Power grid dispatching scene abnormal information reporting method and device, terminal equipment and storage medium

Abstract

The invention discloses a power grid dispatching scene abnormal information reporting method and device, terminal equipment and a storage medium. The method comprises the following steps: acquiring abnormal information in a power grid dispatching process; performing vector conversion according to the abnormal information to obtain an abnormal information vector; matching is carried out according to the abnormal information vector and a preset report requirement document vector, and a report requirement document of the abnormal information is determined; according to the abnormal information and the report requirement document, determining a target report format of the abnormal information; inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document; and sending the target report document to a report unit corresponding to the report unit code according to the report unit code in the target report format and the target report document. According to the invention, the accuracy and efficiency of reporting the abnormal information are improved.

Description

Technical Field

[0001] The present invention relates to the field of power grid dispatching technology, and in particular to a method, device, terminal equipment and storage medium for reporting abnormal information in a power grid dispatching scenario. Background Art

[0002] In the dispatching scenario of the power market, timely reporting of abnormal information is a key link to ensure the safe, stable and efficient operation of the power system. As the complexity and scale of the power system continue to expand, the types and number of abnormal situations are also increasing. However, traditional information reporting methods mainly rely on manual monitoring and reporting, which has obvious limitations. The efficiency of manual reporting is relatively low, and it is difficult to quickly identify all abnormal information for reporting, which is prone to missed reports or false reports. Summary of the invention

[0003] The embodiments of the present invention provide a method, device, terminal device and storage medium for reporting abnormal information in a power grid dispatching scenario, which can effectively solve the problems of relatively low efficiency of manual reporting of abnormal information in the prior art, difficulty in quickly identifying all abnormal information for reporting, and easy to cause missed reports or false reports.

[0004] An embodiment of the present invention provides a method for reporting abnormal information in a power grid dispatching scenario, comprising:

[0005] Obtain abnormal information during power grid dispatching;

[0006] Performing vector conversion according to the abnormal information to obtain an abnormal information vector;

[0007] Matching the abnormal information vector with a preset reporting requirement document vector to determine a reporting requirement document for the abnormal information;

[0008] Determining a target reporting format for the exception information according to the exception information and the reporting requirement document;

[0009] Inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document;

[0010] According to the reporting unit code in the target reporting format and the target reporting document, the target reporting document is sent to the reporting unit corresponding to the reporting unit code.

[0011] Further, vector conversion is performed according to the abnormal information to obtain an abnormal information vector, including:

[0012] Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information;

[0013] Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information;

[0014] The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

[0015] Furthermore, the reporting requires that the document vector be stored in a preset vector database;

[0016] Matching the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document for the abnormal information includes:

[0017] Calculate the vector similarity between the abnormal information and each reporting requirement document vector according to the abnormal information vector and the reporting requirement document vector in the preset vector database;

[0018] The vector similarities are sorted in descending order, and the reporting requirement document vector corresponding to the vector similarity with the highest similarity is used as the target reporting requirement document vector;

[0019] According to the target reporting requirement document vector, a reporting requirement document corresponding to the target reporting requirement document vector is determined as the reporting requirement document for the abnormal information.

[0020] Further, according to the abnormal information and the reporting requirement document, determining a target reporting format of the abnormal information includes:

[0021] Determining, according to the abnormal information, an information level of the abnormal information;

[0022] Determine the reporting unit code and reporting unit level according to the reporting requirement document;

[0023] A target reporting format of the abnormal information is determined according to the information level, the reporting unit code, the reporting unit level and the reporting requirement document.

[0024] Furthermore, the abnormal information and the target reporting format are input into a preset large language model to generate a target reporting document, including:

[0025] Inputting the target reporting format into a preset large language model for framework construction to generate a target reporting format framework;

[0026] Inputting the abnormal information into a preset large language model, filling the abnormal information in the generated target reporting format framework, and obtaining an initial reporting document;

[0027] Verifying the initial reporting document according to the target reporting format to determine whether the initial reporting document complies with the target reporting format;

[0028] If yes, the initial report document is used as the target report document;

[0029] If not, regenerate the target reporting format framework.

[0030] Furthermore, the training of the large language model includes:

[0031] Obtain historical exception information, historical reporting requirements documents, and historical reporting documents;

[0032] Determine the historical reporting format of the historical abnormal information according to the historical reporting requirement document;

[0033] Inputting the historical abnormal information and the historical reporting format into the large language model to be trained for training, and generating a historical target reporting document;

[0034] Calculating a loss function value according to the historical target reporting document and the historical reporting document, and comparing the loss function value with a preset loss function threshold;

[0035] When the loss function value is greater than a preset loss function threshold, adjusting the model training parameters according to the loss function value;

[0036] When the loss function value is less than or equal to a preset loss function threshold, a trained large language model is obtained.

[0037] As an improvement of the above solution, another embodiment of the present invention provides a device for reporting abnormal information of a power grid dispatching scenario, including:

[0038] Abnormal data acquisition module, used to obtain abnormal information during power grid dispatching;

[0039] An exception vector conversion module, used for performing vector conversion according to the exception information to obtain an exception information vector;

[0040] A reporting requirement determination module, used to match the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document of the abnormal information;

[0041] A reporting format determination module, used to determine a target reporting format of the abnormal information according to the abnormal information and the reporting requirement document;

[0042] A reporting document generation module, used for inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document;

[0043] The document reporting module is used to send the target reporting document to the reporting unit corresponding to the reporting unit code according to the reporting unit code in the target reporting format and the target reporting document.

[0044] Further, the abnormal vector conversion module is used to perform vector conversion according to the abnormal information to obtain the abnormal information vector; performing vector conversion according to the abnormal information to obtain the abnormal information vector includes:

[0045] Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information;

[0046] Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information;

[0047] The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

[0048] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a method for reporting abnormal information in a power grid dispatching scenario as described in the above embodiment.

[0049] Another embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for reporting abnormal information in a power grid dispatching scenario as described in the above embodiment.

[0050] By implementing the present invention, at least the following beneficial effects are achieved:

[0051] The present invention provides a method, device, terminal device and storage medium for reporting abnormal information in a power grid dispatching scenario. The method can obtain abnormal information in the power grid dispatching process; perform vector conversion according to the abnormal information to obtain an abnormal information vector; match the abnormal information vector and a preset reporting requirement document vector to determine the reporting requirement document of the abnormal information; determine the target reporting format of the abnormal information according to the abnormal information and the reporting requirement document; input the abnormal information and the target reporting format into a preset large language model to generate a target reporting document; according to the reporting unit code in the target reporting format and the target reporting document, send the target reporting document to the reporting unit corresponding to the reporting unit code. Through the large language model, the target reporting document can be quickly generated and sent to the corresponding reporting unit, which is faster than manual reporting and improves the efficiency of abnormal information reporting; at the same time, the large language model can accurately determine the reporting format of the abnormal information, improve the retrieval efficiency of the target reporting format, and the whole process adopts automatic retrieval generation to avoid errors and omissions in manual reporting, thereby improving the accuracy of abnormal information reporting. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a first flow chart of a method for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of a reporting requirement document retrieval process of a method for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention;

[0054] Figure 3 It is a second flow diagram of a method for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention;

[0055] Figure 4 It is a schematic diagram of a reporting dialogue interface of a method for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention;

[0056] Figure 5 It is a structural diagram of a device for reporting abnormal information in a power grid dispatching scenario provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figure 1 , is a flow chart of a method for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention, comprising:

[0059] S1. Obtain abnormal information during power grid dispatching;

[0060] S2. Perform vector conversion according to the abnormal information to obtain an abnormal information vector;

[0061] S3, matching the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document for the abnormal information;

[0062] S4. Determine a target reporting format of the exception information according to the exception information and the reporting requirement document;

[0063] S5, inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document;

[0064] S6. According to the reporting unit code in the target reporting format and the target reporting document, send the target reporting document to the reporting unit corresponding to the reporting unit code.

[0065] Specifically, the abnormal information includes: abnormal type, abnormal time, abnormal location, abnormal impact range and abnormal event; the reporting requirement document represents a document that specifies the information or file reporting process, format, content requirements and reporting objects in the power grid; the target reporting document represents the final reporting document that forms the abnormal information in the power grid dispatching scenario.

[0066] In a preferred embodiment of the present invention, abnormal information in the power grid dispatching process is first obtained; vector conversion is performed according to the abnormal information to obtain an abnormal information vector; then, the abnormal information vector and a preset reporting requirement document vector are matched to determine the reporting requirement document of the abnormal information; then, based on the abnormal information and the reporting requirement document, a target reporting format of the abnormal information is determined; the abnormal information and the target reporting format are input into a preset large language model to generate a target reporting document; finally, based on the reporting unit code in the target reporting format and the target reporting document, the target reporting document is sent to the reporting unit corresponding to the reporting unit code. For example, the abnormal information of the power dispatching scenario is obtained from the power trading system, "20:09, the 500kV **B line in the ** area was switched from maintenance to operation (in conjunction with the emergency defect handling of the line #1*3 tower *phase hardware, the risk of the second-level event of **B plant losing voltage caused by the tripping of the 500kV **A line was eliminated, and the entire line was maintained by the ** power plant)", and then the abnormal information was converted into a vector to obtain the vectors of these abnormal information, and then matched according to the abnormal information vectors and the preset reporting requirement document vectors to determine the reporting requirement documents of these abnormal information; then, according to the abnormal information and the reporting requirement document, the target reporting format of the abnormal information is determined; then, the abnormal information and the target reporting format are input into the preset large language model to generate the target reporting document; according to the reporting unit code in the target reporting format and the target reporting document, the target reporting document is sent to the reporting unit corresponding to the reporting unit code (for example, the provincial ** unit and the municipal ** unit). Using the understanding ability of the large language model, the abnormal information and the meaning behind the power system can be automatically perceived. By utilizing the text understanding ability of the large model, load power abnormality information reporting files are generated to reduce manual intervention, reduce workload, and achieve a higher degree of automation.

[0067] Preferably, performing vector conversion according to the abnormal information to obtain the abnormal information vector includes:

[0068] Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information;

[0069] Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information;

[0070] The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

[0071] In a preferred embodiment of the present invention, the text of the abnormal information is segmented and the stems are extracted to reduce unnecessary part-of-speech changes. Then, stop-word filtering is performed based on the stems of the abnormal information to remove irrelevant words, such as common words like "of" and "is", to reduce noise and obtain the processed abnormal information. The processed abnormal information is input into a preset text vector encoder to be converted into a vector representation, obtaining an abnormal information vector. This vector representation can capture the text semantics and facilitate similarity comparison with other documents.

[0072] Schematically, the reported requirement document vector is stored in a preset vector database;

[0073] Matching is performed according to the abnormal information vector and the preset reported requirement document vector to determine the reported requirement document for the abnormal information, including:

[0074] According to the abnormal information vector and the reported requirement document vectors in the preset vector database, calculate the vector similarity between the abnormal information and each reported requirement document vector;

[0075] Sort the vector similarities in descending order, and use the reported requirement document vector corresponding to the highest vector similarity as the target reported requirement document vector;

[0076] According to the target reported requirement document vector, determine the reported requirement document corresponding to the target reported requirement document vector as the reported requirement document for the abnormal information.

[0077] In a preferred embodiment of the present invention, through the RAG (Retrieval-Augmented Generation) technology, retrieve the reported requirement document of the superior unit for the abnormal information. Store the reported requirement document vector in the vector database preset by RAG. The reported requirement document vector has been processed and stored in the vector database in the same encoding manner as the abnormal information according to the reported requirement document. Then, according to the abnormal information vector and the reported requirement document vectors in the preset vector database, calculate the vector similarity between the abnormal information and each reported requirement document vector, such as through cosine similarity or other similarity calculation methods; then sort the vector similarities in descending order, and use the reported requirement document vector corresponding to the highest vector similarity as the target reported requirement document vector; according to the target reported requirement document vector, determine the reported requirement document corresponding to the target reported requirement document vector as the reported requirement document for the abnormal information, that is, use the retriever to query the vector database, that is, find the reported requirement document closest to the current abnormal information vector, such as Figure 2The RAG method is used to quickly retrieve the reporting requirement documents of multiple superior units, which improves the retrieval efficiency, reduces the errors and omissions in manual reporting, and improves the accuracy of reporting; at the same time, the RAG retrieval mechanism is used to make this embodiment adapt to the changes in various reporting requirements without the need for frequent manual updates of rules.

[0078] Preferably, determining the target reporting format of the exception information according to the exception information and the reporting requirement document includes:

[0079] Determining, according to the abnormal information, an information level of the abnormal information;

[0080] Determine the reporting unit code and reporting unit level according to the reporting requirement document;

[0081] A target reporting format of the abnormal information is determined according to the information level, the reporting unit code, the reporting unit level and the reporting requirement document.

[0082] In a preferred embodiment of the present invention, different abnormal information has different reporting requirements for different regions and levels. First, according to the abnormal information, the information level of the abnormal information is determined, such as level A, level B, and level C, to indicate the urgency of the abnormal information; then, according to the reporting requirement document, the reporting unit code and the reporting unit level are determined, and the reporting unit level, such as provincial level 1, provincial level 2, municipal level 1, and municipal level 2, is used to indicate the administrative level of the unit receiving the document. Finally, according to the information level, the reporting unit code, the reporting unit level, and the reporting requirement document, the target reporting format of the abnormal information is determined.

[0083] In another preferred embodiment of the present invention, the large language model uses its own text understanding ability and the set prompt to complete the understanding of the abnormal information; then, the large language model outputs according to the context information; finally, the specific reporting file content for the specific abnormal information is output. The reporting document format requirement prompt is designed to stimulate the performance of the large language model, and the abnormal information and the reporting document format file are input into the large language model for processing in combination with specific prompt words, and the event type is analyzed and the unit that needs to be reported is determined. By using prompts, the large model can be guided to identify a variety of different types of abnormal market information types. This mechanism enables the automatic generation of reporting content according to specific requirements, ensuring the normalization and standardization of reporting.

[0084] Specifically, the abnormal information and the target reporting format are input into a preset large language model to generate a target reporting document, including:

[0085] Inputting the target reporting format into a preset large language model for framework construction to generate a target reporting format framework;

[0086] Inputting the abnormal information into a preset large language model, filling the abnormal information in the generated target reporting format framework, and obtaining an initial reporting document;

[0087] Verifying the initial reporting document according to the target reporting format to determine whether the initial reporting document complies with the target reporting format;

[0088] If yes, the initial report document is used as the target report document;

[0089] If not, regenerate the target reporting format framework.

[0090] In a preferred embodiment of the present invention, based on the matched target reporting format, the large language model automatically generates a reporting file that meets the reporting requirements, including event details and copy information of relevant superior units. First, the exception information and the target reporting format are input into the preset large language model. The large language model uses its own text understanding ability and the set prompt to complete the construction of the target reporting format framework. The exception information is then filled in the generated target reporting format framework to obtain an initial reporting document. The initial reporting document is verified according to the target reporting format to determine whether the initial reporting document complies with the target reporting format. If so, the initial reporting document is used as the target reporting document, such as Figure 3 If not, the target reporting format framework will be regenerated. The target reporting document generated in the end will be reported to the designated superior unit through the corresponding channel to ensure that the abnormal event is handled in a timely manner.

[0091] Schematically, the training of the large language model includes:

[0092] Obtain historical exception information, historical reporting requirements documents, and historical reporting documents;

[0093] Determine the historical reporting format of the historical abnormal information according to the historical reporting requirement document;

[0094] Inputting the historical abnormal information and the historical reporting format into the large language model to be trained for training, and generating a historical target reporting document;

[0095] Calculating a loss function value according to the historical target reporting document and the historical reporting document, and comparing the loss function value with a preset loss function threshold;

[0096] When the loss function value is greater than a preset loss function threshold, adjusting the model training parameters according to the loss function value;

[0097] When the loss function value is less than or equal to a preset loss function threshold, a trained large language model is obtained.

[0098] In another preferred embodiment of the present invention, Figure 4 As shown, by inputting abnormal information, a specific abnormal information reporting requirement document is generated. Input "At 22:55, the ** City Bureau 220kV ** line tripped, phase A, and reclosing was successful"; output "

Reporting requirements

[0099] By implementing this embodiment, abnormal information in the power grid dispatching process is obtained; vector conversion is performed according to the abnormal information to obtain an abnormal information vector; matching is performed according to the abnormal information vector and the preset reporting requirement document vector to determine the reporting requirement document of the abnormal information; according to the abnormal information and the reporting requirement document, the target reporting format of the abnormal information is determined; the abnormal information and the target reporting format are input into the preset large language model to generate a target reporting document; according to the reporting unit code in the target reporting format and the target reporting document, the target reporting document is sent to the reporting unit corresponding to the reporting unit code. Through the large language model, the target reporting document can be quickly generated and sent to the corresponding reporting unit, which is faster than manual reporting and improves the efficiency of abnormal information reporting; at the same time, the large language model can accurately determine the reporting format of the abnormal information and improve the retrieval efficiency of the target reporting format. The whole process adopts automatic retrieval generation to avoid errors and omissions in manual reporting and improve the accuracy of abnormal information reporting.

[0100] See also Figure 5 , is a schematic diagram of a structure of a device for reporting abnormal information in a power grid dispatching scenario provided by an embodiment of the present invention, comprising:

[0101] Abnormal data acquisition module, used to obtain abnormal information during power grid dispatching;

[0102] An exception vector conversion module, used for performing vector conversion according to the exception information to obtain an exception information vector;

[0103] A reporting requirement determination module, used to match the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document of the abnormal information;

[0104] A reporting format determination module, used to determine a target reporting format of the abnormal information according to the abnormal information and the reporting requirement document;

[0105] A reporting document generation module, used for inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document;

[0106] The document reporting module is used to send the target reporting document to the reporting unit corresponding to the reporting unit code according to the reporting unit code in the target reporting format and the target reporting document.

[0107] Specifically, the abnormal vector conversion module is used to perform vector conversion according to the abnormal information to obtain the abnormal information vector; performing vector conversion according to the abnormal information to obtain the abnormal information vector includes:

[0108] Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information;

[0109] Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information;

[0110] The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

[0111] The present invention provides a device for reporting abnormal information in a power grid dispatching scenario. The device acquires abnormal information in a power grid dispatching process according to an abnormal data acquisition module; performs vector conversion according to the abnormal information in an abnormal vector conversion module to obtain an abnormal information vector; matches the abnormal information vector with a preset reporting requirement document vector in a reporting requirement determination module to determine a reporting requirement document for the abnormal information; determines a target reporting format for the abnormal information according to the abnormal information and the reporting requirement document in a reporting format determination module; inputs the abnormal information and the target reporting format into a preset large language model according to a reporting document generation module to generate a target reporting document; and finally, in a document reporting module, sends the target reporting document to a reporting unit corresponding to the reporting unit code according to a reporting unit code in the target reporting format and the target reporting document. Through the large language model, the target reporting document can be quickly generated and sent to the corresponding reporting unit. Compared with manual reporting, it is more efficient and improves the efficiency of reporting abnormal information. At the same time, the large language model can accurately determine the reporting format of abnormal information and improve the retrieval efficiency of the target reporting format. The entire process uses automatic retrieval and generation to avoid errors and omissions in manual reporting and improve the accuracy of abnormal information reporting.

[0112] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0113] Those skilled in the art can clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0114] Another embodiment of the present invention further provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a method for reporting abnormal information in a power grid dispatching scenario as described in the above embodiment is implemented. The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0115] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0116] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory 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 required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device or other volatile solid-state storage device.

[0117] Another embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for reporting abnormal information in a power grid dispatching scenario as described in the above embodiment.

[0118] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium.

[0119] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for reporting abnormal information in a power grid dispatching scenario, characterized in that: include: Obtain abnormal information during power grid dispatching; Performing vector conversion according to the abnormal information to obtain an abnormal information vector; Matching the abnormal information vector with a preset reporting requirement document vector to determine a reporting requirement document for the abnormal information; Determining a target reporting format for the exception information according to the exception information and the reporting requirement document; Inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document; According to the reporting unit code in the target reporting format and the target reporting document, the target reporting document is sent to the reporting unit corresponding to the reporting unit code.

2. A method for reporting abnormal information of a power grid dispatching scenario according to claim 1, characterized in that: Performing vector conversion according to the abnormal information to obtain an abnormal information vector includes: Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information; Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information; The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

3. A method for reporting abnormal information of a power grid dispatching scenario according to claim 1, characterized in that: The reporting requirement document vector is stored in a preset vector database; Matching the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document for the abnormal information includes: Calculate the vector similarity between the abnormal information and each reporting requirement document vector according to the abnormal information vector and the reporting requirement document vector in the preset vector database; The vector similarities are sorted in descending order, and the reporting requirement document vector corresponding to the vector similarity with the highest similarity is used as the target reporting requirement document vector; According to the target reporting requirement document vector, a reporting requirement document corresponding to the target reporting requirement document vector is determined as the reporting requirement document for the abnormal information.

4. A method for reporting abnormal information of a power grid dispatching scenario as claimed in claim 1, characterized in that: Determining a target reporting format of the exception information according to the exception information and the reporting requirement document includes: Determining, according to the abnormal information, an information level of the abnormal information; Determine the reporting unit code and reporting unit level according to the reporting requirement document; A target reporting format of the abnormal information is determined according to the information level, the reporting unit code, the reporting unit level and the reporting requirement document.

5. A method for reporting abnormal information of a power grid dispatching scenario according to claim 1, characterized in that: Input the abnormal information and the target reporting format into a preset large language model to generate a target reporting document, including: Inputting the target reporting format into a preset large language model for framework construction to generate a target reporting format framework; Inputting the abnormal information into a preset large language model, filling the abnormal information in the generated target reporting format framework, and obtaining an initial reporting document; Verifying the initial reporting document according to the target reporting format to determine whether the initial reporting document complies with the target reporting format; If yes, the initial report document is used as the target report document; If not, regenerate the target reporting format framework.

6. A method for reporting abnormal information of a power grid dispatching scenario according to claim 1, characterized in that: The training of the large language model includes: Obtain historical exception information, historical reporting requirements documents, and historical reporting documents; Determine the historical reporting format of the historical abnormal information according to the historical reporting requirement document; Inputting the historical abnormal information and the historical reporting format into the large language model to be trained for training, and generating a historical target reporting document; Calculating a loss function value according to the historical target reporting document and the historical reporting document, and comparing the loss function value with a preset loss function threshold; When the loss function value is greater than a preset loss function threshold, adjusting the model training parameters according to the loss function value; When the loss function value is less than or equal to a preset loss function threshold, a trained large language model is obtained.

7. A device for reporting abnormal information in a power grid dispatching scenario, characterized in that: include: Abnormal data acquisition module, used to obtain abnormal information during power grid dispatching; An exception vector conversion module, used for performing vector conversion according to the exception information to obtain an exception information vector; A reporting requirement determination module, used to match the abnormal information vector with a preset reporting requirement document vector to determine the reporting requirement document of the abnormal information; A reporting format determination module, used to determine a target reporting format of the abnormal information according to the abnormal information and the reporting requirement document; A reporting document generation module, used for inputting the abnormal information and the target reporting format into a preset large language model to generate a target reporting document; The document reporting module is used to send the target reporting document to the reporting unit corresponding to the reporting unit code according to the reporting unit code in the target reporting format and the target reporting document.

8. A device for reporting abnormal information of a power grid dispatching scenario according to claim 7, characterized in that: The abnormal vector conversion module is used to perform vector conversion according to the abnormal information to obtain an abnormal information vector; performing vector conversion according to the abnormal information to obtain an abnormal information vector includes: Perform word segmentation processing according to the abnormal information to extract the stem of the abnormal information; Perform stop word filtering processing according to the stem of the abnormal information to obtain processed abnormal information; The processed exception information is input into a preset text vector encoder for vector conversion to obtain an exception information vector.

9. A terminal device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a method for reporting abnormal information in a power grid dispatching scenario as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for reporting abnormal information in a power grid dispatching scenario as described in any one of claims 1 to 6.

Images