Method and apparatus for generating simulation report of power system

Abstract

The application provides a method and a device for generating a simulation report of a power system. The method comprises: preprocessing power system simulation result data by using a data preprocessing method to obtain target simulation data; processing the target simulation data based on a natural language processing model to obtain target natural language data, wherein the natural language processing model is obtained by training a target data set, and the target data set is obtained by performing data enhancement on an initial data set by using a preset data enhancement method; and filling the target natural language data into a target simulation report template by using a text automatic filling method to obtain a target simulation report, wherein the target simulation report template is a simulation report template corresponding to a target simulation scenario and having the highest score. The method solves the problem of low efficiency in generating a simulation report based on the simulation result of a power system in the prior art.

Description

Technical Field

[0001] This application relates to the field of power systems, and more specifically, to a method, apparatus and computer-readable storage medium for generating simulation reports of power systems. Background Technology

[0002] As the scale of power systems continues to increase, their complexity also increases. Therefore, simulation analysis of power systems using simulation software is becoming increasingly important in the field of power engineering.

[0003] However, the simulation results obtained by using simulation software to analyze power systems usually involve a large amount of data and complex analysis, requiring professional knowledge and experience to interpret correctly.

[0004] Traditional power system simulation reports are typically generated manually through analysis and writing, making the entire process of obtaining the corresponding simulation report based on the power system simulation results quite time-consuming. Therefore, there is an urgent need for a method that can quickly and accurately obtain power system simulation results and corresponding simulation reports. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, and computer-readable storage medium for generating simulation reports of power systems, so as to at least solve the problem of low efficiency in generating simulation reports based on power system simulation results in the prior art.

[0006] To achieve the above objectives, according to one aspect of this application, a method for generating a power system simulation report is provided, comprising: preprocessing power system simulation result data using a data preprocessing method to obtain target simulation data, wherein the power system simulation result data is data obtained by simulating a power system using simulation software, and the data preprocessing method includes data cleaning, data normalization, and feature extraction; processing the target simulation data based on a natural language processing model to obtain target natural language data, wherein the natural language processing model is trained using a target dataset, and the target dataset is obtained by augmenting an initial dataset using a preset data augmentation method, wherein the preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection; and filling the target natural language data into a target simulation report template using an automatic text filling method to obtain a target simulation report, wherein the target simulation report template is a simulation report template corresponding to a target simulation scenario and having the highest score, and the target simulation scenario is the simulation scenario corresponding to the power system simulation result data.

[0007] Optionally, the process of training the natural language processing model based on the target dataset includes: fusing multiple initial natural language processing models to obtain a preset natural language processing model, wherein the multiple initial natural language processing models are constructed using different neural networks; and training the preset natural language processing model using the target dataset to obtain the natural language processing model.

[0008] Optionally, the process of augmenting the initial dataset to obtain the target dataset using the preset data augmentation method includes: performing synonym replacement processing on the sentences containing the target keywords in the initial dataset to obtain the initial dataset after synonym replacement processing; adjusting the sentence structure of the sentences in the initial dataset to obtain the initial dataset after structural adjustment; processing the initial dataset using a generative adversarial network to obtain a predetermined dataset; performing interpolation processing on the initial dataset to obtain the initial dataset after interpolation processing; and combining the initial dataset after synonym replacement processing, the initial dataset after structural adjustment, the predetermined dataset, and the initial dataset after interpolation processing to obtain the target dataset.

[0009] Optionally, an auto-fill method is used to fill the target natural language data into a target simulation report template to obtain a target simulation report. This includes: adjusting the layout and structure of the target simulation report template based on the target natural language data to obtain an adjusted target simulation report; filling the target natural language data into the adjusted target simulation report using the auto-fill method to obtain a predetermined simulation report; generating corresponding visualization elements based on the power system simulation result data corresponding to the target natural language data, the visualization elements including charts, images, and formulas; filling the visualization elements into the predetermined simulation report, and performing contextual analysis on the predetermined simulation report after filling in the visualization elements to obtain the target simulation report.

[0010] Optionally, the process of scoring the simulation report template using scoring rules includes: determining a first weight based on the content relevance of the simulation report template; determining a second weight based on the structural rationality of the simulation report template; determining a third weight based on user feedback; multiplying the content relevance score of the simulation report template by the first weight to obtain a first score value; multiplying the structural rationality score of the simulation report template by the second weight to obtain a second score value; multiplying the user score of the simulation report template by the third weight to obtain a third score value; and determining the first score value, the second score value, and the third score value as the corresponding score of the simulation report template.

[0011] Optionally, the process of determining the target simulation report template that matches the target simulation scenario from a plurality of simulation report templates includes: determining the simulation report template corresponding to the meta tag information that matches the target simulation scenario as the target simulation report template, wherein one simulation report template corresponds to one meta tag information, and the meta tag information is used to characterize the simulation scenario to which the corresponding simulation report template is applicable.

[0012] Optionally, the target dataset is a dataset with label information, and the target dataset includes sentence-type data, chart-type data, and image-type data.

[0013] Optionally, after using an autofill method to fill the target natural language data into the target simulation report template to obtain the target simulation report, the generation method further includes: responding to a predetermined operation on the display screen, receiving adjustment information, the adjustment information being information for adjusting the font, font size, font color, and paragraph format in the target simulation report; and adjusting the target simulation report based on the adjustment information to obtain the adjusted target simulation report.

[0014] According to another aspect of this application, an apparatus for generating a simulation report of a power system is provided, comprising: a preprocessing unit, configured to preprocess power system simulation result data using a data preprocessing method to obtain target simulation data, wherein the power system simulation result data is data obtained by simulating a power system using simulation software, and the data preprocessing method includes data cleaning, data normalization, and feature extraction; a processing unit, configured to process the target simulation data based on a natural language processing model to obtain target natural language data, wherein the natural language processing model is trained using a target dataset, and the target dataset is obtained by augmenting an initial dataset using a preset data augmentation method, wherein the preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection; and a generation unit, configured to fill the target natural language data into a target simulation report template using an automatic text filling method to obtain a target simulation report, wherein the target simulation report template is a simulation report template corresponding to a target simulation scenario and having the highest score, and the target simulation scenario is the simulation scenario corresponding to the power system simulation result data.

[0015] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to execute any of the methods for generating simulation reports of the power system described above.

[0016] The technical solution of this application firstly involves data preprocessing of the power system simulation result data, including data cleaning, data normalization, and feature extraction, to obtain target simulation data. Then, the target simulation data is input into a natural language processing model to obtain target natural language data; that is, the target simulation data is interpreted into target natural language data through the natural language processing model. Finally, an automatic text filling method is used to fill the target natural language data into the target simulation report template, resulting in the target simulation report. This achieves the automatic generation of professional target simulation reports based on power system simulation result data, reducing the user's analysis and writing burden. It also allows users to quickly obtain the target simulation report after receiving the power system simulation result data, reducing the time spent writing simulation reports and thus solving the problem of low efficiency in generating simulation reports based on power system simulation results in existing technologies. Attached Figure Description

[0017] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 A hardware block diagram of a mobile terminal for performing a method for generating a simulation report of a power system, according to an embodiment of this application, is shown.

[0019] Figure 2 A flowchart illustrating a method for generating a simulation report of a power system according to an embodiment of this application is shown.

[0020] Figure 3 A flowchart illustrating a specific power system simulation report generation scheme according to an embodiment of this application is shown.

[0021] Figure 4 A schematic diagram of a device for generating simulation reports of a power system according to an embodiment of this application is shown.

[0022] The above figures include the following reference numerals:

[0023] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application 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 for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover 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.

[0027] As described in the background section, the efficiency of generating simulation reports based on power system simulation results in the prior art is low. To solve the above problems, embodiments of this application provide a method, apparatus and computer-readable storage medium for generating power system simulation reports.

[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0029] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of generating a power system simulation report according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0030] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the power system simulation report generation method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0031] This embodiment provides a method for generating a simulation report of a power system running on a mobile terminal, computer terminal, or similar computing device. 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.

[0032] Figure 2 This is a flowchart of a method for generating a simulation report of a power system according to an embodiment of this application. Figure 2 As shown, the generation method includes the following steps:

[0033] Step S201: The power system simulation result data is preprocessed using a data preprocessing method to obtain the target simulation data. The power system simulation result data is obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction.

[0034] Specifically, data preprocessing involves cleaning, normalizing, and extracting features from power system simulation results. This removes irrelevant data, standardizes data units and ranges, and extracts key parameters and indicators, resulting in a simpler data structure that is more suitable for subsequent Natural Language Processing (NLP) models. In practical applications, data preprocessing methods are not limited to data cleaning, normalization, and feature extraction; any feasible data preprocessing method available in the art can be used. In one specific embodiment, the data preprocessing method also includes principal component analysis and handling missing values.

[0035] Step S202: Based on the natural language processing model, the above target simulation data is processed to obtain target natural language data. The above natural language processing model is trained using the target dataset. The above target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0036] In step S202 above, a trained natural language processing model is used to interpret (i.e., process) the target simulation data obtained after data preprocessing. This allows for the extraction of key information from the target simulation data and its conversion into easily understandable natural language. In one specific embodiment, parameters such as active power, reactive power, and voltage in the load flow calculation results are converted into textual descriptions to help users better understand the power system simulation results data.

[0037] In practical applications, the data augmentation of the initial dataset is not limited to the aforementioned methods such as synonym replacement, sentence structure adjustment, and noise injection. Any feasible pre-defined data augmentation method in the existing technology can also be used to augment the initial dataset and obtain the target dataset, thereby improving the generalization ability of the natural language processing model. In one specific embodiment, the aforementioned pre-defined data augmentation methods may include random insertion, random adjustment, and random deletion, etc.

[0038] In addition, during the construction of the target dataset, experts in the field of power systems can be invited to participate in the annotation process. This can determine the quality and accuracy of the target dataset, ensuring that the trained natural language processing model is more accurate and reliable.

[0039] Step S203: Using the text autofill method, the above target natural language data is filled into the target simulation report template to obtain the target simulation report. The above target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The above target simulation scenario is the simulation scenario corresponding to the above power system simulation result data.

[0040] In one specific embodiment, the target simulation scenario can be a load flow calculation simulation scenario, a short-circuit fault analysis simulation scenario, and a power system stability analysis simulation scenario.

[0041] This embodiment first performs data preprocessing on the power system simulation results data, including data cleaning, data normalization, and feature extraction, to obtain target simulation data. Then, the target simulation data is input into a natural language processing (NLP) model to obtain target natural language data; that is, the NLP model interprets the target simulation data into target natural language data. Finally, an auto-fill method is used to fill the target natural language data into the target simulation report template, resulting in the target simulation report. This achieves the automatic generation of professional target simulation reports based on power system simulation results data, reducing the user's analysis and writing burden. It also allows users to quickly obtain target simulation reports after receiving power system simulation results data, reducing the time spent writing simulation reports and thus solving the problem of low efficiency in generating simulation reports based on power system simulation results in existing technologies.

[0042] 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, and 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.

[0043] In the specific implementation process, the process of training the natural language processing model based on the target dataset in step S202 includes: fusing multiple initial natural language processing models to obtain a preset natural language processing model, wherein the multiple initial natural language processing models are constructed using different neural networks; and training the preset natural language processing model using the target dataset to obtain the natural language processing model. In this embodiment, by fusing multiple initial natural language processing models to obtain the preset natural language processing model, the advantages of multiple initial natural language processing models can be combined, ensuring that the obtained natural language model has high accuracy and robustness. Then, training the preset natural language processing model using the target dataset, i.e., fine-tuning the preset natural language processing model, not only significantly shortens the model training time but also improves the overall performance of the model.

[0044] In one specific embodiment of this application, the initial natural language processing model can be BERT or GPT, etc.

[0045] Of course, during the training of the pre-defined natural language processing (NLP) model using the target dataset, various forms of power system simulation results data (such as numerical data, charts, and images) can be integrated together to train the model, thereby further improving its understanding and expressive capabilities. Simultaneously, interpretability optimization can be introduced during the training process to make the NLP model's conversion results more interpretable, facilitating understanding and verification by power engineers.

[0046] To further improve the generalization ability of the natural language processing model obtained through subsequent training and to further ensure the robustness of the obtained natural language processing model, the process of using the aforementioned preset data augmentation method in step S202 of this application to augment the initial dataset to obtain the target dataset can be implemented through the following steps: performing synonym replacement processing on the sentences containing the target keywords in the initial dataset to obtain the initial dataset after synonym replacement processing; adjusting the sentence structure of the sentences in the initial dataset to obtain the initial dataset after structural adjustment; processing the initial dataset using a generative adversarial network to obtain a predetermined dataset; performing interpolation processing on the initial dataset to obtain the initial dataset after interpolation processing; and combining the initial dataset after synonym replacement processing, the initial dataset after structural adjustment, the predetermined dataset, and the initial dataset after interpolation processing to obtain the target dataset.

[0047] In one specific embodiment, one sentence contains the target keyword "LCC-HVDC"; another sentence contains the target keyword "high-voltage direct current transmission"; and yet another sentence includes the target keyword "conventional direct current transmission". To ensure consistency in terminology and further guarantee the accuracy of the resulting natural language processing model, and since LCC-HVDC, high-voltage direct current transmission, and conventional direct current transmission essentially represent the same meaning, both LCC-HVDC and conventional direct current transmission can be replaced with high-voltage direct current transmission.

[0048] In the above embodiments, by performing synonym replacement on the target keywords in the sentences of the initial dataset, the natural language processing model (NLP) can learn the same meaning under different expressions, thus improving its comprehension ability. Adjusting the sentence structure, such as changing the active / passive voice or adjusting word order, helps the NLP model adapt to different sentence structures, improving its robustness. Using a Generative Adversarial Network (GAN) to generate a predetermined dataset with a certain level of difficulty (the content of the predetermined dataset has the same meaning as the content of the initial dataset) forces the NLP model to learn to solve complex problems more effectively during training, improving its generalization ability. Interpolating the initial dataset to generate an interpolated initial dataset that lies between the original dataset in semantic space helps the NLP model better learn the structure of the semantic space, improving its generalization ability to unseen data.

[0049] Of course, the initial dataset can also be combined with external data (such as domain-related text, knowledge graphs, etc.) to generate an initial dataset with richer background information, which helps the natural language processing model better understand the relevant knowledge in the power system field. Adding some noise to the initial dataset (such as randomly shuffling word order, adding spelling errors, etc.) allows the natural language processing model to learn to ignore irrelevant information during training, improving the model's robustness to interference.

[0050] To further improve the efficiency of target simulation report generation and reduce the incidence of human error, in some embodiments, step S203 can be implemented through steps S2031, S2032, S2033, and S2034. Specifically, step S2031 involves adjusting the layout and structure of the target simulation report template based on the target natural language data to obtain an adjusted target simulation report; step S2032 involves using the automatic text filling method to fill the target natural language data into the adjusted target simulation report to obtain a predetermined simulation report; step S2033 involves generating corresponding visualization elements based on the power system simulation result data corresponding to the target natural language data, including charts, images, and formulas; and step S2034 involves filling the visualization elements into the predetermined simulation report and performing contextual analysis on the pre-filled simulation report to obtain the target simulation report.

[0051] In one specific embodiment, based on numerical data from power system simulation results, such as the load, power, and voltage values ​​of various devices, corresponding charts can be generated. Specifically, the chart type can include one or more of line charts, bar charts, and pie charts. For example, based on load data from different time periods in the power system simulation results, a line chart can be generated to show how the load changes over time.

[0052] In another specific embodiment, based on spatial information such as topology and equipment distribution in the power system simulation results data, a corresponding image can be generated. Specifically, this image can include a topology map of the power system, a heat map of equipment distribution, etc. For example, a topology map can be generated based on the connection relationships of the power system to visually display the connection between various devices in the system.

[0053] In another specific embodiment, based on the calculation process and mathematical model in the power system simulation results data, corresponding formulas can be generated. These formulas may include equipment parameter calculation formulas, power flow calculation formulas, and so on. For example, for the output power of a generator, a calculation formula that includes generator parameters and operating status can be generated.

[0054] In the above embodiments, the style of the target simulation report template, such as font, font size, color, and paragraph formatting, is automatically adjusted based on the target natural language data and user needs. Simultaneously, the chapter order, hierarchical structure, and content distribution of the target simulation report template are automatically adjusted according to the importance and relevance of the target natural language data to improve the logic and readability of the target simulation report. This ensures that the resulting target simulation report has visual consistency and professionalism, improving the user's reading experience. Based on the target natural language data, visualization elements such as charts, images, and formulas are automatically created and inserted into corresponding positions within the predetermined simulation report. These visualization elements help users more intuitively understand the power system simulation results data and improve the readability of the target simulation report. When generating the target simulation report, the contextual relationships between different parts can be analyzed to ensure the logical relationship and coherence between the interpreted target natural language data and other content in the target simulation report. This improves the overall quality and readability of the report.

[0055] Furthermore, during the generation of the target simulation report, users can be allowed to view and modify its content in real time. Through interaction and feedback with users, their personalized needs can be better met, improving the quality and satisfaction of the target simulation report.

[0056] In some specific implementation processes, the process of scoring the simulation report template using scoring rules in step S203 can also be implemented through the following steps: determining a first weight based on the content relevance of the simulation report template, determining a second weight based on the structural rationality of the simulation report template, and determining a third weight based on user feedback; multiplying the content relevance score of the simulation report template by the first weight to obtain a first score value, multiplying the structural rationality score of the simulation report template by the second weight to obtain a second score value, and multiplying the user score of the simulation report template by the third weight to obtain a third score value; and determining the first score value, the second score value, and the third score value as the corresponding scores for the simulation report template. This can help users quickly select a target simulation report template with a more reasonable layout.

[0057] In the above embodiments, a scoring mechanism can also be designed for each simulation report template, scoring the template based on its match with the simulation task requirements. The highest-scoring template can then be selected as the best-matching template. The scoring criteria for each simulation report module can include content relevance and structural rationality. The scoring mechanism can be updated in the following two situations: Template library update: When a simulation report template is added or modified, the scores of all simulation report templates need to be re-evaluated. In this case, all simulation report templates will be scored based on the new metadata tags and scoring criteria; Scoring criteria adjustment: When the user or system adjusts the scoring criteria, the scores of all simulation report templates need to be recalculated. For example, when the importance of a certain type of simulation task changes, the associated scoring weights may need to be adjusted.

[0058] Furthermore, the simulation report generation method of this application also includes a user feedback mechanism. By introducing a user feedback mechanism, users can evaluate the generated simulation report and the target simulation report template used. This helps to understand the user's satisfaction with the target simulation report template and further optimize the scoring criteria and template selection. Simulation report templates can also be clustered, grouping templates with similar characteristics. This reduces computational load and improves template selection efficiency during the template matching stage. The scoring weights can also be automatically adjusted based on user history and feedback, making template selection more aligned with user needs. For example, if a user frequently selects simulation report templates with more detailed content, the system can adjust the scoring weight for content relevance accordingly. Template recommendation is also possible. Based on user history and feedback, simulation report templates that users may be interested in are recommended, which helps improve user satisfaction and enhance the system's user experience. The template library can also be maintained. By regularly maintaining and updating the template library, the content and structure of the templates can be kept consistent with the development of the power system simulation field. Simultaneously, new simulation report templates can be collected and organized to improve the accuracy and diversity of template matching.

[0059] To further simplify the matching of the target simulation report template, the process of determining the target simulation report template that matches the target simulation scenario from multiple simulation report templates in step 203 can also be implemented through the following steps: the simulation report template corresponding to the meta tag information that matches the target simulation scenario is determined as the target simulation report template. One simulation report template corresponds to one meta tag information, and the meta tag information is used to characterize the simulation scenario to which the corresponding simulation report template is applicable.

[0060] In the above embodiments, meta-tag information is added to each simulation report template. This meta-tag information may include simulation task type, analysis method, application scenario, etc. When selecting a simulation report template, the simulation report template with the corresponding meta-tag information is matched according to the specific requirements of the simulation task.

[0061] In one specific embodiment of this application, the target dataset is a dataset with label information, and the target dataset includes sentence-type data, chart-type data, and image-type data, which further ensures that the generalization ability of the natural language training model obtained after subsequent training is strong.

[0062] In practical applications, after using the text autofill method to fill the target natural language data into the target simulation report template to obtain the target simulation report, the generation method further includes: responding to a predetermined operation on the display screen, receiving adjustment information, which is information for adjusting the font, font size, font color, and paragraph format in the target simulation report; and adjusting the target simulation report based on the adjustment information to obtain the adjusted target simulation report. This can further meet the user's personalized needs and further improve the quality and satisfaction of the target simulation report.

[0063] In one specific embodiment of this application, the target simulation report obtained by the generation method of this application can be output in WORD or PDF format.

[0064] To enable those skilled in the art to better understand the technical solution of this application, the implementation process of the power system simulation report generation method of this application will be described in detail below with reference to specific embodiments.

[0065] This embodiment relates to a specific scheme for generating simulation reports for power systems, such as... Figure 3 As shown, it includes the following steps:

[0066] A power company conducted a power system simulation project. This project included tasks such as load flow calculation, short-circuit fault analysis, and power system stability analysis. Upon completion of the project, engineers were required to generate a detailed simulation analysis report.

[0067] Using the method for generating power system simulation reports provided in this application, engineers first obtain power system simulation result data from simulation software.

[0068] Step S1: Perform data preprocessing on the power system simulation results data to obtain the target simulation data.

[0069] Step S2: Input the target simulation data into the trained natural language processing model, so that the natural language processing model can interpret the preprocessed target simulation data, thereby extracting key information from the target simulation data and converting the target simulation data into target natural language data.

[0070] Step S3: The engineer selected a target simulation report template suitable for the needs of the power system simulation project. Using an auto-fill method, the interpreted target natural language data was populated into the target simulation report template to generate a complete simulation analysis report. The simulation analysis report includes a detailed analysis of the load flow calculation results, the fault types and their impacts in the short-circuit fault analysis, and the conclusions of the power system stability analysis.

[0071] Step S4: The engineer adjusted the format of the generated simulation analysis report according to the requirements, such as changing the font and font size, adjusting the paragraph style, etc., to obtain the adjusted simulation analysis report.

[0072] Step S5: Output the adjusted simulation analysis report in PDF format for easy viewing and sharing.

[0073] The method for generating simulation reports for power systems described in this application enables engineers to quickly and accurately generate simulation analysis reports, thereby improving work efficiency and reducing workload.

[0074] This application also provides an apparatus for generating a power system simulation report. It should be noted that this apparatus can be used to execute the method for generating a power system simulation report provided in this application. This apparatus is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0075] The following describes the apparatus for generating simulation reports of power systems provided in the embodiments of this application.

[0076] Figure 4 This is a schematic diagram of the structure of a power system simulation report generation device according to an embodiment of this application. Figure 4 As shown, the generating apparatus includes:

[0077] The preprocessing unit 10 is used to preprocess the power system simulation result data using data preprocessing methods to obtain target simulation data. The power system simulation result data is data obtained by simulating the power system using simulation software. The data preprocessing methods include data cleaning, data normalization, and feature extraction.

[0078] Specifically, data preprocessing involves cleaning, normalizing, and extracting features from power system simulation results. This removes irrelevant data, standardizes data units and ranges, and extracts key parameters and indicators, resulting in a simpler data structure that is more suitable for subsequent Natural Language Processing (NLP) models. In practical applications, data preprocessing methods are not limited to data cleaning, normalization, and feature extraction; any feasible data preprocessing method available in the art can be used. In one specific embodiment, the data preprocessing method also includes principal component analysis and handling missing values.

[0079] Processing unit 20 is used to process the target simulation data based on a natural language processing model to obtain target natural language data. The natural language processing model is trained using the target dataset. The target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0080] In the aforementioned processing unit, a trained natural language processing model is used to interpret (i.e., process) the target simulation data obtained after data preprocessing. This allows for the extraction of key information from the target simulation data and its conversion into easily understandable natural language. In one specific embodiment, parameters such as active power, reactive power, and voltage in the load flow calculation results are converted into textual descriptions to help users better understand the power system simulation results data.

[0081] In practical applications, the data augmentation of the initial dataset is not limited to the aforementioned methods such as synonym replacement, sentence structure adjustment, and noise injection. Any feasible pre-defined data augmentation method in the existing technology can also be used to augment the initial dataset and obtain the target dataset, thereby improving the generalization ability of the natural language processing model. In one specific embodiment, the aforementioned pre-defined data augmentation methods may include random insertion, random adjustment, and random deletion, etc.

[0082] In addition, during the construction of the target dataset, experts in the field of power systems can be invited to participate in the annotation process. This can determine the quality and accuracy of the target dataset, ensuring that the trained natural language processing model is more accurate and reliable.

[0083] The generation unit 30 is used to fill the target natural language data into the target simulation report template using an automatic text filling method to obtain the target simulation report. The target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The target simulation scenario is the simulation scenario that corresponds to the power system simulation result data.

[0084] In one specific embodiment, the target simulation scenario can be a load flow calculation simulation scenario, a short-circuit fault analysis simulation scenario, and a power system stability analysis simulation scenario.

[0085] In this embodiment, the preprocessing unit performs data cleaning, data normalization, and feature extraction on the power system simulation result data to obtain target simulation data. The processing unit inputs the target simulation data into a natural language processing model to obtain target natural language data; that is, the natural language processing model interprets the target simulation data into target natural language data. The generation unit uses an automatic text filling method to fill the target natural language data into the target simulation report template to obtain the target simulation report. This achieves the automatic generation of professional target simulation reports based on power system simulation result data, reducing the user's analysis and writing burden, and allowing users to quickly obtain the target simulation report after obtaining the power system simulation result data, reducing the time users spend writing simulation reports. This solves the problem of low efficiency in generating simulation reports based on power system simulation results in the prior art.

[0086] In the specific implementation process, the aforementioned processing unit includes a model fusion module and a training module. The fusion module fuses multiple initial natural language processing (NLP) models to obtain a preset NLP model. These initial NLP models are constructed using different neural networks. The training module trains the preset NLP model using the target dataset to obtain the final NLP model. In this embodiment, by fusing multiple initial NLP models to obtain the preset NLP model, the advantages of each model can be combined, ensuring high accuracy and robustness of the resulting NLP model. Then, training the preset NLP model using the target dataset—that is, fine-tuning the preset NLP model—significantly shortens the training time and improves the overall performance of the model.

[0087] In one specific embodiment of this application, the initial natural language processing model can be BERT or GPT, etc.

[0088] Of course, during the training of the pre-defined natural language processing (NLP) model using the target dataset, various forms of power system simulation results data (such as numerical data, charts, and images) can be integrated together to train the model, thereby further improving its understanding and expressive capabilities. Simultaneously, interpretability optimization can be introduced during the training process to make the NLP model's conversion results more interpretable, facilitating understanding and verification by power engineers.

[0089] To further improve the generalization ability of the natural language processing model obtained through subsequent training and to further ensure the robustness of the obtained natural language processing model, the above-mentioned processing unit further includes a first processing module, a first adjustment module, a second processing module, a third processing module, and a combination module. Specifically, the first processing module performs synonym replacement processing on the sentences containing the target keywords in the initial dataset to obtain the initial dataset after synonym replacement processing; the first adjustment module adjusts the sentence structure of the sentences in the initial dataset to obtain the initial dataset after structural adjustment; the second processing module processes the initial dataset using a generative adversarial network to obtain a predetermined dataset; the third processing module performs interpolation processing on the initial dataset to obtain the initial dataset after interpolation processing; and the combination module combines the initial dataset after synonym replacement processing, the initial dataset after structural adjustment, the predetermined dataset, and the initial dataset after interpolation processing to obtain the target dataset.

[0090] In one specific embodiment, one sentence contains the target keyword "LCC-HVDC"; another sentence contains the target keyword "high-voltage direct current transmission"; and yet another sentence includes the target keyword "conventional direct current transmission". To ensure consistency in terminology and further guarantee the accuracy of the resulting natural language processing model, and since LCC-HVDC, high-voltage direct current transmission, and conventional direct current transmission essentially represent the same meaning, both LCC-HVDC and conventional direct current transmission can be replaced with high-voltage direct current transmission.

[0091] In the above embodiments, by performing synonym replacement on the target keywords in the sentences of the initial dataset, the natural language processing model (NLP) can learn the same meaning under different expressions, thus improving its comprehension ability. Adjusting the sentence structure, such as changing the active / passive voice or adjusting word order, helps the NLP model adapt to different sentence structures, improving its robustness. Using a Generative Adversarial Network (GAN) to generate a predetermined dataset with a certain level of difficulty (the content of the predetermined dataset has the same meaning as the content of the initial dataset) forces the NLP model to learn to solve complex problems more effectively during training, improving its generalization ability. Interpolating the initial dataset to generate an interpolated initial dataset that lies between the original dataset in semantic space helps the NLP model better learn the structure of the semantic space, improving its generalization ability to unseen data.

[0092] Of course, the initial dataset can also be combined with external data (such as domain-related text, knowledge graphs, etc.) to generate an initial dataset with richer background information, which helps the natural language processing model better understand the relevant knowledge in the power system field. Adding some noise to the initial dataset (such as randomly shuffling word order, adding spelling errors, etc.) allows the natural language processing model to learn to ignore irrelevant information during training, improving the model's robustness to interference.

[0093] To further improve the efficiency of target simulation report generation and reduce the incidence of human error, in some embodiments, the generation unit includes a second adjustment module, a filling module, a generation module, and an analysis module. The second adjustment module adjusts the layout and structure of the target simulation report template based on the target natural language data to obtain an adjusted target simulation report. The filling module uses the automatic text filling method to fill the target natural language data into the adjusted target simulation report to obtain a predetermined simulation report. The generation module generates corresponding visualization elements based on the power system simulation result data corresponding to the target natural language data. These visualization elements include charts, images, and formulas. The analysis module fills the visualization elements into the predetermined simulation report and performs contextual analysis on the pre-filled simulation report to obtain the target simulation report.

[0094] In one specific embodiment, based on numerical data from power system simulation results, such as the load, power, and voltage values ​​of various devices, corresponding charts can be generated. Specifically, the chart type can include one or more of line charts, bar charts, and pie charts. For example, based on load data from different time periods in the power system simulation results, a line chart can be generated to show how the load changes over time.

[0095] In another specific embodiment, based on spatial information such as topology and equipment distribution in the power system simulation results data, a corresponding image can be generated. Specifically, this image can include a topology map of the power system, a heat map of equipment distribution, etc. For example, a topology map can be generated based on the connection relationships of the power system to visually display the connection between various devices in the system.

[0096] In another specific embodiment, based on the calculation process and mathematical model in the power system simulation results data, corresponding formulas can be generated. These formulas may include equipment parameter calculation formulas, power flow calculation formulas, and so on. For example, for the output power of a generator, a calculation formula that includes generator parameters and operating status can be generated.

[0097] In the above embodiments, the style of the target simulation report template, such as font, font size, color, and paragraph formatting, is automatically adjusted based on the target natural language data and user needs. Simultaneously, the chapter order, hierarchical structure, and content distribution of the target simulation report template are automatically adjusted according to the importance and relevance of the target natural language data to improve the logic and readability of the target simulation report. This ensures that the resulting target simulation report has visual consistency and professionalism, improving the user's reading experience. Based on the target natural language data, visualization elements such as charts, images, and formulas are automatically created and inserted into corresponding positions within the predetermined simulation report. These visualization elements help users more intuitively understand the power system simulation results data and improve the readability of the target simulation report. When generating the target simulation report, the contextual relationships between different parts can be analyzed to ensure the logical relationship and coherence between the interpreted target natural language data and other content in the target simulation report. This improves the overall quality and readability of the report.

[0098] Furthermore, during the generation of the target simulation report, users can be allowed to view and modify its content in real time. Through interaction and feedback with users, their personalized needs can be better met, improving the quality and satisfaction of the target simulation report.

[0099] In some specific implementation processes, the aforementioned generation unit includes a first determining module, a second determining module, and a third determining module. The first determining module is used to determine a first weight based on the content relevance of the simulation report template, a second weight based on the structural rationality of the simulation report template, and a third weight based on user feedback. The second determining module is used to determine the product of the content relevance score and the first weight to obtain a first score value, the product of the structural rationality score and the second weight to obtain a second score value, and the product of the user score and the third weight to obtain a third score value. The third determining module is used to determine the first, second, and third score values ​​as the corresponding scores for the simulation report template, thus helping users quickly select a target simulation report template with a reasonable layout.

[0100] In the above embodiments, a scoring mechanism can also be designed for each simulation report template, scoring the template based on its match with the simulation task requirements. The highest-scoring template can then be selected as the best-matching template. The scoring criteria for each simulation report module can include content relevance and structural rationality. The scoring mechanism can be updated in the following two situations: Template library update: When a simulation report template is added or modified, the scores of all simulation report templates need to be re-evaluated. In this case, all simulation report templates will be scored based on the new metadata tags and scoring criteria; Scoring criteria adjustment: When the user or system adjusts the scoring criteria, the scores of all simulation report templates need to be recalculated. For example, when the importance of a certain type of simulation task changes, the associated scoring weights may need to be adjusted.

[0101] Furthermore, the simulation report generation method of this application also includes a user feedback mechanism. By introducing a user feedback mechanism, users can evaluate the generated simulation report and the target simulation report template used. This helps to understand the user's satisfaction with the target simulation report template and further optimize the scoring criteria and template selection. Simulation report templates can also be clustered, grouping templates with similar characteristics. This reduces computational load and improves template selection efficiency during the template matching stage. The scoring weights can also be automatically adjusted based on user history and feedback, making template selection more aligned with user needs. For example, if a user frequently selects simulation report templates with more detailed content, the system can adjust the scoring weight for content relevance accordingly. Template recommendation is also possible. Based on user history and feedback, simulation report templates that users may be interested in are recommended, which helps improve user satisfaction and enhance the system's user experience. The template library can also be maintained. By regularly maintaining and updating the template library, the content and structure of the templates can be kept consistent with the development of the power system simulation field. Simultaneously, new simulation report templates can be collected and organized to improve the accuracy and diversity of template matching.

[0102] To further simplify the matching of the target simulation report template, the generation unit further includes a fourth determining module, which determines the simulation report template corresponding to the meta tag information that matches the target simulation scenario as the target simulation report template. One simulation report template corresponds to one meta tag information, and the meta tag information is used to characterize the simulation scenario to which the corresponding simulation report template is applicable.

[0103] In the above embodiments, meta-tag information is added to each simulation report template. This meta-tag information may include simulation task type, analysis method, application scenario, etc. When selecting a simulation report template, the simulation report template with the corresponding meta-tag information is matched according to the specific requirements of the simulation task.

[0104] In one specific embodiment of this application, the target dataset is a dataset with label information, and the target dataset includes sentence-type data, chart-type data, and image-type data, which further ensures that the generalization ability of the natural language training model obtained after subsequent training is strong.

[0105] In practical applications, the aforementioned generation device further includes a receiving unit and an adjustment unit. The receiving unit, after using an auto-fill method to fill the target natural language data into the target simulation report template to obtain the target simulation report, receives adjustment information in response to a predetermined operation on the display screen. This adjustment information includes information for adjusting the font, font size, font color, and paragraph formatting in the target simulation report. The adjustment unit, based on this adjustment information, adjusts the target simulation report to obtain an adjusted target simulation report. This further satisfies users' personalized needs and further improves the quality and satisfaction of the target simulation report.

[0106] In one specific embodiment of this application, the target simulation report obtained by the generation method of this application can be output in WORD or PDF format.

[0107] The device for generating the simulation report of the aforementioned power system includes a processor and a memory. The preprocessing unit, processing unit, and generation unit are all stored as program units in the memory, and the processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules reside in the same processor; alternatively, the modules may be located in different processors in any combination.

[0108] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can address the low efficiency of generating simulation reports from power system simulation results in existing technologies.

[0109] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0110] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to execute a method for generating a simulation report of the power system.

[0111] Specifically, the methods for generating simulation reports for power systems include:

[0112] Step S201: The power system simulation result data is preprocessed using a data preprocessing method to obtain the target simulation data. The power system simulation result data is obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction.

[0113] Specifically, data preprocessing involves cleaning, normalizing, and extracting features from power system simulation results. This removes irrelevant data, standardizes data units and ranges, and extracts key parameters and indicators, resulting in a simpler data structure that is more suitable for subsequent Natural Language Processing (NLP) models. In practical applications, data preprocessing methods are not limited to data cleaning, normalization, and feature extraction; any feasible data preprocessing method available in the art can be used. In one specific embodiment, the data preprocessing method also includes principal component analysis and handling missing values.

[0114] Step S202: Based on the natural language processing model, the above target simulation data is processed to obtain target natural language data. The above natural language processing model is trained using the target dataset. The above target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0115] In step S202 above, a trained natural language processing model is used to interpret (i.e., process) the target simulation data obtained after data preprocessing. This allows for the extraction of key information from the target simulation data and its conversion into easily understandable natural language. In one specific embodiment, parameters such as active power, reactive power, and voltage in the load flow calculation results are converted into textual descriptions to help users better understand the power system simulation results data.

[0116] In practical applications, the data augmentation of the initial dataset is not limited to the aforementioned methods such as synonym replacement, sentence structure adjustment, and noise injection. Any feasible pre-defined data augmentation method in the existing technology can also be used to augment the initial dataset and obtain the target dataset, thereby improving the generalization ability of the natural language processing model. In one specific embodiment, the aforementioned pre-defined data augmentation methods may include random insertion, random adjustment, and random deletion, etc.

[0117] In addition, during the construction of the target dataset, experts in the field of power systems can be invited to participate in the annotation process. This can determine the quality and accuracy of the target dataset, ensuring that the trained natural language processing model is more accurate and reliable.

[0118] Step S203: Using the text autofill method, the above target natural language data is filled into the target simulation report template to obtain the target simulation report. The above target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The above target simulation scenario is the simulation scenario corresponding to the above power system simulation result data.

[0119] In one specific embodiment, the target simulation scenario can be a load flow calculation simulation scenario, a short-circuit fault analysis simulation scenario, and a power system stability analysis simulation scenario.

[0120] This invention provides an electronic device including a memory and a processor. The memory stores a computer program, and the processor is configured to execute the method for generating a simulation report of a power system through the computer program.

[0121] Specifically, the methods for generating simulation reports for power systems include:

[0122] Step S201: The power system simulation result data is preprocessed using a data preprocessing method to obtain the target simulation data. The power system simulation result data is obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction.

[0123] Specifically, data preprocessing involves cleaning, normalizing, and extracting features from power system simulation results. This removes irrelevant data, standardizes data units and ranges, and extracts key parameters and indicators, resulting in a simpler data structure that is more suitable for subsequent Natural Language Processing (NLP) models. In practical applications, data preprocessing methods are not limited to data cleaning, normalization, and feature extraction; any feasible data preprocessing method available in the art can be used. In one specific embodiment, the data preprocessing method also includes principal component analysis and handling missing values.

[0124] Step S202: Based on the natural language processing model, the above target simulation data is processed to obtain target natural language data. The above natural language processing model is trained using the target dataset. The above target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0125] In step S202 above, a trained natural language processing model is used to interpret (i.e., process) the target simulation data obtained after data preprocessing. This allows for the extraction of key information from the target simulation data and its conversion into easily understandable natural language. In one specific embodiment, parameters such as active power, reactive power, and voltage in the load flow calculation results are converted into textual descriptions to help users better understand the power system simulation results data.

[0126] In practical applications, the data augmentation of the initial dataset is not limited to the aforementioned methods such as synonym replacement, sentence structure adjustment, and noise injection. Any feasible pre-defined data augmentation method in the existing technology can also be used to augment the initial dataset and obtain the target dataset, thereby improving the generalization ability of the natural language processing model. In one specific embodiment, the aforementioned pre-defined data augmentation methods may include random insertion, random adjustment, and random deletion, etc.

[0127] In addition, during the construction of the target dataset, experts in the field of power systems can be invited to participate in the annotation process. This can determine the quality and accuracy of the target dataset, ensuring that the trained natural language processing model is more accurate and reliable.

[0128] Step S203: Using the text autofill method, the above target natural language data is filled into the target simulation report template to obtain the target simulation report. The above target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The above target simulation scenario is the simulation scenario corresponding to the above power system simulation result data.

[0129] In one specific embodiment, the target simulation scenario can be a load flow calculation simulation scenario, a short-circuit fault analysis simulation scenario, and a power system stability analysis simulation scenario.

[0130] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps:

[0131] Step S201: The power system simulation result data is preprocessed using a data preprocessing method to obtain the target simulation data. The power system simulation result data is obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction.

[0132] Step S202: Based on the natural language processing model, the above target simulation data is processed to obtain target natural language data. The above natural language processing model is trained using the target dataset. The above target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0133] Step S203: Using the text autofill method, the above target natural language data is filled into the target simulation report template to obtain the target simulation report. The above target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The above target simulation scenario is the simulation scenario corresponding to the above power system simulation result data.

[0134] The devices mentioned in this article can be servers, PCs, tablets, mobile phones, etc.

[0135] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps:

[0136] Step S201: The power system simulation result data is preprocessed using a data preprocessing method to obtain the target simulation data. The power system simulation result data is obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction.

[0137] Step S202: Based on the natural language processing model, the above target simulation data is processed to obtain target natural language data. The above natural language processing model is trained using the target dataset. The above target dataset is obtained by augmenting the initial dataset using a preset data augmentation method. The preset data augmentation method includes synonym replacement, sentence structure adjustment, and noise injection.

[0138] Step S203: Using the text autofill method, the above target natural language data is filled into the target simulation report template to obtain the target simulation report. The above target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The above target simulation scenario is the simulation scenario corresponding to the above power system simulation result data.

[0139] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0140] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0144] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0145] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0146] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0147] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0148] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0149] 1) In the power system simulation report generation method of this application, firstly, the power system simulation result data is preprocessed by data cleaning, data normalization, and feature extraction to obtain target simulation data; then, the target simulation data is input into a natural language processing model to obtain target natural language data, that is, the target simulation data is interpreted into target natural language data through the natural language processing model; finally, an automatic text filling method is used to fill the target natural language data into the target simulation report template to obtain the target simulation report. This achieves automatic generation of professional target simulation reports based on power system simulation result data, reducing the user's analysis and writing burden, and allowing users to quickly obtain the target simulation report after obtaining the power system simulation result data, reducing the time users spend writing simulation reports, thereby solving the problem of low efficiency in generating simulation reports based on power system simulation results in existing technologies.

[0150] 2) In the power system simulation report generation device of this application, the preprocessing unit is used to perform data preprocessing such as data cleaning, data normalization, and feature extraction on the power system simulation result data to obtain target simulation data; the processing unit is used to input the target simulation data into a natural language processing model to obtain target natural language data, that is, to interpret the target simulation data into target natural language data through the natural language processing model; the generation unit is used to fill the target natural language data into the target simulation report template using an automatic text filling method to obtain the target simulation report. This achieves automatic generation of professional target simulation reports based on power system simulation result data, reducing the user's analysis and writing burden, and allowing users to obtain the target simulation report relatively quickly after obtaining the power system simulation result data, reducing the time users spend writing simulation reports, thereby solving the problem of low efficiency in generating simulation reports based on power system simulation results in the prior art.

[0151] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for generating a simulation report of a power system, characterized in that, include: A data preprocessing method is used to preprocess the power system simulation result data to obtain the target simulation data. The power system simulation result data is the data obtained by simulating the power system using simulation software. The data preprocessing method includes data cleaning, data normalization, and feature extraction. Based on a natural language processing model, the target simulation data is processed to obtain target natural language data. The natural language processing model is trained using the target dataset. The target dataset is obtained by augmenting the initial dataset using a preset data augmentation method, which includes synonym replacement, sentence structure adjustment, and noise injection. The target natural language data is filled into the target simulation report template using an automatic text filling method to obtain the target simulation report. The target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The target simulation scenario is the simulation scenario that corresponds to the power system simulation result data. An automatic text completion method is used to fill the target natural language data into a target simulation report template to obtain a target simulation report. This includes: adjusting the layout and structure of the target simulation report template based on the target natural language data to obtain an adjusted target simulation report; filling the target natural language data into the adjusted target simulation report using the automatic text completion method to obtain a predetermined simulation report; generating corresponding visualization elements based on the power system simulation result data corresponding to the target natural language data, the visualization elements including charts, images, and formulas; filling the visualization elements into the predetermined simulation report, and performing contextual analysis on the predetermined simulation report after filling with visualization elements to obtain the target simulation report. The process of scoring the simulation report template using scoring rules includes: determining a first weight based on the content relevance of the simulation report template; determining a second weight based on the structural rationality of the simulation report template; determining a third weight based on user feedback; multiplying the content relevance score of the simulation report template by the first weight to obtain a first score value; multiplying the structural rationality score of the simulation report template by the second weight to obtain a second score value; multiplying the user score of the simulation report template by the third weight to obtain a third score value; and determining the first score value, the second score value, and the third score value as the corresponding score for the simulation report template.

2. The generation method according to claim 1, characterized in that, The process of training the natural language processing model based on the target dataset includes: Multiple initial natural language processing models are fused to obtain a preset natural language processing model, wherein the multiple initial natural language processing models are constructed using different neural networks; The target dataset is used to train the preset natural language processing model to obtain the natural language processing model.

3. The generation method according to claim 1, characterized in that, The process of augmenting the initial dataset to obtain the target dataset using the preset data augmentation method includes: The sentences containing the target keywords in the initial dataset are subjected to synonym replacement processing to obtain the initial dataset after synonym replacement processing; The sentences in the initial dataset are structurally adjusted to obtain the structurally adjusted initial dataset. The initial dataset is processed using a generative adversarial network to obtain a predetermined dataset; The initial dataset is interpolated to obtain the interpolated initial dataset; The target dataset is obtained by combining the initial dataset after synonym replacement processing, the initial dataset after structural adjustment, the predetermined dataset, and the initial dataset after interpolation processing.

4. The generation method according to any one of claims 1 to 3, characterized in that, The process of determining the target simulation report template that matches the target simulation scenario from a plurality of simulation report templates includes: The simulation report template corresponding to the meta tag information that matches the target simulation scenario is determined as the target simulation report template. Each simulation report template corresponds to one meta tag information, and the meta tag information is used to characterize the simulation scenario to which the corresponding simulation report template is applicable.

5. The generation method according to any one of claims 1 to 3, characterized in that, The target dataset is a dataset with label information, and the target dataset includes sentence-type data, chart-type data, and image-type data.

6. The generation method according to any one of claims 1 to 3, characterized in that, After using an auto-fill method to fill the target natural language data into the target simulation report template to obtain the target simulation report, the generation method further includes: In response to a predetermined operation performed on the display screen, adjustment information is received, which is information for adjusting the font, font size, font color and paragraph format in the target simulation report; Based on the adjustment information, the target simulation report is adjusted to obtain the adjusted target simulation report.

7. A device for generating simulation reports of a power system, characterized in that, include: The preprocessing unit is used to preprocess the power system simulation result data using data preprocessing methods to obtain target simulation data. The power system simulation result data is data obtained by simulating the power system using simulation software. The data preprocessing methods include data cleaning, data normalization, and feature extraction. The processing unit is used to process the target simulation data based on a natural language processing model to obtain target natural language data. The natural language processing model is trained using the target dataset. The target dataset is obtained by augmenting the initial dataset using a preset data augmentation method, which includes synonym replacement, sentence structure adjustment, and noise injection. The generation unit is used to fill the target natural language data into the target simulation report template using an automatic text filling method to obtain the target simulation report. The target simulation report template is the simulation report template that corresponds to the target simulation scenario and has the highest score. The target simulation scenario is the simulation scenario that corresponds to the power system simulation result data. The generation unit includes a second adjustment module, a filling module, a generation module, and an analysis module. The second adjustment module adjusts the layout and structure of the target simulation report template based on the target natural language data to obtain an adjusted target simulation report. The filling module uses the automatic text filling method to fill the target natural language data into the adjusted target simulation report to obtain a predetermined simulation report. The generation module generates corresponding visualization elements based on the power system simulation result data corresponding to the target natural language data. These visualization elements include charts, images, and formulas. The analysis module fills the visualization elements into the predetermined simulation report and performs contextual analysis on the pre-filled simulation report to obtain the target simulation report. The generation unit includes a first determining module, a second determining module, and a third determining module. The first determining module is used to determine a first weight based on the content relevance of the simulation report template, a second weight based on the structural rationality of the simulation report template, and a third weight based on user feedback. The second determining module is used to determine the product of the content relevance score and the first weight of the simulation report template to obtain a first score value, the product of the structural rationality score and the second weight of the simulation report template to obtain a second score value, and the product of the user score and the third weight of the simulation report template to obtain a third score value. The third determining module is used to determine the first score value, the second score value, and the third score value as the corresponding score for the simulation report template.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the method for generating a simulation report of the power system according to any one of claims 1 to 6.

Images