Power grid real-time operation weak link identification and emergency screening method

Abstract

The invention discloses a power grid real-time operation weak link identification and emergency screening method, and relates to the technical field of power grids. Comprising the following steps: information acquisition: acquiring a power grid real-time operation state, past event information, feedback address information and user requirements, wherein the power grid real-time operation state is state information; through the set monitoring processing method, the operation information of the power grid is monitored in real time, the abnormal information is extracted in real time, the abnormal information is alarmed, and before the abnormal information is alarmed, the abnormal information is graded, so that different alarm intensities are selected to remind a user, and through the set weakness prediction method, the user experience is improved. According to the real-time power grid operation state information and the previous event information, whether weak phenomena are generated and exist in the future of the operation of the equipment is predicted, and abnormal information is provided for a user in advance, so that the user has enough events to deal with and solve possible abnormal phenomena.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and specifically to a method for identifying weak links in the real-time operation of power grids and screening emergency events. Background Technique

[0002] A power grid is a complex system whose main function is to transmit electric power from power plants to users, including residential, commercial, and industrial applications. The components of a power grid include power generation, power transmission, power distribution, and end-users. During the operation of the power grid, effectively identifying the weak links in the real-time operation of the power grid and quickly screening out possible emergency events are crucial for ensuring the safety and stability of the power system.

[0003] A method for identifying weak links in a power grid and screening emergency events during the real-time operation of the power grid with the patent publication number CN117708576A includes the following steps: S1: Collect power grid operation dynamic data based on the real-time operation of the power grid; S2: Preprocess the power grid operation dynamic data to determine the power grid operation dynamic characterization data; S3: Conduct an analysis on identifying weak links in the power grid for the power grid operation dynamic characterization data to determine the result of the analysis on identifying weak links in the power grid; S4: Deeply mine the result of the analysis on identifying weak links in the power grid, and establish a power grid emergency event set for different power grid operation dynamic data. The present invention solves the problem that in the existing power grid during real-time operation, the weak links in the power grid cannot be quickly identified, resulting in the inability to quickly screen out emergency events, making the control effect of the real-time operation of the power grid poor. The weak links in the power grid of the present invention can be quickly identified, and emergency events can be quickly screened out, making the control effect of the real-time operation of the power grid good.

[0004] For the method for identifying weak links in the real-time operation of the power grid and screening emergency events with the above and similar principles, it lacks predictability. During a large-scale urban music festival that attracts hundreds of thousands of audiences, the demand for the power grid surges instantly. In this case, the power grid needs to monitor the real-time load, ensure stable power supply, and predict any possible events to ensure the timeliness of handling events. The methods with the above and similar principles are not convenient for predicting future possible abnormal phenomena based on known characteristics, and then it will lead to the drawback of lag in obtaining abnormal phenomena during the operation of the power grid, which is not conducive to improving the management efficiency of the power grid operation. Therefore, the present invention is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for identifying weak links in the real-time operation of the power grid and screening emergency events to solve the problems raised in the above background technique.

[0006] To achieve the above purpose, the present invention provides the following technical solution: A method for identifying weak links in the real-time operation of the power grid and screening emergency events, the method includes: Information acquisition: Obtain the real-time operation status of the power grid, past event information, feedback address information, and user requirements. The real-time operation status of the power grid is status information; Threshold establishment: Preset a target threshold and a grade range based on the status information; Information processing: Based on the target threshold, monitor and process the status information in real time through a monitoring and processing method to obtain monitoring information. Based on the monitoring information, obtain the cause and impact of generating the monitoring information through an analysis method to obtain the target cause and target impact; Abnormality classification: Classify the target cause through a classification method based on the past event information and in cooperation with the user requirements to obtain a classification result; Real-time prediction: Obtain a prediction result through a vulnerability prediction method based on the status information; Result output: Send the monitoring information, target cause, target impact, classification result, and prediction result to the user based on the feedback address information; The monitoring and processing method includes: Split the status information into several sub-information, split the target threshold into several sub-thresholds, establish the relevance between the sub-information and the sub-thresholds, judge whether the sub-information exceeds the sub-threshold based on the relevance to obtain a judgment result, extract the judgment result to feedback the size of the sub-information exceeding the sub-threshold and the sub-information to obtain an abnormal value and abnormal information, judge the level of the abnormal value through a level classification method based on the grade range and generate an alarm information to obtain the level information of the abnormal information, and integrate the abnormal information, abnormal value, and level information to obtain the monitoring information.

[0007] Furthermore, the process of presetting the target threshold and the grade range based on the status information is: Split the status information into several information items, determine the equipment items based on the information items, formulate the target threshold based on the equipment items, preset an added value, obtain a first value based on the target threshold combined with the added value, obtain a second value based on the first value combined with the added value, obtain a third value based on the second value combined with the added value. The range between the target threshold and the first value forms the first grade, the range between the first value and the second value forms the second grade, and the range between the second value and the third value forms the third grade.

[0008] Furthermore, the level classification method includes: The grade range is divided into the first grade, the second grade, and the third grade. Reduce the target threshold based on the boundaries of multiple grade items in the grade range to obtain a reduced grade. The reduced grade is divided into the first reduced grade, the second reduced grade, and the third reduced grade. Establish the relevance between the reduced grade and the grade range, judge the inclusion relationship between the abnormal value and multiple grade items in the reduced grade to obtain a classification result, obtain the target grade of the abnormal value based on the classification result in cooperation with the relevance, and generate alarm information of different intensities based on the target grade through an alarm method and output it to the user.

[0009] Further, the alarm method includes: presetting a basic alarm intensity and an increment value, adding the increment value to the basic alarm intensity to obtain a medium alarm intensity, adding the increment value to the medium alarm intensity to obtain a high alarm intensity, where the basic alarm intensity, the medium alarm intensity, and the high alarm intensity respectively correspond to the first level, the second level, and the third level, integrating the basic alarm intensity, the medium alarm intensity, and the high alarm intensity to obtain an alarm intensity set, selecting a corresponding alarm intensity from the alarm intensity set based on the target level to generate an alarm signal, and feeding back the alarm signal to the user based on the feedback address information.

[0010] Further, the analysis method includes: determining an abnormal device based on the abnormal information, obtaining the abnormal data and the cause of occurrence of the past event information, selecting corresponding abnormal data based on the abnormal device to obtain sub-data, presetting a similarity value, traversing the sub-data based on the abnormal information for similarity matching to obtain a matching result, when the matching result indicates that there is sub-data in the sub-data whose similarity to the abnormal information exceeds the similarity value, extracting the sub-data in the sub-data whose similarity exceeds the abnormal information to obtain selected data, determining the cause of occurrence based on the selected data to obtain a target cause, and obtaining the target impact based on the target cause.

[0011] Further, the classification method includes: extracting the types of weak links and emergency events in the past event information to obtain the reasons for weak links and emergency events, establishing a weak reason library for storing the reasons for weak links, establishing an emergency reason library for storing the reasons for emergency events, listening to the user's needs to obtain a demand result, when the demand result indicates that there is added information, processing the weak reason library and the emergency reason library based on the added information, traversing the processed weak reason library and emergency reason library based on the target cause to obtain a traversal result, and obtaining a classification result based on the traversal result.

[0012] Further, the weak prediction method includes: intercepting the information in the past event information where there are weak situations in the device operation to obtain reference information, presetting a time range, intercepting the operation information of the device items before the weakness based on the time range to obtain characteristic information, establishing an association relationship between the reference information and the characteristic information and the device, establishing a reference sub-library for storing the reference information and the characteristic information of a single device, establishing a reference library for storing the reference sub-libraries named after the devices, intercepting the operation information of the device items in the status information based on the time range to obtain matching information, presetting a similarity threshold, traversing the reference library based on the matching information to perform data matching with the characteristic information to obtain a matching result, when the matching result indicates that there is characteristic information in the reference library whose similarity to the matching information exceeds the similarity threshold, extracting the reference information to obtain a prediction result, and when the matching result indicates that there is no characteristic information in the reference library whose similarity to the matching information exceeds the similarity threshold, generating a prediction result that no weakness will occur.

[0013] Further, the target reasons and target impacts with the classification result of weak links are collected in real time to obtain pre-reference information and target devices. Based on the target devices, the corresponding reference sub-library is determined to obtain the storage path. Based on the time range, the events before the occurrence of weak links are intercepted to obtain reference events. Based on the storage path, the reference time and reference information are stored in the reference sub-library.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This method for identifying weak links and screening emergency events in the real-time operation of the power grid monitors the operation information of the power grid in real time through the set monitoring and processing method, extracts abnormal information in real time, alarms the abnormal information, and classifies the abnormal information before alarming it, so as to select different alarm intensities to remind the user. Through the set weak prediction method, according to the real-time operation state information of the power grid combined with the past event information, it predicts whether there will be and exist weak phenomena in the future operation of the equipment, and provides abnormal information to the user in advance, so that the user has enough time to deal with and solve the possible abnormal phenomena.

[0015] At the same time, in the weak prediction method, the past event information is used as a reference, and whether there is weakness is judged by matching the similarity. The overall implementation process is relatively simple and convenient. Through the set analysis method, the reasons and impacts for generating the monitoring information are obtained, so as to be fed back to the user later for the user to immediately process the abnormality. According to the classification of weak links and emergency events in the past event information, the reasons for generating the abnormality are classified through the classification method to obtain the classification result, so as to facilitate the user to judge and solve the abnormality.

[0016] At the same time, by collecting the target reasons and target impacts with the classification result of weak links and storing them in the corresponding reference sub-library according to the device, the number of information in the reference library in this stage of the weak prediction method can be enriched, so as to facilitate the weak prediction method to predict more information on possible weak situations, which is beneficial for use. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is the overall process structure diagram of the present invention; Figure 2 is the structure diagram of the monitoring and processing method of the present invention; Figure 3 is the structure diagram of the grade range of the present invention; Figure 4 is the structure diagram of the weak prediction method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] Timely identification of weak links can effectively reduce the probability of equipment failures, power outages and other incidents, ensuring the safety of power grid operation. Rapid screening and handling of emergency events can take prompt measures when problems occur, reducing the impact of accidents on the power grid and users. By monitoring and identifying potential risks, it is ensured that the power system can stably supply power under peak load and extreme weather conditions. Weak links can be identified to timely maintain and replace equipment, extend the service life of equipment, and reduce unexpected outages.

[0020] As Figures 1 - 4 shown, the present invention provides a technical solution: a method for identifying weak links in real-time operation of a power grid and screening emergency events, the method comprising: Information acquisition: acquiring the real-time operation state of the power grid, past event information, feedback address information, and user requirements, where the real-time operation state of the power grid is state information; It should be noted that in the information acquisition stage, the state information is the real-time operation state of the power grid, which includes parameters such as the load, voltage, frequency, and power supply quality of relevant equipment, and the data obtained by using intelligent sensors and monitoring devices can be used. The past event information is the log information during the past operation of the power grid, which includes state information, weak link information, and emergency event information. The feedback address information is the address information for sending information to the user, and the user requirements are the usage requirements of the user for weak links and emergency events in the power grid.

[0021] Embodiment 1: A large-scale urban music festival attracted hundreds of thousands of audiences, and the demand for the power grid increased sharply instantly. In this case, the power grid needs to monitor the real-time load, ensure stable power supply, and monitor any possible emergency events. In the above scenario, the method is used to identify weak links and screen emergency events. The state information is the real-time operation state of the power grid, including power supply information and the operation information of various devices during the music festival. Through the SCADA system and Internet of Things sensors, the operation states of various devices in the power grid are continuously monitored to obtain state information, including parameters such as voltage, frequency, load, temperature, and vibration. The past event information is the log information during past music festivals, and the feedback address information is the address information of the user. An information receiving interface can be established to receive the user's feedback, and at the same time, the feedback address information is recorded. The user requirements are fed back to the system through the information receiving interface.

[0022] Threshold establishment: Preset a target threshold and a level range based on status information; Information processing: Based on the target threshold, monitor and process the status information in real time through a monitoring and processing method to obtain monitoring information. Based on the monitoring information, obtain the cause and impact of generating the monitoring information through an analysis method to obtain the target cause and target impact; Abnormality classification: Classify the target cause through a classification method based on past event information in cooperation with user requirements to obtain a classification result; Real-time prediction: Obtain a prediction result through a vulnerability prediction method based on status information; Result output: Send the monitoring information, target cause, target impact, classification result, and prediction result to the user based on the feedback address information; It should be noted that in the threshold establishment stage, a target threshold and a level range are preset according to the status information. Through the target threshold, it is convenient to judge whether the status information is in an abnormal state. By setting the level range, the abnormal state is classified. In the information processing stage, through the set monitoring and processing method, the status information is monitored in real time to judge whether an abnormality occurs, and an alarm feedback corresponding to the abnormality is given to obtain the monitoring information. When an abnormality occurs in the status information feedback of the monitoring information, the cause and impact of generating the monitoring information are obtained through the set analysis method, so as to be subsequently fed back to the user for the user to immediately handle the abnormality. In the abnormality classification stage, based on the past event information, the weak links and emergency events are divided as the basis, and the cause of the abnormality is classified through the classification method to obtain the classification result, so as to facilitate the user to judge and solve the abnormality. In the real-time prediction stage, through the set vulnerability prediction method, according to the status information combined with the past event information, it is predicted whether there will be and exist weak phenomena in the future operation of the device, and abnormal information is provided to the user in advance, so as to have enough events to respond to and solve the possible weak abnormal phenomena. In the result output stage, the collected information is sent to the user according to the feedback address information, so as to facilitate the user to understand the results of the real-time monitoring of the power grid operation, and at the same time facilitate the user to respond to the weak links and emergency events.

[0023] Embodiment 2: Set the target threshold according to the operating status of the device load. The target threshold can also be obtained by retrieving the threshold of the same device in the past event information. The level range is the range further composed of the target threshold, and the specific size of the level range is determined according to the actual usage. For example, the rated load of a transformer is usually designed within 80% of its rated power. Its target threshold can be set to 80% of the transformer's rated power, and the level range can be set to 85%, 90%, and 95% of the transformer's rated power, corresponding to three levels in the low-level range. Through the SCADA system and Internet of Things sensors, continuously monitor the operating status of each device in the power grid, including parameters such as voltage, frequency, load, temperature, and vibration, to obtain real-time updated status information. Process the status information through the application of monitoring processing methods to obtain monitoring information, and process the detection information through analysis methods to obtain the causes and impacts of anomalies. Classify the anomalies according to the target reasons to determine whether they are weak links or emergency events to obtain classification results. Predict whether future devices will exhibit weak phenomena according to the weak prediction method to obtain prediction results. Send the collected information to the user according to the user's IP address or email to complete the closed-loop.

[0024] The monitoring processing method includes: splitting the status information into several sub-informations, splitting the target threshold into several sub-thresholds, establishing the correlation between the sub-informations and the sub-thresholds, judging whether the sub-informations exceed the sub-thresholds based on the correlation to obtain judgment results, extracting the judgment results to feedback the size of the sub-informations exceeding the sub-thresholds and the sub-informations to obtain anomaly values and anomaly information, judging the level of the anomaly values based on the level range through the level classification method and generating alarm information to obtain the level information of the anomaly information, and integrating the anomaly information, anomaly values, and level information to obtain monitoring information.

[0025] It should be noted that when the monitoring processing method is running, a time factor needs to be added, specifically the time node for real-time monitoring when splitting the status information, that is, the monitoring frequency. This process can be carried out through a preset monitoring frequency. This method has a relatively fast information processing speed, but there are flaws in accuracy. It is also possible to monitor the device in real time. In this way, the amount of information processed is relatively large, and thus the information processing speed will be slow and the efficiency is low. However, the accuracy of the results obtained is relatively high. Through the set monitoring processing method, continuously monitor the operating information of the power grid in real time, extract anomaly information in real time, and alarm the anomaly information. Before alarming the anomaly information, the anomaly information will be classified by level to facilitate selecting different alarm intensities to remind the user.

[0026] Such as Figure 3As shown in the figure, the process of presetting the target threshold and the level range based on the status information is as follows: the status information is split into several information items, the device items are determined based on the information items, the target threshold is formulated based on the device items, the added value is preset, the first value is obtained by combining the target threshold with the added value, the second value is obtained by combining the first value with the added value, the third value is obtained by combining the second value with the added value, the range between the target threshold and the first value forms the first level, the range between the first value and the second value forms the second level, and the range between the second value and the third value forms the third level.

[0027] It should be noted that the status information is composed of the information of multiple devices in the power grid. By splitting the status information, the information items of multiple device items are obtained, and the target threshold is formulated according to the device items. Specifically, the device item exceeding 50% of the rated power can be selected as the target threshold, and the added value is preset according to the actual use situation. For example, the added value is 5% of the rated power. At this time, the range of the first level is between exceeding 50% of the rated power and exceeding 55% of the rated power, the range of the second level is between exceeding 55% of the rated power and exceeding 60% of the rated power, and the range of the third level is between exceeding 60% of the rated power and exceeding 65% of the rated power. The specific size of the added value and the target threshold can be formulated by the staff according to the actual use requirements. Figure 3 In (a), it represents the target threshold and the added value. Figure 3 In (b), it represents that the first value is the target threshold plus the added value, and the first level is composed of the range between the target threshold and the first value. Figure 3 In (c), it represents that the second value is the first value plus the added value, and the second level is composed of the range between the first value and the second value. Figure 3 In (d), it represents that the third value is the second value plus the added value, and the third level is composed of the range between the second value and the third value.

[0028] As Figure 2 shown, the level classification method includes: the level range is divided into the first level, the second level and the third level. The reduced level is obtained by reducing the target threshold based on the boundaries of multiple level items in the level range. The reduced level is divided into the first reduced level, the second reduced level and the third reduced level. The relevance between the reduced level and the level range is established. The inclusion relationship between the outlier and multiple level items in the reduced level is judged to obtain the classification result. Based on the classification result and the relevance, the target level of the outlier is obtained. Different intensity alarm information is generated through the alarm method based on the target level and output to the user.

[0029] It should be noted that the process of obtaining the reduced level based on the target threshold of the reduction of multiple level item boundaries in the level range is to remove the target threshold from the ranges representing the first level, the second level, and the third level. That is, when the range of the first level is between 50% above the rated power and 55% above the rated power, the range of the second level is between 55% above the rated power and 60% above the rated power, the range of the third level is between 60% above the rated power and 65% above the rated power, and the target threshold is 50% above the rated power as the target threshold. At this time, the range of the first reduced level is between 0 and 5% of the rated power, the range of the second reduced level is between 5% and 10% of the rated power, and the range of the second reduced level is between 10% and 15% of the rated power. According to the inclusion relationship between the outlier and the reduced level, the classification result can be obtained. The level items represent the first reduced level, the second reduced level, and the third reduced level. Through the set level classification method, the abnormality can be classified according to the size of the abnormal data, so as to understand the degree of the abnormality.

[0030] The alarm method includes: presetting the basic alarm intensity and the increment value. The basic alarm intensity plus the increment value obtains the medium alarm intensity, and the medium alarm intensity plus the increment value obtains the high alarm intensity. The basic alarm intensity, the medium alarm intensity, and the high alarm intensity correspond to the first level, the second level, and the third level respectively. Integrating the basic alarm intensity, the medium alarm intensity, and the high alarm intensity to obtain the alarm intensity set. Selecting the corresponding alarm intensity from the alarm intensity set based on the target level to generate an alarm signal, and feeding back the alarm signal to the user based on the feedback address information.

[0031] It should be noted that the specific methods of the basic alarm intensity and the increment value can be set according to the actual usage situation. For example, the basic alarm intensity is to send a single alarm message to the user, and the increment value is to increase two alarm messages. At the same time, it can also be that the basic alarm intensity is to send alarm messages to 50 users within the power grid range, and the increment value is 10 users. The process of generating the alarm signal is specifically to extract the monitoring information for generation, so that the user can quickly understand the abnormal information and the abnormal part, and facilitate the quick solution of the abnormality.

[0032] The analysis method includes: determining the abnormal device based on the abnormal information, obtaining the abnormal data and the cause of generation of the past event information, selecting the corresponding abnormal data based on the abnormal device to obtain sub-data, presetting the similarity value, traversing the sub-data based on the abnormal information for similarity matching to obtain the matching result. When the matching result feedback indicates that there is sub-data in the sub-data whose similarity to the abnormal information exceeds the similarity value, extracting the sub-data in the sub-data whose similarity exceeds the abnormal information to obtain the selected data, determining the cause of generation based on the selected data to obtain the target cause, and obtaining the impact result based on the target cause to obtain the target impact.

[0033] It should be noted that the similarity value is set according to the actual usage. The higher the similarity value, the higher the accuracy of the matching result. The abnormal information is generated by the abnormal device. The abnormal device can be determined through the abnormal information. The process of obtaining the abnormal data and the cause of the past event information can be achieved by obtaining the working log of the past power grid operation.

[0034] As Figure 1 shown, the classification method includes: extracting the types of weak links and emergency events in the past event information to obtain the reasons for weak links and emergency events, establishing a weak reason library for storing the reasons for weak links, establishing an emergency reason library for storing the reasons for emergency events, listening to the user's needs to obtain the demand result. When the demand result feedback indicates the existence of added information, the weak reason library and the emergency reason library are processed based on the added information, the processed weak reason library and emergency reason library are traversed based on the target reason to obtain the traversal result, and the classification result is obtained based on the traversal result.

[0035] It should be noted that the past event information is the working log of the past operation of the power grid, which records the weak link information and emergency event information that occurred in the past, including the reasons for weak links and emergency events. At the same time, the user's needs are listened to, and the user's needs are added to the classification criteria. The classification result is obtained by traversing the classification criteria composed of the emergency reason library and the weak reason library based on the target reason to obtain the traversal result. Specifically, the traversal is to perform matching or similarity matching. When the matched information item is in the emergency reason library, the target reason is an emergency event. When the matched information item is in the weak reason library, the target reason is a weak link.

[0036] The weak prediction method includes: intercepting the information indicating that there are weak situations in the operation of the device in the past event information to obtain the reference information, presetting the time range, intercepting the operation information of the device item before the weakness based on the time range to obtain the characteristic information, establishing the association relationship between the reference information and the characteristic information and the device, establishing a reference sub-library for storing the reference information and the characteristic information of a single device, establishing a reference library for storing the reference sub-libraries named after the device, intercepting the operation information of the device item in the status information based on the time range to obtain the matching information, presetting the similarity threshold, traversing the reference library based on the matching information to perform data matching with the characteristic information to obtain the matching result. When the matching result feedback indicates that there is characteristic information in the reference library whose similarity to the matching information exceeds the similarity threshold, the reference information is extracted to obtain the prediction result. When the matching result feedback indicates that there is no characteristic information in the reference library whose similarity to the matching information exceeds the similarity threshold, it is predicted that no weakness will occur to obtain the prediction result.

[0037] It should be noted that the process of obtaining reference information by intercepting the information on the weak operation of the device in the past event information is to intercept the relevant information on the weak link in the past work log to obtain reference information. The time range is determined according to the actual use situation and represents the time length. The specific time range is the length of obtaining the characteristic information, which can be specifically the 10 minutes before the occurrence of the weak situation. The characteristic information is the operation situation of the device in the 10 minutes before the occurrence of the situation. By obtaining the past information as a reference and intercepting the information with the same length as the characteristic situation according to the real-time status information for similarity matching, it is judged whether there will be a weak situation according to the similarity matching result. The specific similarity threshold is set by itself according to the actual use situation. Generally, the higher the similarity threshold is set, the more accurate the obtained prediction result is. Figure 4 In the device 1, device 2, device 3... device n, they all represent specific device names, and the reference sub-library named after the device name.

[0038] Such as Figure 4 As shown, the target causes and target impacts with the classification result of weak links are collected in real time to obtain pre-reference information and target devices. Based on the target devices, the corresponding storage paths of the reference sub-libraries are determined. Based on the time range, the events before the occurrence of the weak link are intercepted to obtain reference events. Based on the storage paths, the reference time and reference information are stored in the reference sub-libraries.

[0039] It should be noted that by collecting the target causes and target impacts with the classification result of weak links and storing them in the corresponding reference sub-libraries according to the devices, the information quantity of the reference library in this stage of the weak prediction method can be enriched, so as to facilitate the weak prediction method to predict more information on possible weak situations, which is beneficial for use.

[0040] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirits of the present invention. The scope of the present invention is defined by the appended embodiments and their equivalents.

Claims

1. A method for identifying weak links in the real-time operation of a power grid and screening emergency events, characterized in that: The method includes: Information acquisition: acquiring the real-time operation status of the power grid, past event information, feedback address information, and user requirements, where the real-time operation status of the power grid is status information; Establishing thresholds: presetting a target threshold and a level range based on the status information; Information processing: monitoring and processing the status information in real time through a monitoring and processing method based on the target threshold to obtain monitoring information, and obtaining the cause and impact of the monitoring information through an analysis method based on the monitoring information to obtain the target cause and target impact; Abnormality classification: classifying the target cause through a classification method in combination with the user requirements based on the past event information to obtain a classification result; Real-time prediction: obtaining a prediction result through a vulnerability prediction method based on the status information; Result output: sending the monitoring information, target cause, target impact, classification result, and prediction result to the user based on the feedback address information; The monitoring and processing method includes: splitting the status information into several sub-information, splitting the target threshold into several sub-thresholds, establishing the relevance between the sub-information and the sub-thresholds, judging whether the sub-information exceeds the sub-threshold based on the relevance to obtain a judgment result, extracting the judgment result to obtain the size of the sub-information exceeding the sub-threshold and the sub-information as the abnormal value and abnormal information, judging the level of the abnormal value through a level classification method based on the level range and generating an alarm information to obtain the level information of the abnormal information, and integrating the abnormal information, abnormal value, and level information to obtain the monitoring information.

2. The method for identifying weak links in real-time operation of power grid and screening emergency events according to claim 1, wherein: The process of presetting the target threshold and the level range based on the status information is: splitting the status information into several information items, determining the equipment items based on the information items, formulating the target threshold based on the equipment items, presetting an added value, obtaining the first value by combining the target threshold with the added value, obtaining the second value by combining the first value with the added value, obtaining the third value by combining the second value with the added value, the range between the target threshold and the first value forms the first level, the range between the first value and the second value forms the second level, and the range between the second value and the third value forms the third level.

3. The method for identifying weak links in real-time operation of power grid and screening emergency events according to claim 1, characterized in that: The level classification method includes: the level range is divided into the first level, the second level, and the third level, reducing the target threshold based on multiple level item boundaries in the level range to obtain a reduced level, the reduced level is divided into the first reduced level, the second reduced level, and the third reduced level, establishing the relevance between the reduced level and the level range, judging the inclusion relationship between the abnormal value and multiple level items in the reduced level to obtain a classification result, obtaining the target level of the abnormal value based on the classification result in combination with the relevance, and generating alarm information of different intensities through an alarm method based on the target level and outputting it to the user.

4. The method for identifying weak links in real-time grid operation and screening emergency events according to claim 3, characterized in that: The alarm method includes: presetting a basic alarm intensity and an increment value, adding the increment value to the basic alarm intensity to obtain a medium alarm intensity, adding the increment value to the medium alarm intensity to obtain a high alarm intensity, the basic alarm intensity, medium alarm intensity, and high alarm intensity respectively correspond to the first level, the second level, and the third level, integrating the basic alarm intensity, medium alarm intensity, and high alarm intensity to obtain an alarm intensity set, selecting the corresponding alarm intensity in the alarm intensity set based on the target level to generate an alarm signal, and feeding back the alarm signal to the user based on the feedback address information.

5. The method for identifying weak links in real-time operation of power grid and screening emergency events according to claim 1, characterized in that: The analysis method includes: determining abnormal devices based on abnormal information, obtaining abnormal data and causes of past events, selecting corresponding abnormal data based on the abnormal devices to obtain sub-data, presetting a similarity value, traversing the sub-data based on the abnormal information for similarity matching to obtain a matching result. When the matching result indicates that there is sub-data in the sub-data whose similarity to the abnormal information exceeds the similarity value, extracting the sub-data in the sub-data whose similarity exceeds the abnormal information to obtain selected data, determining the cause based on the selected data to obtain the target cause, and obtaining the impact result based on the target cause to obtain the target impact.

6. The method for identifying weak links in the real-time operation of the power grid and screening emergency events according to claim 1, wherein: The classification method includes: extracting the types of weak links and emergency events in past event information to obtain weak link causes and emergency event causes, establishing a weak cause library for storing weak link causes, establishing an emergency cause library for storing emergency event causes, listening to user requirements to obtain a requirement result. When the requirement result indicates that there is added information, processing the weak cause library and the emergency cause library based on the added information, traversing the processed weak cause library and emergency cause library based on the target cause to obtain a traversal result, and obtaining a classification result based on the traversal result.

7. The method for identifying weak links in real-time grid operation and screening emergency events according to claim 1, characterized in that: The weak prediction method includes: intercepting the information indicating that there are weak situations in the operation of devices in past event information to obtain reference information, presetting a time range, intercepting the operation information of device items before the weakness occurs based on the time range to obtain feature information, establishing an association relationship between the reference information and the feature information and the device, establishing a reference sub-library for storing the reference information and feature information of a single device, establishing a reference library for storing reference sub-libraries named after devices, intercepting the operation information of device items in the status information based on the time range to obtain matching information, presetting a similarity threshold, traversing the reference library based on the matching information to perform data matching with the feature information to obtain a matching result. When the matching result indicates that there is feature information in the reference library whose similarity to the matching information exceeds the similarity threshold, extracting the reference information to obtain a prediction result. When the matching result indicates that there is no feature information in the reference library whose similarity to the matching information exceeds the similarity threshold, generating a prediction result indicating that no weakness will occur.

8. The method for identifying weak links in real-time grid operation and screening emergency events according to claim 1, wherein: Real-time collect the target cause and target impact whose classification result is a weak link to obtain pre-reference information and target devices, determine the corresponding reference sub-library based on the target devices to obtain a storage path, intercept the events before the weak link occurs based on the time range to obtain reference events, and store the reference time and reference information in the reference sub-library based on the storage path.

Images