Power grid maintenance training system and method based on augmented reality

Abstract

The invention relates to the technical field of power system operation and maintenance training, and discloses a power grid maintenance training system and method based on augmented reality, and the system comprises a cloud rendering and distribution system which is used for rendering an augmented reality scene, generating real-time rendering output, and transmitting the real-time rendering output to terminal equipment through an edge computing node and a content distribution network, and the cloud rendering and distribution system is connected with the real-time dynamic content updating module and the data processing module, and is used for receiving the real-time state data of the equipment and the student operation behavior data, and is responsible for embedding the data into a rendering scene. The performance requirement of the terminal equipment is remarkably reduced through the cloud-based XR service, real-time rendering and transmission of high-quality content are realized by combining cloud rendering and edge computing with the content distribution network, and low-delay and smooth experience can be kept under the condition of network fluctuation, so that the network adaptability of the system is improved, and the hardware cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and maintenance training, and in particular to an extended reality-based power grid maintenance training system and method. Background Art

[0002] In the power industry, grid maintenance is a complex and high-risk job. Traditional grid maintenance training methods use field teaching and textbook training, which has limitations. Field teaching is costly and risky, especially for operation training in extreme weather or emergency situations. In addition, textbooks or video training cannot provide real operation experience and feedback, making it impossible for trainees to grasp the feel and details of actual operation. At the same time, the update speed of grid equipment and technology is fast, and traditional training methods cannot update content in time, causing trainees' knowledge and skills to lag behind the latest technology status.

[0003] Existing XR training systems rely on high-performance terminal devices to ensure rendering quality and smooth operation, resulting in high hardware costs. In addition, the real-time nature of XR content places high demands on the network. Network fluctuations can easily lead to increased latency or screen freezes, affecting the trainee experience. In traditional XR training systems, virtual device status and task settings are mostly preset static content, which cannot reflect the status changes of actual power grid equipment in a timely manner. There is a lack of real-time updates, and trainees cannot obtain training content that is synchronized with the actual power grid status, which causes the training content to be out of touch with actual work and reduces the authenticity of the training effect. The existing XR training system provides the same training path and task settings for trainees, and cannot be dynamically adjusted according to the trainees' actual operation level and learning progress, resulting in personalization, which in turn affects the learning effect; Existing XR training systems lack precise physical simulation in terms of touch and force feedback, which results in trainees being unable to obtain a tactile sensation similar to that of real power grid equipment during operation. At the same time, complex environmental factors such as temperature, humidity, wind force, etc. cannot be simulated, and trainees lack training to operate in complex environments, reducing the practicality and immersion of the training.

[0004] Therefore, those skilled in the art provide a power grid maintenance training system and method based on extended reality to solve the above-mentioned problems. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a power grid maintenance training system and method based on extended reality to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a power grid maintenance training system based on extended reality, comprising: Cloud rendering and distribution system: used to render the extended reality scene, generate real-time rendering output, and transmit it to the terminal device through the edge computing node and the content distribution network. The cloud rendering and distribution system is connected to the real-time dynamic content update module and the data processing module, and is used to receive the real-time status data of the device and the student's operation behavior data, and is responsible for embedding the data into the rendering scene, so as to update the virtual device status and student interaction status in the extended reality scene in real time, and ensure that the scene seen by the student on the terminal device reflects the actual status in real time and accurately; Data processing module: used to collect data from sensors in the Internet of Things of power grid equipment, clean, convert and filter the data, and generate a real-time status data set of power grid equipment. In addition, the data processing module receives and processes the trainees' operation behavior data to provide input data for the AI-driven intelligent evaluation and personalized learning path modules; Real-time dynamic content update module: It includes a digital twin technology module, which is used to map the real-time status data set of power grid equipment generated by the data processing module to the virtual equipment model in the extended reality scene, and update the training tasks and fault cases through the content management system. The real-time dynamic content update module is connected to the cloud rendering and distribution system to receive the real-time data of the equipment and embed it into the extended reality scene, so as to realize the real-time update of the equipment status in the virtual scene. In addition, the real-time dynamic content update module ensures the synchronous transmission of the trainee operation and environmental data, so that the content of the scene is consistent with the actual situation; AI-driven intelligent assessment and personalized learning path module: Analyze the trainees' operation behavior data and operation accuracy, generate personalized learning paths, and adjust the task difficulty. The AI-driven intelligent assessment and personalized learning path module obtains the behavior data in the data processing module in real time, and recommends corresponding personalized training tasks through the task management module; Task management module: responsible for managing the creation, allocation and difficulty adjustment of training tasks. By connecting with the real-time dynamic content update module and the AI-driven intelligent assessment and personalized learning path module, it generates adaptive tasks based on the status of power grid equipment and trainee performance and transmits them to the trainee's terminal device; Virtual reality and physical simulation fusion module: The weight, friction and operating resistance of the device are simulated through force feedback devices and physical engines. The virtual reality and physical simulation fusion module is connected to the cloud rendering and distribution system to receive real-time rendering content and provide simulation data of force feedback devices.

[0007] Preferably, the cloud rendering and distribution system includes: Rendering server: responsible for generating rendering output of each frame of the extended reality scene and performing video encoding on the rendering output. The rendering server sends the encoded rendering output to the content distribution network for caching and distribution; Content delivery network: used to cache and distribute rendered content. The content delivery network receives encoded rendered content from the rendering server and caches it to quickly respond to terminal requests. At the same time, the content delivery network maintains a connection with the edge computing node and pushes the rendered content to the edge node close to the user to optimize transmission latency. Edge computing nodes: deployed close to the student terminals to handle low-latency tasks. Edge computing nodes obtain rendering content from the content distribution network and perform additional processing when necessary to ensure low-latency transmission, and finally transmit the rendering content to the student terminals.

[0008] Preferably, the data processing module includes: The data cleaning unit performs preliminary processing on the data collected by the sensor to filter out abnormal values ​​and fill in missing data, providing reliable data input for subsequent data standardization processing; The data conversion unit receives the data processed by the data cleaning unit and standardizes it into a unified format to ensure the consistency and comparability of the data. The output of the data conversion unit provides formatted data input for the data integration unit. The data integration unit receives standardized data from the data conversion unit and integrates the data from different sensors into a unified power grid equipment status data set, providing complete equipment status information for real-time updates of extended reality scenarios.

[0009] Preferably, the content management system of the real-time dynamic content update module includes: Training task database, used to store preset fault cases and operation tasks; A dynamic task updating unit, used to automatically generate corresponding training tasks according to the real-time status of power grid equipment; The real-time dynamic content update module realizes real-time status mapping through the digital twin technology module.

[0010] Preferably, the AI-driven intelligent assessment and personalized learning path module includes: Behavior analysis unit, used to analyze the trainees' operational behavior sequences Calculate the behavior score S in real time: , in, is the evaluation value of a single operation, is the behavior score, n is the total number of operations, The specific operations of the i-th step performed by the trainees in the task; When the operation complies with the standard, =1, when not in compliance =0; The behavior analysis unit calculates the behavior score Pass it to the personalized learning path generation unit for dynamic generation of subsequent learning paths; Personalized learning path generation unit generates learning paths based on students' operation accuracy and behavior scores : , in, is the path recommendation function, For personalized learning paths, score the behavior; The personalized learning path generation unit calculates the behavior score Recommend training tasks for students and personalize learning paths Passed to the task management module so that the recommended tasks can be pushed to the student terminal.

[0011] Preferably, the task management module includes: The task generation unit generates basic training tasks based on the real-time status data of power grid equipment and fault cases preset in the system; The task difficulty adjustment unit dynamically adjusts the difficulty coefficient of the task according to the student's behavior score and the learning path recommended by the task generation unit; A task allocation unit allocates tasks to trainee terminals based on personalized learning paths and adjusted task difficulty; The task feedback and recording unit is used to collect students' feedback and performance data during task execution, and return the data to the AI-driven intelligent assessment and personalized learning path module to further adjust the personalized learning path.

[0012] Preferably, the task generation unit generates the task based on the device status data received by the real-time dynamic content update module. When the real-time status data set D of the equipment and the fault case parameters are Create task content: , in, For the generated tasks, Generate a function for the task, It is a collection of real-time status data of the equipment. Fault case parameters preset for the system; The task difficulty adjustment unit calculates the final difficulty of the task based on the student's behavior score S : , in, is the basic difficulty of the task. is the adjustment factor, Rate students' behavior; If S decreases, the task difficulty d will automatically decrease, making the task more suitable for the current level of the students; If S increases, the task difficulty d increases to increase the challenge.

[0013] Preferably, the virtual reality and physical simulation fusion module includes: The force feedback unit is used to simulate the real force and resistance feedback that trainees receive during operation, so that the power grid equipment in virtual operation has real mechanical characteristics; A tactile feedback unit is used to provide tactile feedback so that trainees can feel the surface characteristics and vibration feedback of power grid equipment during virtual operation; High-precision physics engine unit, used to simulate the physical characteristics of power grid equipment, so that the equipment in the virtual environment exhibits realistic physical behaviors; The environmental condition simulation unit is used to simulate different environmental conditions, so that trainees can operate power grid equipment in a variety of scenarios.

[0014] Preferably, the force feedback unit simulates the resistance, pulling force and pushing force during operation through a force feedback glove or controller, and the feedback is based on the operation intensity. Adjusting feedback force , force feedback relationship formula: , in, is the feedback force, is the feedback force coefficient, For operational intensity; The high-precision physical engine unit calculates the friction, reaction force and gravity of the device based on the physical engine, so that the virtual device has real physical feedback during operation. In the process of simulating cable connection, the friction of the device The calculation formula is: , in, is the friction coefficient, The weight of the equipment.

[0015] A power grid maintenance training method based on extended reality includes the following steps: Step 1: The cloud rendering and distribution system generates high-quality rendering output of the power grid maintenance scene, and transmits the content to the trainees' terminal devices through edge computing nodes and content distribution networks. During this process, the real-time dynamic content update module embeds the real-time data of the device into the extended reality scene to provide trainees with a realistic simulation environment. Step 2: The data processing module collects data from the IoT sensors of power grid equipment, cleans, converts and integrates them, and generates a real-time status data set of the equipment; Step 3: Through the real-time dynamic content update module, the system uses digital twin technology to map the real-time status of power grid equipment to the virtual equipment model. At the same time, the content management system retrieves fault cases and operation tasks from the preset database, or generates corresponding training tasks according to the status of the equipment; Step 4: The AI-driven intelligent assessment and personalized learning path module analyzes the students' operation behavior and operation accuracy, conducts real-time assessment of the students, generates personalized learning paths, and dynamically adjusts the difficulty of tasks based on the students' performance, recommending adaptive tasks; Step 5: The task management module creates, assigns and adjusts tasks according to the student's personalized learning path, dynamically generates adaptive tasks based on the real-time status of the device and the student's performance, and pushes the tasks to the student's terminal device; Step 6: During the training process, the virtual reality and physical simulation fusion module provides trainees with real tactile feedback through force feedback devices and physics engines. At the same time, tactile feedback and environmental condition simulation enable trainees to experience the operation of power grid equipment in various scenarios. Step 7: During the task execution, the task management module collects students’ feedback and operation data, and transmits the data to the AI-driven intelligent evaluation module. Based on the students’ real-time operation performance, the system dynamically adjusts the learning path and task difficulty; Step 8: The virtual reality and physical simulation fusion module provides force feedback according to the trainee's operation intensity during the trainee's operation, so that the virtual operation has real physical feedback; Step 9: Based on the students' behavioral scores and feedback, the system continuously optimizes the personalized learning path and task difficulty, so that students can gradually adapt to more challenging tasks and achieve the goal of personalized training.

[0016] The present invention provides a power grid maintenance training system and method based on extended reality, which has the following beneficial effects: 1. The present invention significantly reduces the performance requirements of terminal devices through cloud-based XR services, and utilizes cloud rendering and edge computing combined with content distribution networks to achieve real-time rendering and transmission of high-quality content. In the case of network fluctuations, it can maintain low latency and smooth experience, thereby improving the network adaptability of the system and reducing hardware costs.

[0017] 2. The present invention uses real-time data of power grid equipment collected by IoT devices, combined with data processing and digital twin technology, to synchronize the status of power grid equipment in a virtual environment in real time, dynamically update training tasks and fault cases, ensure that the trainees' training content is consistent with the actual equipment status, and greatly improve the real-time and authenticity of the training, so that trainees can be exposed to the latest power grid operation conditions and fault scenarios in a virtual environment.

[0018] 3. The AI ​​driving module in the present invention analyzes the trainees' operating behaviors in real time, calculates behavioral scores, and generates personalized learning paths and dynamically adjusts task difficulty based on the trainees' performance, so that the trainees' training content is adapted to their current skill level, achieving targeted training, improving learning efficiency, and ensuring that trainees continue to improve at an appropriate level of challenge.

[0019] 4. The present invention reproduces the real physical properties and operational touch of power grid equipment through high-precision physics engines, force feedback and tactile feedback devices, so that trainees can obtain force, friction and vibration feedback similar to actual operations in a virtual environment, and provide operational training in a variety of complex situations through environmental condition simulation, thereby effectively improving trainees' operational stability and ability to cope with complex environments, and enhancing the practicality and immersion of training. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a system framework diagram of the present invention; Figure 2 is a flow chart of the present invention; Figure 3 This is a schematic diagram of the steps of implementing real-time state mapping by the real-time dynamic content update module of the present invention through the digital twin technology module. DETAILED DESCRIPTION

[0021] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0022] The present invention is described in detail below in conjunction with the accompanying drawings: Example: Please see attached Figure 1 -Attached Figure 3 , an embodiment of the present invention provides a power grid maintenance training system based on extended reality, including: Cloud rendering and distribution system: used to render the extended reality scene, generate real-time rendering output, and transmit it to the terminal device through the edge computing node and the content distribution network. The cloud rendering and distribution system is connected to the real-time dynamic content update module and the data processing module, and is used to receive the real-time status data of the device and the student's operation behavior data, and is responsible for embedding the data into the rendering scene, so as to update the virtual device status and student interaction status in the extended reality scene in real time, and ensure that the scene seen by the student on the terminal device reflects the actual status in real time and accurately; Data processing module: used to collect data from sensors in the Internet of Things of power grid equipment, clean, convert and filter the data, and generate a real-time status data set of power grid equipment. In addition, the data processing module receives and processes the trainees' operation behavior data to provide input data for the AI-driven intelligent evaluation and personalized learning path modules; Real-time dynamic content update module: It includes a digital twin technology module, which is used to map the real-time status data set of power grid equipment generated by the data processing module to the virtual equipment model in the extended reality scene, and update the training tasks and fault cases through the content management system. The real-time dynamic content update module is connected to the cloud rendering and distribution system to receive the real-time data of the equipment and embed it into the extended reality scene, so as to realize the real-time update of the equipment status in the virtual scene. In addition, the real-time dynamic content update module ensures the synchronous transmission of the trainee operation and environmental data, so that the content of the scene is consistent with the actual situation; AI-driven intelligent assessment and personalized learning path module: Analyze the trainees' operation behavior data and operation accuracy, generate personalized learning paths, and adjust the task difficulty. The AI-driven intelligent assessment and personalized learning path module obtains the behavior data in the data processing module in real time, and recommends corresponding personalized training tasks through the task management module; Task management module: responsible for managing the creation, allocation and difficulty adjustment of training tasks. By connecting with the real-time dynamic content update module and the AI-driven intelligent assessment and personalized learning path module, it generates adaptive tasks based on the status of power grid equipment and trainee performance and transmits them to the trainee's terminal device; Virtual reality and physical simulation fusion module: The weight, friction and operating resistance of the device are simulated through force feedback devices and physical engines. The virtual reality and physical simulation fusion module is connected to the cloud rendering and distribution system to receive real-time rendering content and provide simulation data of force feedback devices.

[0023] Benefits of cloud rendering and distribution system Reduce hardware costs and the complexity of equipment maintenance, ensure low latency in content transmission, improve system network adaptability and user experience, and improve the real-time and practical nature of training; Benefits of the data processing module: Ensure the accuracy and consistency of data, avoid misleading information caused by noise or outliers, provide high-quality input data for subsequent real-time dynamic content updates and AI intelligent evaluation, improve the overall performance of the system, provide basic support for real-time content updates and the generation of personalized learning paths, and ensure the real-time response of the system; Benefits of real-time dynamic content update modules: Improve the authenticity of training, ensure that trainees are exposed to the latest equipment status and situations, improve the relevance and effectiveness of learning, and enhance the practicality and emergency response capabilities of training; The benefits of AI-driven intelligent assessment and personalized learning path modules enable students to gradually improve their skills in the most suitable way, improve learning efficiency, avoid students' learning effects being affected by tasks that are too difficult or too easy, and ensure the pertinence and effectiveness of training; Benefits of the task management module: It can generate adaptive tasks based on the status of power grid equipment and the trainees' operating performance, ensuring that the trainees' training content meets their current capabilities and needs, reducing the need for manual intervention, ensuring that trainees continue to progress on the appropriate learning curve, and improving training effectiveness; The benefits of the virtual reality and physical simulation fusion module are that trainees can get a tactile experience similar to actual operations in a virtual environment, which enhances the realism and immersion of operations, helps to apply the skills learned in virtual training to actual work, and improves the practicality of training.

[0024] The cloud rendering and distribution system includes: Rendering server: responsible for generating rendering output of each frame of the extended reality scene and performing video encoding on the rendering output. The rendering server sends the encoded rendering output to the content distribution network for caching and distribution; Content delivery network: used to cache and distribute rendered content. The content delivery network receives encoded rendered content from the rendering server and caches it to quickly respond to terminal requests. At the same time, the content delivery network maintains a connection with the edge computing node and pushes the rendered content to the edge node close to the user to optimize transmission latency. Edge computing nodes: deployed close to the student terminals to handle low-latency tasks. Edge computing nodes obtain rendering content from the content distribution network and perform additional processing when necessary to ensure low-latency transmission, and finally transmit the rendering content to the student terminals.

[0025] Benefits of the rendering server: Reduce the performance requirements for terminal devices, ensure the visual quality and rendering efficiency of XR content, and enable students to obtain high-quality immersive experience on low-performance devices; Benefits of content distribution network: Reduce the reliance on long-distance data transmission, effectively reduce transmission delay, enable students to obtain fast access speed in different regions, reduce the bandwidth pressure of rendering servers, and improve the overall stability of the system and user experience; The benefits of edge computing nodes are to reduce the network transmission time and total rendering delay from the cloud to the terminal, so that students can have a smooth and instant interactive experience, improve the network adaptability of the system, and ensure that the student experience is not limited by the network environment.

[0026] The data processing module includes: The data cleaning unit performs preliminary processing on the data collected by the sensor to filter out abnormal values ​​and fill in missing data, providing reliable data input for subsequent data standardization processing; The data conversion unit receives the data processed by the data cleaning unit and standardizes it into a unified format to ensure the consistency and comparability of the data. The output of the data conversion unit provides formatted data input for the data integration unit. The data integration unit receives standardized data from the data conversion unit and integrates the data from different sensors into a unified power grid equipment status data set, providing complete equipment status information for real-time updates of extended reality scenarios.

[0027] Benefits of the data cleaning unit: Removes error data caused by sensor failure or noise interference, ensures that the data set is accurate and reliable, avoids inaccurate data from misleading the training content, avoids interruption of device status or display abnormality in the XR scene due to data loss, and ensures that trainees get a smooth operation experience and complete training information; The data conversion unit enables subsequent processing modules to perform consistent operations on data, eliminating data compatibility issues caused by inconsistent data formats and providing more diverse data support for XR training content; The benefit of the data integration unit is that the system can accurately simulate the operation of the equipment in the XR scenario, ensuring that each module of the system obtains accurate equipment status information, and improving the reliability and real-time nature of training.

[0028] The content management system of the real-time dynamic content update module includes: Training task database, used to store preset fault cases and operation tasks; A dynamic task updating unit, used to automatically generate corresponding training tasks according to the real-time status of power grid equipment; The real-time dynamic content update module realizes real-time status mapping through the digital twin technology module.

[0029] The real-time dynamic content update module of the present invention ensures the real-time update of device status and dynamic generation of task content in the XR training environment through the combination of the content management system and the digital twin technology module. The training task database and dynamic task update unit in the content management system realize rich task reserves and real-time task generation, making the training content more comprehensive and flexible. The digital twin technology module maps the actual device status to the virtual device model through the state mapping step and dynamically displays it in the XR scene, providing trainees with real device status feedback, enhancing the real-time, authenticity and practicality of the training, and enabling trainees to obtain experience and skill improvement similar to real operations in a virtual environment.

[0030] AI-driven intelligent assessment and personalized learning path modules include: Behavior analysis unit, used to analyze the trainees' operational behavior sequences Calculate the behavior score S in real time: , in, is the evaluation value of a single operation, is the behavior score, n is the total number of operations, The specific operations of the i-th step performed by the trainees in the task; When the operation complies with the standard, =1, when not in compliance =0; The behavior analysis unit calculates the behavior score Pass it to the personalized learning path generation unit for dynamic generation of subsequent learning paths; Personalized learning path generation unit generates learning paths based on students' operation accuracy and behavior scores : , in, is the path recommendation function, For personalized learning paths, score the behavior; The personalized learning path generation unit calculates the behavior score Recommend training tasks for students and personalize learning paths Passed to the task management module so that the recommended tasks can be pushed to the student terminal.

[0031] The real-time feedback mechanism of the behavior analysis unit helps students correct their mistakes in time and enhances the effectiveness of learning; Generate personalized learning paths and task recommendations based on students' actual performance, so that students can train at a suitable challenge level, avoid "one size fits all" training methods, and greatly improve learning efficiency; The system can adjust the difficulty of tasks in real time according to the students' learning progress and operating performance, ensuring that the students' learning path is in line with their current operating level and helping them to make steady progress; Quantified behavioral scoring and dynamic recommendation paths enable the system to make personalized recommendations based on data support, ensuring that the training content received by trainees is in line with their actual level, thereby improving the effectiveness and pertinence of training.

[0032] The task management module includes: The task generation unit generates basic training tasks based on the real-time status data of power grid equipment and fault cases preset in the system; The task difficulty adjustment unit dynamically adjusts the difficulty coefficient of the task according to the student's behavior score and the learning path recommended by the task generation unit; A task allocation unit allocates tasks to trainee terminals based on personalized learning paths and adjusted task difficulty; The task feedback and recording unit is used to collect students' feedback and performance data during task execution, and return the data to the AI-driven intelligent assessment and personalized learning path module to further adjust the personalized learning path.

[0033] Each unit of the task management module forms an intelligent and automated task management process by generating tasks, adjusting difficulty, assigning tasks and recording feedback, ensuring that the training content of the power grid maintenance training system can be adjusted in real time according to the level of trainees and equipment status: The task management module realizes the automatic generation, allocation and adjustment of tasks, reduces manual intervention, improves the operating efficiency of the system, and makes training smoother and more efficient; According to the trainees' operation level and personalized learning path, the task management module can dynamically adjust the task difficulty and content, so that each trainee's training task matches his or her actual level, improving the personalization and pertinence of the training; By collecting students' operational feedback and performance data, the system can further optimize personalized learning paths, achieve more accurate task recommendations, and enhance students' gradual progress and learning efficiency during the training process; The task generation unit dynamically provides adaptive tasks based on the actual status of power grid equipment and preset fault cases, helping trainees to practice continuously in real situations and improve their practical operation skills and emergency handling capabilities.

[0034] In summary, the task management module provides efficient and intelligent task management functions for the power grid maintenance training system through its multi-level task management process, making the training more targeted and adaptable, and ensuring that trainees can continuously improve their operating level in a dynamically adjusted task environment.

[0035] The task generation unit receives the device status data from the real-time dynamic content update module and generates tasks When the real-time status data set D of the equipment and the fault case parameters are Create task content: , in, For the generated tasks, Generate a function for the task, It is a collection of real-time status data of the equipment. Fault case parameters preset for the system; The task difficulty adjustment unit calculates the final difficulty of the task based on the student's behavior score S : , in, is the basic difficulty of the task. is the adjustment factor, Rate students' behavior; If S decreases, the task difficulty d will automatically decrease, making the task more suitable for the current level of the students; If S increases, the task difficulty d increases to increase the challenge.

[0036] The task generation unit and task difficulty adjustment unit in the present invention realize automatic generation and difficulty adjustment of tasks through dynamic analysis of real-time device status data and student behavior scores: The task generation unit dynamically generates tasks based on equipment status and fault cases, making the task content more in line with actual needs, enhancing trainees' emergency response capabilities and the practicality of training; The task difficulty adjustment unit dynamically adjusts the task difficulty based on the trainees' performance, ensuring that the task difficulty is always suitable for the trainees' level, so that the trainees can make continuous progress at an appropriate difficulty level, thereby improving learning effects and training efficiency; Dynamically adjusted task difficulty ensures that skilled trainees can obtain more challenging tasks, avoiding simple repetition of task content, and enabling trainees to achieve continuous progress and strengthen their operational skills in moderate challenges; The task generation and difficulty adjustment unit realizes automatic generation and dynamic adjustment of tasks, reduces the complexity of manual operations, significantly improves the system's automation level and operating efficiency, and enables trainees to obtain efficient and personalized training experience.

[0037] The virtual reality and physical simulation fusion module includes: The force feedback unit is used to simulate the real force and resistance feedback that trainees receive during operation, so that the power grid equipment in virtual operation has real mechanical characteristics; A tactile feedback unit is used to provide tactile feedback so that trainees can feel the surface characteristics and vibration feedback of power grid equipment during virtual operation; High-precision physics engine unit, used to simulate the physical characteristics of power grid equipment, so that the equipment in the virtual environment exhibits realistic physical behaviors; The environmental condition simulation unit is used to simulate different environmental conditions, so that trainees can operate power grid equipment in a variety of scenarios.

[0038] The virtual reality and physical simulation fusion module provides trainees with a close-to-real operating experience through the synergy of force feedback, tactile feedback, high-precision physical engine and environmental condition simulation, making the training effect more significant: Feedback and tactile feedback simulate the physical touch and texture of real equipment, allowing students to experience the real feeling of operating the equipment and enhance the immersiveness of learning; The high-precision physics engine's simulation of physical characteristics ensures that trainees acquire practical skills during training, which helps them apply their learning outcomes to real-world operating environments. The environmental condition simulation unit enables trainees to conduct operation training in complex environments, enhance their operational stability and emergency response capabilities under extreme or harsh conditions, and ensure that trainees can adapt to a variety of working environments; The physics engine provides precise physical parameters for force feedback and tactile feedback, ensuring that the operation feedback is consistent with the actual physical characteristics of the device, improving students' operation accuracy and skill mastery.

[0039] Through the above improvements, the virtual reality and physical simulation fusion module provides a high-quality operation experience for the power grid maintenance training system, allowing trainees to accumulate real operation skills in a virtual environment and meet the power industry's demand for high-standard skills training.

[0040] The force feedback unit simulates the resistance, pulling force and pushing force during operation through force feedback gloves or controllers. Adjusting feedback force , force feedback relationship formula: , in, is the feedback force, is the feedback force coefficient, For operational intensity; The high-precision physics engine unit calculates the friction, reaction force and gravity of the device based on the physics engine, so that the virtual device has real physical feedback during operation. In the process of simulating cable connection, the friction of the device The calculation formula is: , in, is the friction coefficient, The weight of the equipment.

[0041] The combination of force feedback unit and high-precision physics engine unit provides highly immersive operation feedback for power grid maintenance training system: Through force feedback and high-precision physics engines, students can experience real operating forces, friction and reaction forces in a virtual environment, increasing the sense of immersion in learning; The feedback force simulation of the force feedback unit and the friction calculation of the physics engine help students better master operating skills in a virtual environment and ensure that the learning outcomes can be transferred to actual work applications. By simulating physical properties such as friction and gravity through the physics engine, students can understand the mechanical properties of the equipment during operation, which helps students to flexibly handle various operating situations in practice and improve their practical skills. The physical parameters provided by the physics engine make the force feedback highly consistent with the actual physical characteristics of the device, ensuring the accuracy and stability of the feedback force, further improving the precision of the operation and the students' control over the operation process.

[0042] A power grid maintenance training method based on extended reality includes the following steps: Step 1: The cloud rendering and distribution system generates high-quality rendering output of the power grid maintenance scene, and transmits the content to the trainees' terminal devices through edge computing nodes and content distribution networks. During this process, the real-time dynamic content update module embeds the real-time data of the device into the extended reality scene to provide trainees with a realistic simulation environment. Step 2: The data processing module collects data from the IoT sensors of power grid equipment, cleans, converts and integrates them, and generates a real-time status data set of the equipment; Step 3: Through the real-time dynamic content update module, the system uses digital twin technology to map the real-time status of power grid equipment to the virtual equipment model. At the same time, the content management system retrieves fault cases and operation tasks from the preset database, or generates corresponding training tasks according to the status of the equipment; Step 4: The AI-driven intelligent assessment and personalized learning path module analyzes the students' operation behavior and operation accuracy, conducts real-time assessment of the students, generates personalized learning paths, and dynamically adjusts the difficulty of tasks based on the students' performance, recommending adaptive tasks; Step 5: The task management module creates, assigns and adjusts tasks according to the student's personalized learning path, dynamically generates adaptive tasks based on the real-time status of the device and the student's performance, and pushes the tasks to the student's terminal device; Step 6: During the training process, the virtual reality and physical simulation fusion module provides trainees with real tactile feedback through force feedback devices and physics engines. At the same time, tactile feedback and environmental condition simulation enable trainees to experience the operation of power grid equipment in various scenarios. Step 7: During the task execution, the task management module collects students’ feedback and operation data, and transmits the data to the AI-driven intelligent evaluation module. Based on the students’ real-time operation performance, the system dynamically adjusts the learning path and task difficulty; Step 8: The virtual reality and physical simulation fusion module provides force feedback according to the trainee's operation intensity during the trainee's operation, so that the virtual operation has real physical feedback; Step 9: Based on the students' behavioral scores and feedback, the system continuously optimizes the personalized learning path and task difficulty, so that students can gradually adapt to more challenging tasks and achieve the goal of personalized training.

[0043] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A power grid maintenance training system based on extended reality, characterized in that: include: Cloud rendering and distribution system: used to render the extended reality scene, generate real-time rendering output, and transmit it to the terminal device through the edge computing node and the content distribution network. The cloud rendering and distribution system is connected to the real-time dynamic content update module and the data processing module, and is used to receive the real-time status data of the device and the student's operation behavior data, and is responsible for embedding the data into the rendering scene, so as to update the virtual device status and student interaction status in the extended reality scene in real time, and ensure that the scene seen by the student on the terminal device reflects the actual status in real time and accurately; Data processing module: used to collect data from sensors in the Internet of Things of power grid equipment, clean, convert and filter the data, and generate a real-time status data set of power grid equipment. In addition, the data processing module receives and processes the trainees' operation behavior data to provide input data for the AI-driven intelligent evaluation and personalized learning path modules; Real-time dynamic content update module: It includes a digital twin technology module, which is used to map the real-time status data set of power grid equipment generated by the data processing module to the virtual equipment model in the extended reality scene, and update the training tasks and fault cases through the content management system. The real-time dynamic content update module is connected to the cloud rendering and distribution system to receive the real-time data of the equipment and embed it into the extended reality scene, so as to realize the real-time update of the equipment status in the virtual scene. In addition, the real-time dynamic content update module ensures the synchronous transmission of the trainee operation and environmental data, so that the content of the scene is consistent with the actual situation; AI-driven intelligent assessment and personalized learning path module: Analyze the trainees' operation behavior data and operation accuracy, generate personalized learning paths, and adjust the task difficulty. The AI-driven intelligent assessment and personalized learning path module obtains the operation behavior data in the data processing module in real time, and recommends corresponding personalized training tasks through the task management module; Task management module: responsible for managing the creation, allocation and difficulty adjustment of training tasks. By connecting with the real-time dynamic content update module and the AI-driven intelligent assessment and personalized learning path module, it generates adaptive tasks based on the status of power grid equipment and trainee performance and transmits them to the trainee's terminal device; Virtual reality and physical simulation fusion module: The weight, friction and operating resistance of the device are simulated through force feedback devices and physical engines. The virtual reality and physical simulation fusion module is connected to the cloud rendering and distribution system to receive real-time rendering content and provide simulation data of force feedback devices.

2. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The cloud rendering and distribution system includes: Rendering server: responsible for generating rendering output of each frame of the extended reality scene and performing video encoding on the rendering output. The rendering server sends the encoded rendering output to the content distribution network for caching and distribution; Content distribution network: used to cache and distribute rendering content. The content distribution network receives the encoded rendering content from the rendering server and caches it to quickly respond to terminal requests. At the same time, the content distribution network maintains a connection with the edge computing node and pushes the rendering content to the edge computing node close to the user to optimize transmission latency. Edge computing nodes: deployed close to the student terminals to handle low-latency tasks. Edge computing nodes obtain rendering content from the content distribution network and perform additional processing when necessary to ensure low-latency transmission, and finally transmit the rendering content to the student terminals.

3. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The data processing module comprises: The data cleaning unit performs preliminary processing on the data collected by the sensor to filter out abnormal values ​​and fill in missing data, providing reliable data input for subsequent data standardization processing; The data conversion unit receives the data processed by the data cleaning unit and standardizes it into a unified format to ensure the consistency and comparability of the data. The output of the data conversion unit provides formatted data input for the data integration unit. The data integration unit receives standardized data from the data conversion unit and integrates the data from different sensors into a unified power grid equipment status data set, providing complete equipment status information for real-time updates of extended reality scenarios.

4. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The content management system of the real-time dynamic content update module includes: Training task database, used to store preset fault cases and operation tasks; A dynamic task updating unit, used to automatically generate corresponding training tasks according to the real-time status of power grid equipment; The real-time dynamic content update module realizes real-time status mapping through the digital twin technology module.

5. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The AI-driven intelligent assessment and personalized learning path module includes: Behavior analysis unit, used to analyze the trainees' operational behavior sequences Calculate the behavior score S in real time: ， in, is the evaluation value of a single operation, is the behavior score, n is the total number of operations, The specific operations of the i-th step performed by the trainees in the task; When the operation complies with the standard, =1, when not in compliance =0; The behavior analysis unit calculates the behavior score Pass it to the personalized learning path generation unit for dynamic generation of subsequent learning paths; Personalized learning path generation unit generates learning paths based on students' operation accuracy and behavior scores : , in, is the path recommendation function, For personalized learning paths, score the behavior; The personalized learning path generation unit calculates the behavior score Recommend training tasks for students and personalize learning paths Passed to the task management module so that the recommended tasks can be pushed to the student terminal.

6. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The task management module includes: The task generation unit generates basic training tasks based on the real-time status data of power grid equipment and fault cases preset in the system; The task difficulty adjustment unit dynamically adjusts the difficulty coefficient of the task according to the student's behavior score and the learning path recommended by the task generation unit; A task allocation unit allocates tasks to trainee terminals based on personalized learning paths and adjusted task difficulty; The task feedback and recording unit is used to collect students' feedback and performance data during task execution, and return the data to the AI-driven intelligent assessment and personalized learning path module to further adjust the personalized learning path.

7. The power grid maintenance training system based on extended reality according to claim 6, characterized in that: The task generation unit generates a task based on the device status data received by the real-time dynamic content update module. When the real-time status data set D of the equipment and the fault case parameters are Create task content: ， in, For the generated tasks, Generate a function for the task, It is a collection of real-time status data of the equipment. Fault case parameters preset for the system; The task difficulty adjustment unit calculates the final difficulty of the task based on the student's behavior score S : ， in, is the basic difficulty of the task. is the adjustment factor, Rate students' behavior; If S decreases, the task difficulty d will automatically decrease, making the task more suitable for the current level of the students; If S increases, the task difficulty d increases to increase the challenge.

8. The power grid maintenance training system based on extended reality according to claim 1, characterized in that: The virtual reality and physical simulation fusion module includes: The force feedback unit is used to simulate the real force and resistance feedback that trainees receive during operation, so that the power grid equipment in virtual operation has real mechanical characteristics; A tactile feedback unit is used to provide tactile feedback so that trainees can feel the surface characteristics and vibration feedback of power grid equipment during virtual operation; High-precision physics engine unit, used to simulate the physical characteristics of power grid equipment, so that the equipment in the virtual environment exhibits realistic physical behaviors; The environmental condition simulation unit is used to simulate different environmental conditions, so that trainees can operate power grid equipment in a variety of scenarios.

9. The power grid maintenance training system based on extended reality according to claim 8, characterized in that: The force feedback unit simulates the resistance, pulling force and pushing force during operation through the force feedback glove or controller. Adjusting feedback force , force feedback relationship formula: ， in, is the feedback force, is the feedback force coefficient, For operational intensity; The high-precision physical engine unit calculates the friction, reaction force and gravity of the device based on the physical engine, so that the virtual device has real physical feedback during operation. In the process of simulating cable connection, the friction of the device The calculation formula is: ， in, is the friction coefficient, The weight of the equipment.

10. A power grid maintenance training method based on extended reality, according to a power grid maintenance training system based on extended reality according to any one of claims 1 to 9, characterized in that: The steps include: Step 1: The cloud rendering and distribution system generates high-quality rendering output of the power grid maintenance scene, and transmits the content to the trainees' terminal devices through edge computing nodes and content distribution networks. During this process, the real-time dynamic content update module embeds the real-time data of the device into the extended reality scene to provide trainees with a realistic simulation environment. Step 2: The data processing module collects data from the IoT sensors of power grid equipment, cleans, converts and integrates them, and generates a real-time status data set of the equipment; Step 3: Through the real-time dynamic content update module, the system uses digital twin technology to map the real-time status of power grid equipment to the virtual equipment model. At the same time, the content management system retrieves fault cases and operation tasks from the preset database, or generates corresponding training tasks according to the status of the equipment; Step 4: The AI-driven intelligent assessment and personalized learning path module analyzes the students' operation behavior and operation accuracy, conducts real-time assessment of the students, generates personalized learning paths, and dynamically adjusts the difficulty of tasks based on the students' performance, recommending adaptive tasks; Step 5: The task management module creates, assigns and adjusts tasks according to the student's personalized learning path, dynamically generates adaptive tasks based on the real-time status of the device and the student's performance, and pushes the tasks to the student's terminal device; Step 6: During the training process, the virtual reality and physical simulation fusion module provides trainees with real tactile feedback through force feedback devices and physics engines. At the same time, tactile feedback and environmental condition simulation enable trainees to experience the operation of power grid equipment in various scenarios. Step 7: During the task execution, the task management module collects students’ feedback and operation data, and transmits the data to the AI-driven intelligent evaluation module. Based on the students’ real-time operation performance, the system dynamically adjusts the learning path and task difficulty; Step 8: The virtual reality and physical simulation fusion module provides force feedback according to the trainee's operation intensity during the trainee's operation, so that the virtual operation has real physical feedback; Step 9: Based on the students' behavioral scores and feedback, the system continuously optimizes the personalized learning path and task difficulty, so that students can gradually adapt to more challenging tasks and achieve the goal of personalized training.

Images