Electric power think tank knowledge base management system

Abstract

The invention provides an electric power think tank knowledge base management system, and relates to the technical field of document retrieval, and the system comprises a document storage module which is used for storing electric power document data; the vector calculation module is used for calculating a document vector of each power document in the power document data; the query retrieval module is used for acquiring similarity between a problem vector of a document query problem and a document vector of each power document; and retrieving the power document corresponding to the document vector when the similarity with the problem vector is the highest from the document storage module, and outputting the power document. Document vectors of all power documents are obtained through the vector calculation module, document query problems are received through the query and retrieval module, and the query and retrieval module is trained based on a training data set, so that the power documents can be retrieved from the document storage module according to similarity between problem vectors and text vectors during query and retrieval; and the efficiency and the accuracy during power document retrieval are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of document retrieval, and in particular to an electric power think tank knowledge base management system. Background Art

[0002] With the continuous development of society, the power industry is one of the pillar industries in modern society. Various companies deeply involved in the power industry are often faced with the challenge of managing large amounts of power document data. Managing power industry document data can be understood as the management of maintenance, archiving, and review of power document data. Usually, power documents stored in databases have problems such as dispersion, disorganization, mixed quality, large quantity, and single copies. Traditional knowledge base management is often limited to simple document storage and retrieval functions, lacking the ability to understand complex document knowledge structures and efficient retrieval methods. When retrieving power documents, there are technical problems such as low retrieval efficiency and low accuracy. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide an electric power think tank knowledge base management system that can solve the technical problems of low retrieval efficiency and low accuracy when retrieving electric power documents in the prior art.

[0004] In order to achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a power think tank knowledge base management system, which includes:

[0006] Document storage module: used to store power document data;

[0007] Vector calculation module: used to calculate the document vector of each power document in the power document data;

[0008] Query retrieval module: includes a question vector acquisition unit, a similarity calculation unit and a document retrieval unit; the question vector acquisition unit is used to receive the document query question raised by the user and obtain the question vector of the document query question; the similarity calculation unit is used to calculate the similarity between the question vector and the document vector of each of the power documents; the document retrieval unit is used to retrieve and output the power document corresponding to the document vector with the highest similarity to the question vector from the document storage module; the query retrieval module is trained by a training data set marked with the similarity between the document query question and the power documents.

[0009] Furthermore, an embodiment of the present invention provides a first possible implementation of the first aspect, wherein calculating the document vector of each power document in the power document data includes:

[0010] Performing text preprocessing on the power document data to obtain preprocessed power document data; wherein the text preprocessing includes: removing stop words, stemming, and formatting;

[0011] A document vector of each power document is obtained based on the preprocessed power document data.

[0012] Furthermore, an embodiment of the present invention provides a second possible implementation of the first aspect, wherein obtaining a question vector for the document query question includes:

[0013] Tokenizing the document query question to obtain multiple question text tokens;

[0014] Each question text tag is embedded to obtain the question vector.

[0015] Furthermore, the embodiment of the present invention provides a third possible implementation of the first aspect, which further includes: a reasoning analysis module;

[0016] The reasoning analysis module is used to perform reasoning analysis based on the stored power document data to provide a power solution; wherein the power solution includes power decision support and power trend prediction.

[0017] Furthermore, the embodiment of the present invention provides a fourth possible implementation of the first aspect, which further includes: a collaborative working module;

[0018] The collaborative working module is used to extract the power document data stored in the document storage module and enable users to collaboratively edit the power document data.

[0019] Furthermore, the embodiment of the present invention provides a fifth possible implementation of the first aspect, which further includes: a data protection module;

[0020] The data protection module is used to encrypt the power document data stored in the document storage module to ensure the security of the power document data.

[0021] Furthermore, the embodiment of the present invention provides a sixth possible implementation of the first aspect, which further includes: a user management module;

[0022] The user management module is used to establish a user account for a user when using the power think tank knowledge base management system, and to assign permissions to the user account.

[0023] Furthermore, the embodiment of the present invention provides a seventh possible implementation of the first aspect, which further includes: an operation log module;

[0024] The operation log module is used to record the operation behavior of each user account when using the power think tank knowledge base management system and generate an operation log.

[0025] Furthermore, the embodiment of the present invention provides an eighth possible implementation of the first aspect, which further includes: a training module;

[0026] The training module is used to obtain multiple document query questions and corresponding power documents, mark the similarity of the vectors between each document query question and each power document, and establish a sample data set based on each marked document query question and each power document; wherein the sample data set includes a training data set;

[0027] The training data set is input into a large language model for training to obtain the query retrieval module.

[0028] Furthermore, the embodiment of the present invention provides a ninth possible implementation of the first aspect, which further includes: a performance testing module;

[0029] The sample data set also includes a test data set;

[0030] The performance testing module is used to input the test data set into the query retrieval module to obtain performance testing parameters to evaluate the performance of the query retrieval module; wherein the performance testing parameters include: accuracy and precision.

[0031] An embodiment of the present invention provides an electric power think tank knowledge base management system, which includes: a document storage module: used to store electric power document data; a vector calculation module: used to calculate the document vector of each electric power document in the electric power document data; a query retrieval module: including a question vector acquisition unit, a similarity calculation unit and a document retrieval unit; the question vector acquisition unit is used to receive a document query question raised by a user and obtain the question vector of the document query question; the similarity calculation unit is used to calculate the similarity between the question vector and the document vector of each electric power document; the document retrieval unit is used to retrieve and output the electric power document corresponding to the document vector with the highest similarity to the question vector from the document storage module; the query retrieval module is trained by a training data set marked with the similarity between the document query question and the electric power document. The vector calculation module in the present invention is used to calculate the document vector of each power document in the power document data. The query retrieval module has been trained using a training data set marked with the similarity between the document query question and the power documents. When the query retrieval module receives a document query question, it can quickly and accurately obtain the similarity between the question vector and the document vector, and by screening the document vector corresponding to the document with the highest similarity to the question vector, accurately retrieve the power document that best meets the document query question from the document storage module and output it, thereby improving the efficiency and accuracy of retrieving power documents.

[0032] Other features and advantages of the embodiments of the present invention will be described in the following description, or some features and advantages can be inferred or determined without doubt from the description, or can be learned by implementing the above-mentioned technologies of the embodiments of the present invention.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 A schematic diagram of the main modules of a power think tank knowledge base management system provided by an embodiment of the present invention is shown;

[0036] Figure 2 A schematic diagram of a unit of a query and retrieval module in a power think tank knowledge base management system provided by an embodiment of the present invention is shown;

[0037] Figure 3 A schematic diagram showing the processing steps of a vector calculation module in an electric power think tank knowledge base management system provided by an embodiment of the present invention is shown;

[0038] Figure 4 A schematic diagram showing the processing steps of a question vector acquisition unit in an electric power think tank knowledge base management system provided by an embodiment of the present invention is shown;

[0039] Figure 5 A full module schematic diagram of an electric power think tank knowledge base management system provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0041] This embodiment provides a power think tank knowledge base management system. For details, see Figure 1 The following is a schematic diagram of the main modules of a power think tank knowledge base management system, which mainly includes:

[0042] Document storage module 11: used to store power document data;

[0043] A distributed crawler architecture is used to acquire various power documents, including power project documents, power technical reports, and power research results. Specifically, these documents include power equipment technical manuals, transmission line design specifications, substation operation and maintenance manuals, power grid dispatching procedures, and other power document files; power equipment CAD drawings, power grid topology diagrams, substation layout diagrams, and other power drawing files; as well as sensor log data such as transformer temperature monitoring and line load rate. The distributed crawler architecture supports incremental collection and scheduled updates of power document data, ensuring the real-time nature of power document data in the power think tank knowledge base management system.

[0044] The collected power document data is preprocessed. Specifically, the collected power text document data is processed by OCR recognition, deduplication, segmentation, and removal of special characters; the collected power drawing data is processed by format unification, coordinate system conversion, layer separation, etc.; the sensor log data is processed by time alignment, outlier filtering, data completion, etc.

[0045] Vector calculation module 12: used to calculate the document vector of each power document in the power document data;

[0046] Obtaining the document vector of each power document facilitates subsequent similarity calculation.

[0047] See also Figure 2 A schematic diagram of a unit of a query and retrieval module in a power think tank knowledge base management system is shown;

[0048] Query retrieval module 13: includes a question vector acquisition unit 131, a similarity calculation unit 132, and a document retrieval unit 133; the question vector acquisition unit 131 is used to receive a document query question raised by a user and obtain a question vector for the document query question; the similarity calculation unit 132 is used to calculate the similarity between the question vector and the document vectors of each power document; the document retrieval unit 133 is used to retrieve from the document storage module 11 the power document corresponding to the document vector with the highest similarity to the question vector and output it; the query retrieval module 13 is trained using a training data set marked with similarities between the document query question and the power documents;

[0049] The query retrieval module 13 uses a large language model (such as chat-GPT4 or DeepSeek) to retrieve the corresponding power documents from the document storage module 11 according to the document query question input by the user. Specifically, the degree of conformity between the power document and the document query question is determined based on the similarity between the question vector and the document vector. The power document corresponding to the document vector with the highest similarity is the optimal answer. Specifically, the calculation formula between the question vector and the document vector is:

[0050] Similarity=Sim(Q,A),

[0051] Wherein, Q represents the question vector, A represents the document vector, and Sim() represents the similarity calculation function; wherein, the similarity calculation function includes the cosine similarity calculation function and the Jaccard similarity calculation function;

[0052] The document vectors are sorted according to the calculated similarity, and the power document corresponding to the document vector with the highest similarity is ranked first as the optimal answer.

[0053] The vector calculation module in the electric power think tank knowledge base management system provided by the present invention is used to calculate the document vector of each electric power document in the electric power document data. The query retrieval module has been trained through a training data set marked with the similarity between the document query question and the electric power documents. When the query retrieval module receives a document query question, it can quickly and accurately obtain the similarity between the question vector and the document vector, and by screening the document vector corresponding to the highest similarity with the question vector, accurately retrieve the electric power document that best meets the document query question from the document storage module and output it, thereby improving the efficiency and accuracy of retrieving electric power documents.

[0054] In one embodiment, see Figure 3 The figure shows a schematic diagram of the processing steps of the vector calculation module in the power think tank knowledge base management system. This embodiment provides a specific implementation method for calculating the document vector of each power document in the power document data:

[0055] Step S301: performing text preprocessing on the power document data to obtain preprocessed power document data; wherein the text preprocessing includes: removing stop words, stemming, and formatting;

[0056] The electric power document data is subjected to text preprocessing to obtain preprocessed electric power document data. Specifically, the text preprocessing includes stop word removal: removing words that do not affect the semantics in the electric power document data; stemming: simplifying words in different states in the electric power document data, for example, removing plural nouns and different tenses of verbs in English; encoding conversion: converting words in the electric power document data into a unified format.

[0057] Step S303, obtaining a document vector for each power document based on the preprocessed power document data;

[0058] The preprocessed power document data is converted into document vectors based on the vector conversion model; among them, the vector conversion model includes the Sentence-BERT model and the Doc2Vec model.

[0059] In one embodiment, see Figure 4 The figure shows a schematic diagram of the processing steps of a question vector acquisition unit in a power think tank knowledge base management system. This embodiment provides a specific implementation method for obtaining the question vector of a document query question:

[0060] Step S401: tokenize the document query question to obtain multiple question text tokens;

[0061] The document query question is segmented into multiple question text blocks (tokens), and each token is mapped to a pre-set internal vocabulary to obtain a unique number corresponding to each token (i.e., question text tag, also recorded as tokens_id).

[0062] Step S403: embed each question text tag to obtain a question vector;

[0063] Each question text token (i.e., the unique number corresponding to each token) is converted into a high-dimensional vector to obtain a question vector.

[0064] In one embodiment, see Figure 5 The schematic diagram of the entire module of a power think tank knowledge base management system shown in the figure, the power think tank knowledge base management system provided by this embodiment further includes: a reasoning analysis module 14;

[0065] The reasoning analysis module 14 is used to perform reasoning analysis based on the stored power document data and provide power solutions; wherein the power solution includes power decision support and power trend forecasting

[0066] Based on the large language model, historical data in power document data is mined and analyzed to provide power decision support for the power industry. Specifically, historical data such as grid load and power consumption are analyzed to determine peak power consumption periods and factors affecting power consumption, providing power decision support for grid planning and scheduling.

[0067] Based on the large language model, historical data and industry knowledge in the power document data are analyzed to predict future power trends. Specifically, based on historical electricity consumption data, economic development trends, climate change and other factors in the power document data, the changing trend of electricity consumption in the future is predicted.

[0068] In one embodiment, Figure 5 As shown, the electric power think tank knowledge base management system provided in this embodiment further includes: a collaborative work module 15;

[0069] The collaborative work module 15 is used to extract the power document data stored in the document storage module 11 and enable users to collaboratively edit the power document data;

[0070] The collaborative work module 15 supports multiple users to edit the same power document in the power document data at the same time, realizes the synchronous change of the power document, and meets the needs of team collaboration. Specifically, the collaborative work module 15 uses the OperationalTransformation algorithm to handle the concurrent editing conflicts generated when multiple users collaboratively edit the power document at the same time, ensuring the consistency of the power document data. For example, when two users modify the same power document at the same time, the collaborative work module 15 can intelligently merge the changes to avoid data overwriting; at the same time, the power document edited by the collaborative work module 15 will be transmitted back to the document storage module 11 for storage. The document storage module 11 can realize the backtracking and viewing of historical versions of the power document, support version comparison, and intuitively display the modified content of each power document.

[0071] In one embodiment, Figure 5 As shown, the power think tank knowledge base management system provided in this embodiment further includes: a data protection module 16;

[0072] The data protection module 16 is used to encrypt the power document data stored in the document storage module 11 to ensure the security of the power document data;

[0073] The power document data in the document storage module 11 is stored and encrypted through the data protection module 16. Specifically, the sensitive data in the power document data (such as unpublished power policy drafts, power grid topology maps, and key equipment parameters) is encrypted and protected. The SM4 national encryption algorithm is used for end-to-end encryption, and the key length is set to 128 bits. It supports multiple encryption modes such as CBC and CTR, and has excellent encryption performance. The data protection module 16 also supports field-level encryption, and can adopt different encryption strategies for power document data of different sensitivity levels to balance security and ease of use; at the same time, the data protection module 16 can also adjust the content and format of the power document data in the document storage module 11, so that the power document data meets the compliance requirements of the power industry (including the corporate standards of the State Grid Corporation of China, power safety production regulations, power facility protection regulations and other regulatory requirements).

[0074] In one embodiment, Figure 5 As shown, the power think tank knowledge base management system provided by this embodiment further includes: a user management module 17;

[0075] The user management module 17 is used to establish user accounts for users to use the power think tank knowledge base management system and to assign permissions to user accounts;

[0076] The user management module 17 can establish multiple user accounts based on users, and assign different permissions to user accounts based on user roles (such as administrators, ordinary employees, equipment engineers, etc.). For example, only administrators are allowed to modify core power document data; ordinary employees can only query power documents and have no editing permissions; the user management module 17 adopts a role-based access control (RBAC) model to ensure that users only have permissions related to their responsibilities; at the same time, the minimum permissions of each user role can be set according to the work requirements of the user role. For example, R&D personnel can only access and write power technical documents in the power document data, and the sales department can only view and write power market information in the power document data.

[0077] In one embodiment, Figure 5 As shown, the power think tank knowledge base management system provided in this embodiment further includes: an operation log module 18;

[0078] The operation log module 18 is used to record the operation behavior of each user account when using the power think tank knowledge base management system and generate an operation log;

[0079] The operation log module 18 records all operations of each user account when using the power think tank knowledge base management system (including query, editing, downloading, equipment parameter modification, safety procedure update, etc.). The operation log generated based on the operation is stored through blockchain technology to ensure that the operation log cannot be tampered with. At the same time, the operation log module 18 supports log analysis and auditing, and can generate operation statistical reports and abnormal behavior warnings to assist in the security monitoring of the power think tank knowledge base management system.

[0080] In one embodiment, Figure 5 As shown, the power think tank knowledge base management system provided by this embodiment further includes: a training module 19;

[0081] The training module 19 is used to obtain multiple document query questions and corresponding power documents, mark the similarity of the vectors between each document query question and each power document, and establish a sample data set based on the marked document query questions and each power document; wherein the sample data set includes a training data set;

[0082] Obtain multiple document query questions and corresponding power documents, calculate the question vector of each document query question, calculate the document vector of each power document, calculate the similarity between each question vector and each document vector, and mark each document query question and power document based on each similarity. After marking, establish a sample data set for each document query question and each power document.

[0083] Input the training data set into the large language model for training to obtain a query retrieval module 13;

[0084] The large language model is trained using a training data set, so that the large language model can determine the similarity between the document query question and the power document based on the two. Retrieving documents based on similarity helps improve the accuracy of retrieval.

[0085] In one embodiment, Figure 5 As shown, the power think tank knowledge base management system provided by this embodiment further includes: a performance testing module 20;

[0086] The sample dataset also includes a test dataset;

[0087] The performance testing module 20 is used to input the test data set into the query retrieval module 13 to obtain performance testing parameters to evaluate the performance of the query retrieval module 13; wherein the performance testing parameters include: accuracy and precision;

[0088] After obtaining the test data set and inputting it into the query retrieval module 13, the predicted value of the similarity between each document query question and the power document is obtained, and the accuracy and precision are determined based on the predicted value of the similarity and the actual value of the similarity between each document query question and the power document; specifically, the higher the accuracy obtained, the better the performance of the query retrieval module 13; the higher the precision obtained can reflect to a certain extent that the performance of the query retrieval module 13 is better.

[0089] The electric power think tank knowledge base management system provided in the embodiment of the present invention is also provided with an API integration function, which connects with the power industry-specific systems (such as SCADA, ERP) through a standardized interface to realize the linkage analysis between the electric power think tank knowledge base management system and other external devices; among them, the API supports two calling methods, REST and GraphQL, and provides functions such as query, push, and subscription of electric power document data.

[0090] The power think tank knowledge base management system provided by the embodiment of the present invention is mainly aimed at power industry enterprises (for example, power companies, energy suppliers, equipment manufacturers, etc.), power industry companies (for example, companies that provide power policy interpretation and consulting services), institutions and universities engaged in power industry research, government departments and regulatory agencies; the above-mentioned power industry enterprises and power industry companies need to efficiently manage a large amount of project knowledge and technical documents to support decision-making and service delivery; the above-mentioned institutions and universities need to store and share power research results and power technical reports, etc.; the above-mentioned government departments and regulatory agencies need to access, query, and analyze the power industry knowledge base;

[0091] The electric power think tank knowledge base management system can perform similarity analysis between the question vector and the document vector of the document query question input by the user based on the trained large language model, retrieve the corresponding electric power document with the highest similarity and use it as output, thereby improving the efficiency of electric power document retrieval and ensuring the accuracy of electric power document retrieval; the data protection module 16 realizes encryption protection of electric power document data, thereby improving the security of the electric power think tank knowledge base management system; the user management module 17 sets the permissions of each user account, so that each user can only use the electric power think tank knowledge base management system within the scope of his or her own permissions, thereby avoiding arbitrary modification of the electric power think tank knowledge base management system; the collaborative work module 15 realizes collaborative editing of electric power documents by multiple users, thereby improving team work efficiency; the reasoning analysis module 14 provides electric power decision support and electric power trend forecasting; and the API integration function is provided to realize the integration of the electric power think tank knowledge base management system with other systems, thereby realizing linkage analysis.

[0092] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0093] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0094] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0095] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A power think tank knowledge base management system, characterized in that: include: Document storage module: used to store power document data; Vector calculation module: used to calculate the document vector of each power document in the power document data; Query retrieval module: includes a question vector acquisition unit, a similarity calculation unit and a document retrieval unit; the question vector acquisition unit is used to receive the document query question raised by the user and obtain the question vector of the document query question; the similarity calculation unit is used to calculate the similarity between the question vector and the document vector of each of the power documents; the document retrieval unit is used to retrieve and output the power document corresponding to the document vector with the highest similarity to the question vector from the document storage module; the query retrieval module is trained by a training data set marked with the similarity between the document query question and the power documents.

2. The electric power think tank knowledge base management system according to claim 1 is characterized in that: The calculating of the document vector of each power document in the power document data includes: Performing text preprocessing on the power document data to obtain preprocessed power document data; wherein the text preprocessing includes: removing stop words, stemming, and formatting; A document vector of each power document is obtained based on the preprocessed power document data.

3. The electric power think tank knowledge base management system according to claim 1 is characterized in that: The step of obtaining a question vector for the document query question includes: Tokenizing the document query question to obtain multiple question text tokens; Each question text tag is embedded to obtain the question vector.

4. The electric power think tank knowledge base management system according to claim 1 is characterized in that: Also includes: Reasoning analysis module; The reasoning analysis module is used to perform reasoning analysis based on the stored power document data to provide a power solution; wherein the power solution includes power decision support and power trend prediction.

5. The electric power think tank knowledge base management system according to claim 1 is characterized in that: Also includes: Collaborative work module; The collaborative working module is used to extract the power document data stored in the document storage module and enable users to collaboratively edit the power document data.

6. The electric power think tank knowledge base management system according to claim 1 is characterized in that: Also includes: Data protection module; The data protection module is used to encrypt the power document data stored in the document storage module to ensure the security of the power document data.

7. The electric power think tank knowledge base management system according to claim 1, characterized in that: Also includes: User management module; The user management module is used to establish a user account for a user when using the power think tank knowledge base management system, and to assign permissions to the user account.

8. The electric power think tank knowledge base management system according to claim 7 is characterized in that: Also includes: Operation log module; The operation log module is used to record the operation behavior of each user account when using the power think tank knowledge base management system and generate an operation log.

9. The electric power think tank knowledge base management system according to claim 1, characterized in that: Also includes: Training modules; The training module is used to obtain multiple document query questions and corresponding power documents, mark the similarity of the vectors between each document query question and each power document, and establish a sample data set based on each marked document query question and each power document; wherein the sample data set includes a training data set; The training data set is input into a large language model for training to obtain the query retrieval module.

10. The electric power think tank knowledge base management system according to claim 1, characterized in that: Also includes: Performance testing module; The sample data set also includes a test data set; The performance testing module is used to input the test data set into the query retrieval module to obtain performance testing parameters to evaluate the performance of the query retrieval module; wherein the performance testing parameters include: accuracy and precision.

Images