Method for automatically generating data analysis report of hydropower main equipment

Abstract

The invention provides a method for automatically generating a data analysis report of water and electricity main equipment, which is based on a mother set (minimum business unit), and realizes flexible combined report template construction by connecting the mother set (minimum business unit) in series through report outline nodes. Various report template contents are quickly constructed through mother set data instantiation, and a timed task function is provided to automatically embed data and generate a visual report. According to the method, the defects in the prior art are overcome, the report configuration process is standardized, the data analysis report generation efficiency is improved, and meanwhile, the core problem of knowledge precipitation and reuse of the report content block level in a specific service scene is compatible.

Description

Technical Field

[0001] The invention relates to the field of automatic report generation, and in particular to a method for automatically generating a data analysis report for a hydropower main equipment. Background Art

[0002] At present, report generation technology is widely used, mostly for monitoring reports and data report generation. With the construction and development of energy Internet and smart grid, higher requirements are put forward for the status data information of hydropower main equipment. Data analysis report is an important basis for equipment status information and evaluation. Report generation technology can effectively reduce manual labor and improve work efficiency, so it has received great attention. However, most report generation systems have weak report configuration capabilities and poor reusability for generating reports for specific business scenarios, which causes many inconveniences.

[0003] In addition, the existing various report generation methods mainly provide report generation tools for general technology and professional services, each with its own advantages. Although general technical services cover the combination of general technical components and solve some basic problems at the IT level, such as data visualization related technologies, natural language processing technologies, and text automatic generation technologies, they lack customized configuration capabilities for specific industries or business scenarios. Professional services provide fixed templates for specific industries or business scenarios. For example, inspection reports for hydropower plants, electrical test reports, fault location and analysis reports, etc. These solve the problem of efficient report generation in specific fields. However, the report templates lack relative universality and do not utilize the knowledge accumulation of report content and the reuse of templates, which greatly reduces the overall significance of the platform. Summary of the invention

[0004] The main purpose of the present invention is to provide a method for automatically generating a data analysis report of a hydropower main equipment, so as to solve the problem that the visibility meter needs to be cleaned manually, which is inconvenient and easy to scratch the lens surface.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for automatically generating a data analysis report of a hydropower main equipment, the generating method comprising: Master template production: breaking down the data analysis report content into the smallest business units and configuring the content based on the breaking down definition; Template construction, listing the outline of the data analysis report, filling in the content according to each outline node definition, and realizing the goal of automatically loading the corresponding time range data according to different time ranges through global variables; Generate a report. Based on business needs, select a configured report template to generate a static report. Online preview and / or download are supported.

[0006] In the preferred solution, the content of the master page is customized according to business needs, including: Text description: including summary description, which is universal and does not change with time or data; Statistical text: content that has a certain time period and changes with time; suitable for periodic statistics, used when presented in text form, and used in combination with the above text description; Visual display: including business graphics, tables and other visual charts; suitable for periodic monitoring; Model result diagram: graphical output of model calculation results under specific scenarios; Picture: Supports inserting picture function.

[0007] In the preferred solution, the accuracy of the text description is ensured, and the importance of the text is measured using TF-IDF in natural language processing. The formula is: (1); (2); (3); Among them, TF(t, d) is the word frequency, which means the word t In the documentation d The frequency of occurrence in , IDF(t) is the inverse document frequency, indicating the word t The rarity of a word, TF-IDF(t, d) is the weight that comprehensively considers the word frequency and inverse document frequency; In the process of mastering, the TF-IDF formula is applied to determine which words in the text description are the most important, and then decide which words should be emphasized and which can be simplified or omitted.

[0008] In the preferred solution, after the master content is configured and published, all accounts on the platform can use the master content; if it is not published, only this account can use it.

[0009] In a preferred embodiment, the template construction includes: Content configuration: fill in template content according to outline nodes; customize the target directory outline according to the data analysis report theme; and / or directly fill in node content: Configure variables: After configuring local variables or global variables according to the report template content, you can preview the effect of the report template after loading data; Save or publish: save or publish after the template content, data parameters and global variables are configured; Template version management: supports version rollback, you can view the contents of different versions, and choose to open the version according to actual needs.

[0010] In the preferred solution, during the template construction process, compression encoding technology is used to optimize loading speed and reduce memory consumption; Among them, the compression coding technology adopts Huffman coding, which is a lossless data compression method. The construction process of Huffman coding can be expressed as: (4); in, C is a set of characters, C left and C right They are the character sets corresponding to the left subtree and right subtree of the Huffman tree respectively.

[0011] In the preferred embodiment, the report generation includes manual triggering and automatic generation; The manual triggering method selects the report type and time range according to the business needs of the data analysis report; After viewing the preview effect, save it as a static report; The automatic generation method is as follows: select a corresponding report template, select a report type and a time range, and then start a scheduled report generation task, and generate and save the report as a static report according to the time schedule.

[0012] In the preferred solution, when generating a report, MAPE is used to evaluate the accuracy of the model prediction, and MAPE is expressed as: (5); in, A t Indicates the actual value, F t represents the predicted value, N is the sample size. MAPE is used to measure the percentage difference between the predicted value and the actual value. The closer it is to 0, the more accurate the prediction.

[0013] The present invention provides a method for automatically generating data analysis reports for hydropower main equipment. The method is based on the master template (minimum business unit) and connects the master template (minimum business unit) in series through the report outline node to realize flexible combination construction of report templates. It realizes rapid construction of various report template contents through master template data instantiation and provides a scheduled task function to automatically embed data and generate visual reports. It overcomes the shortcomings of the prior art, standardizes the report configuration process, improves the efficiency of data analysis report generation, and is compatible with the core issues of knowledge precipitation and reuse of report content blocks in specific business scenarios.

[0014] This application focuses on the report itself, rather than a specific report based on a specific scenario in a specific industry. In particular, when the hierarchy of main equipment in the hydropower industry is relatively clear and standard, the hydropower main equipment can be taken as the main body, and the data of each main equipment in each scenario can be analyzed in various dimensions to automatically form a data analysis visualization report, which has a certain universality and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a flow chart of the method for automatically generating a data analysis report of a hydropower main equipment according to the present invention; Figure 2 It is a flow chart of the method for automatically generating a data analysis report of the present invention; Figure 3 It is a flow chart of template construction of the present invention; Figure 4 It is a flow chart of generating a report according to the present invention. DETAILED DESCRIPTION Example 1 like Figures 1 to 4 As shown, a method for automatically generating a data analysis report of a hydropower main equipment includes: Master template production: breaking down the data analysis report content into the smallest business units and configuring the content based on the breaking down definition; Template construction, listing the outline of the data analysis report, filling in the content according to each outline node definition, and realizing the goal of automatically loading the corresponding time range data according to different time ranges through global variables; Generate a report. Based on business needs, select a configured report template to generate a static report. Online preview and / or download are supported.

[0016] In the preferred solution, the content of the master page is customized according to business needs, including: Text description: including summary description, which is universal and does not change with time or data; Statistical text: content that has a certain time period and changes with time; suitable for periodic statistics, used when presented in text form, and used in combination with the above text description; Visual display: including business graphics, tables and other visual charts; suitable for periodic monitoring; Model result diagram: graphical output of model calculation results under specific scenarios; Picture: Supports inserting picture function.

[0017] In the preferred solution, the accuracy of the text description is ensured, and the importance of the text is measured using TF-IDF in natural language processing. The formula is: (1); (2); (3); Among them, TF(t, d) is the word frequency, which means the word t In the documentation d The frequency of occurrence in , IDF(t) is the inverse document frequency, indicating the word t The rarity of a word, TF-IDF(t, d) is the weight that comprehensively considers the word frequency and inverse document frequency; In the process of mastering, the TF-IDF formula is applied to determine which words in the text description are the most important, and then decide which words should be emphasized and which can be simplified or omitted.

[0018] In the automatic generation algorithm of hydropower main equipment data analysis report, the main purpose of the master template production stage is to create a standardized basic template so that reports that meet specific business needs can be quickly generated later. In this process, although it mainly involves business logic and technical implementation, some mathematical methods can also be used to assist in improving the quality and accuracy of text descriptions.

[0019] The role of TF-IDF in mastering (1) Text optimization: By calculating the importance of each word in the document, it can help determine which words should be emphasized and which can be simplified or omitted, thereby optimizing the content of the text description. When setting the text description, TF-IDF can be used to identify key terms and concepts to ensure that these contents are fully emphasized in the report, while those less important words can be appropriately reduced in mention.

[0020] (2) Keyword extraction: Helps extract the most representative and relevant keywords, which can be used in report titles, abstracts or indexes. When making a master page, you can pre-set a set of core keywords and use TF-IDF to dynamically update the list of these keywords to ensure that they always reflect the latest business focus and data trends.

[0021] (3) Document summary: Generate concise and clear document summaries to improve report readability. During the mastering stage, TF-IDF can be used to identify the most important sentences in each paragraph of text to build a short and comprehensive summary.

[0022] (4) Content filtering: Remove redundant information and keep the report concise. By calculating the TF-IDF value of each term, you can filter out common words that are irrelevant to the current report topic, ensuring that the report content is more focused on key information.

[0023] Specific steps Define the smallest business unit: First, clarify the basic components of the report, such as an overview description, statistical information, chart presentation, etc.

[0024] Text description: Write an overview description and optimize the text using TF-IDF to ensure that important information is highlighted.

[0025] Statistical text: defines the format and display method of periodic statistical information, such as the maximum and minimum values ​​of equipment operation data.

[0026] Visual presentation: Design the layout and style of visualization elements such as charts and graphics.

[0027] Model result graph: defines the graphical output method of model calculation results.

[0028] Image insertion: Determine the insertion position and size of the image.

[0029] Master configuration: After completing all the above content configurations, choose to publish or limit it to this account.

[0030] example: Create a master page about the operation of a turbine, which needs to include an overview description. To ensure that the description is both comprehensive and focused, use the TF-IDF method to help us decide which words should be emphasized: 1. Collect text: Collect text that describes the operation of the turbine.

[0031] 2. Calculate TF-IDF: Calculate the TF-IDF value for each word in the text.

[0032] 3. Keyword sorting: Sort keywords according to TF-IDF value.

[0033] 4. Text optimization: Based on the ranking results, optimize the description text to ensure that key information is highlighted.

[0034] If the TF-IDF values ​​of words such as "turbine", "operation", and "maintenance" are high, these words should be emphasized in the text description. As a result, the text description in the master will more accurately reflect the key information of the turbine's operating status, thereby improving the overall quality of the report.

[0035] In the preferred solution, after the master content is configured and published, all accounts on the platform can use the master content; if it is not published, only this account can use it.

[0036] Example 2 Further described in conjunction with Example 1, the template construction includes: Content configuration: fill in template content according to outline nodes; customize the target directory outline according to the data analysis report theme; and / or directly fill in node content: Configure variables: After configuring local variables or global variables according to the report template content, you can preview the effect of the report template after loading data; Save or publish: save or publish after the template content, data parameters and global variables are configured; Template version management: supports version rollback, you can view the contents of different versions, and choose to open the version according to actual needs.

[0037] In the preferred solution, during the template construction process, compression encoding technology is used to optimize loading speed and reduce memory consumption; Among them, the compression coding technology adopts Huffman coding, which is a lossless data compression method. The construction process of Huffman coding can be expressed as: (4); in, C is a set of characters, C left and C right They are the character sets corresponding to the left subtree and right subtree of the Huffman tree respectively.

[0038] The main purpose of Huffman coding in the template building process is to optimize data loading speed and reduce memory consumption. By using Huffman coding to compress static content in the template, the size of the template file can be effectively reduced, thereby speeding up loading and reducing server and client resource consumption. The following explains in more detail how Huffman coding helps achieve this goal.

[0039] The role of Huffman coding is: (1) Reduce storage space: Huffman coding can assign different length codes to characters according to their frequency of occurrence in the text. Characters that appear more frequently are assigned shorter codes, while characters that appear less frequently are assigned longer codes. This encoding method can significantly reduce the size of template files because common text elements (such as titles, fixed paragraphs, etc.) will occupy fewer bits.

[0040] (2) Improve transmission efficiency: When templates need to be transmitted over the network, smaller file sizes mean faster transmission speeds. This is especially important for reports that need to be updated and loaded in real time, as it can reduce user waiting time.

[0041] (3) Fast decoding: Huffman coding is a prefix coding, that is, no code is a prefix of another code. This feature makes the decoding process very efficient, and only needs to traverse down the Huffman tree to determine the specific character represented by each code.

[0042] The application in template construction is: In the process of building templates, Huffman coding can help optimize the following aspects: Text description compression: Fixed text descriptions in reports, such as titles, standard terms, etc., can be compressed using Huffman coding.

[0043] Statistical text compression: If the report contains some periodic statistical data, the repeated parts of this data can also be Huffman encoded to reduce storage requirements.

[0044] Template content compression: Static content in a template (such as titles, fixed paragraphs, etc.) can be encoded to reduce the size of the template file itself.

[0045] Data transfer optimization: When templates need to be transferred over the network, compression can significantly reduce transfer time and bandwidth consumption.

[0046] Implementation steps Collect data: Collect all the content that needs to be compressed in the template and its frequency of occurrence.

[0047] Construct Huffman tree: Construct a Huffman tree based on the frequency of occurrence.

[0048] Generate Codes: Generate Huffman codes for each character or symbol.

[0049] Replace content: Use Huffman coding to replace the original characters or symbols in the template.

[0050] Save template: Save the compressed template and record how to decode it.

[0051] example: Given a simple report template containing several fixed paragraphs, we can optimize it using Huffman coding by following the steps below: Analyze text: Identify recurring words or phrases in the template.

[0052] Calculate frequency: Count the number of times each word or phrase occurs.

[0053] Build Huffman tree: Build a Huffman tree based on the frequency.

[0054] Generate encoding: Generate encoding for each word or phrase based on the Huffman tree.

[0055] Replace content: Replace the original word or phrase with the code in the template.

[0056] Test: Make sure the template decodes correctly and displays on a variety of devices.

[0057] In this way we can significantly improve the performance and user experience of report templates.

[0058] Example 3 Further described in conjunction with Example 1, the report generation includes manual triggering and automatic generation; The manual triggering method selects the report type and time range according to the business needs of the data analysis report; After viewing the preview effect, save it as a static report; The automatic generation method is as follows: select a corresponding report template, select a report type and a time range, and then start a scheduled report generation task, and generate and save the report as a static report according to the time schedule.

[0059] In the preferred solution, when generating a report, MAPE is used to evaluate the accuracy of the model prediction, and MAPE is expressed as: (5); in, A t Indicates the actual value, F t represents the predicted value, N is the sample size. MAPE is used to measure the percentage difference between the predicted value and the actual value. The closer it is to 0, the more accurate the prediction.

[0060] Mean Absolute Percentage Error (MAPE) plays a key role in generating reports by evaluating the accuracy of the forecast model. MAPE is a commonly used metric that quantifies the difference between the forecasted values ​​and the actual observed values. Here is how MAPE plays a role in generating reports: Model evaluation: MAPE can help evaluate the accuracy of model predictions. By calculating the mean absolute percentage error between the predicted values ​​and the actual values, you can get an intuitive understanding of how the model performs. A low MAPE value indicates that the model predictions are closer to the actual observed values ​​and the model performs well; conversely, if the MAPE value is high, you may need to further adjust the model or use other prediction methods.

[0061] Decision support: Before making important decisions based on forecasts, it is important to understand the reliability of the forecasts. MAPE provides a quantitative indicator to help decision makers determine whether the forecast results can be trusted. For example, in financial reports, if the MAPE of forecasted sales is low, then budgeting and planning based on these forecasts may be more reliable.

[0062] Performance comparison: MAPE is also commonly used to compare the predictive capabilities of different models. By comparing the MAPE of multiple models, the best model can be selected for the final report. This helps to continuously improve the prediction model, thereby improving the quality and credibility of the report.

[0063] Trend Analysis: MAPE can be used to monitor how a forecasting model changes over time. If the MAPE increases over time, it may indicate that the model needs to be retrained or parameters adjusted.

[0064] Report quality assurance: By calculating MAPE as part of quality control during report generation, you can ensure that the forecast information provided in the report has a certain degree of accuracy. This helps build trust in the report content among report users.

[0065] Implementation steps Collecting data: getting actual values and the corresponding predicted values .

[0066] Calculate the error: For each time point t, calculate the absolute percentage error .

[0067] Calculate the mean: Average the absolute percentage errors at all time points to get MAPE.

[0068] Evaluate the model: Evaluate the performance of the model based on the value of MAPE.

[0069] Report generation: The assessment results are incorporated into the final report for reference.

[0070] In this way, MAPE not only helps us select the optimal forecasting model but also improves the credibility and usefulness of the report.

[0071] Example 4 Further illustrate with reference to Example 1, Figure 1-4As shown in the structure, the data analysis reports for different business scenarios in the hydropower industry have different characteristics from the report content structure to the multi-source data analysis dimension. As the business data analysis changes, the report content must also change. It is difficult to use one report structure to meet multiple business needs. Therefore, high requirements are placed on the configuration capabilities and knowledge accumulation at the report content block level. The innovative focus of this solution is to integrate IT technology and professional business in the design of the entire process of automatic report generation: standardize the report generation process through IT technology, explore professional business scenarios, and extract the smallest business unit with universal adaptability and configure it as a master template, providing a unified and rapid instantiation basis for report template construction, so that in different business scenarios, the smallest business unit for data analysis of hydropower main equipment can be quickly reused in different report templates, which is more extensive and universal.

[0072] This application designs a full process of automatically generating data analysis reports for hydropower business, which is suitable for report design and production in any scenario. A method for automatically generating data analysis reports for hydropower main equipment is provided, such as Figure 1 As shown, specifically: like Figure 2 As shown, the master is the smallest business unit and the starting point of the entire data analysis report automatic generation method. When making a data analysis report for any scenario, the report content is broken down into the smallest business unit, which has the characteristics of universality and the content is configured according to the breakdown definition.

[0073] The content of the master page can be customized according to business needs, including: (1) Text description: An overview description that is universal and does not change with time or data.

[0074] (2) Statistical text: content that has a certain time period and changes with time.

[0075] For example: the minimum-maximum value of a device this month; suitable for periodic statistics, used when displayed in text form, generally used in combination with text descriptions.

[0076] (3) Visual display: including business graphics, tables and other visual charts; suitable for use in periodic monitoring.

[0077] (4) Model result diagram: The output is graphical based on the model calculation results in a specific scenario.

[0078] For example: a scatter plot of abnormal distribution of a certain device in a month; suitable for in-depth analysis of specific issues.

[0079] (5) Picture: Supports the function of inserting pictures.

[0080] In particular, after the master content is configured, you can save or publish it. After publishing, all accounts on the platform can use the master content; if it is not published, only this account can use it.

[0081] Template construction such as Figure 3 The template is a combination of standardized report contents and is the core module of the entire data analysis report automatic generation method.

[0082] Before making a data analysis report for any scenario, first list the outline of the data analysis report, fill in the content according to the definition of each outline node, and use global variables to automatically load the corresponding time range data according to different time ranges.

[0083] In particular, the template construction includes: (1) Content configuration: the method of filling the template content according to the outline nodes, which includes two forms: 1) Configure report outline: customize the target directory outline according to the data analysis report theme; 2) Filling node content: There are two ways: Select an existing master template: Select the required master template and drag it directly into the content editing area. Modify the data parameters according to the business requirements of the template. New content: Customize the edited text description, select the display type, including text statistics, graphics, tables, models, pictures, etc., and save after configuring the data.

[0084] (2) Configure variables: After configuring local variables or global variables according to the report template content, you can preview the effect of loading data into the report template.

[0085] (3) Save or publish: After the template content, data parameters, and global variables are configured, save or publish.

[0086] (4) Template version management: supports version rollback, allows you to view the contents of different versions, and select the version to enable based on actual needs.

[0087] Generate reports such as Figure 4 As shown: According to business needs, select a configured report template to generate a static report, which supports online preview or downloading of word documents.

[0088] There are two ways to generate reports: (1) Manual triggering: Select the report type and time range based on the business needs of the data analysis report; after viewing the preview effect, save it as a static report; (2) Automatic generation: Select the corresponding report template, select the report type and time range, and then start the scheduled report generation task. The report will be generated and saved as a static report according to the time schedule.

[0089] The report types include daily reports, monthly reports, weekly reports, quarterly reports, annual reports, special reports, etc.; the time range is automatically matched according to the report type.

[0090] This application designs a variable hierarchy priority strategy based on the report content block level: in order to meet the needs of data analysis for different report types in different business scenarios, the hierarchy priority strategy of local variables and global variables (local variables take precedence over global variables) is used to control the variable levels between report content block levels, so as to customize the corresponding variables according to the needs of different report content block levels, thereby improving the flexible configuration capabilities of various business report templates.

[0091] A knowledge system at the report content block level has been constructed: by deeply exploring professional business scenarios, the smallest business unit with universal adaptability is extracted, and data parameters are configured for it to form a business master system. Based on different classification methods such as business scenarios and master content sources, a business master directory is formed to facilitate quick search; at the same time, the association between outline nodes and report content blocks is used to automatically extract the directory hierarchy, providing a content basis for quickly building report templates.

[0092] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for automatically generating data analysis reports for hydropower main equipment, Its characteristics are: The generated methods include: Master template production: breaking down the data analysis report content into the smallest business units and configuring the content based on the breaking down definition; Template construction, listing the outline of the data analysis report, filling in the content according to each outline node definition, and realizing the goal of automatically loading the corresponding time range data according to different time ranges through global variables; Generate a report. Based on business needs, select a configured report template to generate a static report. Online preview and / or download are supported.

2. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 1, Its characteristics are: The content of the master page is customized according to business needs, including: Text description: including summary description, which is universal and does not change with time or data; Statistical text: content that has a certain time period and changes with time; suitable for periodic statistics, used when presented in text form, and used in combination with the above text description; Visual display: including business graphics, tables and other visual charts; suitable for periodic monitoring; Model result diagram: graphical output of model calculation results under specific scenarios; Picture: Supports inserting picture function.

3. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 2, Its characteristics are: To ensure the accuracy of text description, TF-IDF in natural language processing is used to measure the importance of text. The formula is: （1）； （2）； （3）； Among them, TF(t, d) is the word frequency, which means the word t In the documentation d The frequency of occurrence in , IDF(t) is the inverse document frequency, indicating the word t The rarity of a word, TF-IDF(t, d) is the weight that comprehensively considers the word frequency and inverse document frequency; In the process of mastering, the TF-IDF formula is applied to determine which words in the text description are the most important, and then decide which words should be emphasized and which can be simplified or omitted.

4. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 1, Its characteristics are: After the master content is configured and published, all accounts on the platform can use the master content; if it is not published, only this account can use it.

5. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 1, Its characteristics are: The template construction includes: Content configuration: fill in template content according to outline nodes; customize the target directory outline according to the data analysis report theme; and / or directly fill in node content: Configure variables: After configuring local variables or global variables according to the report template content, you can preview the effect of the report template after loading data; Save or publish: save or publish after the template content, data parameters and global variables are configured; Template version management: supports version rollback, you can view the contents of different versions, and choose to open the version according to actual needs.

6. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 5, Its characteristics are: During the template building process, compression encoding technology is used to optimize loading speed and reduce memory consumption; Among them, the compression coding technology adopts Huffman coding, which is a lossless data compression method. The construction process of Huffman coding can be expressed as: （4）； in, C is a set of characters, C left and C right They are the character sets corresponding to the left subtree and right subtree of the Huffman tree respectively.

7. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 1, Its characteristics are: The report generation includes manual triggering and automatic generation; The manual triggering method selects the report type and time range according to the business needs of the data analysis report; After viewing the preview effect, save it as a static report; The automatic generation method is as follows: select a corresponding report template, select a report type and a time range, and then start a scheduled report generation task, and generate and save the report as a static report according to the time schedule.

8. According to the method for automatically generating a data analysis report of a hydropower main equipment as described in claim 7, Its characteristics are: When generating reports, MAPE is used to evaluate the accuracy of model predictions. MAPE is expressed as: （5）； in, A t Indicates the actual value, F t represents the predicted value, N is the sample size. MAPE is used to measure the percentage difference between the predicted value and the actual value. The closer it is to 0, the more accurate the prediction.

Images