programming Interface (API), you can obtain data by calling the API through a programming language. For data sources that provide file downloads, you can download the latest data files regularly. For internal databases, you can obtain data through Structured Query Language (SQL).

[0074] The collected raw data can be stored in a central data warehouse, which can be a relational database or a non-relational database.

[0075] The collected raw data can be cleaned to remove duplicate, erroneous or incomplete data entries. The specific data cleaning steps include: deduplication, missing value processing, outlier processing and data consistency check. Among them, unique identifiers (such as event identifiers or timestamps) can be used to detect and delete duplicate records. Missing value processing can be done in the following ways: delete records with missing values; fill missing values ​​with mean, median, mode or interpolation; predict missing values ​​using machine learning algorithms. Outlier processing can be done in the following ways: use statistical methods such as standard deviation and quartiles to identify and process outliers; use visualization tools such as box plots and scatter plots to identify outliers; judge outliers based on domain knowledge and process them. Data consistency checks can ensure data consistency between different data sources, for example, check whether the timestamp format is unified and whether the units are consistent.

[0076] The cleaned raw data can be formatted, dimensioned, encoded, and merged. Format unification includes timestamp format unification and string format unification. For example, convert all timestamps to ISO 8601 format; convert all place names to standard place name codes. Dimension conversion is to convert data in different units to the same unit, and to normalize or standardize numerical data. For example, convert temperature from Fahrenheit to Celsius; standardize numerical data using standard deviation (Z-score) standardization or deviation (Min-Max) standardization. Data encoding is to encode categorical variables. For example, use One-Hot Encoding or Label Encoding to encode categorical variables. Data merging is to merge data from different data sources into a unified data set.

[0077] The initial impact features of emergency treatment that have an impact on the handling of historical power emergency events and target power emergency events can be obtained by power emergency personnel based on their knowledge and experience in the field of power emergency, as well as various types of original data related to historical power emergency events and target power emergency events after cleaning and standardization. Statistical methods and machine learning techniques can be used to evaluate the feature importance of the initial impact features of emergency treatment, and obtain the basic impact features of emergency treatment for historical power emergency events and target power emergency events. For example, the correlation analysis method can be used to calculate the correlation coefficient between the initial impact features of emergency treatment and the target variable (such as the severity of power emergency events, recovery time, etc.). The importance score of the initial impact features of emergency treatment can be calculated using models such as decision trees and random forests. The most important features of the initial impact features of emergency treatment can be identified by the principal component analysis (PCA) method.

[0078] According to the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events, the combined impact characteristics of emergency treatment of historical power emergency events and target power emergency events and the interactive impact characteristics of emergency treatment can be obtained.

[0079] The time series features of the historical power emergency events and the target power emergency events can be obtained based on the time series data in various raw data of the historical power emergency events and the target power emergency events. For example, the time interval between adjacent power emergency events can be calculated as the event interval time feature, and the event interval time feature is a time series feature.

[0080] The geographic characteristics of historical power emergency events and target power emergency events can be obtained based on the geographic information system data in various raw data of historical power emergency events and target power emergency events. For example, the amplification or mitigation effect of the terrain on the power emergency event can be evaluated as a terrain slope feature, which is a geographical feature.

[0081] In this embodiment, basic impact features of emergency treatment that have an important impact on the emergency treatment of historical power emergency events and target power emergency events can be extracted from various types of original data of historical power emergency events and target power emergency events, and new emergency treatment combination impact features and emergency treatment interaction impact features can be created. The implicit information in various types of original data can be captured, and multi-factor disaster loss features can be integrated when calculating the similarity between power emergency events. The complexity and diversity of power emergency events can be fully considered, and the accurate calculation of the similarity between power emergency events can be achieved.

[0082] In one of the embodiments, based on the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events, the combined impact characteristics of emergency treatment and the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events are obtained. The specific steps are as follows: two or more basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events are combined to obtain the combined impact characteristics of emergency treatment of historical power emergency events and target power emergency events; based on the interactive relationship between the basic impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events are obtained.

[0083] Two or more basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events can be combined to obtain the combined impact characteristics of emergency treatment of historical power emergency events and target power emergency events. For example, the combined impact characteristic of emergency treatment can be a restoration difficulty index, which is obtained by combining two basic impact characteristics of emergency treatment (the size of the affected area and the structure of the power network).

[0084] According to the interactive relationship between the basic impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events can be obtained. For example, according to the interactive relationship between meteorological conditions and duration, meteorological conditions and duration can be obtained to evaluate the impact of meteorological conditions on the duration of power emergency events. Among them, meteorological conditions and duration are the basic impact characteristics of emergency treatment, and meteorological conditions and duration are the interactive impact characteristics of emergency treatment.

[0085] In this embodiment, new emergency treatment combined impact characteristics and emergency treatment interactive impact characteristics can be created based on the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events. When calculating the similarity between power emergency events, multi-factor disaster loss characteristics can be integrated to fully consider the complexity and diversity of power emergency events, so as to achieve accurate calculation of the similarity between power emergency events.

[0086] In one of the embodiments, before inputting the disaster loss characteristics of historical power emergency events and target power emergency events into the similarity calculation formula recommendation model, the method provided by the present application also includes: screening and obtaining a number of candidate models to be trained based on the calculation characteristics of the similarity between power emergency events and the data characteristics of the disaster loss characteristics; obtaining a training feature set based on the disaster loss characteristics of historical power emergency events; training each candidate model to be trained based on the training feature set to obtain a number of candidate models; and selecting the candidate model with the best performance as the similarity calculation formula recommendation model.

[0087] According to the calculation characteristics of the similarity between power emergency events and the data characteristics of disaster loss characteristics, several candidate models to be trained can be screened from common machine learning algorithms. Among them, common machine learning algorithms include decision trees, random forests, support vector machines (SVM), neural networks, K-nearest neighbor (KNN) and clustering algorithms. Clustering algorithms include k-means clustering algorithm (K-means) and density-based spatial clustering with noise algorithm (DBSCAN).

[0088] Various raw data related to historical power emergency events can be collected from multiple data sources, and the collected raw data can be cleaned and standardized. The features of the cleaned and standardized raw data can be extracted to obtain the disaster loss features of historical power emergency events. The disaster loss features of 70% of historical power emergency events can be used as a training feature set, and the disaster loss features of 30% of historical power emergency events can be used as a test feature set.

[0089] Each candidate model to be trained can be trained according to the training feature set to obtain several candidate models. Specifically, the grid search method, random search method or Bayesian optimization method can be used to adjust the hyperparameters of the candidate model to be trained and find the best parameter combination of the candidate model to be trained to obtain several candidate models.

[0090] The generalization ability and stability of several candidate models can be evaluated based on the cross-validation method (such as K-fold cross-validation) and the test feature set, and various performance indicators of each candidate model can be calculated to determine the performance of each candidate model. The candidate model with the best performance can be selected as the recommendation model for the similarity calculation formula.

[0091] In this embodiment, according to the calculation characteristics of the similarity between power emergency events and the data characteristics of disaster characteristics, several candidate models to be trained are screened; through the disaster characteristics of historical power emergency events, a similarity calculation formula recommendation model is trained to calculate the similarity between the target power emergency event and the historical power emergency events. It can not only process large-scale data, but also mine valuable patterns and laws from various types of data of historical power emergency events, so that the similarity calculation formula recommendation model can have good generalization ability and prediction accuracy, improve the accuracy and reliability of similarity calculation, so as to quickly identify the historical power emergency event that is most similar to the target power emergency event in the calculation of the similarity between power emergency events, provide verified response measures and resource allocation plans, and improve the efficiency and effectiveness of handling power emergency events.

[0092] In one of the embodiments, after determining the list of most similar historical power emergency events to the target power emergency event, the method provided by the present application also includes: obtaining response measures and resource allocation plans corresponding to each most similar historical power emergency event based on the list of most similar historical power emergency events to the target power emergency event; adjusting the response measures and resource allocation plans corresponding to each most similar historical power emergency event based on the characteristics and basic information of the target power emergency event to obtain the initial response measures and initial resource allocation plans corresponding to the target power emergency event.

[0093] According to the most similar historical power emergency event list of the target power emergency event, the corresponding response measures and resource allocation plans of each most similar historical power emergency event are obtained. For example, the communication coordination measures, repair team dispatch plan and material allocation plan of each most similar historical power emergency event in the most similar historical power emergency event list can be extracted.

[0094] According to the characteristics and basic information of the target power emergency event, the response measures and resource allocation schemes corresponding to each most similar historical power emergency event are adjusted to obtain the initial response measures and initial resource allocation scheme corresponding to the target power emergency event. For example, according to the characteristics and basic information of the target power emergency event, the particularity of the target power emergency event is determined, such as the difference in the affected area, the change in meteorological conditions, etc.; according to the particularity of the target power emergency event, the response measures and resource allocation schemes corresponding to each most similar historical power emergency event are adjusted to obtain the initial response measures and initial resource allocation scheme corresponding to the target power emergency event. According to the initial response measures and initial resource allocation scheme corresponding to the target power emergency event, the emergency response strategy corresponding to the target power emergency event can be obtained. The initial response measures and initial resource allocation scheme corresponding to the target power emergency event can be organized into a structured strategy report, which can include specific action steps, division of responsibilities and expected results.

[0095] In this embodiment, according to the characteristics and basic information of the target power emergency event, the response measures and resource allocation plans corresponding to each most similar historical power emergency event in the most similar historical power emergency event list are adjusted to obtain the initial response measures and initial resource allocation plans corresponding to the target power emergency event.

[0096] In one of the embodiments, after obtaining the initial response measures and initial resource allocation plan corresponding to the target power emergency event, the method provided in the present application also includes: obtaining real-time data on the progress of the target power emergency event; processing and analyzing the real-time data on the progress of the target power emergency event to obtain evaluation results of the initial response measures and the initial resource allocation plan; adjusting the initial response measures and the initial resource allocation plan based on the evaluation results of the initial response measures and the initial resource allocation plan to obtain adjusted response measures and resource allocation plans for the target power emergency event.

[0097] Real-time data on the progress of target power emergency events can be obtained in real time through sensors or monitoring systems, where the real-time data may include information such as repair progress, weather changes, and resource consumption.

[0098] The real-time data of the progress of the target power emergency event can be processed and analyzed to obtain the evaluation effect of the initial response measures and the initial resource allocation plan; according to the evaluation effect of the initial response measures and the initial resource allocation plan, the initial response measures and the initial resource allocation plan are adjusted to obtain the adjusted response measures and resource allocation plan for the target power emergency event. For example, if it is found that the repair progress of a certain area is lagging behind, the repair team can be increased or more resources can be deployed. The adjusted response measures and resource allocation plan can be updated to the strategy report, and the relevant personnel can be notified to make corresponding adjustments and executions.

[0099] In this embodiment, the evaluation effect of the initial response measures and the initial resource allocation plan is obtained according to the real-time data of the progress of the target power emergency event; the initial response measures and the initial resource allocation plan are adjusted according to the evaluation effect of the initial response measures and the initial resource allocation plan to obtain the adjusted response measures and resource allocation plan for the target power emergency event. The initial response measures and the initial resource allocation plan can be adjusted in real time according to the progress of the target power emergency event to ensure the effectiveness and timeliness of the response measures and the resource allocation plan, and can quickly respond to changes in the target power emergency event, promptly discover and correct problems in the response measures, ensure the reasonable allocation and effective use of resources, reduce delays and losses caused by information lags, and improve the overall adaptability and effectiveness of the emergency response to power emergencies.

[0100] After obtaining the initial response measures and initial resource allocation plan corresponding to the target power emergency event, as well as the adjusted response measures and resource allocation plan, they are presented to the user in an intuitive form, and user feedback is collected. Specifically, visualization tools such as charts and maps can be used to display the initial response measures and initial resource allocation plan corresponding to the target power emergency event, as well as the adjusted response measures and resource allocation plan. For example, a map can be used to display the distribution and movement path of the repair team, and a bar chart can be used to display resource consumption.

[0101] You can also collect users' opinions and suggestions on countermeasures and resource allocation plans through questionnaires, online feedback, etc. You can record user feedback, analyze common problems and improvement suggestions. Based on user feedback, you can analyze the shortcomings and improvement directions of the similarity calculation formula recommendation model to update the parameters and algorithms of the similarity calculation formula recommendation model and improve the accuracy and robustness of the similarity calculation formula recommendation model.

[0102] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0103] Based on the same inventive concept, the embodiment of the present application also provides a device for calculating the similarity between power emergency events for implementing the method for calculating the similarity between power emergency events involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the device for calculating the similarity between one or more power emergency events provided below can be referred to the limitations of the method for calculating the similarity between power emergency events above, and will not be repeated here.

[0104] In an exemplary embodiment, Figure 4 As shown, a device for calculating the similarity between power emergency events is provided, wherein:

[0105] The similarity calculation formula acquisition module 401 is used to input the disaster loss characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains the similarity calculation formula;

[0106] A similarity score acquisition module 402 is used to obtain a similarity score between the target power emergency event and each historical power emergency event according to the disaster loss characteristics of the historical power emergency event and the target power emergency event and the similarity calculation formula;

[0107] The event list acquisition module 403 is used to determine the most similar historical power emergency event list to the target power emergency event according to the similarity score.

[0108] In one embodiment, the disaster damage characteristics include at least any one of emergency treatment basic impact characteristics, emergency treatment combined impact characteristics, emergency treatment interactive impact characteristics, time series characteristics, geographical characteristics and domain knowledge characteristics.

[0109] In one embodiment, the device also includes a feature acquisition module, which is used to: obtain the initial impact characteristics of emergency treatment input by power emergency personnel that have an impact on the handling of historical power emergency events and target power emergency events; perform feature importance evaluation on the initial impact characteristics of emergency treatment to obtain basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events; obtain combined impact characteristics of emergency treatment and interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events based on the basic impact characteristics of emergency treatment of historical power emergency events and target power emergency events; obtain time series characteristics and geographic characteristics of historical power emergency events and target power emergency events based on time series data and geographic information system data of historical power emergency events and target power emergency events.

[0110] In one of the embodiments, the feature acquisition module is also used to: combine two or more of the emergency handling basic impact features of the historical power emergency events and the target power emergency events to obtain the emergency handling combined impact features of the historical power emergency events and the target power emergency events; and obtain the emergency handling interactive impact features of the historical power emergency events and the target power emergency events based on the interactive relationship between the emergency handling basic impact features.

[0111] In one embodiment, the device also includes a model acquisition module, which is used to: screen and obtain a number of candidate models to be trained based on the calculation characteristics of the similarity between power emergency events and the data characteristics of disaster characteristics; obtain a training feature set based on the disaster characteristics of historical power emergency events; train each candidate model to be trained based on the training feature set to obtain a number of candidate models; select the candidate model with the best performance as the recommendation model for the similarity calculation formula.

[0112] In one of the embodiments, the device also includes a measures and plan acquisition module, which is used to: obtain response measures and resource allocation plans corresponding to each most similar historical power emergency event based on a list of the most similar historical power emergency events of the target power emergency event; adjust the response measures and resource allocation plans corresponding to each most similar historical power emergency event based on the characteristics and basic information of the target power emergency event, and obtain initial response measures and initial resource allocation plans corresponding to the target power emergency event.

[0113] In one of the embodiments, the measures and plans acquisition module is also used to: acquire real-time data on the progress of the target power emergency event; process and analyze the real-time data on the progress of the target power emergency event to obtain evaluation results of the initial response measures and the initial resource allocation plan; adjust the initial response measures and the initial resource allocation plan based on the evaluation results of the initial response measures and the initial resource allocation plan to obtain adjusted response measures and resource allocation plans for the target power emergency event.

[0114] Each module in the above-mentioned device for calculating the similarity between power emergency events can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0115] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data of an embodiment of a method for calculating the similarity between power emergency events. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for calculating the similarity between power emergency events is implemented.

[0116] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0117] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0118] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0119] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0120] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0121] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0123] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for calculating the similarity between power emergency events, characterized in that: The method comprises: Inputting the disaster loss characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains the similarity calculation formula; According to the disaster loss characteristics of the historical power emergency events and the target power emergency events, and the similarity calculation formula, a similarity score between the target power emergency event and each historical power emergency event is obtained; According to the similarity scores, a list of historical power emergency events most similar to the target power emergency event is determined.

2. The method according to claim 1, characterized in that The disaster damage characteristics include at least any one of emergency treatment basic impact characteristics, emergency treatment combined impact characteristics, emergency treatment interactive impact characteristics, time series characteristics, geographical characteristics and domain knowledge characteristics.

3. The method according to claim 2, characterized in that Before inputting the basic impact characteristics of emergency treatment, combined impact characteristics of emergency treatment, interactive impact characteristics of emergency treatment, time series characteristics, geographical characteristics and domain knowledge characteristics of historical power emergency events and target power emergency events into the similarity calculation formula recommendation model, the method further includes: Obtaining initial impact characteristics of emergency response input by power emergency personnel that have an impact on the handling of historical power emergency events and target power emergency events; Performing a feature importance assessment on the initial impact features of the emergency treatment to obtain basic impact features of the emergency treatment of historical power emergency events and target power emergency events; According to the emergency treatment basic impact characteristics of the historical power emergency events and the target power emergency events, the emergency treatment combined impact characteristics and the emergency treatment interactive impact characteristics of the historical power emergency events and the target power emergency events are obtained; According to the time series data and geographic information system data of historical power emergency events and target power emergency events, the time series characteristics and geographic characteristics of the historical power emergency events and target power emergency events are obtained.

4. The method according to claim 3, characterized in that: The emergency treatment combined impact characteristics and emergency treatment interactive impact characteristics of the historical power emergency events and the target power emergency events are obtained according to the emergency treatment basic impact characteristics of the historical power emergency events and the target power emergency events, including: Combining two or more of the emergency handling basic impact features of the historical power emergency events and the target power emergency events to obtain the emergency handling combined impact features of the historical power emergency events and the target power emergency events; According to the interactive relationship between the basic impact characteristics of emergency treatment, the interactive impact characteristics of emergency treatment of historical power emergency events and target power emergency events are obtained.

5. The method according to claim 1, characterized in that Before inputting the disaster loss characteristics of the historical power emergency events and the target power emergency events into the similarity calculation formula recommendation model, the method further includes: According to the calculation characteristics of the similarity between power emergency events and the data characteristics of disaster damage characteristics, several candidate models to be trained are screened; According to the damage characteristics of historical power emergency events, a training feature set is obtained; According to the training feature set, each candidate model to be trained is trained to obtain a plurality of candidate models; The candidate model with the best performance is selected as the recommendation model for the similarity calculation formula.

6. The method according to claim 1, characterized in that After determining the most similar historical power emergency event list to the target power emergency event, the method further includes: According to the most similar historical power emergency event list of the target power emergency event, corresponding response measures and resource allocation plans for each most similar historical power emergency event are obtained; According to the characteristics and basic information of the target power emergency event, the response measures and resource allocation plans corresponding to the most similar historical power emergency events are adjusted to obtain the initial response measures and initial resource allocation plans corresponding to the target power emergency event.

7. The method according to claim 6, characterized in that After obtaining the initial response measures and initial resource allocation plan corresponding to the target power emergency event, the method further includes: Obtain real-time data on the progress of target power emergency events; Processing and analyzing the real-time data of the progress of the target power emergency event to obtain the evaluation effect of the initial response measures and initial resource allocation plan; According to the evaluation results of the initial response measures and the initial resource allocation plan, the initial response measures and the initial resource allocation plan are adjusted to obtain the adjusted response measures and resource allocation plan for the target power emergency event.

8. A device for calculating the similarity between power emergency events, characterized in that: The device comprises: A similarity calculation formula acquisition module is used to input the disaster loss characteristics of historical power emergency events and target power emergency events into a similarity calculation formula recommendation model, so that the similarity calculation formula recommendation model determines the similarity measurement method and weight of the disaster loss characteristics according to the data type, importance and impact degree of the disaster loss characteristics, and obtains a similarity calculation formula; A similarity score acquisition module, used to obtain a similarity score between the target power emergency event and each historical power emergency event according to the disaster loss characteristics of the historical power emergency event and the target power emergency event and the similarity calculation formula; The event list acquisition module is used to determine the most similar historical power emergency event list of the target power emergency event according to the similarity score.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images