A wind power intelligent interaction integrated system based on a large language model

Abstract

The present disclosure provides a wind power intelligent interaction integrated system based on a large language model, belonging to the technical field of intelligent interaction. The wind power intelligent interaction integrated system based on the large language model comprises a view component, a scene construction component, a rendering component, a large language model component and a knowledge graph component; the rendering component is connected with the scene construction component and is used for rendering the input view of a target user to obtain a new rendering view; the scene construction component is connected with the view component and is used for stylizing the input view to obtain a stylized view, and fusing the stylized view and the new rendering view to obtain a scene view; the view component is connected with the large language model component and is used for receiving an interaction request of the target user; the large language model component is connected with the knowledge graph component and is used for determining response data corresponding to the interaction request based on data in the knowledge graph component in response to the interaction request. The present disclosure can realize efficient interaction operation.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent interaction technology, and in particular to an intelligent interactive integrated system for wind power generation based on a large language model. Background Technology

[0002] With the rapid development of intelligent technologies, interactive environments and intelligent systems are becoming increasingly important in fields such as virtual reality, robotics, and automation systems. However, existing intelligent interaction technologies often lack integration and have limited functionality, resulting in inefficient interactive operations and difficulty in meeting diverse user needs. Summary of the Invention

[0003] This disclosure provides a wind power generation intelligent interactive integration system based on a large language model to solve the problem of inefficient interactive operation.

[0004] This disclosure provides a wind power generation intelligent interactive integration system based on a large language model, including: a view component, a scene construction component, a rendering component, a large language model component, and a knowledge graph component;

[0005] The rendering component, connected to the scene building component, is used to render the target user's input view to obtain a new rendered view;

[0006] The scene building component connects with the view component to stylize the input view, obtain a stylized view, and then merges the stylized view with the rendered new view to obtain the scene view.

[0007] The view component is used to display the scene view;

[0008] The view component, connected to the large language model component, is used to receive interaction requests from the target user;

[0009] A large language model component, connected to a knowledge graph component, is used to respond to interaction requests and determine the response data corresponding to the interaction request based on the data in the knowledge graph component.

[0010] The view component is also used to display response data.

[0011] In one exemplary embodiment of this disclosure, a wind power generation intelligent interactive integration system based on a large language model further includes a robot assistant component;

[0012] The robot assistant component connects to the large language model component to respond to commands from the target user. It performs retrieval based on the large language model component, determines the corresponding control instructions based on the retrieval results, and controls the external robot to execute the corresponding actions according to the control instructions.

[0013] In one exemplary embodiment of this disclosure, the robot assistant component is connected to the knowledge graph component;

[0014] The robot assistant component is also used to collect environmental data and send the environmental data to the knowledge graph component;

[0015] The knowledge graph component is used to update the knowledge base based on environmental data.

[0016] In one exemplary embodiment of this disclosure, a wind power generation intelligent interactive integration system based on a large language model further includes a wind power generation scenario component;

[0017] The wind power generation scenario component is connected to the large-scale language model component;

[0018] The view component is also used to receive the target user's first request;

[0019] The large language model component is also used to determine wind power generation scenario data from the wind power generation scenario component in response to the first request;

[0020] The view component is also used to display wind power generation scenario data.

[0021] In one exemplary embodiment of this disclosure, the wind power generation scenario components include a wind turbine, a wind sensor, a control module, a data monitoring module, and a maintenance module;

[0022] The wind turbine is connected to the wind sensor, the wind sensor is connected to the control module, and the control module is connected to the data monitoring module and the maintenance module respectively; the control module, the data monitoring module, and the maintenance module are all connected to the large language model component.

[0023] The wind sensor is used to collect wind condition data and send the wind condition data to the control module;

[0024] The data monitoring module is used to monitor operational data, while the maintenance module is used to manage and maintain data.

[0025] The data for wind power generation scenarios includes operational data or maintenance data.

[0026] In one exemplary embodiment of this disclosure, a wind power generation intelligent interactive integration system based on a large language model further includes:

[0027] The view component is also used to receive a second request from the target user;

[0028] The large language model component is also used in response to a second request to determine the target scenario based on the scenario-based component construction.

[0029] The view component is also used to display the target scene.

[0030] In one exemplary embodiment of this disclosure, a wind power generation intelligent interactive integration system based on a large language model further includes:

[0031] The view component is also used to receive third-party requests from the target user;

[0032] The large language model component is also used to update scene data based on scene components in response to third-party requests;

[0033] The view component is also used to display updated scene data.

[0034] In one exemplary embodiment of this disclosure, the rendering component is specifically used to render the input view based on a 3D Gaussian sputtering algorithm to obtain a new rendered view.

[0035] In one exemplary embodiment of this disclosure, the knowledge graph component includes a knowledge base, which includes external data, historical data, and industry data;

[0036] Large language model components connect to knowledge graph components via a knowledge graph interface.

[0037] In one exemplary embodiment of this disclosure, the view component is also used to display a newly rendered view.

[0038] The beneficial effects of the wind power generation intelligent interactive integration system based on a large language model provided in this disclosure are as follows:

[0039] This embodiment of the disclosure provides a visual scene view for the target user by rendering the input view through a rendering component and merging the stylized view and the rendered new view through a scene building component, greatly enhancing the visual experience. Secondly, the connection between the view component and the large language model component enables timely responses to the target user's interaction requests. Leveraging the rich data in the knowledge graph component, the large language model component can accurately determine the response data corresponding to the interaction request, thereby achieving efficient interactive operations and meeting diverse user needs. Therefore, the close collaboration and efficient integration between the components optimizes the overall performance of the system and greatly improves the efficiency of interactive operations. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the structure of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure;

[0042] Figure 2 This is a schematic diagram of another intelligent interactive integrated system for wind power generation based on a large language model provided in this embodiment of the present disclosure;

[0043] Figure 3 This is a use case diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure;

[0044] Figure 4 This is a component diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure;

[0045] Figure 5 This is a sequence diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure;

[0046] Figure 6 This is an activity diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure;

[0047] Figure 7 This is a state diagram of a wind power generation intelligent interactive integrated system based on a large language model, provided in an embodiment of this disclosure. Detailed Implementation

[0048] To enable those skilled in the art to better understand this solution, the technical solutions in the embodiments of this solution will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this solution, not all of them. Based on the embodiments of this solution, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this solution.

[0049] The term "comprising" and any other variations thereof in the specification, claims, and accompanying drawings of this invention mean "including but not limited to," and are intended to cover a non-exclusive inclusion, not limited to the examples listed herein. Furthermore, the terms "first" and "second," etc., are used to distinguish different objects, not to describe a specific order.

[0050] The implementation of this disclosure will be described in detail below with reference to the specific accompanying drawings:

[0051] Figure 1 This is a schematic diagram of the structure of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure. Figure 3 This is a use case diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure. Figure 4This is a component diagram of a wind power generation intelligent interactive integration system based on a large language model, provided in an embodiment of this disclosure. (Refer to...) Figure 1 , Figure 3 and Figure 4 The wind power generation intelligent interactive integration system based on a large language model includes:

[0052] View components, scene building components, rendering components, large language model components, and knowledge graph components;

[0053] The rendering component, connected to the scene building component, is used to render the target user's input view to obtain a new rendered view;

[0054] The scene building component connects with the view component to stylize the input view, obtain a stylized view, and then merges the stylized view with the rendered new view to obtain the scene view.

[0055] The view component is used to display the scene view;

[0056] The view component, connected to the large language model component, is used to receive interaction requests from the target user;

[0057] A large language model component, connected to a knowledge graph component, is used to respond to interaction requests and determine the response data corresponding to the interaction request based on the data in the knowledge graph component.

[0058] The view component is also used to display response data.

[0059] In this embodiment, the target users include, but are not limited to, users and administrators.

[0060] The rendering component is responsible for rendering the input view of the target user to obtain a new rendered view. For example, by utilizing various graphics technologies and algorithms, it adds realistic lighting, materials, textures, and other effects to the input view, generating a new rendered view that is more vivid and realistic, thus enhancing the overall visual experience. After generating the new rendered view, it can be sent to the scene building component.

[0061] The scene building component is responsible for receiving the input view, selecting a style (i.e., allowing the target user to choose different styles), and obtaining a stylized view, giving the input view a specific visual effect and presentation. Then, it also receives a new rendered view generated by the rendering component, and merges the stylized view with the new rendered view to obtain the scene view.

[0062] In the scene building component, style selection allows the target user to choose different styles for the input view, giving the final generated view a specific visual effect and presentation. Styles can be used to stylize the input view, and style types can include:

[0063] The realist style pursues visual effects that highly reproduce the real world, emphasizing realistic lighting and shadow effects and fine details, and is suitable for scenes that require high realism.

[0064] Cartoon style emphasizes simplification and exaggeration, often using bright colors and smooth lines.

[0065] Sketch style, using monochrome or minimal tones, mimicking the lines and shading effects of pencil or charcoal.

[0066] Watercolor style, simulating the effect of watercolor painting, with elegant colors and soft edges.

[0067] Cyberpunk style typically features a futuristic, technological, and dark color palette.

[0068] Retro style, through the use of faded colors and old textures, creates a nostalgic atmosphere, suitable for historical reproduction, artifact display and nostalgic product design.

[0069] Impressionism is characterized by its emphasis on changes in light and color; images often appear slightly blurred but are rich in color.

[0070] Low-poly style simplifies the geometry of objects by using fewer polygons, creating a unique visual effect.

[0071] Abstract style emphasizes the free expression of form and color rather than the representation of reality, and is suitable for artistic creation, modern art exhibitions, and design thinking inspiration.

[0072] The realistic style, which pursues extremely high image detail and lighting processing, makes the generated images close to the effect of photographs, making it suitable for product display, virtual reality scene reproduction, and high-end advertising production.

[0073] Through these style selections, scene building components can generate visually stylized effects based on the needs of the target user and the application scenario, thereby enhancing the expressiveness and appeal of the view.

[0074] The view component is the main interface through which the target user interacts with the system. It is responsible for displaying the final generated scene view to the target user, providing an intuitive visualization. The view component also receives interaction requests from the target user, such as querying information or performing operations, and passes these requests to the large language model component.

[0075] For example, the view component is used to receive and process the input view of the target user; 3D Gaussian uses Gaussian representation to process 3D data; 3D Gaussian distribution is used to visualize 3D Gaussian; style selection is used to select different styles for visual presentation; generalizable Gaussian sputtering style is used to incorporate stylization selection into Gaussian sputtering; present novel view is used to present new views based on the processed data; stylized 3D Gaussian is used to apply stylization to 3D Gaussian representation; stylized novel view is used to present novel views through the applied style.

[0076] The large language model component is responsible for receiving interaction requests from the target user through the view component. Then, by connecting to the knowledge graph component, it retrieves and queries data related to the interaction request from the knowledge graph component, and leverages the powerful language understanding and generation capabilities of the large language model component to determine and generate corresponding response data. Simultaneously, the view component displays the corresponding response data, ensuring that the target user can obtain the necessary information feedback in a timely manner.

[0077] The knowledge graph component stores a large amount of data and information, providing data support for large language model components, enabling them to obtain accurate and comprehensive knowledge and information when processing interactive requests.

[0078] The scene building component includes "processing input view", and the large language model component includes "storing session data", "generating natural language response", "processing natural language query", and "querying data from knowledge graph". The specific interaction relationships and processes are as follows:

[0079] First, a preliminary analysis of the input view.

[0080] After receiving the input view (such as a 3D image, scene description, or related parameters) from the target user, the scene building component first performs preliminary analysis and processing, which may include identifying key elements and user intent in the input view and extracting this information into structured data or natural language description.

[0081] For example, if a target user uploads an image of a wind power generation scene, key information such as the wind turbine and environmental conditions can be extracted.

[0082] Second, process natural language queries.

[0083] After initial analysis, the extracted information can be submitted to a large language model component in the form of natural language for further processing and understanding.

[0084] Large language model components parse natural language input and identify the specific needs of the target user (e.g., the target user wants to know the optimal operating settings for the current wind turbine).

[0085] Third, query data from knowledge graphs.

[0086] Based on the results of processing natural language queries, large language model components need to retrieve relevant data from knowledge graphs. Knowledge graphs store a wide range of domain knowledge, including optimal operating parameters for wind farms, equipment maintenance records, and environmental impact analyses.

[0087] Large language model components will issue requests to retrieve necessary knowledge data. The knowledge graph interface will return optimization information or operational suggestions related to the input view.

[0088] Fourth, generate natural language responses.

[0089] Once the required data is retrieved from the knowledge graph, large language model components transform it into natural language descriptions or instructions that are user-friendly. These responses can include detailed analysis reports, operational suggestions, or answers to the target user's queries.

[0090] For example, a large language model component can generate any suggestion to guide the target user on how to adjust turbine settings based on current wind conditions to maximize power generation efficiency.

[0091] Fifth, store session data.

[0092] Throughout the interaction, all inputs, processing steps, and outputs are recorded. Large language model components store this data for future analysis, model performance optimization, or to provide a reference for subsequent queries. Storing session data can also be used to personalize the user experience, enabling the system to remember the target user's preferences and historical queries, providing more accurate suggestions.

[0093] The interaction process between the knowledge graph component and the large language model component is as follows:

[0094] After receiving a natural language query from a target user, the large language model component first uses the natural language understanding module to parse the user's intent and identify the entities, relationships, and context involved in the query.

[0095] The structured information parsed by large language model components can contain requirements for specific information in knowledge graph components, such as retrieving the attributes of an entity, finding a certain relationship, or obtaining rules for a specific domain.

[0096] Then, data is retrieved from the knowledge graph component. The knowledge graph component contains information about entities (such as people, places, objects, etc.) and their attributes and relationships, and can retrieve relevant information from the rich semantic network of the large language model component based on requests.

[0097] Large language model components send query requests to knowledge graph components based on the parsed requirements. These requests are typically structured, such as SPARQL queries, and are used to retrieve specific entity information, attributes, or relationships from the knowledge graph components.

[0098] Next, data fusion and natural language generation are performed. The knowledge graph component returns the query results, which are highly structured data, such as the attribute values ​​of an entity, the relationship between two entities, or a set of predefined rules.

[0099] After receiving structured data from the knowledge graph component, the large language model component combines this data with the context of the natural language query to generate a natural language response suitable for user understanding. This step is typically performed by the natural language generation module.

[0100] For example, a user inputs "What is the optimal setting for the wind turbine at the current wind speed?"; the large language model component parses the query and identifies the key entities (wind speed, wind turbine) and the required information type (optimal setting); the large language model component sends a query request to the knowledge graph component to obtain rules or historical data related to wind speed and wind turbine operation; the knowledge graph component returns data on the relationship between wind speed and turbine settings; the large language model component converts this data into a natural language response, such as "Based on the current wind speed, it is recommended to adjust the blade angle of the wind turbine to 30 degrees to maximize power generation efficiency."

[0101] Finally, after completing the interaction, the large language model component can store the query results and generated responses in its memory module, enabling faster and more accurate responses to future related queries. These memories can further enrich the content of the knowledge graph, thereby improving the overall intelligence level of the system.

[0102] Knowledge graph components can also update their internal data structures through feedback from large language model components, such as adding new entity relationships or updating existing data.

[0103] Therefore, the interaction between the knowledge graph component and the large language model component is bidirectional. The large language model component uses the structured information of the knowledge graph component to enhance its language understanding and generation capabilities, while the knowledge graph component obtains contextual information from the queries of the large language model component to continuously update and optimize its internal data.

[0104] The large language model component includes a large language model. The interaction process between natural language understanding and the large language model includes:

[0105] Natural Language Understanding (NLE) first receives natural language input from the target user and parses out key intents and information. This parsed information is then passed to a large language model as input for further processing. The large language model uses this structured information to generate contextual understanding, building a more comprehensive semantic model. This structured information helps the large language model accurately identify user needs and generate corresponding natural language responses in subsequent processing.

[0106] As can be seen from the above, the rendering component's rendering of the input view, and the scene building component's fusion of the stylized view and the rendered new view, provide the target user with a visual scene view, greatly enhancing the visual experience. Secondly, the connection between the view component and the large language model component ensures timely responses to the target user's interactive requests. Leveraging the rich data in the knowledge graph component, the large language model component accurately determines the response data corresponding to the interactive request, thereby achieving efficient interactive operations and meeting diverse user needs. Therefore, the close collaboration and efficient integration between the components optimizes the overall system performance and significantly improves the efficiency of interactive operations.

[0107] In one embodiment of this disclosure, reference is made to Figure 2 , Figure 3 and Figure 4 A wind power generation intelligent interactive integrated system based on a large language model also includes a robot assistant component;

[0108] The robot assistant component connects to the large language model component to respond to commands from the target user. It performs retrieval based on the large language model component, determines the corresponding control instructions based on the retrieval results, and controls the external robot to execute the corresponding actions according to the control instructions.

[0109] In this embodiment, the target user's command can be either voice or text. When the target user issues a command, the robot assistant component will respond immediately.

[0110] First, the robot assistant component performs retrieval based on a large language model component. This large language model component stores a wealth of information and data, supporting the robot assistant component's retrieval process. Then, based on the retrieval results, the robot assistant component determines the control instructions corresponding to the target user's command. This process requires analyzing and understanding the retrieved information to accurately translate the target user's command into specific, executable control instructions. Finally, the robot assistant component transmits the generated control instructions to the external robot, enabling it to perform the corresponding actions.

[0111] In wind power generation scenarios, the application of robot assistant components can significantly improve operational efficiency, maintenance quality, and the overall system intelligence. Specific application scenarios for robot assistant components include the following:

[0112] (1) Routine inspection and maintenance of wind turbines.

[0113] The robotic assistant component can be equipped with high-resolution cameras and sensors to periodically inspect the blades, nacelles, and tower structures of wind turbines, detecting potential problems such as wear, cracks, and corrosion.

[0114] The robot assistant component integrates with wind sensors and a data monitoring module to collect real-time data and transmit the inspection results to the control module. If an anomaly is detected, the robot assistant component can automatically generate a maintenance report and suggest or perform necessary maintenance operations.

[0115] (2) Calibration and debugging of wind sensors

[0116] Wind sensors require regular calibration to ensure the accuracy of their measurements. Robotic assistant components can perform automated calibration and adjustment of wind sensors, reducing the need for human intervention.

[0117] The robot assistant component uses a standardized calibration procedure to work in conjunction with the control module, adjusting wind sensor settings and verifying their accuracy under different wind speed conditions. Upon completion, the robot assistant component generates a calibration report and uploads it to the database.

[0118] (3) Real-time monitoring and data analysis of wind fields

[0119] The robot assistant component monitors the operation status of the wind farm in real time through an integrated data monitoring module, including wind speed, wind direction, power generation, and equipment status.

[0120] The robot assistant component collects and analyzes various types of data in real time, using machine learning algorithms to predict equipment performance trends and identify potential failure risks. It can also interface with a knowledge graph component to optimize wind farm operating parameters using industry standards and historical data.

[0121] (4) Rapid response to emergency situations

[0122] When a wind farm encounters extreme weather or sudden equipment failure, the robot assistant component can automatically activate the emergency response program to ensure the safe operation of the wind farm.

[0123] Based on real-time data and pre-set emergency plans, the robot assistant component automatically executes shutdown or restart procedures for wind turbines and assists on-site personnel in troubleshooting and emergency repairs. The robot assistant component can also report detailed event information to the control module to ensure that subsequent response measures are in place.

[0124] (5) Remote control and optimization of power generation equipment

[0125] Through remote connection, the robot assistant component can control wind power generation equipment from any location outside the control room, enabling remote operation and system optimization.

[0126] The robotic assistant component analyzes current power generation efficiency based on collected real-time data and optimizes the overall output of the wind farm by adjusting wind turbine operating parameters, such as blade angle and generator load. Furthermore, the robotic assistant component can remotely start, stop, or adjust the operating mode of the wind power equipment upon receiving commands from the control module.

[0127] (6) Intelligent maintenance and autonomous learning

[0128] The robot assistant component learns autonomously by continuously collecting and analyzing wind field operation data, thereby improving the intelligence level of maintenance operations.

[0129] The robotic assistant component combines historical maintenance data with real-time data, using machine learning algorithms to optimize maintenance strategies. For example, the robotic assistant component can adjust maintenance cycles based on equipment usage frequency and failure rate, thereby extending equipment lifespan and reducing maintenance costs.

[0130] (7) Predictive maintenance of wind power generation scenarios

[0131] By leveraging the predictive capabilities of the robot assistant component, potential equipment malfunctions can be identified in advance, preventing unplanned downtime.

[0132] The robot assistant component continuously analyzes the equipment's operating status through the data monitoring module, combines industry standards and historical fault data in the knowledge graph to predict potential equipment failures, and arranges maintenance work in advance to ensure the continuity of wind power generation.

[0133] By deploying robotic assistant components in wind power generation scenarios, the operational efficiency, equipment maintenance quality, and overall management level of wind farms can be significantly improved, while also reducing the risks and costs of manual operation.

[0134] For example, the target user's command is voice. The robot assistant component processes the target user's voice into text through speech recognition; then, it uses natural language understanding to understand and interpret the target user's voice, enabling dialogue management, i.e., managing the interaction and dialogue flow with the target user; next, it performs knowledge retrieval by querying relevant information based on the target user's voice and a large language model component, thereby making decisions based on the retrieval results and the target user's voice; finally, it controls the external physical robot based on the decisions and the target user's voice.

[0135] External physical robots refer to physical robots with autonomous mobility, operational capabilities, and sensory abilities, capable of performing various tasks in the physical world, including inspection, maintenance, monitoring, and troubleshooting. Specifically, the application of external physical robots in wind power generation includes:

[0136] Inspection robots, equipped with cameras, sensors, and other detection devices, can climb the towers, blades, or nacelles of wind turbines to perform physical inspections and detect structural damage or wear.

[0137] Maintenance robots can carry tools and spare parts to perform complex maintenance tasks on-site, such as replacing turbine components, repairing cracks, or adjusting equipment settings.

[0138] Inspection robots typically have automatic navigation and obstacle avoidance capabilities, enabling them to patrol wind fields, monitor equipment operation in real time, and identify and report abnormalities.

[0139] Emergency response robots can react quickly in the event of an emergency at a wind farm (such as a fire or extreme weather), performing emergency shutdowns, cooling, or other protective measures to ensure the safety of the wind farm.

[0140] Remotely operated robots allow operators to control and operate them remotely to complete tasks that are unsuitable or impossible to perform manually, such as maintenance work at high altitudes, in harsh weather, or in dangerous environments.

[0141] Robotic assistant components can significantly reduce manual workload, lower maintenance costs, and improve the safety and reliability of wind power generation facilities.

[0142] The large language model component integrates a backend, a large language model, a knowledge graph interface, memory, a dialogue manager, natural language generation, an inference engine, data processing, and training modules. The backend manages backend operations and communication between components. The large language model is the core language model of the large language model component and can be an instance such as ChatGPT-4 or Llama 3. The knowledge graph interface connects to the knowledge graph for information retrieval and integration. Memory manages the storage and retrieval of information within the system. The dialogue manager manages dialogue with the target user, ensuring coherent and context-appropriate responses. Natural language generation generates natural language responses based on the target user's commands and system data. The inference engine uses the large language model to perform inference tasks. Data processing provides data for system use. The training module manages the training of the large language model, integrating new data and updates. The integration of the various modules within the large language model component facilitates interaction with other components.

[0143] As can be seen from the above, the introduction of the robot assistant component significantly enhances the system's execution capabilities. The target user's commands can be directly translated into actual actions of the external robot, improving work efficiency. Through connection and precise retrieval with a large language model component, the accuracy and rationality of control commands are ensured, reducing the risk of erroneous operations. Simultaneously, it provides the target user with a more convenient operating experience; no complex settings are required, and control of the external robot can be achieved simply by issuing commands, saving time and effort.

[0144] In one embodiment of this disclosure, reference is made to Figure 2 and Figure 4 The robot assistant component is connected to the knowledge graph component;

[0145] The robot assistant component is also used to collect environmental data and send the environmental data to the knowledge graph component;

[0146] The knowledge graph component is used to update the knowledge base based on environmental data.

[0147] In this embodiment, the robot assistant component not only determines control commands based on information from the large language model component, but also has the function of collecting environmental data. The robot assistant component can collect various real-time data about its environment through various sensors. Then, it sends the collected environmental data to the knowledge graph component. Upon receiving the environmental data, the knowledge graph component analyzes and processes it, integrating the new data into the existing knowledge base to update the knowledge base.

[0148] In wind power generation scenarios, robotic assistant components can collect various real-time environmental data through multiple sensors, helping to optimize wind farm operation, perform predictive maintenance, and improve safety and efficiency. Environmental data types and corresponding sensors include:

[0149] (1) Wind speed and wind direction

[0150] Data types: wind speed, wind direction;

[0151] Sensors: anemometer, wind vane;

[0152] Application: Used for real-time monitoring of wind turbine operating conditions to help adjust turbine blade angles and rotational speeds to optimize power generation efficiency.

[0153] (2) Temperature

[0154] Data types: ambient temperature, equipment surface temperature;

[0155] Sensor: Temperature sensor;

[0156] Applications: To monitor ambient and equipment temperatures, prevent equipment overheating, and take protective measures under extreme weather conditions.

[0157] (3) Humidity

[0158] Data type: Air humidity;

[0159] Sensor: Humidity sensor;

[0160] Application: To monitor humidity in the air and prevent equipment corrosion or material deterioration caused by high humidity.

[0161] (4) Vibration

[0162] Data types: vibration frequency, amplitude;

[0163] Sensors: Accelerometer, vibration sensor;

[0164] Application: To detect abnormal vibrations in wind turbines and other mechanical components, and to identify potential mechanical failures in advance.

[0165] (5) Sound

[0166] Data types: environmental noise, equipment operating noise;

[0167] Sensors: microphone, acoustic sensor;

[0168] Purpose: To monitor the sound of equipment during operation and detect abnormal noise, thereby identifying faults or damage in advance.

[0169] (6) Light intensity

[0170] Data type: Light intensity;

[0171] Sensor: Optical sensor;

[0172] Application: To monitor the lighting conditions of wind farms, which is especially important when maintenance or visual inspection is required.

[0173] (7) Pressure

[0174] Data types: pneumatic pressure, hydraulic pressure;

[0175] Sensor: Pressure sensor;

[0176] Application: To monitor pressure changes inside equipment or in the environment to ensure the system operates within a safe range.

[0177] (8) Distance and position

[0178] Data type: Distance to the object, position coordinates;

[0179] Sensors: laser rangefinder, ultrasonic sensor, GPS module;

[0180] Applications: To measure the tilt or deformation of wind turbines, and to monitor the real-time position and navigation of external physical robots.

[0181] (9) Rainfall

[0182] Data type: Precipitation;

[0183] Sensor: Rain gauge;

[0184] Applications: Monitoring rainfall to help predict potential impacts on wind power systems, such as equipment waterproofing requirements.

[0185] (10) Electromagnetic field

[0186] Data type: Electromagnetic field strength;

[0187] Sensor: Electromagnetic field sensor;

[0188] Purpose: To monitor potential electromagnetic interference in wind fields, ensure normal equipment operation, and protect the safety of personnel.

[0189] (11) Inclination angle

[0190] Data type: Device tilt angle;

[0191] Sensor: Tilt sensor;

[0192] Application: To monitor the tilt angle of wind turbine towers to prevent collapse or damage caused by structural instability.

[0193] For example, data is collected from various sensors and preprocessed for further analysis and use. Then, data integration is performed, incorporating the processed data into the system. Finally, a knowledge base is generated to store knowledge specifically for the robot assistant component. Simultaneously, continuous learning ensures the robot assistant component continuously learns and improves, providing analysis and feedback to refine decision-making and system performance. The model of the robot assistant component is updated based on new data and feedback, resulting in continuous optimization of its performance.

[0194] As can be seen from the above, the robot assistant component collects environmental data and transmits it to the knowledge graph component, enabling the knowledge base to be updated in real time. This ensures the timeliness and accuracy of information, allowing the system to adapt more accurately to environmental changes and make reasonable decisions. Simultaneously, it provides a powerful impetus for the intelligent development of the system, enabling the robot assistant component to continuously learn and evolve, better serving users and meeting various complex needs.

[0195] In one embodiment of this disclosure, reference is made to Figure 2 , Figure 5 , Figure 6 and Figure 7 A wind power generation intelligent interactive integration system based on a large language model also includes wind power generation scenario components;

[0196] The wind power generation scenario component is connected to the large-scale language model component;

[0197] The view component is also used to receive the target user's first request;

[0198] The large language model component is also used to determine wind power generation scenario data from the wind power generation scenario component in response to the first request;

[0199] The view component is also used to display wind power generation scenario data.

[0200] In this embodiment, the first request is "requesting wind power generation scenario data." When the target user issues the first request through the view component, the large language model component will respond. The large language model component will obtain relevant data from the wind power generation scenario component it is connected to, which may include information such as the real-time power generation of wind power, the operating status of wind turbines, wind speed, and wind direction. The view component will then receive the wind power generation scenario data from the large language model component and display it to the user, allowing the user to intuitively understand the relevant information about wind power generation.

[0201] For example, the target user sends a first request to the view component, then the large language model component responds to the first request. The large language model component requests data from the wind power generation scenario component, the data monitoring module collects the data, and returns the collected data to the wind power generation scenario component. The wind power generation scenario component can also request maintenance data from the maintenance module, which returns the maintenance data to the wind power generation scenario component. The wind power generation scenario component returns scenario data to the large language model component, and simultaneously, the large language model component returns the wind power generation scenario data to the view component.

[0202] For example, the system can be in various states. Idle indicates the system is in a waiting state, ready to receive input or instructions from the target user; Target user input received means the system has received the input and is processing the request; Scene construction initiated means the system begins building the scene based on the target user input; 3D Gaussian object created means the system creates a 3D Gaussian object based on the input view; Style selection applied means the system applies the selected style to the scene; Gaussian sputtering means the system performs sputtering on the 3D Gaussian; Rendering means the system generates novel and stylized novel views; Large language model processing means the system interacts with a large language model to process data and generate responses; Inference execution means the system uses the large language model to perform inference tasks; Knowledge graph query means the system queries the knowledge graph to obtain relevant information; Memory management means the system manages memory operations, such as storing and retrieving information; Data processing means the system processes data input for further analysis; Robot assistant operation means the system performs tasks related to the robot assistant; System analyzes feedback to improve decision-making; System updates the model based on new data and feedback; System monitors and manages wind power scenarios; System performs maintenance activities on wind power systems; Process completes, and the system returns to the idle state.

[0203] As can be seen from the above, the wind power generation scenario component provides target users with a convenient way to obtain wind power generation scenario data. Target users can quickly obtain the required information simply by issuing a request through the view component, improving the user experience and allowing users to grasp the status of wind power generation in real time and intuitively. This helps to identify problems in a timely manner and make decisions, thereby improving power generation efficiency and safety.

[0204] In one embodiment of this disclosure, reference is made to Figure 2 and Figure 4 The components for wind power generation include wind turbines, wind sensors, control modules, data monitoring modules, and maintenance modules.

[0205] The wind turbine is connected to the wind sensor, the wind sensor is connected to the control module, and the control module is connected to the data monitoring module and the maintenance module respectively; the control module, the data monitoring module, and the maintenance module are all connected to the large language model component.

[0206] The wind sensor is used to collect wind condition data and send the wind condition data to the control module;

[0207] The data monitoring module is used to monitor operational data, while the maintenance module is used to manage and maintain data.

[0208] The data for wind power generation scenarios includes operational data or maintenance data.

[0209] In this embodiment, a wind sensor is connected to the wind turbine to collect real-time wind data, such as wind speed and direction, and sends the collected data to the control module. The control module receives the data from the wind sensor and adjusts the operation of the wind turbine. Simultaneously, the control module connects to both a data monitoring module and a maintenance module to facilitate information exchange. The data monitoring module closely monitors the system's operational data, including power generation and equipment temperature, to ensure normal system operation. The maintenance module manages maintenance data, such as equipment maintenance records and repair plans. By connecting to a large-scale language model component, relevant data is transmitted to the component, providing data support for the system's intelligent decision-making and interaction.

[0210] For example, a wind sensor monitors wind conditions and provides data to the control module. A power converter converts wind energy into usable electrical energy, which is then connected to the power grid for power supply. The control module manages the control of the wind power generation scenario components, thereby monitoring their data and managing their maintenance activities. The wind power generation scenario components are integrated into the overall system by interacting with a large language model component to perform control and monitoring tasks.

[0211] The specific process by which the wind power generation scenario component performs control and monitoring tasks through interaction with the large language model component is as follows:

[0212] First, the dynamic adjustment of wind turbines.

[0213] Wind speed and wind direction sensors in wind power generation scenarios monitor current wind speed and direction data in real time.

[0214] The wind power generation scenario component transmits this data to the large language model component, which analyzes the optimal wind turbine blade angle and speed settings through its built-in inference engine and knowledge graph interface.

[0215] The large language model component generates a set of optimized control commands that instruct the wind turbine to adjust its blade angle and rotational speed to maximize power generation efficiency.

[0216] The adjusted power generation efficiency and equipment status data are transmitted back to the large language model component for further optimization and adjustment of future control strategies.

[0217] Second, the detection and response to abnormal vibrations.

[0218] Vibration sensors in wind power generation systems monitor the vibration frequency and amplitude of turbines and other mechanical components in real time. This data is transmitted to a large-scale language model component, which analyzes the data using a pre-trained model and knowledge graph to determine whether the current vibration is abnormal.

[0219] If abnormal vibration is detected, the large language model component generates a detailed alarm message, including possible causes of the failure and recommended maintenance measures.

[0220] Alarm information is sent to on-site operators via the view component and simultaneously transmitted to the robot assistant component, triggering it to perform automated inspection or maintenance tasks.

[0221] The robot assistant component performs specific maintenance operations based on the analysis results of the large language model component, such as checking the mechanical components of a wind turbine or adjusting the equipment settings.

[0222] Third, predicting the impact of weather changes on power generation.

[0223] The wind power generation scenario component collects real-time weather data from environmental sensors (such as temperature, humidity, and light intensity). This weather data is then transmitted to a large-scale language model component, where it is analyzed using a knowledge graph and weather model to predict the impact of future weather changes on wind power generation efficiency.

[0224] Large language model components generate adjustment strategies, including possible power generation regulation and equipment protection measures (such as power generation suspension under extreme weather conditions).

[0225] The control module executes adjustment strategies and performs necessary equipment checks and preventative maintenance through the robot assistant component.

[0226] Continuously monitor the actual impact of weather changes and transmit the feedback data back to a large language model component for optimization of future strategies.

[0227] Fourth, remote monitoring and maintenance.

[0228] Sensors and data monitoring modules within the wind power generation scenario components continuously provide data streams to the large-scale language model component, including equipment status, environmental conditions, and power generation performance. The large-scale language model component integrates this data and uses machine learning algorithms to analyze the overall operational status of the power plant and identify potential problems.

[0229] If a problem is detected, the large language model component generates a detailed diagnostic report and provides guidance to the remote operations team via the view component, outlining the specific actions to be taken. Simultaneously, the large language model component can also generate automated maintenance instructions, enabling the robot assistant component to perform necessary on-site inspections and maintenance operations.

[0230] The execution results are fed back to the large language model component in real time to further optimize the model's diagnostic and predictive capabilities.

[0231] Through these interactive processes, the connection between the wind power generation scenario and the entire system is enhanced, making it not just an independent component, but tightly integrated into a large language model-driven intelligent control and monitoring system, which can significantly improve the automation level and operational efficiency of wind farms.

[0232] As can be seen from the above, wind sensors collect wind condition data and transmit it to the control module, enabling wind turbines to optimize operation based on real-time wind conditions and improve power generation efficiency. The data monitoring module monitors operational data, helping to promptly identify potential faults, ensuring stable system operation, and reducing downtime losses. The maintenance module manages maintenance data, allowing for the rational scheduling of equipment maintenance plans, extending equipment lifespan, and reducing maintenance costs. Wind power generation scenario data encompasses both operational and maintenance data, providing comprehensive information for large-scale language model components, facilitating intelligent power generation management and decision-making, and improving the reliability of the entire wind power generation system.

[0233] In one embodiment of this disclosure, reference is made to Figure 2 , Figure 5 , Figure 6 and Figure 7 A wind power generation intelligent interactive integration system based on a large language model also includes:

[0234] The view component is also used to receive a second request from the target user;

[0235] The large language model component is also used in response to a second request to determine the target scenario based on the scenario-based component construction.

[0236] The view component is also used to display the target scene.

[0237] In this embodiment, the second request is "request for scene visualization." When the target user issues the second request through the view component, the large language model component responds. Simultaneously, the large language model component collaborates with the scene building component to determine the target scene. The scene building component utilizes its functions and data to generate or select a suitable target scene based on the target user's needs, system settings, and relevant conditions and rules. Subsequently, the view component receives the information about the target scene and displays it to the user in a visual format. The target user can intuitively see the requested scene, thereby better understanding and interacting with the system.

[0238] For example, the target user sends a second request to the view component, which then calls the scene visualization to visualize the scene. The scene visualization sends a command to the scene building component to request scene data. The scene building component sends a "Get Rendered View" command to the rendering view component, which then sends a "Return Rendered View" response back to the scene building component. The scene building component then sends a "Get Stylized View" command to the stylized view, which returns the stylized view to the scene building component. Finally, the scene building component returns the scene data to the scene visualization, and the scene is ultimately displayed through the view component.

[0239] As can be seen from the above, by issuing a second request, the target user can obtain the specific scenario display they need, which can improve the target user's ability to understand and grasp complex information. The intuitive presentation of the target scenario makes it easier for the target user to understand the system's operating status. Therefore, this embodiment not only helps to improve the accuracy and efficiency of decision-making, but also enhances the system's interactivity and user experience.

[0240] In one embodiment of this disclosure, reference is made to Figure 2 , Figure 5 , Figure 6 and Figure 7 A wind power generation intelligent interactive integration system based on a large language model also includes:

[0241] The view component is also used to receive third-party requests from the target user;

[0242] The large language model component is also used to update scene data based on scene components in response to third-party requests;

[0243] The view component is also used to display updated scene data.

[0244] In this embodiment, the third request is "reload scene". When the target user issues this third request through the view component, the large language model component responds immediately. The large language model component works in conjunction with the scene component. The scene component is responsible for storing and managing scene-related data. The large language model component sends instructions to the scene component, requesting an update to the scene data. This may include obtaining the latest real-time data, recalculating certain parameters, loading new configuration information, or correcting previous data errors. After the update is complete, the view component promptly displays the updated scene data, ensuring that the user sees the latest and most accurate scene information.

[0245] For example, the target user sends a third request to the view component to reload the scene. Then, the scene reload sends a "Request scene data" command to the scene building component, which in turn sends a "Get render view" command to the render view component. The render view component then sends a "Return render view" response to the scene building component. After that, the scene building component sends a "Get stylized view" command to the stylized view, which returns the stylized view to the scene building component. Finally, the scene building component returns the scene data to the scene reload, and the scene data is ultimately displayed through the view component.

[0246] As can be seen from the above, the view component ensures that the target user always has access to the latest scene data, maintaining the accuracy and timeliness of information in a rapidly changing environment. For complex scenarios, it helps users make timely and accurate decisions. Furthermore, it not only enhances the target user experience, allowing them to easily refresh the scene and stay up-to-date without complex operations, but also strengthens the system's flexibility and adaptability, responding quickly to data updates and scene configuration adjustments to meet the ever-changing needs of users.

[0247] In one embodiment of this disclosure, reference is made to Figure 2 and Figure 4 The rendering component is specifically used to render the input view based on the 3D Gaussian sputtering algorithm to obtain a new rendered view.

[0248] In this embodiment, the steps of the 3D Gaussian sputtering algorithm include: preparing raw data for further processing through a preprocessing module; extracting relevant features from the preprocessed data through a feature extraction module; normalizing the data to ensure consistency through a normalization module; dividing the spatial data into manageable parts through a spatial subdivision module; defining the Gaussian kernel used in sputtering through a Gaussian kernel definition module; calculating the weights of the sputtering process through a weight calculation module; applying splatting technology to the 3D data through a splatting module; integrating the dispersed data to generate the final output through a fusion module; storing the final processed data for rendering; and finally, outputting a rendered view of the processed data. The 3D Gaussian sputtering algorithm processes data for rendering and visualization tasks, which are then visualized by a scene building component.

[0249] As can be seen from the above, the 3D Gaussian sputtering algorithm can bring a more immersive experience to the target users. At the same time, it improves rendering efficiency, reducing the rendering time and computational resources required while maintaining high quality. Therefore, it contributes to the system's rapid response, allowing users to obtain exquisite rendering results in a timely manner, and better engage in interaction and analysis.

[0250] In one embodiment of this disclosure, reference is made to Figure 2 The knowledge graph component includes a knowledge base, which contains external data, historical data, and industry data.

[0251] Large language model components connect to knowledge graph components via a knowledge graph interface.

[0252] In this embodiment, the knowledge base refers to a central repository of structured knowledge used by the system. External data refers to interfaces with external data sources to obtain more information. Historical data refers to storing and managing historical data for reference and analysis. Industry standards refer to industry standards and guidelines related to system operation. Knowledge graph output refers to providing the output and results of knowledge graph queries. Large language model components connect to knowledge graph components through the knowledge graph interface to perform knowledge graph operations, the process of which includes: knowledge base query, external data integration, historical data access, and industry standard reference. Specifically, knowledge base query refers to the system querying the knowledge base to obtain structured information. External data integration refers to the system integrating external data sources. Historical data access refers to the system accessing historical data for reference. Industry standard reference refers to the system referencing industry standards.

[0253] As can be seen from the above, the rich knowledge base provides the system with comprehensive and structured knowledge, covering external data, historical data, and industry data, enhancing the system's information reserves and analytical capabilities. Large language model components connect to the base via interfaces, enabling operations such as knowledge base queries and external data integration, allowing access to broader and more accurate information and improving the quality and accuracy of responses. Access to historical data facilitates the analysis of trends and patterns, providing a reference for decision-making. External data integration enriches the data sources, and industry standard references ensure that system operations comply with regulations. Therefore, this embodiment enhances the system's intelligence and practicality, making it suitable for various scenarios.

[0254] In one embodiment of this disclosure, reference is made to Figure 2 and Figure 4 The view component is also used to display and render new views.

[0255] In this embodiment, the view component displays not only a stylized new view but also a rendered new view. Rendering a new view involves generating a new view of the scene from different angles or perspectives, which can be obtained using a 3D Gaussian sputtering algorithm.

[0256] As can be seen from the above, the scene views displayed by the view components are rich and diverse, providing excellent support for scene visualization and greatly enhancing the user experience.

[0257] This invention relates to an integrated system that combines scene construction, large-scale language model interaction, and robot assistance. The system utilizes a 3D Gaussian sputtering algorithm, a large-scale language model, retrieval-enhanced generation, and knowledge graphs to create an interactive intelligent environment. The system enhances the user experience by providing real-time data visualization, responsive robot interaction, and seamless integration with various data sources. Simultaneously, the system also offers efficient data processing capabilities.

[0258] The above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit it. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A wind power generation intelligent interactive integration system based on a large language model, characterized in that, include: View components, scene building components, rendering components, large language model components, and knowledge graph components; The rendering component, connected to the scene building component, is used to render the target user's input view to obtain a new rendered view; The scene building component connects with the view component to stylize the input view, obtain a stylized view, and then merges the stylized view with the rendered new view to obtain the scene view. The view component is used to display the scene view; The view component, connected to the large language model component, is used to receive interaction requests from the target user; A large language model component, connected to a knowledge graph component, is used to respond to interaction requests and determine the response data corresponding to the interaction request based on the data in the knowledge graph component. The view component is also used to display response data; It also includes a robot assistant component; the robot assistant component is connected to the large language model component and is used to respond to the commands of the target user, perform retrieval based on the large language model component, determine the control instructions corresponding to the commands based on the retrieval results, and control the external robot to perform the actions corresponding to the control instructions according to the control instructions. The robot assistant component is connected to the knowledge graph component; the robot assistant component is also used to collect environmental data and send the environmental data to the knowledge graph component. The knowledge graph component is used to update the knowledge base based on environmental data; It also includes wind power generation scenario components; the wind power generation scenario components are connected to large language model components; The view component is also used to receive the target user's first request; The large language model component is also used to determine wind power generation scenario data from the wind power generation scenario component in response to the first request; The view component is also used to display wind power generation scenario data; The wind power generation scenario components include a wind turbine, a wind sensor, a control module, a data monitoring module, and a maintenance module. The wind turbine is connected to the wind sensor, the wind sensor is connected to the control module, and the control module is connected to both the data monitoring module and the maintenance module. The control module, data monitoring module, and maintenance module are all connected to a large-scale language model component. The wind sensor is used to collect wind condition data and send the wind condition data to the control module; The data monitoring module is used to monitor operational data, while the maintenance module is used to manage and maintain data. The data for wind power generation scenarios includes operational data or maintenance data; It also includes: the view component is also used to receive a second request from the target user; The large language model component is also used in response to a second request to determine the target scenario based on the scenario-based component construction. The view component is also used to display the target scene; It also includes: the view component is also used to receive third requests from the target user; The large language model component is also used to update scene data based on scene components in response to third-party requests; The view component is also used to display updated scene data; The rendering component is specifically used to render the input view based on the 3D Gaussian sputtering algorithm to obtain a new rendered view. The steps of the 3D Gaussian sputtering algorithm include: preparing the raw data for further processing through a preprocessing module; extracting relevant features from the preprocessed data through a feature extraction module; normalizing the data through a normalization module to ensure consistency; dividing the spatial data into manageable parts through a spatial subdivision module; defining the Gaussian kernel used in sputtering through a Gaussian kernel definition module; calculating the weights of the sputtering process through a weight calculation module; applying splatting technology to the 3D data through a splatting module; integrating the dispersed data through a fusion module to generate the final output; storing the final processed data for rendering; and finally, outputting the processed data rendered view through a rendered view. The 3D Gaussian sputtering algorithm processes the data for rendering and visualization tasks, which are then visualized by the scene building component. The knowledge graph component includes a knowledge base, which contains external data, historical data, and industry data; the large language model component connects to the knowledge graph component through the knowledge graph interface. View components are also used to display and render new views; The aforementioned intelligent interactive integrated system for wind power generation provides a visually appealing scene view for the target user by rendering the input view through a rendering component and merging the stylized view with the rendered new view through a scene building component, greatly enhancing the visual experience. Secondly, the connection between the view component and the large language model component ensures timely responses to the target user's interactive requests. The large language model component, leveraging the rich data in the knowledge graph component, accurately determines the response data corresponding to the interactive request, thereby achieving efficient interactive operations and meeting diverse user needs. Therefore, the close collaboration and efficient integration between the components optimize the overall system performance and significantly improve the efficiency of interactive operations. As an integrated system combining scene building, large language model interaction, and robot assistance, the aforementioned intelligent interactive integrated system for wind power generation applies 3D Gaussian sputtering algorithms, large language models, retrieval-enhanced generation, and knowledge graphs to create an interactive intelligent environment. It enhances the user experience by providing real-time data visualization, responsive robot interaction, and seamless integration of various data sources. Simultaneously, the system also provides efficient data processing capabilities.

Images