Power grid model and method, device, equipment and medium for generating associated sample data thereof

Abstract

The present application belongs to the field of electric power automation, and discloses a power grid model and a method, device, equipment and medium for generating associated sample data of the power grid model. The method comprises: obtaining the labeling requirement of the power grid model, the required number of power grid models, and the sample and verification case requirements; matching the power grid model in the pre-constructed power grid model library according to the obtained labeling requirement of the power grid model and the required number of power grid models, and obtaining the matched power grid model; generating a sample pre-configuration requirement table according to the sample and verification case requirements; extracting sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and obtaining the associated sample data of the matched power grid model; and outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table. The present application can dynamically generate training samples and verification cases according to the actual requirements of power grid regulation and control applications.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power automation, and particularly relates to a power grid model and a method, device and equipment for generating associated sample data of the power grid model and a medium. BACKGROUND

[0002] In recent years, emerging ICT (information and communication technology) technologies such as cloud, big data, migration, intelligence and chain have made great progress and been widely applied, and the same is true in the field of power grid regulation. The power grid model and its associated sample data are often an important basis for the application of these emerging ICT technologies. Taking the application of artificial intelligence technology as an example, artificial intelligence technology is one of the most important emerging technologies in the 21st century. At present, countries around the world are actively researching and deploying applications. In the field of power grid regulation, research and application of artificial intelligence technology are also in the ascendant. Power grid regulation applications involve national economy and people's livelihood safety and belong to the field of high reliability requirements. However, the mainstream artificial intelligence technology is often a "black box" technology. Although it may achieve a very high level in terms of precision (accuracy) and efficiency, it may sometimes make low-level errors. Due to its low interpretability, the process of correcting errors often brings new errors, so testing is particularly important, which requires a large amount of effective power grid model and its associated sample data.

[0003] The data in the field of power grid regulation is closely related to the power grid model and has certain complexity. The data required by different regulation applications (even different artificial intelligence models of the same type of regulation application) is generally not exactly the same. The training samples and verification cases for artificial intelligence applications in the field of power grid regulation are often independently constructed and difficult to reuse. It often takes a lot of effort to construct a set of training samples and supporting verification cases suitable for a specific application in the field of power grid regulation. Sometimes, it is difficult to obtain a large number of such samples and cases. SUMMARY

[0004] The purpose of the present application is to provide a power grid model and a method, device and equipment for generating associated sample data of the power grid model, which can realize dynamic on-demand generation of the power grid model and its associated sample data, and solve the technical problem that the training samples and verification cases for artificial intelligence applications in the field of power grid regulation are often independently constructed and difficult to reuse, and a lot of manpower is required.

[0005] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a method for generating a power grid model and its associated sample data, comprising:

[0007] Obtaining the labeling requirement of the power grid model, the required number of power grid models, and the sample and verification case requirements;

[0008] According to the obtained labeling requirement of the power grid model and the required number of power grid models, matching the power grid model in the pre-constructed power grid model library to obtain the matched power grid model;

[0009] According to the sample pre-configuration requirement table, the matched power grid model, and the sample database associated with the matched power grid model, extracting sample data to obtain the associated sample data of the matched power grid model;

[0010] According to the output sample format in the sample pre-configuration requirement table, outputting the matched power grid model and the associated sample data.

[0011] The further improvement of the present application is that the pre-constructed power grid model library is a labeled power grid model library; the construction method of the labeled power grid model library comprises:

[0012] Collecting and storing power grid models, establishing feature labels for each power grid model, and storing according to the preset power grid model storage format and labels to obtain the labeled power grid model library.

[0013] The further improvement of the present application is that the preset power grid model storage format is CIM / E format or CIM / XML format;

[0014] The feature label comprises one or more of power grid model ID, time, place, scale, longitude range, dimension range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale, and node number.

[0015] The further improvement of the present application is that the sample database associated with the power grid model is constructed by the following method: collecting sample data corresponding to the corresponding power grid model for each power grid model, and storing according to the preset storage format to obtain the sample database associated with the power grid model;

[0016] The sample database preset storage format is a data file or a database; the storage content comprises one or more of the corresponding power grid model ID, time point, signal point ID, signal point name, whether the power grid model is a device signal, and signal value.

[0017] The further improvement of the present application is that the step of matching the power grid model in the pre-constructed power grid model library according to the obtained labeling requirement of the power grid model and the required number of power grid models to obtain the matched power grid model specifically comprises:

[0018] According to the tag requirement, the tags of each power grid model in the pre-constructed power grid model library are matched, and N power grid models with the highest matching degrees are selected, wherein N is the number of required power grid models.

[0019] The matching step specifically includes:

[0020] A matching degree variable array V[i] is established for all power grid models in the power grid model library, and the initial values of each component of the array are all set to 0.

[0021] The tag requirement for the power grid model and the weight of each requirement are read in, and for each tag requirement, the corresponding tag of each power grid model in the power grid model library is processed in a loop, and the value of the model matching degree array is calculated according to the following formula:

[0022] Wherein V[i] represents the model matching degree value of the i-th power grid model in the power grid model library, n represents the total number of tag requirements for the power grid model, j represents the j-th requirement, w j represents the weight of the j-th tag requirement, 0≤w j ≤1, and x i,j represents the score of the i-th power grid model in the j-th tag requirement, according to the difference between the tag requirement and the value or range of the tag requirement, the scoring standard is as follows:

[0023] If the tag requirement is a value, if the two tags are equal, the corresponding power grid model scores 1 for this item of tag requirement, otherwise 0;

[0024] If the tag requirement is a numerical range area, if the corresponding tag value of the power grid model is exactly equal to the set tag requirement numerical range, the corresponding power grid model scores 1 for this item of tag requirement, otherwise, if the corresponding tag value of the power grid model completely covers the numerical range area of the tag requirement, the corresponding power grid model scores 0.5 for this item of tag requirement, otherwise 0;

[0025] The matching degree variables of all power grid models are sorted from large to small, and if the matching degree variable values of two power grid models are equal, the sorting is performed according to the order of reading in;

[0026] The N power grid models at the front are selected as the matched power grid models.

[0027] The further improvement of the application is that: according to the sample and the verification case requirement, a sample pre-configuration requirement table is generated; according to the sample pre-configuration requirement table and the matched power grid model, sample data is extracted from the sample database associated with the matched power grid model to obtain the associated sample data of the matched power grid model.

[0028] generate a sample pre-configuration requirement table according to the sample and verification case requirements; the sample pre-configuration requirement table comprises one of the following two forms: the first form: time range, signal point ID; the second form: time range, comprehensive signal ID group, comprehensive signal calculation formula; the comprehensive signal ID group comprises at least two signal point IDs;

[0029] extract signal point data in a specified time range or new data calculated by comprehensive signal calculation formula from several signal point data from a sample database associated with the matched power grid model as the associated sample data of the matched power grid model according to the sample pre-configuration requirement table.

[0030] Further improvement of the present application is that: in the step of outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table, the output sample format is one or more of CSV format, txt format, CIM / E format and Excel file format.

[0031] In a second aspect, the present application provides a power grid model and associated sample data generation device, comprising:

[0032] The acquisition module is configured to acquire the labeling requirements for the power grid model, the required number of power grid models, and the sample and verification case requirements.

[0033] The model matching module is configured to match the power grid model in the pre-constructed power grid model library according to the acquired labeling requirements for the power grid model and the required number of power grid models, and obtain the matched power grid model.

[0034] The data matching module is configured to generate a sample pre-configuration requirement table according to the sample and verification case requirements, and extract sample data from a sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, to obtain the associated sample data of the matched power grid model.

[0035] The output module is configured to output the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table.

[0036] Further improvement of the present application is that: the pre-constructed power grid model library is a labeled power grid model library; and the construction method of the labeled power grid model library comprises:

[0037] Collect and store the power grid models, establish a feature label for each power grid model, and store according to a preset power grid model storage format and label to obtain the labeled power grid model library.

[0038] The further improvement of the present application is that the preset power grid model storage format is CIM / E format or CIM / XML format.

[0039] The feature label comprises one or more of a power grid model ID, time, location, scale, longitude range, latitude range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale and node number.

[0040] The further improvement of the present application is that the sample database associated with the power grid model is constructed by the following method: for each power grid model, sample data corresponding to the corresponding power grid model is collected, and the sample data is stored according to a preset storage format to obtain the sample database associated with the power grid model.

[0041] The preset storage format of the sample database is a data file or a database; and the storage content comprises one or more of a corresponding power grid model ID, a time point, a signal point ID, a signal point name, whether it is a device signal in the power grid model, and a signal value.

[0042] The further improvement of the present application is that the model matching module matches the power grid model in the pre-constructed power grid model library according to the obtained labeling requirement for the power grid model and the required number of power grid models, and obtains the matched power grid model, and the step specifically comprises:

[0043] According to the labeling requirement, the labels of each power grid model in the pre-constructed power grid model library are matched, and N power grid models with the highest matching degree are selected; the N is the required number of power grid models.

[0044] The matching step specifically comprises:

[0045] A matching degree variable array V[i] is established for all power grid models in the power grid model library, and the initial value of each component of the array is 0.

[0046] The labeling requirement for the power grid model and the weight of each requirement are read in, and for each labeling requirement, the corresponding label of each power grid model in the power grid model library is processed in a loop, and the value of the model matching degree array is calculated according to the following formula: Wherein V[i] represents the model matching degree value of the i-th power grid model in the power grid model library; n represents the total number of the labeling requirements for the power grid model; j represents the j-th requirement; w j represents the weight of the j-th labeling requirement, 0≤w j ≤1, and x i,jThe score of the i-th power grid model under the j-th labeled demand is represented, and according to the difference of the labeled demand being a value or a range, the labeled value of the labeled demand and the power grid model are compared, and the scoring standard is as follows: if the labeled demand is a value, if the two labels are equal, the score of the corresponding labeled demand of the power grid model is 1, otherwise 0; if the labeled demand is a numerical range area, if the corresponding label value of the power grid model is exactly equal to the set labeled demand numerical range, the score of the corresponding labeled demand of the power grid model is 1, if the corresponding label value of the power grid model completely covers the numerical range area of the labeled demand, the score of the corresponding labeled demand of the power grid model is 0.5, otherwise 0;

[0047] The matching degree variables of all power grid models are sorted from large to small, and if the matching degree variable values of two power grid models are equal, the sorting is performed according to the order of the circular reading;

[0048] The first N power grid models are selected as the matched power grid models.

[0049] The further improvement of the present application is that the data matching module generates a sample pre-configuration demand table according to sample and verification case requirements; and the step of extracting sample data from a sample database associated with the matched power grid model to obtain the associated sample data of the matched power grid model according to the sample pre-configuration demand table and the matched power grid model, specifically includes:

[0050] The sample pre-configuration demand table is generated according to sample and verification case requirements; the sample pre-configuration demand table includes one of the following two forms: the first form: time range, signal point ID; the second form: time range, comprehensive signal ID group, comprehensive signal calculation formula; the comprehensive signal ID group includes at least two signal point IDs;

[0051] According to the sample pre-configuration demand table, the signal point data in the specified time range or the new data calculated by the comprehensive signal calculation formula from the signal point data are extracted from the sample database associated with the matched power grid model as the associated sample data of the matched power grid model.

[0052] The further improvement of the present application is that the output module outputs the matched power grid model and the associated sample data in the step of outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration demand table, and the output sample format is one or more of CSV format, txt format, CIM / E format and Excel file format.

[0053] In a third aspect, the present application provides an electronic device including a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to realize the power grid model and the associated sample data generation method.

[0054] In a fourth aspect, the present application provides a computer readable storage medium storing at least one instruction, which, when executed by a processor, implements the power grid model and its associated sample data generation method.

[0055] Compared with the prior art, the present application has the following beneficial effects:

[0056] Compared with the prior art, the present application has the following beneficial effects:

[0057] The present application provides a power grid model and its associated sample data generation method, comprising: obtaining the labeling requirements for the power grid model, the required number of power grid models, and the sample and verification case requirements; matching the power grid model in the pre-constructed power grid model library according to the obtained labeling requirements for the power grid model and the required number of power grid models, to obtain the matched power grid model; generating a sample pre-configuration requirement table according to the sample and verification case requirements; extracting sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, to obtain the associated sample data of the matched power grid model; and outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table. The present application can dynamically generate training samples and verification cases (including power grid models and corresponding power grid data) according to the actual requirements of power grid regulation and control applications. The training samples and verification cases (i.e. the matched power grid model + the extracted sample data) finally generated by the present application are generated according to the power grid model labeling requirements and the sample pre-configuration requirements, and the output format of the sample is customized.

[0058] The present application can maintain the effectiveness of the original data. The data in the sample database are all corresponding to the corresponding power grid model, ensuring the effectiveness of the sample data. For the power grid model, although there are functions such as model splicing and model extraction (cutting), considering that the model may be changed (expanded or reduced or modified), the original sample data may be distorted (because the physical characteristics of the power grid will change with the model), so the present application adopts a label matching method for the power grid model, and finally matches a complete original power grid model (this power grid model and the corresponding sample data are generally collected by the actual power grid system SCADA system, or calculated by power grid simulation). The sample data extracted subsequently is a part of the sampling data corresponding to the original power grid model. Through the above method, the present application maximizes the effectiveness of the generated training samples and verification cases.

[0059] The power grid model and data reusability are maximized. After the tagged power grid model library and the sample database are established, the power grid model in the power grid model library and the data in the sample database can be dynamically matched according to the actual demand of the power grid regulation and control field for the power grid model, and the data in the same sample database can be extracted according to different pre-configuration requirements, so as to realize the reusability of the power grid model and the sample data. BRIEF DESCRIPTION OF DRAWINGS

[0060] The accompanying drawings, which form a part of the specification, are included to provide a further understanding of the application and are incorporated herein in conjunction with the description of the application. The embodiments of the present application and its description are used to explain the present application and are not used to limit the present application. In the drawings:

[0061] Figure 1 A flowchart of a power grid model and its associated sample data generation method according to the present application;

[0062] Figure 2 A flowchart of another power grid model and its associated sample data generation method according to the present application;

[0063] Figure 3 A block diagram of a power grid model and its associated sample data generation device according to the present application;

[0064] Figure 4 A block diagram of an electronic device according to the present application. DETAILED DESCRIPTION

[0065] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0066] The following detailed description is exemplary and is intended to provide further detailed description of the present application. Unless otherwise specified, all technical terms used in the present application have the same meaning as generally understood by those skilled in the art to which the present application belongs. The terms used in the present application are only for the purpose of describing the specific embodiments and are not intended to limit the exemplary embodiments according to the present application.

[0067] Embodiment 1

[0068] Please refer to Figure 1 The present application provides a power grid model and its associated sample data generation method, comprising the following steps:

[0069] S1, a tagged power grid model library and a sample database associated with the power grid model are constructed;

[0070] S2, obtaining a labeling requirement for the power grid model and a required number of power grid models; and automatically matching the power grid models in the power grid model library constructed in step S1 according to the obtained labeling requirement for the power grid model and the required number of power grid models;

[0071] S3, obtaining a sample and a verification case requirement; filling a sample pre-configuration requirement according to the obtained sample and the verification case requirement; and automatically extracting sample data from a sample database associated with the matched power grid model according to the sample pre-configuration requirement and the matched power grid model;

[0072] S4, outputting the power grid model and the associated sample data according to an output sample format in the pre-configuration requirement.

[0073] In an embodiment, the step S1 of constructing the labeled power grid model library and the sample database associated with the power grid model specifically includes:

[0074] Collecting and storing the power grid models, and establishing a feature label for each power grid model, typical labels including but not limited to: power grid model ID, time, location, scale, longitude range, latitude range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale, and node number.

[0075] Storing a commonly used power grid model storage format and its label; the model storage format is unified, and typical commonly used storage formats include CIM / E format and CIM / XML format, and each stored power grid model has a series of feature labels.

[0076] For each collected power grid model, sample data corresponding to the power grid model is collected and stored in a unified format. The storage format can be a data file or a database. The storage content includes but is not limited to the following key fields: corresponding power grid model ID, time point, signal point ID, signal point name, whether the signal point is a device signal in the power grid model, signal value, etc. When the value of "whether the signal point is a device signal in the power grid model" is "yes", it indicates that the signal point is a related signal of the power grid device in the corresponding power grid model; when the value of "whether the signal point is a device signal in the power grid model" is "no", it indicates that the signal point is not directly corresponding to the power grid device in the corresponding power grid model. For example, some artificial intelligence prediction models require holiday information and local weather information, etc. Although these information has no direct correspondence with the power grid device in the corresponding power grid model, it should still be saved in the corresponding sample data of the power grid model.

[0077] In an embodiment, the sample database is associated, each sample database uniquely corresponds to one power grid model in the model library, and the sample data is stored in the form of a two-dimensional table of signal points and time points, and the difference between the sampling time points is uniformly set, for example, in the form of Table 1 below:

[0078] Table 1

[0079]

[0080] In a specific embodiment, the sample data is recorded at a uniform interval of 5 minutes, and in actual application, the sampling point time interval in the sample library can be uniformly and flexibly set. The signal point sample data at specific time points is stored in the "00", "05", "10",..., "55" fields, and here a value is sampled every 5 minutes.

[0081] In a specific embodiment, the step S2 acquires the labeling requirements for the power grid model and the required number of power grid models; and according to the acquired labeling requirements for the power grid model and the required number of power grid models, the step of automatically matching the power grid models in the constructed power grid model library specifically includes:

[0082] The labeling requirements for the power grid model and the number of required power grid models are acquired, the labeling requirements are matched with the labels of each power grid model in the power grid model library, and a specified number of power grid models with the highest matching degree are selected.

[0083] The matching process is as follows:

[0084] S21, a matching degree variable array V[i] is established for all models in the power grid model library, and the initial value of each component of the array is set to 0.

[0085] S22, the labeling requirements for the power grid model and the weight of each requirement are read in, and for each labeling requirement, the corresponding label of each power grid model in the power grid model library is processed in a loop, and the value of the model matching degree array is calculated according to the following formula:

[0086] Where V[i] represents the model matching degree value of the i-th power grid model in the power grid model library; n represents the total number of labeling requirements for the power grid model; j represents the j-th requirement; w j represents the weight of the j-th labeling requirement, 0≤w j ≤1, and x i,j represents the score of the i-th power grid model in the j-th labeling requirement, according to the difference between the value and the range of the labeling requirement, the labeling requirement and the label value of the power grid model are compared, and the scoring standard is as follows:

[0087] If the labeling requirement is a value, if the two labels are equal, the corresponding power grid model scores 1 for this item of labeling requirement, otherwise 0;

[0088] If the labeling requirement is a numerical range area, if the corresponding label value of the power grid model is exactly equal to the set labeling requirement numerical range, then the corresponding power grid model has a labeling requirement score of 1, otherwise, if the corresponding label value of the power grid model completely covers the numerical range area of the labeling requirement, then the corresponding power grid model has a labeling requirement score of 0.5, otherwise, 0;

[0089] S23, sort the matching degree variables of all power grid models from large to small, and if the matching degree variable values of two power grid models are equal, then sort them according to the order in the S22 step cycle (i.e., the first processed is placed in front).

[0090] Select the specified number of power grid models placed in the front as the power grid models with the highest matching degree.

[0091] In a specific embodiment, the step S3 acquires sample and verification case requirements; according to the acquired sample and verification case requirements, fill in the sample pre-configuration requirements, and according to the sample pre-configuration requirements and the matched power grid model, automatically extract sample data from the sample database associated with the matched power grid model.

[0092] According to the acquired sample and verification case requirements, specify which data in the sample database corresponding to the selected power grid model is needed. Typical pre-configuration requirements include the following two forms:

[0093] Time range, signal point ID;

[0094] Time range, comprehensive signal ID group (1st signal point ID, 2nd signal point ID, 3rd signal point ID, …), comprehensive signal calculation formula;

[0095] The reason for including the above two forms of pre-configuration is that it is possible to apply part of the required sample data, which is not directly included in the sample database associated with the selected power grid model, but can be obtained by calculating a number of data in the sample database, so it is necessary to define a set of signal points in the sample database (i.e., “comprehensive signal ID group”) and how they are calculated (i.e., “comprehensive signal calculation formula”).

[0096] According to the pre-configuration requirements, extract the signal point data within the specified time range or the new data calculated from a number of signal point data from the sample database associated with the selected power grid model.

[0097] In a specific embodiment, the step S4 outputs the power grid model and its associated sample data according to the output sample format in the pre-configuration requirements, which specifically includes:

[0098] The extracted sample data is formed into an output data file according to a data file format specified in the pre-configuration requirement, and is output as a training sample or a verification case of the related application together with the corresponding power grid model. There are many optional data text formats, such as CSV format, txt format, CIM / E format, Excel file format, etc.

[0099] The present application realizes the dynamic on-demand generation of the power grid model and its associated sample data by constructing the power grid model and the associated sample database, automatically matching the power grid model labeling, automatically extracting sample data according to the pre-configuration requirement, and customizing the output, and designs a corresponding implementation system. These data can be used as sample data for training of artificial intelligence models and the like, and can also be used as verification cases of related applications.

[0100] The present application solves the following key problems: 1) to realize the effective storage of the power grid model and its associated sample data, and to provide on demand according to the application requirement when used. The data in the power grid regulation field is closely related to the power grid model and has certain complexity, and the data required by different regulation applications (even different artificial intelligence models of the same type of regulation application) is generally not exactly the same. It often takes a lot of effort to build a set of training samples and supporting verification cases suitable for a specific application in the power grid regulation field, and sometimes it is difficult to obtain a large number of such samples and cases. 2) to improve the reusability of the power grid model and its associated sample data. At present, the training samples and verification cases of artificial intelligence applications in the power grid regulation field are often independently constructed and cannot be reused (unless the samples and cases required by two artificial intelligence models are exactly the same).

[0101] The present application can provide a large number of applicable training samples and verification cases for ICT technology applications related to the power grid regulation field, which is of great significance for the application promotion of these technologies in the power grid regulation field and the improvement of the safety and reliability of the power grid regulation.

[0102] In the face of the problem of lack of training samples and verification cases for ICT technology applications related to the power grid regulation field, the present application proposes to dynamically generate the power grid model and its associated sample data on demand by constructing the power grid model and the associated sample database, automatically matching the power grid model labeling, automatically extracting sample data according to the pre-configuration requirement, and customizing the output, and designs a corresponding implementation system. Thus, the landing and popularization of emerging ICT technology applications in the power grid regulation field are promoted, and the safety and reliability of the dispatching automation system in the power grid regulation field are ensured and improved.

[0103] Embodiment 2

[0104] The present application provides a power grid model and its associated sample data generation method, comprising the following steps:

[0105] 1) Power grid model library and associated sample database construction case

[0106] In a specific embodiment, the following information is collected:

[0107] 1) Power grid model of a certain provincial power grid in western China and its full set of adoption data from 2018 to 2021.

[0108] 1) Power grid model of a certain second-tier city in eastern China and its full set of adoption data from 2020 to 2022.

[0109] The power grid model library is then established in the following manner:

[0110] The power grid model of the western provincial power grid is stored in the CIM / E format, and the tag information for this power grid model is established as shown in Table 2:

[0111] Table 2

[0112]

[0113] The power grid model of the eastern second-tier city is stored in the CIM / E format, and the tag information for this power grid model is established as shown in Table 3:

[0114] Table 3

[0115]

[0116] The historical data associated with the provincial power grid model and the municipal power grid model is stored in the form of a database table. The sample data records for the provincial power grid from 0:00 to 1:00 on January 1, 2020 are shown in Table 4. In a specific embodiment, sample data is recorded at 5-minute intervals. In actual applications, the sampling point time interval in the sample library can be uniformly and flexibly set. The signal point sample data at specific time points is stored in the "00", "05", "10",..., "55" fields, where a value is sampled every 5 minutes. The typical form is shown in Table 4:

[0117] Table 4

[0118]

[0119] 2) Dynamic construction of training samples and verification cases for a load prediction neural network model

[0120] Assuming a certain load prediction neural network model is needed for city power grid load prediction, the dynamic construction of a verification case for this model is as follows:

[0121] 2.1) Match the power grid model

[0122] Need 1 power grid model verification case, the power grid model labeling requirements are as follows:

[0123]

[0124] According to the matching rule, the matching degree of power grid model 1 (a certain level of power grid model, ID is 1) is 0x0.5+0.5x0.5=0.25, and the matching degree of power grid model 2 (a certain city level power grid model, ID is 2) is 1x0.5+0.5x0.5=0.75, so the final matching power grid model is power grid model 2.

[0125] 2.2), extract sample data

[0126] The sample pre-configuration requirements are as follows:

[0127]

[0128] In the above pre-configuration requirements, the signal point ID is filled in actual filling, and the signal point name is directly replaced for convenience of explanation; In the above pre-configuration requirements, "east city total load + west city total load + central city total load" is a comprehensive signal calculation formula of several signal point data.

[0129] Based on the sample pre-configuration requirements, the data of the specified signal points in the sample database associated with power grid model 2 (its "power grid model ID" is "2") in 2020 are extracted, and the comprehensive calculation value of part of the signal point data is extracted as output data.

[0130] 2.3), customized output

[0131] According to the pre-configuration requirements, all the extracted data is organized into CSV file format, and the size of a single file does not exceed "3M bytes". The final output is a sequentially numbered CSV sample data.

[0132] These sample data can be used as training samples or verification cases according to the requirements of the artificial intelligence application model.

[0133] 3), a dynamic construction case of active scheduling neural network model training sample and verification case

[0134] Suppose a load forecasting neural network model is needed to replace the traditional power grid active scheduling, and the information it needs is mainly power grid model and load change data, then the dynamic construction of the model verification case process is as follows:

[0135] 3.1), match power grid model

[0136] Need 1 power grid model verification case, the power grid model labeling requirements are as follows:

[0137]

[0138] According to the matching rule, the matching degree of the power grid model 1 (a certain level of power grid model, ID is 1) is 1*0.7+0.5*0.3=0.85, and the matching degree of the power grid model 2 (a certain city level power grid model, ID is 2) is 0*0.7+0.5*0.3=0.15, so the finally matched power grid model is the power grid model 1.

[0139] 3.2), extract sample data

[0140] The sample pre-configuration requirements are as follows:

[0141]

[0142] Based on the sample pre-configuration requirements, the data of the specified signal points in the sample database associated with the power grid model 1 (the "power grid model ID" of which is "1") in 2020 is extracted as output data.

[0143] 3.3), customized output

[0144] According to the pre-configuration requirements, all the extracted data is organized into a CSV file format, and the size of a single file does not exceed "3M bytes". The final output is a sequentially numbered CSV sample data.

[0145] Subsequent sample data can be used as training samples or verification cases according to the requirements of the artificial intelligence application model, combined with relevant power grid simulators and power grid dispatching result evaluation software. (Note: For active power dispatching artificial intelligence models, only power grid models and sample data are not enough. Power grid simulators and power grid dispatching result evaluation software also need to be added to form complete training samples or verification cases.)

[0146] Embodiment 3

[0147] Please refer to Figure 2 As shown in the figure, the present application provides a power grid model and its associated sample data generation method, comprising:

[0148] S101, obtaining the labeling requirements of the power grid model, the required number of power grid models, and the sample and verification case requirements;

[0149] S102, according to the obtained labeling requirements of the power grid model and the required number of power grid models, matching the power grid model in the pre-constructed power grid model library, and obtaining the matched power grid model;

[0150] S103, generating a sample pre-configuration requirement table according to the sample and the verification case requirement; extracting sample data from a sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model to obtain the associated sample data of the matched power grid model;

[0151] S104, outputting the matched power grid model and the associated sample data according to an output sample format in the sample pre-configuration requirement table.

[0152] In an embodiment, the pre-constructed power grid model library is a tagged power grid model library; and a construction method of the tagged power grid model library comprises:

[0153] collecting and storing power grid models, establishing a feature tag for each power grid model, and storing according to a preset power grid model storage format and a tag to obtain the tagged power grid model library.

[0154] In an embodiment, the preset power grid model storage format is a CIM / E format or a CIM / XML format.

[0155] The feature tag comprises one or more of a power grid model ID, a time, a location, a scale, a longitude range, a latitude range, a new energy proportion, a thermal power proportion, a hydropower proportion, a nuclear power proportion, a load scale, and a node number.

[0156] In an embodiment, the sample database associated with the power grid model is constructed by the following method: for each power grid model, collecting sample data corresponding to the corresponding power grid model, and storing according to a preset storage format to obtain the sample database associated with the power grid model.

[0157] The sample database preset storage format is a data file or a database; and the storage content comprises one or more of a corresponding power grid model ID, a time point, a signal point ID, a signal point name, whether a device signal in the power grid model, and a signal value.

[0158] In an embodiment, the step of matching the power grid model in the pre-constructed power grid model library according to the obtained tagged requirement for the power grid model and the required number of power grid models comprises:

[0159] According to the tagged requirement, matching the tags of the power grid models in the pre-constructed power grid model library, and selecting N power grid models with the highest matching degrees; the N is the required number of power grid models.

[0160] The matching step specifically comprises:

[0161] A matching degree variable array V[i] is established for all the power grid models in the power grid model library, and the initial values of the components of the array are all set to 0.

[0162] read in the labeling requirements for the power grid model and the weight of each requirement, for each labeling requirement, loop through the corresponding label of each power grid model in the power grid model library, and calculate the value of the model matching degree array according to the following formula: where V[i] represents the model matching degree value of the i-th power grid model in the power grid model library; n represents the total number of labeling requirements for the power grid model; j represents the j-th requirement; w j represents the weight of the j-th labeling requirement, 0≤w j ≤1, and x i,j represents the score of the i-th power grid model in the j-th labeling requirement. According to the difference between the labeling requirement and the value or range, the labeling requirement and the label value of the power grid model are compared, and the scoring standard is as follows: if the labeling requirement is a value, then if the two labels are equal, the corresponding power grid model scores 1 for this item of labeling requirement, otherwise 0; if the labeling requirement is a numerical range area, then if the corresponding label value of the power grid model is exactly equal to the set numerical range of the labeling requirement, the corresponding power grid model scores 1 for this item of labeling requirement, otherwise, if the corresponding label value of the power grid model completely covers the numerical range area of the labeling requirement, the corresponding power grid model scores 0.5 for this item of labeling requirement, otherwise 0;

[0163] Sort the matching degree variables of all power grid models from large to small, and if the matching degree variable values of two power grid models are equal, sort them according to the order of reading in the loop;

[0164] Select the first N power grid models as the matched power grid models.

[0165] In a specific embodiment, the sample pre-configuration requirement table is generated according to the sample and the verification case requirements; the associated sample data of the matched power grid model is obtained by extracting sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and the step specifically includes:

[0166] According to the sample and the verification case requirements, a sample pre-configuration requirement table is generated; the sample pre-configuration requirement table includes one of the following two forms: the first form: time range, signal point ID; the second form: time range, comprehensive signal ID group, comprehensive signal calculation formula; the comprehensive signal ID group includes at least two signal point IDs;

[0167] According to the sample pre-configuration requirement table, the signal point data within the specified time range or the new data calculated by the comprehensive signal calculation formula from the signal point data is extracted from the sample database associated with the matched power grid model as the associated sample data of the matched power grid model.

[0168] In an embodiment, the output sample format is one or more of CSV format, txt format, CIM / E format, and Excel file format.

[0169] Embodiment 4

[0170] Referring to Figure 3 The power grid model and its associated sample data generation device is characterized in that it comprises:

[0171] The acquisition module 1 is configured to acquire the labeling requirements for the power grid model, the required number of power grid models, and the sample and verification case requirements.

[0172] The model matching module 2 is configured to match the power grid model in the pre-constructed power grid model library according to the acquired labeling requirements for the power grid model and the required number of power grid models, and obtain the matched power grid model.

[0173] The data matching module 3 is configured to generate a sample pre-configuration requirement table according to the sample and verification case requirements, extract sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and obtain the associated sample data of the matched power grid model.

[0174] The output module 4 is configured to output the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table.

[0175] In an embodiment, the pre-constructed power grid model library is a labeled power grid model library, and the construction method of the labeled power grid model library comprises:

[0176] The power grid model is collected and stored, and a feature label is established for each power grid model. The power grid model is stored according to a preset power grid model storage format and label, and a labeled power grid model library is obtained.

[0177] In an embodiment, the preset power grid model storage format is CIM / E format or CIM / XML format.

[0178] The feature label comprises one or more of the power grid model ID, time, location, scale, longitude range, latitude range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale, and node number.

[0179] In an embodiment, the sample database associated with the power grid model is constructed by the following method: for each power grid model, collecting sample data corresponding to the corresponding power grid model, storing according to a preset storage format, and obtaining the sample database associated with the power grid model.

[0180] The preset storage format of the sample database is a data file or a database; and the storage content includes one or more of the corresponding power grid model ID, time point, signal point ID, signal point name, whether the signal is in the power grid model device, and signal value.

[0181] In an embodiment, the step of matching the power grid model in the pre-constructed power grid model library according to the obtained labeling requirements for the power grid model and the required number of power grid models specifically includes:

[0182] According to the labeling requirements, the labels of each power grid model in the pre-constructed power grid model library are matched, and the N power grid models with the highest matching degree are selected; the N is the required number of power grid models;

[0183] The matching step specifically includes:

[0184] A matching degree variable array V[i] is established for all power grid models in the power grid model library, and the initial value of each component of the array is set to 0;

[0185] The labeling requirements for the power grid model and the weight of each requirement are read in, and for each labeling requirement, the corresponding label of each power grid model in the power grid model library is processed in a loop, and the value of the model matching degree array is calculated according to the following formula: Where V[i] represents the model matching degree value of the i-th power grid model in the power grid model library; n represents the total number of labeling requirements for the power grid model; j represents the j-th requirement; w j represents the weight of the j-th labeling requirement, 0≤w j ≤1, and x i,j represents the score of the i-th power grid model in the j-th labeling requirement, according to the difference between the value and the range of the labeling requirement, the labeling requirement and the label value of the power grid model are compared, and the scoring standard is as follows: if the labeling requirement is a value, then if the two labels are equal, the corresponding power grid model scores 1 for this item of labeling requirement, otherwise 0; if the labeling requirement is a numerical range area, then if the corresponding label value of the power grid model is exactly equal to the set labeling requirement numerical range, the corresponding power grid model scores 1 for this item of labeling requirement, otherwise, if the corresponding label value of the power grid model completely covers the numerical range area of the labeling requirement, the corresponding power grid model scores 0.5 for this item of labeling requirement, otherwise 0;

[0186] The matching degree variables of all power grid models are sorted from large to small, and if the matching degree variable values of two power grid models are equal, the two power grid models are sorted according to the order of reading in circulation.

[0187] The first N power grid models are selected as the matched power grid models.

[0188] In an embodiment, the step of generating a sample pre-configuration requirement table according to sample and verification case requirements, and extracting sample data from a sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model to obtain associated sample data of the matched power grid model specifically includes:

[0189] The sample pre-configuration requirement table includes one of the following two forms: the first form: time range, signal point ID; the second form: time range, comprehensive signal ID group, comprehensive signal calculation formula; the comprehensive signal ID group includes at least two signal point IDs.

[0190] According to the sample pre-configuration requirement table, signal point data in a specified time range or new data calculated by a comprehensive signal calculation formula from a plurality of signal point data are extracted from a sample database associated with the matched power grid model as associated sample data of the matched power grid model.

[0191] In an embodiment, the output sample format in the step of outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table is one or more of CSV format, txt format, CIM / E format, and Excel file format.

[0192] Embodiment 5

[0193] Please refer to Figure 4 As shown in the figure, the application further provides an electronic device 100 for implementing the power grid model and associated sample data generation method; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0194] The memory 101 can be used to store the computer program 103, and the processor 102 can realize the steps of the power grid model and the associated sample data generation method of embodiments 1, 2 or 3 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data (such as audio data) created according to the use of the electronic device 100. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.

[0195] The at least one processor 102 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 102 can be a microprocessor or can also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and is connected to various parts of the entire electronic device 100 through various interfaces and lines.

[0196] The memory 101 in the electronic device 100 stores a plurality of instructions to realize a power grid model and an associated sample data generation method, and the processor 102 can execute the plurality of instructions to realize:

[0197] Obtaining the labeling requirement of the power grid model, the required number of power grid models, and the sample and verification case requirements;

[0198] According to the obtained labeling requirement of the power grid model and the required number of power grid models, matching the power grid model in the pre-constructed power grid model library to obtain the matched power grid model;

[0199] According to the sample and verification case requirements, a sample pre-configuration requirement table is generated; according to the sample pre-configuration requirement table and the matched power grid model, sample data is extracted from a sample database associated with the matched power grid model to obtain associated sample data of the matched power grid model;

[0200] According to the output sample format in the sample pre-configuration requirement table, the matched power grid model and the associated sample data are output.

[0201] Embodiment 6

[0202] The modules / units integrated in the electronic device 100, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory and read-only memory (ROM).

[0203] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0204] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for performing the functions specified in one flow or multiple flows and / or blocks Figure 1 The device for performing the functions specified in one flow or multiple flows and / or blocks

[0205] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 one or more flow or block Figure 1 one or more blocks or blocks specified in the flow.

[0206] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 one or more flow or block Figure 1 one or more blocks or blocks specified in the flow.

[0207] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limiting the same. Even though the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced equivalently, and any modification or replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. A method for generating a power grid model and its associated sample data, characterized in that, The application comprises the following steps: acquiring the labeling requirements of the power grid model, the required number of power grid models, and sample and verification case requirements; matching the power grid model in the pre-constructed power grid model library according to the acquired labeling requirements of the power grid model and the required number of power grid models, and obtaining the matched power grid model; generating a sample pre-configuration requirement table according to the sample and verification case requirements; extracting sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and obtaining the associated sample data of the matched power grid model; outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table; the step of matching the power grid model in the pre-constructed power grid model library according to the acquired labeling requirements of the power grid model and the required number of power grid models, and obtaining the matched power grid model, specifically comprises the following steps: matching the labels of each power grid model in the pre-constructed power grid model library according to the labeling requirements, and selecting the N power grid models with the highest matching degrees; N is the required number of power grid models; the matching step specifically comprises the following steps: establishing a matching degree variable array V[i] for all power grid models in the power grid model library, and setting the initial value of each component of the array to 0; Read in the labeling requirements for the power grid model and the weight of each requirement. For each labeling requirement, iteratively process the corresponding label for each power grid model in the power grid model library, and calculate the value of the model matching degree array according to the following formula: Where V[i] represents the i-th element in the power grid model library. i The model matching degree value of each power grid model; n represents the total number of labeling requirements for the power grid model; j Indicates the first j One requirement; Indicates the first j The weight of each labeled requirement ,and ; Indicates the first i The power grid model in the first... j The score for each labeled requirement is determined by comparing the labeled requirement with the label value of the power grid model, depending on whether the labeled requirement is a value or a range. The scoring criteria are as follows: If the labeled requirement is a value, then if the two labels are equal, the corresponding labeled requirement of the power grid model scores 1; otherwise, it scores 0. If the labeled requirement is a numerical range, then if the corresponding label value of the power grid model is exactly equal to the set numerical range of the labeled requirement, the corresponding labeled requirement of the power grid model scores 1; if the corresponding label value of the power grid model completely covers the numerical range of the labeled requirement, the corresponding labeled requirement of the power grid model scores 0.5; otherwise, it scores 0. sorting the matching degree variables of all power grid models from large to small, and if the matching degree variable values of two power grid models are equal, sorting them according to the order of cyclic reading; selecting the N power grid models at the front as the matched power grid models.

2. The power grid model and its associated sample data generation method according to claim 1, characterized in that, The pre-constructed power grid model library is a labeled power grid model library; the construction method of the labeled power grid model library comprises the following steps: collecting and storing power grid models, establishing characteristic labels for each power grid model, and storing according to the pre-set power grid model storage format and labels to obtain the labeled power grid model library.

3. The method of claim 2, wherein the grid model and its associated sample data are generated by: The pre-set power grid model storage format is CIM / E format or CIM / XML format; The characteristic labels include one or more of the power grid model ID, time, location, scale, longitude range, latitude range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale, and node number.

4. The power grid model and its associated sample data generation method according to claim 1, characterized in that, The sample database associated with the power grid model is constructed by the following method: for each power grid model, collecting sample data corresponding to the corresponding power grid model, and storing according to the pre-set storage format to obtain the sample database associated with the power grid model; The sample database pre-set storage format is a data file or a database; the storage content includes one or more of the corresponding power grid model ID, time point, signal point ID, signal point name, whether it is a device signal in the power grid model, and signal value.

5. The method of claim 1, wherein, The steps of generating a sample pre-configuration requirement table according to the sample and verification case requirements; extracting sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and obtaining the associated sample data of the matched power grid model, specifically comprise the following steps: According to the sample and verification case requirements, a sample pre-configuration requirement table is generated; the sample pre-configuration requirement table comprises one of the following two forms: the first form: time range, signal point ID; the second form: time range, comprehensive signal ID group, comprehensive signal calculation formula; the comprehensive signal ID group comprises at least two signal point IDs; According to the sample pre-configuration requirement table, signal point data in a specified time range or new data calculated by a comprehensive signal calculation formula from a plurality of signal point data is extracted from a sample database associated with the matched power grid model as the associated sample data of the matched power grid model.

6. The method of claim 1, wherein, In the step of outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table, the output sample format is one or more of a CSV format, a txt format, a CIM / E format, and an Excel file format.

7. An electrical grid model and its associated sample data generating apparatus, characterized by, Comprise: The acquisition module is configured to acquire a labeling requirement for a power grid model, a required number of power grid models, and sample and verification case requirements; The model matching module is configured to match power grid models in a pre-constructed power grid model library according to the acquired labeling requirement for the power grid model and the required number of power grid models, and obtain matched power grid models; The data matching module is configured to generate a sample pre-configuration requirement table according to the sample and verification case requirements, and extract sample data from a sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model, and obtain associated sample data of the matched power grid model; The output module is configured to output the matched power grid model and the associated sample data according to an output sample format in the sample pre-configuration requirement table; The model matching module matches power grid models in a pre-constructed power grid model library according to the acquired labeling requirement for the power grid model and the required number of power grid models, and obtains matched power grid models, and the steps specifically comprise: According to the labeling requirement, the labels of each power grid model in the pre-constructed power grid model library are matched, and N power grid models with the highest matching degrees are selected; N is the required number of power grid models; The matching step specifically comprises: A matching degree variable array V[i] is established for all power grid models in the power grid model library, and the initial values of the components of the array are all set to 0; Read in the labeling requirements for the power grid model and the weight of each requirement. For each labeling requirement, iteratively process the corresponding label for each power grid model in the power grid model library, and calculate the value of the model matching degree array according to the following formula: Where V[i] represents the i-th element in the power grid model library. i The model matching degree value of each power grid model; n represents the total number of labeling requirements for the power grid model; j Indicates the first j One requirement; Indicates the first j The weight of each labeled requirement ,and ; Indicates the first i The power grid model in the first... j The score for each labeled requirement is determined by comparing the labeled requirement with the label value of the power grid model, depending on whether the labeled requirement is a value or a range. The scoring criteria are as follows: If the labeled requirement is a value, then if the two labels are equal, the corresponding labeled requirement of the power grid model scores 1; otherwise, it scores 0. If the labeled requirement is a numerical range, then if the corresponding label value of the power grid model is exactly equal to the set numerical range of the labeled requirement, the corresponding labeled requirement of the power grid model scores 1; if the corresponding label value of the power grid model completely covers the numerical range of the labeled requirement, the corresponding labeled requirement of the power grid model scores 0.5; otherwise, it scores 0. The matching degree variables of all power grid models are sorted from large to small, and if the matching degree variable values of two power grid models are equal, the sorting is performed according to the order of the loop reading; The N power grid models at the front are selected as the matched power grid models.

8. The power grid model and its associated sample data generating apparatus according to claim 7, characterized by, The pre-constructed power grid model library is a labeled power grid model library; and a construction method of the labeled power grid model library comprises: Collect and store power grid models, establish characteristic labels for each power grid model, and store according to a pre-set power grid model storage format and labels to obtain the labeled power grid model library.

9. The power grid model and its associated sample data generating apparatus according to claim 8, characterized by, The pre-set power grid model storage format is a CIM / E format or a CIM / XML format; The feature label comprises one or more of a power grid model ID, time, location, scale, longitude range, latitude range, new energy proportion, thermal power proportion, hydropower proportion, nuclear power proportion, load scale, and node number.

10. The power grid model and its associated sample data generating apparatus according to claim 7, characterized by, The sample database associated with the power grid model is constructed by: for each power grid model, collecting sample data corresponding to the corresponding power grid model, and storing according to a preset storage format to obtain the sample database associated with the power grid model. The sample database preset storage format is a data file or a database; and the storage content comprises one or more of a corresponding power grid model ID, a time point, a signal point ID, a signal point name, whether a device signal in the power grid model, and a signal value.

11. The power grid model and its associated sample data generating apparatus according to claim 7, characterized by, The data matching module generates a sample pre-configuration requirement table according to the sample and the verification case requirement; extracts sample data from the sample database associated with the matched power grid model according to the sample pre-configuration requirement table and the matched power grid model to obtain the associated sample data of the matched power grid model, and the step specifically comprises: The sample pre-configuration requirement table comprises one of the following two forms: the first form: a time range and a signal point ID; and the second form: a time range, a comprehensive signal ID group, and a comprehensive signal calculation formula; the comprehensive signal ID group comprises at least two signal point IDs. According to the sample pre-configuration requirement table, signal point data within a specified time range or new data calculated from a plurality of signal point data through a comprehensive signal calculation formula is extracted from the sample database associated with the matched power grid model as the associated sample data of the matched power grid model.

12. The power grid model and its associated sample data generating apparatus according to claim 7, characterized by, In the step of outputting the matched power grid model and the associated sample data according to the output sample format in the sample pre-configuration requirement table, the output sample format is one or more of a CSV format, a txt format, a CIM / E format, and an Excel file format.

13. An electronic device, comprising: The computer readable storage medium stores at least one instruction, and the at least one instruction is executed by the processor to implement the power grid model and the associated sample data generation method.

14. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one instruction, and the at least one instruction is executed by the processor to implement the power grid model and the associated sample data generation method.

Images