A power grid disaster prediction method and device, a storage medium and an electronic device

Abstract

The application discloses a power grid disaster prediction method and device, a storage medium and an electronic device, and relates to the field of power grid disaster prediction.The method comprises the following steps: obtaining disaster parameters; inputting the disaster parameters into a first disaster loss prediction model constructed by using a big data algorithm to obtain a first disaster loss result; based on the disaster parameters, performing scenario deduction on the disaster parameters by using a disaster loss knowledge element model combined with a knowledge graph model to obtain a second disaster loss result; correcting the second disaster loss result according to the deviation between the first disaster loss result and the second disaster loss result to obtain a power grid disaster prediction result. By implementing the application, the first disaster loss prediction model obtained by using the big data algorithm processes the disaster parameters to obtain the first disaster loss result, the disaster loss knowledge element model combined with the knowledge graph model is used to perform scenario deduction on the disaster parameters to obtain the second disaster loss result, and then the second disaster loss result is corrected based on the deviation between the first disaster loss result and the second disaster loss result, so that a more accurate power grid disaster prediction result is obtained.

Description

Technical Field

[0001] This invention relates to the field of emergency decision-making technology, specifically to a method, device, storage medium, and electronic equipment for predicting power grid disasters. Background Technology

[0002] The safe and stable operation of the power grid system is crucial to the national economy and people's livelihood, and rapid response is required after a large-scale disaster. In order to quickly complete the emergency response to power grid disasters and prevent serious failures of high-risk facilities in advance, it is necessary to predict disaster losses in advance.

[0003] Existing methods for predicting power grid disaster losses are mainly divided into predictions based on big data and predictions based on physical models. However, big data prediction methods can only estimate the total amount of loss and cannot infer the specific location, while the prediction results of physical model calculation methods have low accuracy. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, storage medium and electronic device for predicting power grid disasters, in order to solve the technical problem that existing power grid disaster loss prediction methods cannot obtain accurate disaster loss results.

[0005] The technical solution proposed in this invention is as follows:

[0006] The first aspect of this invention provides a power grid disaster prediction method, comprising: acquiring disaster parameters; inputting the disaster parameters into a first disaster loss prediction model to obtain a first disaster loss result, wherein the first disaster loss prediction model is constructed using a big data algorithm; based on the disaster parameters, performing scenario deduction using a disaster loss knowledge element model combined with a knowledge graph model to obtain a second disaster loss result; and correcting the second disaster loss result according to the deviation between the first disaster loss result and the second disaster loss result to obtain a power grid disaster prediction result.

[0007] Optionally, the power grid disaster prediction method further includes: inputting the disaster parameters into a second disaster loss prediction model to obtain a first emergency response result, wherein the second disaster loss prediction model is constructed using a big data algorithm; and / or, based on the power grid disaster prediction result, using an emergency knowledge element model combined with a knowledge graph model to perform scenario deduction to obtain a second emergency response result.

[0008] Optionally, the power grid disaster prediction method further includes: based on the power grid disaster prediction results, using a causal knowledge element model combined with a knowledge graph model to perform scenario deduction, and obtaining a causal relationship diagram of power facility failures in the power grid; and converting the causal relationship diagram to obtain a fault text report.

[0009] Optionally, the first disaster loss result includes a first number of power facilities experiencing any type of failure; the second disaster loss result includes the probability of each power facility failing and the type of failure.

[0010] Optionally, the second disaster loss result is corrected based on the deviation between the first disaster loss result and the second disaster loss result to obtain the power grid disaster prediction result, including: summing and calculating the second number of power facilities that have experienced any type of fault based on the second disaster loss result; determining whether the difference between the second number and the first number is greater than a first threshold; and deleting power facilities with a fault probability less than the second threshold in the corresponding fault type when the difference is greater than the first threshold.

[0011] Optionally, the power grid disaster prediction method further includes: determining whether the difference between the first quantity and the second quantity is greater than a third threshold; when it is greater than the third threshold, adding power facilities with a fault probability greater than a fourth threshold from other fault types to the corresponding fault type.

[0012] Optionally, the first disaster loss prediction model is constructed in the following manner: obtaining historical disaster parameters and disaster loss results; training multiple big data algorithms based on the historical disaster parameters and disaster loss results to obtain a disaster loss prediction model corresponding to each big data algorithm; testing the disaster loss prediction model corresponding to each big data algorithm to obtain multiple test results; comparing multiple test results and selecting the disaster loss prediction model corresponding to the most accurate test result as the first disaster loss prediction model.

[0013] A second aspect of this invention provides a power grid disaster prediction device, comprising: a parameter acquisition module for acquiring disaster parameters; a first prediction module for inputting the disaster parameters into a first disaster loss prediction model to obtain a first disaster loss result, wherein the first disaster loss prediction model is constructed using a big data algorithm; a second prediction module for performing scenario deduction based on the disaster parameters using a disaster loss knowledge element model combined with a knowledge graph model to obtain a second disaster loss result; and a correction module for correcting the second disaster loss result based on the deviation between the first disaster loss result and the second disaster loss result to obtain a power grid disaster prediction result.

[0014] A third aspect of the present invention provides a computer-readable storage medium storing computer instructions for causing the computer to perform the power grid disaster prediction method as described in the first aspect and any one of the first aspects of the present invention.

[0015] A fourth aspect of the present invention provides an electronic device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the power grid disaster prediction method as described in the first aspect and any one of the first aspects of the present invention.

[0016] The technical solution provided by this invention has the following effects:

[0017] The power grid disaster prediction method, device, storage medium, and electronic equipment provided in this invention process disaster parameters using a first disaster loss prediction model obtained through big data algorithms to obtain a first disaster loss result. Then, a disaster loss knowledge element model combined with a knowledge graph model is used to perform scenario extrapolation on the disaster parameters to obtain a second disaster loss result. Subsequently, the second disaster loss result is corrected based on the deviation between the first and second disaster loss results, thereby obtaining a more accurate power grid disaster prediction result. This solves the technical problem of inaccurate power grid disaster prediction in existing technologies. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a power grid disaster prediction method according to an embodiment of the present invention;

[0020] Figure 2 This is a structural block diagram of a power grid disaster prediction device according to an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of the structure of a computer-readable storage medium provided according to an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] The terms "first," "second," "third," "fourth," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] According to an embodiment of the present invention, a method for predicting power grid disasters is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] This embodiment provides a power grid disaster prediction method, which can be used in electronic devices such as computers, mobile phones, and tablets. Figure 1 This is a flowchart of a power grid disaster prediction method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0027] Step S101: Obtain disaster parameters. Specifically, disaster parameters include data at the time of the disaster and forecast data. Data at the time of the disaster includes data on the natural disaster itself and data on power facilities in the power grid affected by the natural disaster. Forecast data mainly includes meteorological forecast data for a preset time period after the natural disaster and data on the affected power facilities. For example, if the natural disaster is a typhoon, the disaster parameters obtained are the specific meteorological parameters of the typhoon's formation and landfall, as well as the data parameters of the power facilities affected by the typhoon; it is also necessary to obtain the typhoon's movement path and impact range after landfall, as well as data on power facilities within the impact range.

[0028] Step S102: Input the disaster parameters into the first disaster loss prediction model to obtain the first disaster loss result. The first disaster loss prediction model is constructed using a big data algorithm.

[0029] The first disaster loss prediction model can be trained using big data algorithms such as BP neural networks, random forest algorithms, and support vector machines. Specifically, to improve the accuracy of disaster loss prediction, multiple big data algorithms can be combined, and the model with the best performance can be selected.

[0030] In one embodiment, the first disaster loss prediction model is constructed as follows: historical disaster parameters and disaster loss results are obtained; based on the historical disaster parameters and disaster loss results, multiple big data algorithms are used for training to obtain a disaster loss prediction model corresponding to each big data algorithm; the disaster loss prediction model corresponding to each big data algorithm is tested to obtain multiple test results; the multiple test results are compared, and the disaster loss prediction model corresponding to the most accurate test result is selected as the first disaster loss prediction model.

[0031] When constructing the first disaster loss prediction model using the above method, the test results of multiple disaster loss prediction models can be compared using mean squared error, root mean square error, and mean absolute error to select the optimal model. Furthermore, the training process using multiple big data algorithms can be based on existing training processes for the respective algorithms, which will not be elaborated upon here.

[0032] Once the first disaster loss prediction model is constructed, the acquired disaster parameters are input into the model to output the first disaster loss result. This first disaster loss result includes the number of power facilities experiencing any type of fault. Specific fault types include common faults such as breakage and collapse. Therefore, the first disaster loss result specifically includes the number of power facilities that broke, the number of power facilities that collapsed, and so on. Furthermore, the first disaster loss result also categorizes the power facilities, such as 110kV grid-level power facilities, 220V grid-level power facilities, etc.

[0033] Step S103: Based on the disaster parameters, scenario simulation is performed using a disaster loss knowledge element model combined with a knowledge graph model to obtain the second disaster loss result.

[0034] A knowledge element refers to an indivisible unit of knowledge with complete knowledge expression; it is the smallest controllable unit of explicit knowledge. This disaster damage knowledge element model can be determined by manually constructing rules for the time of each knowledge element based on knowledge element patterns. In contrast, the knowledge graph model is automatically generated based on large-scale documents. The construction method for the knowledge graph model can be based on historical disaster scenarios and other information using existing methods, which will not be elaborated upon here.

[0035] In constructing the disaster loss knowledge element model and knowledge graph model, the document data of historical cases is first acquired. Then, the document data in the historical cases is converted into processable text data, and disaster elements, power facility elements, and disaster loss result elements are extracted. By mining the logical relationships between the elements, the elements and the logical relationships between the elements are saved in the form of triples. Then, based on the saved triple data, a disaster loss knowledge element model based on historical cases is constructed.

[0036] Specifically, after determining the disaster damage knowledge meta-model and the knowledge graph model, the system first determines whether a disaster damage knowledge meta-model for the current disaster scenario has been constructed based on the current disaster parameters. If it has, the current disaster parameters are input into the disaster damage knowledge meta-model for scenario deduction. That is, based on the input elements and the relationships between the various elements in the knowledge meta-model, the current disaster scenario is deduced and analyzed, thereby obtaining a complete event evolution process through a series of scenarios derived from the deduction. If a disaster damage knowledge meta-model for the current disaster scenario has not been constructed, a second disaster damage result is obtained using the knowledge graph model based on the acquired disaster parameters.

[0037] Since the disaster loss knowledge element model or knowledge graph model performs scenario extrapolation and analysis for each element, the second disaster loss result obtained through the disaster loss knowledge element model or knowledge graph model includes the probability of failure and the type of failure for each power facility.

[0038] Step S104: Correct the second disaster loss result based on the deviation between the first disaster loss result and the second disaster loss result to obtain the power grid disaster prediction result. Specifically, since the first disaster loss result is determined by the first disaster loss prediction model, and the first disaster loss prediction model obtained by big data algorithm can obtain a more accurate number of power facility failures; therefore, the number of failures in the second disaster loss result is corrected using the first disaster loss result, thereby obtaining a more accurate power grid disaster prediction result.

[0039] The power grid disaster prediction method provided in this invention uses a first disaster loss prediction model obtained through big data algorithms to process disaster parameters and obtain a first disaster loss result. Then, it uses a disaster loss knowledge element model combined with a knowledge graph model to perform scenario extrapolation on the disaster parameters and obtain a second disaster loss result. Finally, based on the deviation between the first and second disaster loss results, the second disaster loss result is corrected, thereby obtaining a more accurate power grid disaster prediction result. This solves the technical problem of inaccurate power grid disaster prediction in existing technologies.

[0040] In one embodiment, the second disaster loss result is corrected based on the deviation between the first disaster loss result and the second disaster loss result to obtain a power grid disaster prediction result. This includes: summing and calculating a second number of power facilities that have experienced any type of fault based on the second disaster loss result; determining whether the difference between the second number and the first number is greater than a first threshold; if it is greater than the first threshold, deleting power facilities with a fault probability less than the second threshold in the corresponding fault type; determining whether the difference between the first number and the second number is greater than a third threshold; if it is greater than the third threshold, adding power facilities with a fault probability greater than a fourth threshold in other fault types to the corresponding fault type.

[0041] The second disaster loss result includes the probability and type of failure for each power facility. Therefore, the number of power facilities corresponding to each failure type in the second disaster loss result is summed to obtain the second number of power facilities experiencing any failure type. For example, if the second disaster loss result includes power facilities that have broken and power facilities that have collapsed, the number of power facilities that have broken is obtained by summing the probabilities of all power facilities that have broken, and the number of power facilities that have collapsed is obtained by summing the probabilities of all power facilities that have collapsed.

[0042] Then, the first and second quantities of the same fault type, obtained from the first and second disaster loss results, are compared. For example, the number of power facilities that broke in the first disaster loss result is compared with the number of power facilities that broke in the second disaster loss result. If the difference between the second quantity and the first quantity is greater than a first threshold, that is, the second quantity is much larger than the first quantity, then the power facilities with a lower probability of failure in the second quantity are deleted. For example, if the first quantity is 200 and the second quantity is 300, then the power facilities with a lower probability of failure included in the second quantity are deleted, thus obtaining a result where the second quantity and the first quantity are basically equal. The corrected second disaster loss result is then used as the power grid disaster prediction result. This power grid disaster prediction result includes not only the number of power facilities experiencing each fault type, but also the parameters of the specific power facilities that experienced the fault.

[0043] Furthermore, if the difference between the first and second quantities exceeds a third threshold, meaning the first quantity is significantly larger than the second, then power facilities with a higher probability of other fault types are added to this quantity. For example, if the first quantity is 300 and the second quantity is 200, power facilities with a higher probability of that fault type from the other fault types in the second disaster loss result are added to that fault type, thus obtaining a result where the second quantity and the first quantity are approximately equal. The corrected second disaster loss result is then used as the power grid disaster prediction result.

[0044] In one embodiment, the power grid disaster prediction method further includes: inputting the disaster parameters into a second disaster loss prediction model to obtain a first emergency response result, wherein the second disaster loss prediction model is constructed using a big data algorithm; and / or, based on the power grid disaster prediction result, using an emergency knowledge element model combined with a knowledge graph model to perform scenario deduction to obtain a second emergency response result.

[0045] Specifically, in addition to disaster loss results, pre-built knowledge element models, knowledge graph models, or second disaster loss prediction models can be used to predict the emergency response results of the current disaster. The construction method for the second disaster loss prediction model is the same as that for the first disaster loss prediction model; only the training samples need to be changed. In this implementation, when constructing the second disaster loss prediction model, the training samples can be selected from disaster parameters and corresponding emergency response results from historical cases, enabling the trained second disaster loss prediction model to predict emergency response results based on disaster parameters.

[0046] Furthermore, the same construction method can be used for emergency knowledge meta-models and disaster loss knowledge meta-models, only the selected elements differ. When constructing an emergency knowledge meta-model, loss outcome elements, emergency response elements, and emergency resource elements are extracted from historical cases. By mining the logical relationships between these elements, the elements and their relationships are saved as triples. Then, based on the saved triple data, an emergency knowledge meta-model based on historical cases is constructed.

[0047] In one embodiment, the power grid disaster prediction method further includes: based on the power grid disaster prediction results, using a causal knowledge element model combined with a knowledge graph model to perform scenario deduction, and obtaining a causal relationship diagram of power facility failures in the power grid; and converting the causal relationship diagram to obtain a fault text report.

[0048] Specifically, in constructing causal knowledge meta-models, similar to emergency response knowledge meta-models and disaster loss knowledge meta-models, disaster elements and fault logic elements from historical cases are extracted. By mining the logical relationships between these elements, the elements and their relationships are saved as triples. Then, based on the saved triple data, a causal knowledge meta-model based on historical cases is constructed. Furthermore, the obtained causal relationship diagram can be transformed into corresponding text result paragraphs through preset transformation relationships. Then, according to the needs of different power emergency response companies or departments, the transformed text paragraphs are combined and further transformed into fault text reports.

[0049] Furthermore, the first disaster loss prediction model, the second disaster loss prediction model, the disaster loss knowledge element model, the emergency knowledge element model, the type causal knowledge element model, and the knowledge graph model constructed above can be retrained and updated by continuously updating historical cases after the model construction is completed.

[0050] In one embodiment, when the current disaster is a typhoon disaster, the power grid disaster prediction method is implemented according to the following process:

[0051] Step 1: When a typhoon forms in the Pacific Ocean and approaches the coast, just before making landfall, obtain the expected development and path of the typhoon from meteorological departments.

[0052] Step 2: Based on the typhoon's path and its impact range, obtain information on power facilities within the affected area.

[0053] Step 3: Based on the data obtained in the above steps, use the first disaster loss prediction model to predict the types of damage to power facilities that the typhoon may cause and the quantity of each type of damage.

[0054] Step 4: Based on the data obtained in the above steps, use the disaster loss knowledge element model combined with the knowledge graph model to perform scenario simulation to determine the possible fault types and the probability of each power facility experiencing a fault.

[0055] Step 5: If the overall number of faults obtained in Step 4 deviates from the result in Step 3, then the result in Step 4 is corrected based on the result in Step 3. During the correction, if there are too many scenario prediction results for a certain fault type, power facilities with lower probabilities are deleted; if there are too few scenario prediction results for a certain fault type, the prediction results for facilities with a higher probability of this type of fault are changed to indicate that this type of fault has occurred.

[0056] Step 6: After determining the predicted consequences of each facility during the typhoon disaster in Step 5, the specific causal relationships of the failures are revealed through scenario simulation, and the causal chain is stored as graph structure data.

[0057] Step 7: Convert the graph-structured data into text-structured paragraphs using preset transformation relationships. Simultaneously, based on the needs of different power emergency response companies or departments, combine the converted text paragraphs, and further convert the results of Step 4 into readable text results as a disaster loss overview paragraph, ultimately forming a scenario prediction result document that meets the needs of different units or departments.

[0058] This invention also provides a power grid disaster prediction device, such as... Figure 2 As shown, the device includes:

[0059] The parameter acquisition module is used to acquire disaster parameters; for details, please refer to the corresponding section of the above method embodiment, which will not be repeated here.

[0060] The first prediction module is used to input the disaster parameters into the first disaster loss prediction model to obtain the first disaster loss result. The first disaster loss prediction model is constructed using a big data algorithm. For details, please refer to the corresponding part of the above method embodiment, which will not be repeated here.

[0061] The second prediction module is used to perform scenario deduction based on the disaster parameters, using a disaster loss knowledge element model combined with a knowledge graph model, to obtain the second disaster loss result; for details, please refer to the corresponding part of the above method embodiment, which will not be repeated here.

[0062] The correction module is used to correct the second disaster loss result based on the deviation between the first disaster loss result and the second disaster loss result, thereby obtaining the power grid disaster prediction result. For details, please refer to the corresponding section of the above method embodiment; further elaboration will not be repeated here.

[0063] The power grid disaster prediction device provided in this invention uses a first disaster loss prediction model obtained through big data algorithms to process disaster parameters and obtain a first disaster loss result. Then, it uses a disaster loss knowledge element model combined with a knowledge graph model to perform scenario extrapolation on the disaster parameters and obtain a second disaster loss result. Finally, based on the deviation between the first and second disaster loss results, the second disaster loss result is corrected, thereby obtaining a more accurate power grid disaster prediction result. This solves the technical problem of inaccurate power grid disaster prediction in existing technologies.

[0064] For a detailed description of the functions of the power grid disaster prediction device provided in this embodiment, please refer to the power grid disaster prediction method description in the above embodiments.

[0065] This invention also provides a storage medium, such as... Figure 3 As shown, a computer program 601 is stored on it. When executed by a processor, this program implements the steps of the power grid disaster prediction method described in the above embodiments. The storage medium also stores audio and video stream data, feature frame data, interactive request signaling, encrypted data, and a preset data size. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium may also include combinations of the above types of memory.

[0066] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0067] This invention also provides an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 51 and a memory 52, wherein the processor 51 and the memory 52 may be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0068] Processor 51 can be a central processing unit (CPU). Processor 51 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 gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0069] The memory 52, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the corresponding program instructions / modules in the embodiments of the present invention. The processor 51 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 52, thereby realizing the power grid disaster prediction method in the above method embodiments.

[0070] The memory 52 may include a program storage area and a data storage area. The program storage area may store applications required for operating the device and at least one function; the data storage area may store data created by the processor 51, etc. Furthermore, the memory 52 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 52 may optionally include memory remotely located relative to the processor 51, and these remote memories may be connected to the processor 51 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0071] The one or more modules are stored in the memory 52, and when executed by the processor 51, they perform the following: Figure 1 The power grid disaster prediction method in the illustrated embodiment.

[0072] For specific details regarding the aforementioned electronic devices, please refer to the relevant documentation. Figure 1 The relevant descriptions and effects in the illustrated embodiments are for understanding purposes only and will not be repeated here.

[0073] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for predicting power grid disasters, characterized in that, include: Obtain disaster parameters; The disaster parameters are input into a first disaster loss prediction model to obtain a first disaster loss result. The first disaster loss prediction model is constructed using a big data algorithm. The first disaster loss result includes a first number of power facilities that have experienced any type of fault. Based on the disaster parameters, a disaster loss knowledge element model combined with a knowledge graph model is used to perform scenario simulation to obtain a second disaster loss result; wherein, the second disaster loss result includes the probability of each power facility failing and the failure type; The second disaster loss result is corrected based on the deviation between the first disaster loss result and the second disaster loss result to obtain the power grid disaster prediction result, including: The second number of power facilities that experienced any type of failure is calculated by summing the second disaster damage results. Determine whether the difference between the second quantity and the first quantity is greater than the first threshold; When the probability exceeds the first threshold, power facilities with a fault probability less than the second threshold in the corresponding fault type will be deleted. Also includes: Determine whether the difference between the first quantity and the second quantity is greater than a third threshold; When the probability exceeds the third threshold, power facilities with a probability greater than the fourth threshold from other fault types are added to the corresponding fault type.

2. The power grid disaster prediction method according to claim 1, characterized in that, Also includes: The disaster parameters are input into the second disaster loss prediction model to obtain the first emergency response result. The second disaster loss prediction model is constructed using a big data algorithm; and / or, Based on the power grid disaster prediction results, scenario simulations were performed using an emergency knowledge element model combined with a knowledge graph model to obtain the second emergency response result.

3. The power grid disaster prediction method according to claim 1, characterized in that, Also includes: Based on the power grid disaster prediction results, a causal knowledge element model combined with a knowledge graph model is used to perform scenario simulation and obtain a causal relationship diagram of power facility failures in the power grid. The causal relationship diagram is then transformed to obtain a fault text report.

4. The power grid disaster prediction method according to claim 1, characterized in that, The first disaster loss prediction model is constructed in the following manner: Obtain historical disaster parameters and damage results; Based on the historical disaster parameters and disaster loss results, multiple big data algorithms are used for training to obtain a disaster loss prediction model corresponding to each big data algorithm. Each big data algorithm's corresponding disaster loss prediction model was tested, and multiple test results were obtained. By comparing multiple test results, the disaster loss prediction model corresponding to the most accurate test result is selected as the first disaster loss prediction model.

5. A power grid disaster prediction device, characterized in that, The apparatus is used to implement the method as described in any one of claims 1-4, comprising: The parameter acquisition module is used to acquire disaster parameters; The first prediction module is used to input the disaster parameters into the first disaster loss prediction model to obtain the first disaster loss result. The first disaster loss prediction model is constructed using big data algorithms. The second prediction module is used to perform scenario deduction based on the disaster parameters, using a disaster loss knowledge element model combined with a knowledge graph model, to obtain a second disaster loss result. The correction module is used to correct the second disaster loss result based on the deviation between the first disaster loss result and the second disaster loss result, so as to obtain the power grid disaster prediction result.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the power grid disaster prediction method as described in any one of claims 1-4.

7. An electronic device, characterized in that, include: The system includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the power grid disaster prediction method as described in any one of claims 1-4.

Images