Intelligent energy data management platform front-end interface generation method and system

Abstract

The invention discloses an intelligent energy data management platform front-end interface generation method and system. The energy data management platform front-end interface generation method comprises the following steps: inputting an interface customization requirement into an intelligent model; the intelligent model outputs a preliminary design suggestion based on an interface customization requirement and a knowledge base; performing demand adjustment on the preliminary design suggestion to obtain an actual design demand; the intelligent model generates a corresponding code text; the intelligent model converts the corresponding code text into a visual interface through an intelligent interface display module; and adding a preset interaction effect to the visual interface, and obtaining a final energy data management platform front-end interface. The energy data management platform front-end interface generation system comprises a user behavior module, an intelligent model, an intelligent model auxiliary module, an SQL generator module, a knowledge graph, a vector database and an intelligent interface display module. According to the method, the energy data management platform front-end interface meeting the user requirements can be more efficiently researched and developed.

Description

Technical Field

[0001] The present invention relates to the technical field of platform front-end interface generation, and in particular to a method and system for generating a front-end interface of an intelligent energy data management platform. Background Art

[0002] With the development of Internet technology, more and more management platforms have emerged. Correspondingly, the technology of front-end interface design of various management platforms has become more and more mature. In the process of research and development, in order to save research and development costs to the maximum extent, a "one-size-fits-all" approach is usually adopted when designing the front-end interface. All users of the front-end interface share the same interface, that is, only one interface needs to be generated during the research and development process. Although this design simplifies development and management, it also leads to poor user adaptability. Different users have different operating habits and needs. The unified interface is difficult to meet the personalized needs of all users, resulting in uneven user experience. In addition, the problem of unified interface is also reflected in functional redundancy. The unified interface usually contains all possible functions, but different users may only need some of them. This functional redundancy not only increases the complexity of the interface, but may also cause users to be confused when looking for the required functions. If different interfaces are developed for different users, under the existing research and development technology, each interface needs to be developed separately, which will greatly increase the research and development costs.

[0003] From the above, we can see that how to quickly generate the front-end interface based on a large amount of background data is an urgent problem that needs to be solved. Summary of the invention In view of the above problems, the present invention proposes an intelligent method and system for generating a front-end interface of a data management platform.

[0004] To achieve the purpose of the present invention, a method for generating an intelligent data management platform front-end interface is provided, comprising the following steps: S1: The user communicates with the intelligent model and inputs the interface customization requirements into the intelligent model; the intelligent model calls the knowledge base through the intelligent model auxiliary module; the intelligent model outputs preliminary design suggestions based on the interface customization requirements and the knowledge base; S2: The user makes corresponding demand adjustments to the preliminary design suggestions output by the intelligent model and obtains actual design requirements; the intelligent model generates corresponding code text based on the adjusted actual design requirements; S3: The user sends a code integration request to the intelligent model; the intelligent model transforms the corresponding code text into a visual interface through an intelligent interface display module; S4: Based on the web design and interaction design specifications, a preset interaction effect is added to the visualization interface to obtain a final energy data management platform front-end interface.

[0005] Furthermore, in step S1, the knowledge base includes: a vector database and a knowledge graph; The specific process of communicating with the intelligent model and inputting interface customization requirements into the intelligent model includes: Sending the interface customization requirement to the intelligent model; The intelligent model analyzes the interface customization requirement and returns confirmation information; The confirmation information is evaluated: if the confirmation information meets the user's expectations, a confirmation operation is performed; if the confirmation information does not meet the user's expectations, the interface customization requirements are modified, and then the modified interface customization requirements are sent to the intelligent model again; the intelligent model parses the modified interface customization requirements again and returns the confirmation information again; the confirmation and modification are repeated until the confirmation information meets the user's expectations; The confirmation information is confirmed, the evaluation is completed, and the intelligent model accepts the confirmed interface customization requirements.

[0006] Furthermore, in step S1, the process of acquiring the vector database includes the following steps: S101: Collect knowledge information and form a preliminary knowledge base; S102: performing a parsing operation on the preliminary knowledge base to obtain a parsed knowledge base; S103: performing vectorization processing on the parsed knowledge base to obtain a vector database.

[0007] Furthermore, in step S103, the specific process of vectorizing the parsed knowledge base and obtaining the vector database includes the following: S1031: constructing a bag-of-words model based on the parsed knowledge base, wherein the bag-of-words model includes a vocabulary of all words; S1032: Count the frequency of occurrence of each word in the bag-of-words model: TF(t,d)=(number of times word t appears in document d) / (total number of words in document d), where TF stands for word frequency; IDF(t,D)=ln(N / df(t)), where N is the total number of documents; df(t) is the number of documents containing word t; IDF stands for inverse document frequency, which is used to measure the importance of a word in the entire document collection; TF-IDF(t,d,D)=TF(t,d)*IDF(t,D), which is used to measure the importance of a word in a document; S1033: Constructing vectors corresponding to each text in the bag-of-words model based on the frequency of occurrence of each word: Assume that the training text has a context window size of 2; set the words in the window as positive samples, and the words not in the window as negative samples; combine the positive samples and negative samples to generate the training samples required for training; Initialize two matrices: an Embedding matrix and a Context matrix; the row data of the Embedding matrix and the Context matrix are vectors corresponding to each word in the vocabulary; Select a training data sample pair, and query the vector corresponding to each word in the Embedding matrix and the Context matrix based on a row of data in the training data sample pair; The dot product of the vector corresponding to each word and the vector corresponding to the word in its context is calculated, and then the sigmoid function is used to convert the dot product into probability to evaluate the accuracy of the model prediction.

[0008] Furthermore, in step S1, the process of acquiring the knowledge graph includes the following steps: S1001: Communicate with the intelligent model and send SQL query instructions to the intelligent model; S1002: the intelligent model calls the vector database through the intelligent model auxiliary module, and performs a query operation in the vector database based on the SQL query instruction, and finally returns the queried data table; S1003: The intelligent model stores the queried data table in the knowledge graph through an intelligent model auxiliary module.

[0009] Furthermore, in step S1, the ways of communicating with the intelligent model include: command input, voice input and mouse operation; the preliminary design suggestions include: page layout, icon type and prompt words.

[0010] The present invention also provides an intelligent energy data management platform front-end interface generation system, including: a user behavior module, an intelligent model, an intelligent model auxiliary module, an SQL generator module, a knowledge graph, a vector database and an intelligent interface display module; The user behavior module is used to send the customization requirements of the visualization interface to the intelligent model; The intelligent model is used to accept the customization requirements of various visualization interfaces and intelligently generate visualization interfaces with the help of a knowledge base; The intelligent model auxiliary module is used to call the API interface and knowledge base of the intelligent model; The SQL generator module is used to generate database function codes for extracting target data; The knowledge graph is used to generate various data tables after storing query operations; The vector database is used to store various energy data; The intelligent interface display module is used to convert complex energy data into an easy-to-understand user interface and set the relationship between charts and data.

[0011] Furthermore, the intelligent energy data management platform front-end interface generation system further includes: a process update module; First, the code text of the intelligent model is converted into a browsable form; the process update module is used to update the code text converted into the browsable form; The process update module includes: N function blocks, where N is greater than or equal to 4; Each function block is used to store: a PDF file, a text file or a text file.

[0012] Furthermore, the intelligent energy data management platform front-end interface generation system further includes: a retrieval module; The implementation process of the retrieval module includes: extracting key parts from the code text based on preset keywords to implement text indexing.

[0013] The present invention also provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute any of the methods for generating a front-end interface of an intelligent energy data management platform.

[0014] Compared with the prior art, the present invention has the following beneficial technical effects: 1. The present invention can realize automatic extraction, data extraction and chart generation: by prompting the intelligent model, generating SQL function code, automatically extracting the required data, and generating the most suitable chart for display based on the knowledge graph, simplifying the user's data analysis process and improving the intuitiveness and ease of use of data display.

[0015] 2. Multiple command input support: Supports multiple command input methods, including command input window, voice input and mouse operation, which can more conveniently input user requirements; at the same time, the front-end visual interface can output information in various forms such as charts and texts.

[0016] 3. Intelligent data extraction and display: Combine data from different sources and types with intelligent information extraction technology, sort and display the data according to its importance, and form a diversified visual effect to help operators quickly identify key information and alleviate the problem of information overload.

[0017] 4. Customized interface generation: Based on the demand information input by the user, various display data are analyzed and processed through intelligent models to generate a professional interface that meets user needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of a method for generating a front-end interface of an intelligent data management platform according to an embodiment; Figure 2 is a module diagram of a front-end interface generation system of an intelligent energy data management platform according to an embodiment; Figure 3 is a user question details logic diagram of an embodiment; Figure 4 The figure is a schematic diagram of the overall flow of user asking questions in one embodiment. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0021] refer to Figure 1 As shown, Figure 1 The figure is a flow chart of a method for generating a front-end interface of an intelligent data management platform according to an embodiment. The method for generating a front-end interface of an intelligent energy data management platform comprises the following steps: S1: Communicate with the intelligent model and input interface customization requirements into the intelligent model; the intelligent model calls the knowledge base through the intelligent model auxiliary module; the intelligent model outputs preliminary design suggestions based on the interface customization requirements and the knowledge base; S2: adjusting the preliminary design suggestions accordingly and obtaining actual design requirements; the intelligent model generates corresponding code text based on the actual design requirements; S3: Sending a code integration request to the intelligent model; the intelligent model converts the corresponding code text into a visual interface through an intelligent interface display module; S4: Based on the web design and interaction design specifications, a preset interaction effect is added to the visualization interface to obtain a final energy data management platform front-end interface.

[0022] In one embodiment, in step S1, the knowledge base includes: a vector database and a knowledge graph; The specific process of communicating with the intelligent model and inputting interface customization requirements into the intelligent model includes: Sending the interface customization requirement to the intelligent model; The intelligent model analyzes the interface customization requirement and returns confirmation information; The confirmation information is evaluated: if the confirmation information meets the user's expectations, a confirmation operation is performed; if the confirmation information does not meet the user's expectations, the interface customization requirements are modified, and then the modified interface customization requirements are sent to the intelligent model again; the intelligent model parses the modified interface customization requirements again and returns the confirmation information again; the confirmation and modification are repeated until the confirmation information meets the user's expectations; The confirmation information is confirmed, the evaluation is completed, and the intelligent model accepts the confirmed interface customization requirements.

[0023] In one embodiment, in step S1, the process of acquiring the vector database includes the following steps: S101: Collect knowledge information and form a knowledge base; S102: performing a parsing operation on the knowledge base and obtaining a parsed knowledge base; S103: performing vectorization processing on the parsed knowledge base to obtain a vector database.

[0024] In one embodiment, in step S103, the specific process of vectorizing the parsed knowledge base and obtaining the vector database includes the following: S1031: constructing a bag-of-words model based on the parsed knowledge base, wherein the bag-of-words model includes a vocabulary of all words; S1032: Count the frequency of occurrence of each word in the bag-of-words model: TF(t,d)=(number of times word t appears in document d) / (total number of words in document d), where TF stands for word frequency; IDF(t,D)=ln(N / df(t)), where N is the total number of documents; df(t) is the number of documents containing word t; IDF stands for inverse document frequency, which is used to measure the importance of a word in the entire document collection; If a word appears in many documents, it means that the word has a weak ability to distinguish and a low IDF value. TF-IDF(t,d,D)=TF(t,d)*IDF(t,D), which is used to measure the importance of a word in a document; S1033: Constructing vectors corresponding to each text in the bag-of-words model based on the frequency of occurrence of each word: Assume that the training text has a context window size of 2; set the words in the window as positive samples, and the words not in the window as negative samples; combine the positive samples and negative samples to generate the training samples required for training; Initialize two matrices: an Embedding matrix and a Context matrix; the row data of the Embedding matrix and the Context matrix are vectors corresponding to each word in the vocabulary; Select a training data sample pair, and query the vector corresponding to each word in the Embedding matrix and the Context matrix based on a row of data in the training data sample pair; The dot product of the vector corresponding to each word and the vector corresponding to the word in its context is calculated, and then the sigmoid function is used to convert the dot product into probability to evaluate the accuracy of the model prediction.

[0025] In one embodiment, in step S1, the process of acquiring the knowledge graph includes the following steps: S1001: Communicate with the intelligent model and send SQL query instructions to the intelligent model; S1002: the intelligent model calls the vector database through the intelligent model auxiliary module, and performs a query operation in the vector database based on the SQL query instruction, and finally returns the queried data table; S1003: The intelligent model stores the queried data table in the knowledge graph through an intelligent model auxiliary module.

[0026] In one embodiment, in step S1, the ways of communicating with the intelligent model include: command input, voice input and mouse operation; the preliminary design suggestions include: page layout, icon type and prompt words.

[0027] like Figure 2 The figure shows a module diagram of an intelligent energy data management platform front-end interface generation system according to an embodiment. An intelligent energy data management platform front-end interface generation system includes: a user behavior module, an intelligent model, an intelligent model auxiliary module, an SQL generator module, a knowledge graph, a vector database and an intelligent interface display module; The user behavior module is used to send the customization requirements of the visualization interface to the intelligent model; The intelligent model is used to accept the customization requirements of various visualization interfaces and intelligently generate visualization interfaces with the help of a knowledge base; The intelligent model auxiliary module is used to call the API interface and knowledge base of the intelligent model; The SQL generator module is used to generate database function codes for extracting target data; The knowledge graph is used to generate various data tables after storing query operations; The vector database is used to store various energy data; The intelligent interface display module is used to convert complex energy data into an easy-to-understand user interface and set the relationship between charts and data.

[0028] In one embodiment, the intelligent energy data management platform front-end interface generation system further includes: a process update module; First, the code text of the intelligent model is converted into a browsable form; the process update module is used to update the code text converted into the browsable form; The process update module includes: N function blocks, where N is greater than or equal to 4; Each function block is used to store: a PDF file, a text file or a text file.

[0029] In one embodiment, the intelligent energy data management platform front-end interface generation system further includes: a retrieval module; The implementation process of the retrieval module includes: extracting key parts from the code text based on preset keywords to implement text indexing.

[0030] In one embodiment, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute any of the methods for generating a front-end interface of an intelligent energy data management platform.

[0031] like Figure 3 FIG. 1 is a schematic diagram of a user question detail logic according to an embodiment of the present invention. Figure 3 The following workflows are mainly described.

[0032] Complex problem handling process Enter a complex question: The user asks a question that is complex or requires in-depth analysis.

[0033] Vector database to obtain knowledge embedding: The system first searches the vector database for knowledge embedding related to the problem (i.e., the feature vector representation of the problem).

[0034] Create prompt: Leverage the retrieved knowledge embeddings to generate a prompt suitable for processing by a large language model (LLM).

[0035] Query LLM: Send the generated prompt to the Large Language Model (LLM) and get the model's answer.

[0036] Feedback and storage: Feedback the user's questions and the model's answers to the vector database so that similar questions in the future can be answered directly from the database.

[0037] Processing procedures for similar historical issues: Input historical similar questions: The user asks a question and the system checks whether it is similar to previous questions.

[0038] Cache check: First check if there is a similar query or answer in the cache. If so, return the answer in the cache directly.

[0039] Create prompt and query LLM: If the answer is not in the cache, create a prompt and query the large language model (LLM) to get the model's answer.

[0040] Feedback and storage: Feedback the user's questions and the model's answers into the vector database and store them in the cache so that similar questions in the future can be answered directly from the cache.

[0041] Simple problem handling process: Enter a simple question: The user asks a relatively simple or common question.

[0042] Vector database to obtain relevant files: The system searches the vector database for files or information related to the problem.

[0043] Create prompt: Using the retrieved documents or information, generate a prompt suitable for Large Language Model (LLM) processing.

[0044] Query LLM: Send the generated prompt to the Large Language Model (LLM) and get the model's answer.

[0045] Direct feedback: The model’s answers are fed back directly to the user.

[0046] Summary: Through the above process, the system can adopt different processing strategies for different types of questions to improve the accuracy and efficiency of answers. Complex questions are processed through knowledge embedding and LLM, historical similar questions are processed through caching and LLM, and simple questions are directly processed through vector database and LLM. Each strategy is designed to optimize the user experience and ensure that users can quickly obtain accurate information or answers.

[0047] Figure 4 This is a schematic diagram of the overall process of asking questions by a user in one embodiment. Figure 4 middle, The flow chart of the interaction between the user and the intelligent model is as follows: User sent query content: The user sends a query request to the intelligent model, including a text description of a question, requirement or instruction.

[0048] For example: “I need to analyze power data from an energy storage system.” The intelligent model confirms the answer content: The intelligent model first parses the user's query and tries to understand the user's intent.

[0049] The model generates preliminary answers or suggestions and asks the user to confirm the accuracy of the answers. For example: "Your question is about generating power data charts for energy storage systems. Are you sure?" User confirmed answer content: Users check the answers or suggestions provided by the model to confirm whether they meet their needs.

[0050] Users can confirm or revise the model's understanding by answering "yes" or "no".

[0051] Ask questions about the code and charts based on the determined content: Based on the confirmed content, the user further proposes specific code or diagram requirements.

[0052] For example: "I need to generate a line chart with the title 'Energy storage system power change', the X-axis represents time, and the Y-axis represents power." Intelligent model output chart type and code or requirement code: The intelligent model generates corresponding chart types and codes according to user needs.

[0053] The model output chart type (such as line chart, bar chart, etc.) and the code required to generate the chart.

[0054] The user confirms the type and code: Users check the chart types and codes generated by the model to confirm whether they meet their needs.

[0055] The user can confirm or modify the model's output by answering "yes" or "no".

[0056] After the user confirms the generated graph, the query command is sent to the knowledge graph storage: Once the user confirms that the chart type and code are correct, the model will store the generated chart data and query instructions in the knowledge graph.

[0057] The stored content may include chart types, data sources, code snippets, etc.

[0058] Vector database supports: The vector database provides data support, including database, data table and other information.

[0059] The intelligent model can obtain necessary data from the vector database to generate diagrams or codes.

[0060] For example: "Get the power data table from the energy storage system database." The specific contents of query instruction confirmation (such as SQL type) are as follows: The user sends an SQL query command: The user sends an SQL query command to the smart model, for example: "SELECT * FROM energy_storage_dataWHERE year = 2023;" The intelligent model confirms SQL instructions: The intelligent model parses the user's SQL instructions and confirms the syntax and intent of the instructions.

[0061] The model generates preliminary confirmation information, such as: "You want to query the energy storage data for 2023, are you sure?" User confirms SQL command: The user checks the confirmation information provided by the model to confirm the accuracy of the SQL instructions.

[0062] The user can confirm or modify the instruction by answering "yes" or "no".

[0063] The smart model executes the SQL query and returns the result: The model executes the SQL query instructions confirmed by the user and returns the query results.

[0064] For example: "The query results are as follows: [data list]" The query results are stored in the knowledge graph: After the user confirms that the query results are correct, the model stores the query results and instructions in the knowledge graph.

[0065] During the multiple rounds of question-and-answer sessions between the user and the intelligent model, the user's feedback is not limited to a simple "yes" or "no", but also requires an evaluation of whether the answer meets their needs. If the answer does not meet the user's expectations, the user will correct the text content and continue to ask questions; if it is roughly in line with the user's expectations, the user will slightly adjust the content of the next question; only when it is completely in line with the user will the user give positive feedback.

[0066] In one embodiment, the user dialogues with the intelligent model and inputs prompts. The intelligent model includes ChatGPT, and the knowledge database can be selected from: openai flask pandas numpy SQLAlchemy py2neo graphviz, etc. The knowledge database workflow is as follows: Collect knowledge, such as articles, lectures, conferences, etc. Create a knowledge base, parse the files in the knowledge base, and extract text, tables and other information; Parse text to get smaller semantic units, for example, PDF text contains text PDF and scanned PDF; The parsing methods include segmentation by character, segmentation by code, and segmentation by character size. Among them, the knowledge database first matches the intelligent model and then extracts information from the content parser. In one embodiment, the intelligent model can also be used to implement prompt word clipping, prompt word summarization, and prompt word-based soft compression and retrieval enhancement generation.

[0067] Construct a logical chain of prompt words, including context: role → task → knowledge; input data: title → a paragraph of text; output content: sentence → article → answer the question; In one embodiment, an external knowledge base is introduced into the intelligent model by means of vector retrieval; Data extraction → block division → vectorization → index creation → retrieval → rearrangement → LLM induction; Generate energy storage prompt words, such as energy storage professional vocabulary and vocabulary strings; Write prompt words to capture the semantics and contextual relationships of words, and use intelligent models to map words into low-dimensional vector space to capture semantic relationships; Send the prompt words generated by the vector database to the energy storage intelligent model to generate a response, output the answer and reference; The model calculates the key-value pairs of each token based on the original input token sequence to generate a coherent and semantically relevant output sequence. The details are as follows: (1) The output sentence is generated using an autoregressive method, which outputs token by token.

[0068] (2) Each time a new token is generated, the model predicts and generates the next token based on the currently generated token sequence and the previously calculated key-value pair cache.

[0069] (3) The key and value of the newly generated token will be updated in the cache so that it can be used in the next generation.

[0070] In the autoregressive approach, the model generates output sentences token by token.

[0071] First token: The first token predicted and generated by the model based on the initial token sequence input is usually "battery".

[0072] Generate results: Battery Second token: The model predicts and generates the next token based on the currently generated token sequence (i.e., “battery”) and the previously calculated key-value pair cache.

[0073] Generates result: Voltage Update Cache: Update the key and value of "voltage" into the cache.

[0074] Third token: The model continues to predict and generate the next token based on the currently generated token sequence (i.e., “battery voltage”) and the updated cache.

[0075] Generate results: Update Cache: Update the key and value of "to" into the cache.

[0076] Fourth token: The model continues to generate the next token.

[0077] Generated result: 3.6V Update cache: Update the key and value of "3.6V" into the cache.

[0078] Fifth token: The model continues to generate the next token.

[0079] Generates result: "," Update cache: Update the key and value of "," into the cache.

[0080] Sixth token: The model continues to generate the next token.

[0081] Generate Result: Charging Status Update cache: Update the key and value of "charging status" into the cache.

[0082] Seventh token: The model continues to generate the next token.

[0083] Generate results: Update Cache: Update the key and value of "to" into the cache.

[0084] Eighth token: The model continues to generate the next token.

[0085] Generate results: 50% Update Cache: Updates "50%" of the keys and values ​​into the cache.

[0086] Ninth token: The model continues to generate the next token.

[0087] Generate results: "," Update cache: Update the key and value of "," into the cache.

[0088] Tenth token: The model continues to generate the next token.

[0089] Generate results: Temperature Update Cache: Update the key and value of "temperature" into the cache.

[0090] Eleventh token: The model continues to generate the next token.

[0091] Generate results: Update Cache: Update the key and value of "to" into the cache.

[0092] Twelfth token: The model continues to generate the next token.

[0093] Generated results: 25°C Update cache: Update the key and value of "25°C" into the cache.

[0094] Thirteenth token: The model continues to generate the next token.

[0095] Generates result: "," Update cache: Update the key and value of "," into the cache.

[0096] Fourteenth token: The model continues to generate the next token.

[0097] Generate results: Load demand Update Cache: Update the key and value of "Load Requirement" into the cache.

[0098] Fifteenth token: The model continues to generate the next token.

[0099] Generate results: Update Cache: Update the key and value of "to" into the cache.

[0100] Sixteenth token: The model continues to generate the next token.

[0101] Generation result: 100kW Update cache: Update the key and value of "100kW" into the cache.

[0102] Seventeenth token: The model continues to generate the next token.

[0103] Generates result: "." Update Cache: Update the key and value of "." into the cache.

[0104] The resulting output sequence is: battery voltage is 3.6V, state of charge is 50%, temperature is 25°C, and load demand is 100kW.

[0105] Set different chart functions such as radar chart, heat map, distribution map, stacked bar chart, line chart and waterfall chart; The radar chart generates the return rate, charge and discharge efficiency, charge and discharge depth, number of cycles and attenuation rate for the application of modular products, and expresses them in the form of integral marks. It can be compared with different projects in the same region or the same model products in different regions, and the multi-dimensional performance comparison of products can be made.

[0106] Taking standardized energy storage products as an example, the comparative data that can be used on the radar chart include charging and discharging efficiency, charging and discharging depth, number of cycles, attenuation rate, yield rate under current performance, predicted payback period and other performance.

[0107] Heat maps are used to analyze power fluctuations on different dates within the same period, and power changes at different times on the same day. They show the temporal and spatial distribution of real-time data fluctuations such as power and temperature, and analyze the fluctuations of power and temperature of equipment at the same time on different dates, as well as the changes at different times on the same day.

[0108] Taking standardized energy storage products as an example, the data that can be used in the heat map include real-time parameters such as charging power, discharging power, temperature, battery voltage, and state of charge.

[0109] The waterfall chart is used to simulate the relationship between product value and time, such as profitability, revenue increase or decrease, etc.

[0110] Line charts and stacked line charts provide real-time data visualization and can monitor data changes of different products at different dates and time periods.

[0111] Taking standardized energy storage products as an example, the X-axis is set as the time value and the Y-axis is set as the power value. Each broken line represents a parameter, including real-time parameters such as charging and discharging power, load power, grid power, battery SOC, battery voltage, and state of charge. Each broken line contains different colors, which can show the coupling relationship between the energy storage device and the load and grid power.

[0112] The bar chart and stacked bar chart are used to count the quantity and cumulative amount. The X-axis is set as the time value and the Y-axis is set as the data type. Each column represents a parameter, such as revenue, cost, electricity consumption, etc. They can be displayed individually or in a stacked form. If stacking is used, different colors can be used to represent different parameter contents. The revenue and cost of different projects can be displayed in one column to intuitively understand the proportion of each project.

[0113] Pie charts and waterfall charts are specifically used to analyze the proportion of each part in a data combination.

[0114] Extract target indicators, constraints and progress ratios; The target indicator is extracted by displaying the gap between the actual value and the target value through a bar chart or a stacked bar chart; Constraints are extracted by displaying them in radar charts or line charts with threshold markers; Progress ratio is displayed through stacked bar charts or pie charts to show extraction; Calculate specific values ​​to be used as reference standards, decision-making basis, and forecast targets.

[0115] Calculate the value corresponding to each progress ratio; Find the corresponding value according to the provided progress ratio; Calculate the value corresponding to 0% ratio; Set sorting constraints; Linear interpolation calculates the value corresponding to 0% scale; Calculate a specific value; Here is an example: # Example data absolute_target = 312 boundaries = [20,40,80] achievements = [320,318,314] # Calculate a specific value result = calculate_value(absolute_target, boundaries, achievements) print("Specific value:", result) In one embodiment, the specific operation process is as follows: The user provides query content prompt information to the intelligent model; The prompt information includes project background, target user group, expected visual style and interaction requirements, etc.

[0116] The intelligent model generates corresponding report code text; The intelligent model module generates an outline of a customized report and design suggestions for each sub-module before asking questions.

[0117] The outline includes a sorting index and visual design for knowledge identification; the submodules have standardized code snippets (such as CSS styles, specific component codes) for compilers or interpreters such as Vim, Jupyter, Langchain, Python, and responses to user interaction content.

[0118] The sorting index technology automatically identifies and extracts key functions and variable names from the code text to obtain an efficient text index.

[0119] The user modifies the generated embedded text through text indexing; Users use the generated text index to quickly locate and adjust the text content generated by the intelligent model.

[0120] Users can modify and optimize the report content to meet specific needs and desired results.

[0121] Update the knowledge graph with new text after user modification; After the user completes the modification of the generated text content, the optimized text is applied to the corresponding interface design or functional module to complete the iterative update.

[0122] Users can customize the report content according to actual needs to achieve conversion and integration of code text.

[0123] Using the code generated by the intelligent model, the document-related code text is generated in the compiler or interpreter and converted into a report.

[0124] Finally, the digital interface is output.

[0125] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] It should be noted that the terms "first\second\third" involved in the embodiments of the present application are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the specific order or sequence of "first\second\third" can be interchanged where permitted. It should be understood that the objects distinguished by "first\second\third" can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0127] The terms "including" and "having" and any variations thereof in the embodiments of the present application are intended to cover non-exclusive inclusions. For example, a process, method, device, product, or equipment that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules that are not listed, or may optionally include other steps or modules that are inherent to these processes, methods, products, or equipment.

[0128] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for generating a front-end interface of an intelligent energy data management platform, characterized in that: The steps include: S1: Communicate with the intelligent model and input interface customization requirements into the intelligent model; the intelligent model calls the knowledge base through the intelligent model auxiliary module; The intelligent model outputs preliminary design suggestions based on the interface customization requirements and the knowledge base; S2: adjusting the preliminary design suggestions accordingly and obtaining actual design requirements; the intelligent model generates corresponding code text based on the actual design requirements; S3: Sending a code integration request to the intelligent model; The intelligent model transforms the corresponding code text into a visual interface through an intelligent interface display module; S4: Based on the web design and interaction design specifications, add preset interaction effects to the visualization interface, and obtain the final energy data management platform front-end interface; In step S1, the knowledge base includes: a vector database and a knowledge graph; The specific process of communicating with the intelligent model and inputting interface customization requirements into the intelligent model includes: Sending the interface customization requirement to the intelligent model; The intelligent model analyzes the interface customization requirement and returns confirmation information; The confirmation information is evaluated: if the confirmation information meets the user's expectations, a confirmation operation is performed; if the confirmation information does not meet the user's expectations, the interface customization requirements are modified, and then the modified interface customization requirements are sent to the intelligent model again; the intelligent model parses the modified interface customization requirements again and returns the confirmation information again; the confirmation and modification are repeated until the confirmation information meets the user's expectations; The confirmation information is confirmed, the evaluation is completed, and the intelligent model accepts the confirmed interface customization requirements.

2. According to claim 1, a method for generating a front-end interface of an intelligent energy data management platform is characterized in that: In step S1, the process of acquiring the vector database includes the following steps: S101: Collect knowledge information and form a knowledge base; S102: performing a parsing operation on the knowledge base and obtaining a parsed knowledge base; S103: performing vectorization processing on the parsed knowledge base to obtain a vector database.

3. The method for generating a front-end interface of an intelligent energy data management platform according to claim 2, characterized in that: In step S103, the specific process of vectorizing the parsed knowledge base and obtaining a vector database includes the following: S1031: constructing a bag-of-words model based on the parsed knowledge base, wherein the bag-of-words model includes a vocabulary of all words; S1032: Counting the frequency of occurrence of each word in the bag-of-words model; S1033: Constructing a vector corresponding to each text in the bag-of-words model based on the frequency of occurrence of each word.

4. The method for generating a front-end interface of an intelligent energy data management platform according to claim 1, characterized in that: In step S1, the process of acquiring the knowledge graph includes the following steps: S1001: Communicate with the intelligent model and send SQL query instructions to the intelligent model; S1002: the intelligent model calls the vector database through the intelligent model auxiliary module, and performs a query operation in the vector database based on the SQL query instruction, and finally returns the queried data table; S1003: The intelligent model stores the queried data table in the knowledge graph through an intelligent model auxiliary module.

5. The method for generating a front-end interface of an intelligent energy data management platform according to claim 1, characterized in that: In step S1, the ways of communicating with the intelligent model include: command input, voice input and mouse operation; the preliminary design suggestions include: page layout, icon type and prompt words.

6. An intelligent energy data management platform front-end interface generation system, characterized in that: include: User behavior module, intelligent model, intelligent model auxiliary module, SQL generator module, knowledge graph, vector database and intelligent interface display module; The user behavior module is used to send the customization requirements of the visualization interface to the intelligent model; The intelligent model is used to accept the customization requirements of various visualization interfaces and intelligently generate visualization interfaces with the help of a knowledge base; The intelligent model auxiliary module is used to call the API interface and knowledge base of the intelligent model; The SQL generator module is used to generate database function codes for extracting target data; The knowledge graph is used to generate various data tables after storing query operations; The vector database is used to store various energy data; The intelligent interface display module is used to convert complex energy data into an easy-to-understand user interface and set the relationship between charts and data.

7. The intelligent energy data management platform front-end interface generation system according to claim 6 is characterized in that: Also includes: Process update module; Firstly, the code text of the intelligent model is converted into a browsable form; The process update module is used to update the code text converted into a browsable form; The process update module includes: N function blocks, where N is greater than or equal to 4; Each function block is used to store: a PDF file, a text file or a text file.

8. The intelligent energy data management platform front-end interface generation system according to claim 7 is characterized in that: Also includes: Retrieval module; The implementation process of the retrieval module includes: extracting key parts from the code text based on preset keywords to implement text indexing.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for generating a front-end interface of an intelligent energy data management platform as described in any one of claims 1 to 5.

Images