Power grid accident rule determination method, device, equipment, medium and product

Abstract

The invention discloses a power grid accident rule determination method and device, equipment, a medium and a product, and relates to the technical field of power grid accident safety, and the method comprises the steps: determining the historical operation information of a to-be-detected power grid enterprise; determining historical operation data corresponding to the historical operation type of which the operation type is abnormal as intermediate accident data; processing the intermediate accident data by using a pre-trained target data completion model to obtain target accident data of the to-be-detected power grid enterprise; and processing the at least one accident, the accident time information corresponding to the at least one accident and the environment information to obtain an accident occurrence rule of the to-be-detected power grid enterprise. According to the invention, the missing accident information in the historical operation information of the to-be-detected power grid enterprise can be complemented by using the target data complementation model, and the accident rule of the to-be-detected power grid enterprise is determined according to the complemented accident information, so that the accuracy of the accident rule is improved.

Description

Technical Field

[0001] The present invention relates to the field of power grid accident safety technology, and in particular to a method, device, equipment, medium and product for determining power grid accident rules. Background Art

[0002] Due to the particularity of the power industry, the mortality rate of power accidents is much higher than that of other types of work-related accidents, which seriously endangers the personal safety of power grid practitioners and the safe production of related companies. Therefore, the law of power grid accidents is one of the issues that has attracted much attention in the power field.

[0003] At present, the laws of power grid accidents are determined by analyzing historical accident data. For example, by analyzing the multi-dimensional accident data of the power grid, a personal safety early warning model is established to prevent, stop and deal with power grid accidents in a timely manner. However, an accurate early warning model depends on complete and comprehensive accident data. Once the accident data is missing, the amount of accident data is limited, or there are differences in the dimensions of multiple accident data records, the accuracy of the prediction results of the personal safety early warning model will be reduced, and even the established personal safety early warning model will be unable to predict power grid accidents. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and product for determining the law of power grid accidents. The method first uses a target data completion model to complete the missing accident information in the historical operation information of the power grid enterprise to be detected, and then determines the accident law of the power grid enterprise to be detected based on the completed accident information, thereby effectively improving the accuracy of the accident law.

[0005] According to one aspect of the present invention, a method for determining a power grid accident law is provided, the method comprising:

[0006] Determine historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type and the historical operation data correspond one to one;

[0007] Determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data;

[0008] Using the pre-trained target data completion model, the intermediate accident data is processed to obtain the target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, and the accident time information and environmental information corresponding to at least one accident;

[0009] At least one accident, accident time information and environmental information corresponding to at least one accident are processed to obtain the accident occurrence pattern of the power grid enterprise to be detected.

[0010] Optionally, before using a pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be tested, it also includes: using a preset information extraction method to obtain a power grid accident information corpus, a power accident compilation and a power safety event compilation; processing the intermediate accident data, the power grid accident information corpus, the power accident compilation and the power safety event compilation to obtain a test data set and a verification data set; using the data test set to optimize the initial data completion model to obtain an intermediate data completion model; verifying the intermediate data completion model based on the verification data set, and determining the intermediate data completion model as the target data completion model when the verification result is qualified.

[0011] Optionally, a pre-trained target data completion model is used to process the intermediate accident data to obtain target accident data of the power grid enterprise to be detected, including: determining whether there is a key data missing anomaly in the intermediate accident data, wherein the intermediate accident data includes at least one sub-intermediate accident data; if the intermediate accident data does not have a key data missing anomaly, then determining the intermediate accident data to be the target accident data; if the intermediate accident data has a key data missing anomaly, then determining the sub-intermediate accident data with the key data missing anomaly to be the accident data to be completed, and using the target data completion model to process the accident data to be completed to obtain the target accident data.

[0012] Optionally, the target data completion model is used to process the accident data to be completed to obtain the target accident data, including: determining a data missing identifier and a data completion scheme for the accident data to be completed, wherein the data missing identifier and the data completion scheme correspond one-to-one, and the data missing identifier is at least one of a time information missing identifier and an environmental information missing identifier; determining the missing data of the accident data to be completed using the target data completion model and the data completion scheme; and adding the missing data to the missing position of the data to be completed to obtain the target accident data.

[0013] Optionally, the environmental information includes maximum temperature, minimum temperature, wind direction information and clearness information; at least one accident, the accident time information corresponding to at least one accident and the environmental information are processed to obtain the accident occurrence pattern of the power grid enterprise to be detected, including: sorting at least one accident in order from early to late according to the accident time information; determining the accident occurrence pattern based on the maximum temperature, minimum temperature, wind direction information, clearness information, holiday information, quarterly information and the interval duration of each accident of at least one accident.

[0014] Optionally, after obtaining the accident occurrence pattern of the power grid enterprise to be detected, it also includes: determining the current environmental information and current time information of the power grid enterprise to be detected; based on the current time information, current environmental information and the accident occurrence pattern, determining whether the accident occurrence probability of the power grid enterprise to be detected at the current moment is greater than or equal to a preset probability; if the accident occurrence probability is greater than or equal to the preset probability, generating a warning message.

[0015] According to another aspect of the present invention, a device for determining a power grid accident law is provided. The device for determining a power grid accident law is used to implement the method for determining a power grid accident law in any embodiment of the present invention. The device includes:

[0016] An information acquisition module, used to determine the historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type and the historical operation data correspond one to one;

[0017] An abnormality determination module, used to determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data;

[0018] A data determination module is used to process the intermediate accident data using a pre-trained target data completion model to obtain target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, accident time information and environmental information corresponding to at least one accident;

[0019] The rule determination module is used to process at least one accident, the accident time information and the environmental information corresponding to at least one accident, and obtain the accident occurrence rule of the power grid enterprise to be detected.

[0020] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0021] at least one processor; and a memory communicatively coupled to the at least one processor;

[0022] The memory stores a computer program executable by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the method for determining the law of power grid accidents in any embodiment of the present invention.

[0023] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for implementing the method for determining a power grid accident law in any embodiment of the present invention when executed by a processor.

[0024] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the method for determining a power grid accident law according to any embodiment of the present invention is implemented.

[0025] The method for determining the law of power grid accidents of the present invention comprises: determining the historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data one by one; determining the historical operation data corresponding to the historical operation type with abnormal operation type as the intermediate accident data; using the pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, the accident time information and environmental information corresponding to at least one accident; processing at least one accident, the accident time information and environmental information corresponding to at least one accident to obtain the law of accident occurrence of the power grid enterprise to be detected. The technical scheme of the present invention can use the target data completion model to complete the missing accident information in the historical operation information of the power grid enterprise to be detected, and determine the accident law of the power grid enterprise to be detected according to the completed accident information, so as to improve the accuracy of the determined accident law. It solves the problem that when the accident data is missing, the amount of accident data is limited, or there are differences in the recording dimensions of multiple accident data, the prediction results of the personal safety early warning model established based on the multi-dimensional accident data of the power grid are low in accuracy, and even the power grid accident cannot be predicted.

[0026] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 It is a flow chart of a method for determining a power grid accident law provided by the present invention;

[0029] Figure 2 It is a flow chart of another method for determining the law of power grid accidents provided by the present invention;

[0030] Figure 3 It is a structural schematic diagram of a device for determining a power grid accident law provided by the present invention;

[0031] Figure 4 It is a structural schematic diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] Figure 1 This is a flow chart of a method for determining the law of power grid accidents provided by the present invention. This embodiment can be applied to determining the law of occurrence of power grid accidents, preventing power grid accidents, reducing the impact of power grid accidents, etc. The method can be executed by the device for determining the law of power grid accidents provided by the present invention. The device can be implemented in the form of hardware and / or software. In a specific embodiment, the device can be integrated in an electronic device. The following embodiments will be described by taking the device integrated in an electronic device as an example. Figure 1 , the method specifically comprises the following steps:

[0035] S101. Determine the historical operation information of the power grid enterprise to be detected.

[0036] Among them, the power grid enterprise to be detected can be understood as a power grid company that needs to summarize the accident patterns, and the historical operation information can be understood as the operation data of the power grid enterprise to be detected within a preset time period. The preset time period can be related to the query requirements based on the accident patterns, and this embodiment does not limit this.

[0037] Specifically, the historical operation information may be the operation information of the power grid enterprise to be tested in the past six months, one year, two years, three years or five years. The operation information may be summarized in units of days or hours. In units of days, the operation information includes at least one daily operation state, the time information corresponding to each daily operation state and the environment information.

[0038] The historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data. In a specific example, assuming that the operation information is summarized in days, the historical operation type can be understood as the daily operation status, and the historical operation data can be understood as the time information and environmental information corresponding to the daily operation status. The time information includes but is not limited to date, holiday information, weekend information within the week, quarterly information, and location information in the year. The environmental information includes but is not limited to temperature range, sunny or rainy information, light intensity information, light information, and wind direction information.

[0039] The advantage of this setting is that complete and comprehensive operating data of the power grid enterprise to be tested can be obtained, so as to summarize the accident occurrence patterns of the power grid enterprise to be tested.

[0040] S102: Determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data.

[0041] During operation, a power grid enterprise may or may not have a safety accident. In units of days, the operating status information of the day when a safety accident occurs is abnormal operation, and the operating status information of the day when no safety accident occurs is normal operation. Among them, the operation type of abnormal operation can be understood as the occurrence of a safety accident, and the intermediate accident data can be understood as the accident data in the historical operation information of the power grid enterprise to be detected, that is, the time information and environmental information of the date when the safety accident occurred in the power grid enterprise to be detected.

[0042] Specifically, determining that the historical operation data corresponding to the historical operation type with the operation type of abnormal operation is the intermediate accident data can be understood as sorting out the historical operation types in the historical operation information, determining the historical operation type with the operation type of abnormal operation as the candidate operation type, and determining the candidate operation type and the operation data of the candidate operation type as the intermediate accident data. It is worth noting that the candidate operation type includes at least one abnormal operation type.

[0043] The advantage of this setting is that the accident information of the power grid enterprise to be detected can be quickly determined, thereby improving the efficiency of determining the accident patterns.

[0044] S103: Using the pre-trained target data completion model, the intermediate accident data is processed to obtain the target accident data of the power grid enterprise to be detected.

[0045] Since the requirements of each power grid company for the dimensions of accident records are not exactly the same or the power grid company will regularly clean up the accident information, when integrating the accident data of the power grid company to be detected, there may be a problem of missing data in some dimensions. Among them, the target data completion model can be understood as a pre-trained algorithm for completing missing dimensional data, and the target accident data can be understood as the complete accident data of the power grid company to be monitored.

[0046] Specifically, the target accident data includes at least one accident, the accident time information and environmental information corresponding to at least one accident. The accident time information can be understood as the date of the accident, holiday information on that date, weekend information within the week, quarterly information and location information in the year; the environmental information can be understood as the temperature range, maximum temperature information, minimum temperature information, average temperature information, sunny or rainy information, light intensity information, light information and wind direction information on the day or hour of the accident.

[0047] The target data completion model has the ability to infer contextual semantics and obtain information from web crawlers. It can retrieve the contextual data corresponding to the actual dimension data to infer the missing dimension data, or retrieve the record of the missing data on the web page and parse the queried page information to obtain the specific dimension data.

[0048] For example, assuming that accident data A lacks the temperature information at time A on day X, the target data completion model can be used to retrieve the temperature values ​​at two times before and after time A on day X, and the average of the two temperature values ​​can be determined as the temperature information at time A. The target data completion model can also be used to open a query web page, query the temperature information at time A on day X, and parse the temperature value in the query interface to determine that the parsed temperature value is the temperature information at time A.

[0049] The advantage of this setting is that comprehensive and complete accident data can be obtained, so as to more accurately summarize the accident patterns of the power grid companies to be tested.

[0050] S104: Process at least one accident, the accident time information and the environmental information corresponding to at least one accident to obtain the accident occurrence pattern of the power grid enterprise to be detected.

[0051] Among them, the accident occurrence pattern of the power grid enterprise to be detected can be understood as the correlation relationship between the accidents, including but not limited to time intervals, temperature exceeding a certain temperature value, light intensity exceeding a certain intensity value, rainfall per hour greater than A mm, etc. This embodiment does not limit this.

[0052] For example, assuming that the grid company to be tested has an accident on the Xth day of every month, it is determined that the accident occurrence pattern of the grid company to be tested is once a month, and the time of the next accident can be predicted based on the time of the last accident, thereby helping the grid company to be tested avoid safety accidents. If the temperature information of the accidents of the grid company to be tested is greater than A degrees Celsius, it is determined that the accident occurrence pattern of the grid company to be tested is that safety accidents occur when the temperature is greater than A degrees Celsius, and the way of monitoring meteorological information can be used to help the grid company to be tested avoid safety accidents.

[0053] The advantage of such a setting is that it can assist the power grid enterprises to be inspected to avoid safety accidents or reduce the frequency of safety accidents, thereby improving the safety and efficiency of the power grid enterprises to be inspected.

[0054] The technical solution of this embodiment can use the target data completion model to complete the missing accident information in the historical operation information of the power grid enterprise to be detected, and determine the accident law of the power grid enterprise to be detected based on the completed accident information, thereby improving the accuracy of the determined accident law. This solves the problem that when the accident data is missing, the amount of accident data is limited, or there are differences in the dimensions of multiple accident data records, the prediction results of the personal safety early warning model established based on the multi-dimensional accident data of the power grid are low in accuracy, or even unable to predict power grid accidents.

[0055] Figure 2 This is a flow chart of another method for determining the law of power grid accidents provided by the present invention. This embodiment can be applied to determining the law of occurrence of power grid accidents, preventing power grid accidents, reducing the impact of power grid accidents, etc. The method can be executed by the device for determining the law of power grid accidents provided by the present invention. The device can be implemented in the form of hardware and / or software. In a specific embodiment, the device can be integrated in an electronic device. The following embodiments will be described by taking the device integrated in an electronic device as an example. Figure 2 , the method specifically comprises the following steps:

[0056] S201. Determine the historical operation information of the power grid enterprise to be detected.

[0057] In this embodiment, the power grid company to be detected can be understood as a power grid company that needs to summarize the accident rules, and the historical operation information can be understood as the operation data of the power grid company to be detected within a preset time period. The preset time period can be related to the query requirements of the accident rules, and this embodiment does not limit this. Specifically, the historical operation information can be the operation information of the power grid company to be detected in the past six months, one year, two years, three years or five years. The operation information can be summarized in days or hours. In days, the operation information includes at least one daily operation state, the time information corresponding to each day's operation state, and environmental information.

[0058] The historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data. In a specific example, assuming that the operation information is summarized in days, the historical operation type can be understood as the daily operation status, and the historical operation data can be understood as the time information and environmental information corresponding to the daily operation status. The time information includes but is not limited to date, holiday information, weekend information within the week, quarterly information, and location information in the year. The environmental information includes but is not limited to temperature range, sunny or rainy information, light intensity information, light information, and wind direction information.

[0059] S202: Determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data.

[0060] In this embodiment, the operation type of abnormal operation can be understood as the occurrence of a safety accident, and the intermediate accident data can be understood as the accident data in the historical operation information of the power grid enterprise to be detected, that is, the time information and environmental information of the date when the safety accident occurred in the power grid enterprise to be detected. Specifically, determining the historical operation data corresponding to the historical operation type of abnormal operation as the intermediate accident data can be understood as sorting out the historical operation types in the historical operation information, determining the historical operation type of abnormal operation as the candidate operation type, and determining the candidate operation type and the operation data of the candidate operation type as the intermediate accident data.

[0061] S203: Using the pre-trained target data completion model, the intermediate accident data is processed to obtain the target accident data of the power grid enterprise to be detected.

[0062] In this embodiment, the target data completion model can be understood as a pre-trained algorithm for completing missing dimensional data, and the target accident data can be understood as the complete accident data of the power grid enterprise to be monitored. Specifically, the target accident data includes at least one accident, at least one accident corresponding to the accident time information and environmental information. The accident time information can be understood as the date of the accident, the holiday information of the date, the weekend information within the week, the quarterly information and the location information in the year; the environmental information can be understood as the temperature range, the highest temperature information, the lowest temperature information, the average temperature information, the sunny and rainy information, the light intensity information, the brightness information and the wind direction information of the day or hour of the accident. Furthermore, the target data completion model has the ability to push context semantics and the ability to obtain web crawler information. The context data corresponding to the actual dimensional data can be retrieved to infer the missing dimension data, or the record of the missing data in the web page can be retrieved, and the queried page information can be parsed to obtain specific dimension data.

[0063] In one embodiment, S203 may specifically include: determining whether there is a key data missing anomaly in the intermediate accident data, wherein the intermediate accident data includes at least one sub-intermediate accident data; if the intermediate accident data does not have a key data missing anomaly, determining that the intermediate accident data is the target accident data; if the intermediate accident data has a key data missing anomaly, determining that the sub-intermediate accident data with the key data missing anomaly is the accident data to be completed, and using the target data completion model to process the accident data to be completed to obtain the target accident data.

[0064] Among them, the key data missing anomaly can be understood as the lack of data used to determine the law of accident occurrence in the sub-intermediate accident data. In the present invention, the data used to determine the law of accident occurrence includes the specific time information (i.e. date) of the accident, whether the date of the accident is a weekend, whether it is a holiday, whether it is close to a holiday, which quarter, which month, which week, which day of the year, the highest temperature, the lowest temperature, the weather and the wind direction on the day of the accident. If the above twelve dimensional data are missing, it is determined that the intermediate accident data has a key data missing anomaly. If the above twelve dimensional data are all present, it is determined that the intermediate accident data does not have a key data missing anomaly. The accident data to be completed can be understood as the data value of the dimensional data that needs to be completed in the sub-intermediate accident data. The advantage of this setting is that the intermediate accident data of the power grid enterprise to be detected can be detected first, and only the intermediate accident data with key data missing anomalies can be completed, saving the workload of the processor and improving the efficiency of determining the law of the accident.

[0065] Furthermore, the target data completion model is used to process the accident data to be completed to obtain the target accident data, including: determining the data missing identifier and the data completion scheme of the accident data to be completed, wherein the data missing identifier and the data completion scheme correspond one to one, and the data missing identifier is at least one of the time information missing identifier and the environmental information missing identifier; using the target data completion model and the data completion scheme, determining the missing data of the accident data to be completed; adding the missing data to the missing position of the data to be completed to obtain the target accident data.

[0066] Different missing data have different methods for completion. The missing data mark can be understood as the attribute information of the missing data, including the number, name, abbreviation, etc. of the missing data, which is used to distinguish the type of missing data. The data completion scheme can be understood as the way to complete the missing data, including but not limited to context inference and web crawler determination. The missing time information mark includes but is not limited to the date mark, week mark, quarter mark and holiday mark. The missing environmental information mark includes but is not limited to the maximum temperature mark, the minimum temperature mark, the weather mark and the wind direction mark. In general, each of the twelve data dimensions has its corresponding missing mark.

[0067] Exemplarily, using the target data completion model and the data completion scheme to determine the missing data of the accident data to be completed can be understood as using the target data completion model to determine the missing data of the accident data to be completed according to the data completion scheme. For example, when the data completion scheme is context inference, the context of the accident data to be completed is analyzed using the target data completion model to determine the missing data of the accident data to be completed. When the data completion scheme is web crawler acquisition, the web page address and query content of the accident data to be completed are determined using the target data completion model, the web page address is accessed and the query content is entered to obtain the query interface, and the results of the query interface are analyzed to obtain the missing data of the accident data to be completed. Exemplarily, Table 1 gives examples of several completed target accident data.

[0068] Table 1

[0069] ACCIDENT 1 2 3 … date A B C … Day of the week three six two … Is it a holiday? yes yes no … Holiday Name Qingming Mid-Autumn Festival / … Which quarter two three Four … Month Four Nine eleven … Week 14 35 44 … Day 95 138 305 … Maximum temperature / ℃ 19 26 16 … Minimum temperature / ℃ 11 18 9 … wind direction north South southwest … weather light rain partly cloudy Light Snow …

[0070] The detailed data of the accident can be seen intuitively from Table 1.

[0071] Optionally, before using the pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be detected, it also includes: training the target data completion model. The pre-training method can greatly save the time of the accident law determination process. Training the target data completion model includes: using a preset information extraction method to obtain the power grid accident information corpus, the power accident compilation and the power safety event compilation; processing the intermediate accident data, the power grid accident information corpus, the power accident compilation and the power safety event compilation to obtain a test data set and a verification data set; using the data test set to optimize the initial data completion model to obtain an intermediate data completion model; verifying the intermediate data completion model based on the verification data set, and determining the intermediate data completion model as the target data completion model when the verification result is qualified.

[0072] Among them, the initial data completion model can be understood as a completion algorithm with data filling capabilities, the power grid accident information corpus can be understood as the representation of power grid accidents, the power accident compilation can be understood as the summary information of accidents in the power grid field, and the power safety incident compilation can be understood as the summary information of safety accidents that occurred in the power grid field. The power grid accident information corpus, power accident compilation and power safety incident compilation are used to help the initial data completion model learn knowledge in the power grid field, so that it can interpret the language of the power grid field and have the ability to complete missing data in the power grid field. The test data set can be understood as the samples used to train the initial data completion model, and the verification data set can be understood as the samples used to verify whether the trained initial data completion model is qualified.

[0073] Specifically, the intermediate accident data, power grid accident information corpus, power accident compilation and power safety incident compilation can be summarized to obtain a data set, and part of the data in the data set or part of the information in all the data can be fuzzy processed to obtain a test data set for training the initial data completion model. The data set that has not undergone information fuzzification is the verification data set, which is used to verify whether the data completion capability of the trained initial data completion model (i.e., the intermediate data completion model) is qualified. If the data after completion by the intermediate data completion model is the same as the verification test set, it proves that the data completion capability of the intermediate data completion model is qualified.

[0074] Furthermore, the intermediate accident data, power grid accident information corpus, power accident compilation and power safety incident compilation can be divided into a certain proportion to obtain training samples and test samples. The training samples are used to train the initial data completion model to obtain the intermediate data completion model; the fuzzy data of the test sample is input into the intermediate data completion model for completion, and it is determined whether the completion result is the same as the clear data of the test sample. If they are the same, the intermediate data completion model is determined to be qualified.

[0075] The advantage of this setting is that the initial data completion model can be optimized by combining the accident data of the power grid enterprise to be tested, the power grid accident information corpus, the power accident compilation and the power safety incident compilation, and it can be trained into an algorithm that is more suitable for the power grid enterprise to be tested and can better complete the missing data of the power grid enterprise to be tested.

[0076] S204. Sort at least one accident according to the accident time information in order from early to late.

[0077] Specifically, sorting at least one accident in order from early to late according to the accident time information can be understood as arranging the accidents based on the occurrence time, so as to determine the interval length of each accident and then analyze the time relationship between each accident.

[0078] S205. Determine the pattern of accident occurrence based on the maximum temperature, minimum temperature, wind direction information, clearness information, holiday information, season information and the intervals between accidents of at least one accident.

[0079] In this embodiment, based on the maximum temperature, minimum temperature, wind direction information, clearness information, holiday information, quarterly information and the interval duration between accidents of at least one accident, the pattern of accident occurrence is determined. This can be understood as summarizing the maximum temperature, minimum temperature, wind direction information, clearness information, holiday information, quarterly information and the interval duration between accidents to determine the correlation between the accidents.

[0080] For example, if all accidents occur on holidays, then the accident occurrence pattern is determined to be holidays; if accidents occur on the Xth day of each month, that is, the interval between each accident is 30 days, then the accident occurrence pattern of the power grid enterprise to be tested is determined to be once a month; if the temperature information corresponding to the accidents of the power grid enterprise to be tested is greater than A degrees Celsius, then the accident occurrence pattern of the power grid enterprise to be tested is determined to be that safety accidents occur when the temperature is greater than A degrees Celsius.

[0081] Optionally, after obtaining the accident occurrence pattern of the power grid enterprise to be detected, it also includes: determining the current environmental information and current time information of the power grid enterprise to be detected; based on the current time information, current environmental information and the accident occurrence pattern, determining whether the accident occurrence probability of the power grid enterprise to be detected at the current moment is greater than or equal to a preset probability; if the accident occurrence probability is greater than or equal to the preset probability, generating a warning message.

[0082] Among them, the preset probability can be understood as a quantitative indicator to judge whether an accident will occur, for example, 80%, 90%, etc. If the probability of an accident at the current moment is greater than or equal to the preset probability, it proves that an accident may occur, and a warning message is generated to prompt the power grid enterprise to be tested to take timely response measures, reduce the risk of accidents at any time, and ensure the safety of personnel and equipment of the power grid enterprise to be tested.

[0083] For example, assuming that the accident occurrence pattern of the power grid enterprise to be tested is that an accident occurs on the 4th of each month when the maximum temperature exceeds 35°C, if the current date is the 4th, the current environmental information is 32°C and the current time is morning, and the temperature is on an upward trend, then the probability of an accident is determined to be 93%, and a warning message will be generated to prompt the power grid enterprise to be tested to take timely response measures; if the current date is the 4th, the current environmental information is 32°C, and the current time is morning, but the weather on that day is greater than that, and the temperature is on a downward trend, then the probability of an accident is determined to be 45%, and it will be considered that the power grid enterprise to be tested does not have the conditions for a safety accident to occur, that is, there are no safety hazards.

[0084] The technical solution of this embodiment uses the target data completion model to complete the missing accident information in the historical operation information of the power grid enterprise to be detected, and determines the accident law of the power grid enterprise to be detected based on the highest temperature, lowest temperature, wind direction information, clearness information, holiday information, quarterly information and the interval duration of each accident in the completed accident information, thereby improving the accuracy of the determined accident law. This solves the problem that when accident data is missing, the amount of accident data is limited, or there are differences in multiple accident data record dimensions, the prediction results of the personal safety early warning model established based on the multi-dimensional accident data of the power grid are low in accuracy, or even unable to predict power grid accidents.

[0085] Figure 3Schematic diagram of the structure of a device for determining the law of power grid accidents provided by the present invention. Figure 3 As shown, the device includes: an information acquisition module 301, an abnormality determination module 302, a data determination module 303 and a rule determination module 304.

[0086] The information acquisition module 301 is used to determine the historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data one by one.

[0087] The abnormality determination module 302 is used to determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data.

[0088] The data determination module 303 is used to process the intermediate accident data using the pre-trained target data completion model to obtain the target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, and the accident time information and environmental information corresponding to at least one accident.

[0089] The rule determination module 304 is used to process at least one accident, the accident time information and the environmental information corresponding to at least one accident, and obtain the accident occurrence rule of the power grid enterprise to be detected.

[0090] Optionally, the device also includes a model training module, which is used to use a preset information extraction method to obtain power grid accident information corpus, power accident compilation and power safety event compilation before using a pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be detected; process the intermediate accident data, power grid accident information corpus, power accident compilation and power safety event compilation to obtain a test data set and a verification data set; optimize the initial data completion model using the data test set to obtain an intermediate data completion model; verify the intermediate data completion model based on the verification data set, and determine the intermediate data completion model as the target data completion model when the verification result is qualified.

[0091] Optionally, the data determination module 303 is specifically used to determine whether there is a key data missing anomaly in the intermediate accident data, wherein the intermediate accident data includes at least one sub-intermediate accident data; if the intermediate accident data does not have a key data missing anomaly, the intermediate accident data is determined to be the target accident data; if the intermediate accident data has a key data missing anomaly, the sub-intermediate accident data with the key data missing anomaly is determined to be the accident data to be completed, and the target data completion model is used to process the accident data to be completed to obtain the target accident data.

[0092] Optionally, the data determination module 303 is specifically used to determine the data missing identifier and data completion plan of the accident data to be completed, wherein the data missing identifier and the data completion plan correspond one to one, and the data missing identifier is at least one of the time information missing identifier and the environmental information missing identifier; using the target data completion model and the data completion plan, determine the missing data of the accident data to be completed; add the missing data to the missing position of the data to be completed to obtain the target accident data.

[0093] Optionally, the environmental information includes maximum temperature, minimum temperature, wind direction information and clearness information.

[0094] Optionally, the pattern determination module 304 is specifically used to sort at least one accident in order from morning to night according to the accident time information; based on the maximum temperature, minimum temperature, wind direction information, clearness information, holiday information, quarterly information and the interval duration of each accident of at least one accident, determine the pattern of accident occurrence.

[0095] Optionally, the rule determination module 304 is also used to determine the current environmental information and current time information of the power grid enterprise to be detected after obtaining the accident occurrence law of the power grid enterprise to be detected; based on the current time information, current environmental information and accident occurrence law, determine whether the accident occurrence probability of the power grid enterprise to be detected at the current moment is greater than or equal to the preset probability; if the accident occurrence probability is greater than or equal to the preset probability, generate a warning message.

[0096] The apparatus for determining the law of power grid accidents provided in this embodiment can execute the method for determining the law of power grid accidents provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0097] Figure 4 : is a schematic diagram of the structure of an electronic device provided by the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0098] like Figure 4As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (also called random access memory, RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, ROM 12 and RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0099] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0100] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for determining the law of power grid accidents.

[0101] In some embodiments, the method for determining the law of power grid accidents can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining the law of power grid accidents described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for determining the law of power grid accidents in any other appropriate manner (for example, by means of firmware).

[0102] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0104] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0105] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0106] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server) or a computing system that includes middleware components (e.g., an application server) or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein) or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0107] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0108] In one embodiment, the present invention further includes a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the method for determining the law of power grid accidents of any embodiment of the present invention.

[0109] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0110] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0111] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining the law of power grid accidents, characterized in that: include: Determine historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data in a one-to-one manner; Determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data; Using a pre-trained target data completion model, the intermediate accident data is processed to obtain target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, accident time information and environmental information corresponding to the at least one accident; The at least one accident, the accident time information and the environmental information corresponding to the at least one accident are processed to obtain the accident occurrence pattern of the power grid enterprise to be detected.

2. The method according to claim 1, characterized in that Before using the pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be detected, the method further includes: Using the preset information extraction method, obtain the power grid accident information corpus, power accident compilation and power safety incident compilation; Processing the intermediate accident data, the power grid accident information corpus, the power accident compilation and the power safety event compilation to obtain a test data set and a verification data set; Optimizing the initial data completion model using the data test set to obtain an intermediate data completion model; The intermediate data completion model is verified based on the verification data set, and when a verification result is qualified, the intermediate data completion model is determined to be the target data completion model.

3. The method according to claim 1, characterized in that The method of using the pre-trained target data completion model to process the intermediate accident data to obtain the target accident data of the power grid enterprise to be detected includes: Determining whether there is a key data missing anomaly in the intermediate accident data, wherein the intermediate accident data includes at least one sub-intermediate accident data; If the intermediate accident data does not have the key data missing anomaly, determining the intermediate accident data as the target accident data; If the intermediate accident data has the key data missing anomaly, the sub-intermediate accident data with the key data missing anomaly is determined to be the accident data to be completed, and the target data completion model is used to process the accident data to obtain the target accident data.

4. The method according to claim 3, characterized in that The using the target data completion model to process the accident data to be completed to obtain the target accident data includes: Determine a data missing identifier and a data completion scheme for the accident data to be completed, wherein the data missing identifier and the data completion scheme correspond to each other one by one, and the data missing identifier is at least one of a time information missing identifier and an environmental information missing identifier; Determine the missing data of the accident data to be completed by using the target data completion model and the data completion scheme; The missing data is added to the missing position of the data to be completed to obtain the target accident data.

5. The method according to claim 1, characterized in that The environmental information includes maximum temperature, minimum temperature, wind direction information and clearness information; The processing of the at least one accident, the accident time information and the environment information corresponding to the at least one accident to obtain the accident occurrence rule of the to-be-detected power grid enterprise includes: Sorting the at least one accident according to the order of accident time information from early to late; The accident occurrence pattern is determined based on the maximum temperature, the minimum temperature, the wind direction information, the clearness information, the holiday information, the season information and the interval duration between each accident of the at least one accident.

6. The method according to claim 1, characterized in that After obtaining the accident occurrence rule of the power grid enterprise to be detected, the method further includes: Determine the current environment information and current time information of the power grid enterprise to be detected; Based on the current time information, the current environment information and the accident occurrence law, determining whether the probability of the accident occurrence of the to-be-detected power grid enterprise at the current moment is greater than or equal to a preset probability; If the probability of the accident occurring is greater than or equal to the preset probability, a warning message is generated.

7. A device for determining the law of power grid accidents, characterized in that: A method for determining a power grid accident law according to any one of claims 1 to 6, wherein the power grid accident law determining device comprises: An information acquisition module, used to determine the historical operation information of the power grid enterprise to be detected, wherein the historical operation information includes at least one historical operation data and at least one historical operation type, and the historical operation type corresponds to the historical operation data in a one-to-one manner; An abnormality determination module, used to determine that the historical operation data corresponding to the historical operation type whose operation type is abnormal operation is the intermediate accident data; A data determination module, used to process the intermediate accident data using a pre-trained target data completion model to obtain target accident data of the power grid enterprise to be detected, wherein the target accident data includes at least one accident, accident time information and environmental information corresponding to the at least one accident; The rule determination module is used to process the at least one accident, the accident time information and the environmental information corresponding to the at least one accident to obtain the accident occurrence rule of the power grid enterprise to be detected.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the law of power grid accidents as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a power grid accident law as described in any one of claims 1 to 6 when executed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the method for determining a power grid accident law according to any one of claims 1 to 6.

Images