500kV standardized switching operation method based on VR virtual reality technology

Abstract

The invention discloses a 500kV standardized switching operation method based on a VR virtual reality technology, and aims to solve the problems of unreal environment, incomplete evaluation and low efficiency in the switching operation training of an existing 500kV transformer substation. The method comprises the following steps: constructing a full-immersive virtual reality scene, collecting actual environment data of a transformer substation, and constructing and optimizing a high-precision model; designing multi-modal interaction of the standardized switching operation process, wherein the multi-modal interaction comprises visual guidance, a multi-modal interaction interface and an intelligent sensing function; dynamic action scoring and intelligent behavior evaluation are realized, actions are captured in real time by using high-precision action capture equipment, an instant score is generated by comparing with a standard, and dynamic feedback is provided; and real-time optimization and learning migration of the virtual character model are realized, user operation habits are captured, virtual character actions are optimized, and multi-scene migration and adaptation are realized. The method is suitable for VR-based 500kV standardized switching operation work, and the training authenticity and efficiency are improved.

Description

Technical Field

[0001] It relates to the field of large model construction technology, and specifically relates to 500 kV standardized switching operations based on VR virtual reality technology. Background Art

[0002] In the power industry, the switching operation of a 500 kV substation is a crucial task, and its accuracy and safety are directly related to the stable operation of the power grid. With the continuous development of power technology and the continuous expansion of the power grid scale, higher requirements are put forward for the skill level and training efficiency of switching operation personnel.

[0003] Currently, there are mainly two training methods for 500 kV substation switching operations: one is traditional simulator training, and the other is learning from paper textbooks. Simulator training simulates a real operation environment, allowing trainees to practice operations in a work scenario close to the actual situation. However, although this training method can provide certain operation experience, due to the differences between the simulated environment and the actual environment, it is difficult to fully reproduce all the details and situations in real operations. On the other hand, learning from paper textbooks focuses on the imparting of theoretical knowledge but lacks practical operation links. It is difficult for trainees to combine theoretical knowledge with practical operations, resulting in poor training effects.

[0004] In addition, in recent years, with the continuous development of virtual reality (VR) technology, some power enterprises have also begun to try to apply VR technology to the training of switching operations. These attempts mainly focus on constructing a virtual operation environment through VR technology, allowing trainees to perform simulated operations in the virtual environment. However, most of these applications are still in the initial exploration stage and there are still many deficiencies. For example, the realism of the virtual environment is insufficient, the operation feedback is not real enough, and there is a lack of intelligent evaluation and feedback mechanisms.

[0005] To sum up, the existing technologies have the following main problems in the training of 500 kV substation switching operations:

[0006] The operation environment is not real: Both traditional simulator training and learning from paper textbooks are difficult to fully reproduce all the details and situations in real operations, resulting in trainees being difficult to adapt in actual operations.

[0007] The skill evaluation is not comprehensive: The existing training methods lack intelligent evaluation and feedback mechanisms, and cannot comprehensively and accurately evaluate the operations of trainees, nor can they provide targeted improvement suggestions.

[0008] The training efficiency is low: Due to problems such as the unrealistic operation environment and the incomplete skill evaluation, trainees need to spend more time and energy to master the switching operation skills, resulting in low training efficiency.

[0009] These problems are exactly the technical challenges that this solution aims to address. Through the 500kV standardized switching operation method based on VR virtual reality technology, a panoramic training and assessment system can be constructed to achieve comprehensive simulation, real-time assessment, and continuous optimization of switching operations, thereby improving training efficiency and quality and ensuring the safe and stable operation of the power grid. Summary of the Invention

[0010] To solve the technical deficiencies in the existing 500kV substation switching operation training, such as unrealistic operation environments, incomplete skill assessments, and low training efficiency, the technical solution provided by the present invention is as follows:

[0011] A method for 500kV standardized switching operations based on VR virtual reality technology, including:

[0012] Steps for constructing a fully immersive virtual reality scene, including:

[0013] Collect actual environmental data of the 500kV substation, including equipment layout, line orientation, ground texture, and equipment material characteristics;

[0014] According to the collected data, construct a high-precision scene model in the virtual reality environment, and apply texture mapping and material optimization to simulate a real substation environment;

[0015] Design dynamic environment settings;

[0016] Steps for designing multi-modal interaction in the standardized switching operation process, including:

[0017] In the virtual reality scene, design a visual operation process guide to display the current state and operation path of the equipment through augmented reality technology;

[0018] Provide multi-modal interaction interfaces;

[0019] Introduce an intelligent operation environment perception function to detect the user's operation range and equipment interaction status in real time and record the user's action trajectory;

[0020] Steps for implementing dynamic action scoring and intelligent behavior assessment, including:

[0021] Use high-precision motion capture equipment to capture the user's switching operation actions in real time;

[0022] Compare the captured actions frame by frame with the standard actions, calculate the trajectory error and time difference, and generate an instant score;

[0023] According to the scoring results, provide dynamic feedback, including error prompts, optimization suggestions, and task difficulty adjustment;

[0024] Steps for real-time optimization and learning transfer of virtual character models, including:

[0025] Capture the user's operation habits, dynamically optimize the bone binding and action curves of the virtual character, and make the actions of the virtual character consistent with the user's actual operations;

[0026] Implement multi-scenario migration and adaptation.

[0027] Furthermore, a preferred implementation is provided, which also includes

[0028] Steps for data-driven personalized training and operation improvement, including:

[0029] Collect and analyze the user's operation data, including key metrics such as completion time and error frequency;

[0030] Generate optimization suggestions based on the data analysis results;

[0031] Provide a team collaboration evaluation function, record the team collaboration efficiency and the rationality of task division, and optimize the role allocation strategy.

[0032] Furthermore, a preferred implementation is provided, and the instant scoring is achieved through the following formula:

[0033]

[0034] Where S: comprehensive score, α, β, δ: weight parameters used to balance the proportions of accuracy, correction ability, and fluency in the scoring, n: total number of tasks, ω i : single-task accuracy rate, indicating the matching degree between the operation and the standard action, c i : task complexity coefficient, used to highlight the importance of high-difficulty tasks, t i : single-task completion time, ε: time compensation coefficient to prevent unreasonable magnification of the score when the time is close to zero, e: total number of errors accumulated by the user in the entire task, γ: exponential coefficient of correction ability, reflecting the learning curve of the user's error reduction, e -γ·e : non-linear function used to reward users with strong error correction ability, F l : action fluency score, calculated based on the smoothness of the joint movement curve and the continuity of the action, δ: weight coefficient of action fluency, used to evaluate the overall operation quality of the user.

[0035] Furthermore, a preferred implementation is provided, and the optimization of the action curve is achieved through the following interpolation algorithm:

[0036] P(t) = (1 - λ(t))·P p rev(t) + λ(t)·P n ext(t) + ω(t)·f(t)

[0037] P(t): The interpolation position of the optimized motion curve, representing the target position at time point t, P p rev(t), P n ext(t): Respectively represent the key frame positions before and after time point t, generating the motion trajectory through smooth transitions between key frames. λ(t): The time weighting factor, which controls the interpolation ratio over time. The value range of λ(t) is [0, 1], representing the weighting coefficient from the previous frame to the next frame. The specific calculation formula is as follows:

[0038]

[0039] -κ: The rate factor, controlling the speed of the interpolation transition, t 0 : The reference time of the key frame, identifying the occurrence time of the key action. w(t): The smoothness optimization weight, which is gradually adjusted over time to increase or decrease the smoothness of the action. Its value can be dynamically adjusted according to training feedback, such as increasing or decreasing the adjustment of the action based on the user's proficiency. ω(t) = 1 / (1 + α·(t - t 0 ) 2 )

[0040] α: The smoothness factor, controlling the intensity of action smoothness. A larger α value will enhance the effect of action smoothness and reduce unnecessary rapid action switches. f(t): The smoothness enhancement function, used to further optimize the naturalness and smoothness of the action. This function is weighted based on the user's hand motion trajectory, joint rotation smoothness, and overall operation smoothness:

[0041] f(t) = 1 / (1 + β·cos(2π·t / T))

[0042] β: The smoothness adjustment factor, used to fine-tune the action smoothness. A higher β value will result in higher smoothness feedback. T: The action period, representing the time length of the action during the entire operation process.

[0043] Furthermore, a preferred implementation is provided. In the step of constructing the fully immersive virtual reality scene, it further includes:

[0044] Model the substation equipment using 3D modeling software according to the size, shape, and material of the actual device;

[0045] Design interaction hotspots in the virtual environment to provide device information to users.

[0046] Furthermore, a preferred implementation is provided. In the step of designing the multi-modal interaction of the standardized switching operation process, it further includes:

[0047] Provide a voice command library, allowing users to control device operations in a virtual environment through preset voice commands;

[0048] Provide a gesture recognition algorithm to recognize users' hand movements and convert them into operation instructions in the virtual environment.

[0049] A device for 500kV standardized switching operations based on VR virtual reality technology, including:

[0050] A module for constructing a fully immersive virtual reality scene, including:

[0051] Collect the actual environmental data of a 500kV substation, including equipment layout, line routes, ground textures, and equipment material characteristics;

[0052] Based on the collected data, construct a high-precision scene model in the virtual reality environment, and apply texture mapping and material optimization to simulate a real substation environment;

[0053] Design a dynamic environment setting;

[0054] Design a multi-modal interaction module for the standardized switching operation process, including:

[0055] In the virtual reality scene, design a visual operation process guide to display the current state and operation path of the equipment through augmented reality technology;

[0056] Provide a multi-modal interaction interface;

[0057] Introduce an intelligent perception function for the operation environment to detect the user's operation range and equipment interaction status in real time, and record the user's action trajectory;

[0058] A module for implementing dynamic action scoring and intelligent behavior evaluation, including:

[0059] Use high-precision motion capture equipment to capture the user's switching operation actions in real time;

[0060] Compare the captured actions frame by frame with the standard actions, calculate the trajectory error and time difference, and generate an instant score;

[0061] According to the scoring results, provide dynamic feedback, including error prompts, optimization suggestions, and task difficulty adjustment;

[0062] A module for real-time optimization and learning transfer of virtual character models, including:

[0063] Capture the user's operation habits, dynamically optimize the bone binding and action curves of the virtual character, and make the actions of the virtual character consistent with the user's actual operations;

[0064] Implement multi-scene migration and adaptation.

[0065] Compared with the prior art, the beneficial effects of the technical solution provided by the present invention are as follows:

[0066] First of all, through the construction of a fully immersive scenario based on VR technology, the problem of the unrealistic operation environment in traditional training methods is solved. This method provides a virtual operation environment highly similar to the actual working environment for trainees through technical means such as high-precision scene modeling, dynamic environment setting, and physical feedback and immersion enhancement. Compared with the existing simulator training and paper textbook learning, this method greatly improves the authenticity and immersion of training, enabling trainees to train and assess under conditions closer to the actual operation environment.

[0067] Secondly, the multi-modal interaction design of the standardized switching operation process solves the problem of incomplete skill assessment in traditional training methods. This solution realizes the comprehensive monitoring and real-time assessment of the operation process of trainees through technical means such as visual operation process guidance, multi-modal interaction interfaces, and intelligent perception of the operation environment. This not only includes the accuracy assessment of trainees' equipment operations, but also covers multiple aspects such as the compliance of the operation process, the fluency of the operation, and the trainees' response capabilities in emergency situations. This all-round assessment method can more accurately reflect the actual operation skill level of trainees and provides strong support for subsequent personalized training and optimization.

[0068] In addition, the use of dynamic action scoring and intelligent behavior assessment further improves the accuracy and pertinence of skill assessment. This solution accurately captures the operation actions of trainees through high-precision and compares them frame by frame with the standard actions, thereby realizing the precise scoring of trainees' operation actions. At the same time, the system can also adjust the task difficulty in real time according to the operation performance of trainees and provide optimization suggestions. This intelligent feedback mechanism can help trainees discover and correct their mistakes faster, thereby improving the training effect. Compared with the existing training methods based on simulators or paper textbooks, this method is more flexible and efficient and can better adapt to the learning needs and ability levels of different trainees.

[0069] Finally, through technical means such as the real-time optimization and learning transfer of virtual character models and data-driven personalized training and operation improvement, this solution also realizes the continuous optimization and personalized customization of the training process for trainees. This can not only improve the learning efficiency and skill level of trainees, but also cultivate more switching operation personnel with high quality and high skills for the power industry, providing a more powerful guarantee for the safe and stable operation of the power grid.

[0070] It is suitable for application in the work of 500kV standardized switching operations based on VR virtual reality technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1Flow chart of a method for 500kV standardized switching operation based on VR virtual reality technology. Detailed implementation manners

[0072] To make the advantages and beneficial effects of the technical solution provided by the present invention more clearly reflected, the technical solution provided by the present invention will be further described in detail below in conjunction with the accompanying drawings. Specifically:

[0073] Embodiment 1. This embodiment provides a method for 500kV standardized switching operation based on VR virtual reality technology, including:

[0074] Steps for constructing a fully immersive virtual reality scene, including:

[0075] Collect actual environmental data of a 500kV substation, including equipment layout, line orientation, ground texture, and equipment material characteristics;

[0076] According to the collected data, construct a high-precision scene model in the virtual reality environment, and apply texture mapping and material optimization to simulate the real substation environment;

[0077] Design dynamic environment settings;

[0078] Steps for designing multi-modal interaction of the standardized switching operation process, including:

[0079] In the virtual reality scene, design a visual operation process guide, and display the current state and operation path of the equipment through augmented reality technology;

[0080] Provide multi-modal interaction interfaces;

[0081] Introduce an intelligent perception function for the operation environment, real-time detect the user's operation range and equipment interaction status, and record the user's action trajectory;

[0082] Steps for realizing dynamic action scoring and intelligent behavior evaluation, including:

[0083] Use high-precision motion capture equipment to capture the user's switching operation actions in real time;

[0084] Compare the captured actions frame by frame with the standard actions, calculate the trajectory error and time difference, and generate an instant score;

[0085] According to the scoring results, provide dynamic feedback, including error prompts, optimization suggestions, and task difficulty adjustment;

[0086] Steps for real-time optimization and learning transfer of the virtual character model, including:

[0087] Capture the user's operation habits, dynamically optimize the bone binding and action curves of the virtual character, and make the actions of the virtual character consistent with the user's actual operations;

[0088] Implement multi-scenario migration and adaptation.

[0089] Specifically:

[0090] I. Construction of a fully immersive scenario based on VR technology

[0091] Brief steps:

[0092] First, collect the actual environmental data of a 500kV substation, then construct a high-precision scenario model, and design dynamic environmental settings to enhance the immersion.

[0093] Detailed steps:

[0094] Data collection:

[0095] Use a Lidar device to scan the actual environment of a 500kV substation and collect data on equipment layout, line routing, and ground texture.

[0096] Use a high-resolution camera to capture the material characteristics of the equipment surface for subsequent texture mapping.

[0097] Model construction and optimization:

[0098] Import the collected point cloud data into 3D modeling software (such as Blender) to generate a basic geometric model.

[0099] Apply texture mapping and material optimization to refine key details such as equipment edges, nameplates, and buttons to ensure the authenticity and accuracy of the model.

[0100] Dynamic environmental settings:

[0101] Design a real-time adjustable weather system, including scene changes such as sunny, rainy, and foggy days, to simulate the operating environment under different weather conditions.

[0102] Simulate day-night switching to test the operating ability of trainees under low-light conditions and improve the comprehensiveness and practicality of training.

[0103] II. Multimodal interaction design for the standardized switching operation process

[0104] Brief steps:

[0105] In the constructed fully immersive scenario, design visual operation process guides, multimodal interaction interfaces, and intelligent operation environment perception functions to achieve interaction with trainees.

[0106] Detailed steps:

[0107] Visual operation process guide:

[0108] Utilize augmented reality (AR) technology to overlay and display the current status of the device in real time (such as closed / open), highlight the device to be operated, and display the function description of the device on the side of the screen.

[0109] Mark the task execution path with virtual arrows or flashing indicator lights, and update the navigation prompts in real time according to the operation progress of the trainee.

[0110] Automatically display the operation requirements for the next step after each operation is completed, and incorrect operations will trigger specific corrective prompts.

[0111] Multi-modal interaction interface:

[0112] Built-in standard operation instruction set, allowing trainees to issue commands by voice. The system recognizes the voice in real time and executes the corresponding operations.

[0113] Use LeapMotion to capture hand movements, verify whether the movement path conforms to the standard process, and automatically generate corrective suggestions and display animation examples for incorrect movements.

[0114] Design intelligent handle operations. The handle buttons are partitioned to define quick operations, such as switching scenes, pausing training, etc., and support the action recording function for trainees to review and optimize training.

[0115] Intelligent perception of the operation environment:

[0116] Use sensors and motion capture devices to detect the operation range of trainees and the device interaction status, and record the motion trajectories of trainees.

[0117] Issue a warning when it does not conform to the standard action. Incorrect operations are prompted by sound and visual means (such as red flashing or alarm sound), and the task is automatically paused to provide an operation guide.

[0118] III. Dynamic action scoring and intelligent behavior evaluation

[0119] Brief steps:

[0120] Capture the actions of trainees with high precision, compare them frame by frame with the standard actions, calculate the trajectory error and time difference, generate an instant score, and provide dynamic feedback and adjustment according to the score result.

[0121] Detailed steps:

[0122] Real-time action capture and standardization comparison:

[0123] Use OptiTrack equipment to capture the starting point, movement path, and ending point of each action, and store the action data as a time series.

[0124] Compare the trainee's actions with the standard actions frame by frame, calculate the trajectory error and time difference, and generate an instant score.

[0125] Comprehensive scoring model:

[0126] Consider the action complexity and completion time, dynamically adjust the weights, and formulate a scoring formula.

[0127] Calculate the comprehensive score according to the scoring formula, including the main scoring parts (accuracy, correction ability, and fluency), the error correction part, and the action fluency part.

[0128] Evaluate and classify the trainees according to the scoring results (excellent, good, need improvement).

[0129] Dynamic feedback and adjustment:

[0130] Instantly display optimization suggestions when errors occur, such as adjusting the standing position or gestures.

[0131] Recommend corresponding training modules according to the trainees' weak links, and dynamically adjust the task difficulty and scoring according to the trainees' operation performance, error correction ability, and task execution fluency.

[0132] IV. Real-time optimization and learning transfer of virtual character models

[0133] Brief steps:

[0134] Capture the trainees' operation habits, dynamically optimize the bone binding and action curves of the virtual character, and achieve multi-scenario migration and adaptation.

[0135] Detailed steps:

[0136] Bone optimization driven by actions:

[0137] Capture the trainees' operation habits, dynamically optimize the bone binding of the virtual character, and ensure that the actions of the virtual character are consistent with the trainees' actual operations.

[0138] Optimize the interpolation algorithm of joint movement, reduce jitter phenomenon, and improve the fluency and naturalness of actions.

[0139] Multi-scenario migration and adaptation:

[0140] Reuse training data in different scenarios, adjust the action amplitude through an adaptive algorithm, and ensure that the action performance of the virtual character is consistent in different scenarios.

[0141] Expand new task modules according to existing tasks, such as emergency operation training, to improve the flexibility and practicality of training.

[0142] V. Data-driven personalized training and operation improvement

[0143] Brief steps:

[0144] Collect and analyze the operation data of trainees, generate optimization suggestions, formulate personalized training programs, and conduct team collaboration evaluations.

[0145] Detailed steps:

[0146] Data analysis and optimization suggestions:

[0147] Visualize the operation history of trainees, including key indicators such as completion time and error frequency.

[0148] Automatically generate targeted intensive training plans based on the data and provide personalized training programs for trainees.

[0149] Team collaboration evaluation:

[0150] Multiple people participate in tasks online, and the collaboration efficiency and rationality of task division are recorded in real time.

[0151] Provide team scores and individual contribution rate reports, optimize the role assignment strategy, and improve team collaboration efficiency.

[0152] Through the implementation of the above steps, this solution can achieve a comprehensive simulation, real-time assessment, and continuous optimization of 500kV switching operations, providing a new, efficient, safe, and intelligent training model for the power industry.

[0153] Embodiment 2. This embodiment further describes in detail the above-provided technical solution. Specifically:

[0154] It includes:

[0155] 1. Construction of a fully immersive scenario based on VR technology

[0156] The goal of this step is to use high-precision VR modeling technology to construct a realistic and immersive substation operation environment and provide trainees with a fully simulated operation training platform. The steps include data collection, modeling and optimization, and dynamic environment setting.

[0157] 1.1 High-precision scenario modeling

[0158] Data collection:

[0159] Lidar device: Use lidar to scan the actual environment of a 500kV substation to obtain data on equipment layout, line orientation, and ground texture.

[0160] High-resolution camera: Capture the material characteristics of the equipment surface to generate high-quality texture maps for enhancing the visual effect of the equipment surface in modeling.

[0161] Model construction and optimization:

[0162] 3D Modeling: Import the collected point cloud data into 3D modeling software (such as Blender) to generate a basic geometric model. Ensure that the model has accurate proportions and positions at this time.

[0163] Texture Mapping and Refinement: Apply texture mapping to the model and refine key details such as the edges, nameplates, and buttons of the equipment to ensure visual accuracy of each part of the model.

[0164] Dynamic Environment Settings:

[0165] Weather System Design: Implement real-time adjustable weather conditions, including sunny, rainy, and hazy days, etc., through script control to simulate the operation ability of trainees under different climate conditions.

[0166] Day-Night Switching: Design the environmental change of day-night alternation to test the operation ability of trainees under low light conditions, especially their reactions in insufficient light.

[0167] 1.2 Scene Multi-Task Adaptation

[0168] Task Classification and Modular Design:

[0169] Divide the operation tasks into three categories: daily operations, troubleshooting, and emergency handling. Create independent modules for each category. For example, troubleshooting includes sub-tasks such as circuit breaker inspection and equipment status assessment.

[0170] Task Logic Arrangement:

[0171] Write scripts using Python or C language to define the execution order and triggering conditions of tasks to ensure a reasonable execution order of tasks. For example, the opening operation is only performed after the equipment status is confirmed.

[0172] Abnormal Scene Simulation:

[0173] Design the triggering conditions of equipment failures (such as high-temperature alarms, abnormal currents, etc.) and simulate equipment feedback (such as abnormal switch switching).

[0174] 1.3 Physical Feedback and Immersion Enhancement

[0175] Haptic Feedback:

[0176] Use a force feedback glove to simulate the mechanical resistance of equipment operation and combine with a vibration module to truly restore the mechanical feeling of opening and closing.

[0177] Sound Effect Simulation:

[0178] Add sound effects such as current cut-off sounds and equipment operation sounds, and at the same time restore the background noise of the substation (such as current buzzing, wind sounds, etc.).

[0179] Rendering Optimization:

[0180] Optimize the lighting and shadow effects using real-time ray tracing technology to ensure the authenticity of the visual effects.

[0181] 2. Multimodal Interaction Design for Standardizing the Switching Operation Process

[0182] Through multimodal interaction design, trainees can learn the switching operation process more intuitively and obtain intelligent guidance and real-time feedback.

[0183] 2.1 Visual Operation Process Guide

[0184] Dynamic Equipment Labeling:

[0185] Use augmented reality (AR) technology to display the current state of the equipment (such as closed or open) in real time and show its function description beside the equipment.

[0186] Highlight the equipment to be operated to guide trainees to operate accurately.

[0187] Process Navigation:

[0188] During the trainees' operation, use virtual arrows or flashing indicators to mark the task path and update the navigation prompts in real time according to the trainees' progress.

[0189] Step-by-Step Operation Tips:

[0190] After each step of the operation is completed, the system automatically displays the requirements for the next step. If the operation is incorrect, the system will provide specific correction tips.

[0191] 2.2 Multimodal Interaction Interface

[0192] Voice Interaction:

[0193] The system supports trainees to operate through voice commands, such as "close the disconnecting switch". The system recognizes the commands and performs the corresponding operations.

[0194] Gesture Capture:

[0195] Use LeapMotion to capture the trainees' hand movements and verify whether the movement path conforms to the standard process. If incorrect movements are detected, the system will provide correction suggestions and demonstrate through animations.

[0196] Intelligent Handle Operation:

[0197] Equip with an intelligent handle and define shortcut operations (such as switching scenes, pausing training, etc.) through zone buttons. The handle also supports action recording for trainees to review and optimize training.

[0198] 2.3 Intelligent Perception of the Operating Environment

[0199] Behavior Analysis Module:

[0200] Use sensors and motion capture devices to record the movement trajectories of trainees, and monitor the interaction status between trainees and the devices in real time to ensure that the operations comply with the standard procedures.

[0201] Error prompt and dynamic adjustment:

[0202] The system promptly prompts incorrect operations through means such as sound and vision, and pauses the task to provide guidance for trainees to correct their operations.

[0203] 3. Dynamic action scoring and intelligent behavior assessment

[0204] Through real-time motion capture and comparison with the standard, the system can intelligently evaluate the operation performance of trainees and provide feedback.

[0205] 3.1 Real-time motion capture and comparison with the standard

[0206] High-precision capture:

[0207] Use OptiTrack devices to capture the starting point, movement path, and ending point of each action, and accurately record the action data.

[0208] Standardized comparison and feedback:

[0209] Compare the trainee's actions with the standard actions frame by frame, calculate the trajectory error and time difference, and generate an instant score.

[0210] 3.2 Comprehensive scoring model

[0211] Scoring formula:

[0212] The scoring will consider factors such as action accuracy, correction ability, and fluency. The formula will dynamically adjust the weights to adapt to different task difficulties.

[0213] Error correction part: Evaluate the correction ability based on the number of errors made by the trainee and the correction speed, etc.

[0214] Action fluency part: Evaluate the overall quality of the trainee's operation based on the continuity and smoothness of the actions.

[0215] Graded evaluation:

[0216] Excellent (above 90 points): The actions are fluent and error-free.

[0217] Good (70 - 89 points): There are a small number of trajectory deviations.

[0218] Needs improvement (below 70 points): Key steps need to be practiced again.

[0219] 3.3 Dynamic feedback and adjustment

[0220] Real-time feedback:

[0221] When errors occur, the system will provide instant optimization suggestions to help students adjust their positions, gestures, and other operations.

[0222] Step-by-step adjustment mechanism:

[0223] Based on the trainees' operational performance, the system will recommend targeted training modules and adjust the task difficulty in real time to ensure that trainees can complete the training at an appropriate difficulty.

[0224] 4. Real-time optimization and learning transfer of virtual character models

[0225] By optimizing the virtual character model and learning transfer technology, the students' operational proficiency and skill transfer ability can be improved.

[0226] 4.1 Action-driven Skeleton Optimization

[0227] Joint parameter adjustment:

[0228] The system dynamically adjusts the skeleton binding of the virtual character according to the students' operating habits, making the movements more consistent with the students' actual operations.

[0229] Action curve optimization:

[0230] Optimize the joint motion interpolation algorithm to reduce possible stuttering during movement and ensure natural and smooth movements.

[0231] 4.2 Multi-scenario migration and adaptation

[0232] Action Mapping and Adjustment:

[0233] The trainee's movement range is adjusted through an adaptive algorithm to meet the needs of different scenarios.

[0234] New task generation:

[0235] Expand new task modules (such as emergency operation training) based on existing tasks to enhance trainees' ability to respond.

[0236] 5. Data-driven personalized training and operational improvements

[0237] Through data analysis technology, a personalized training plan is customized for each trainee, and operational improvement suggestions are provided.

[0238] 5.1 Data analysis and optimization suggestions

[0239] Data integration and presentation:

[0240] The system integrates the trainees’ historical operation data and displays key indicators such as completion time and error frequency for trainees to review.

[0241] Optimization solution generation:

[0242] According to the performance of the trainees, the system automatically generates a targeted intensive training plan to improve the weak skills of the trainees.

[0243] 5.2 Teamwork Assessment

[0244] Collaboration Task Simulation:

[0245] Multiple people participate in the collaboration task online, and the system records the collaboration efficiency of the team and the rationality of task division in real time.

[0246] Team Feedback:

[0247] Provide a team score and individual contribution rate report to help optimize role allocation and task execution efficiency.

[0248] Technical Implementation Effect

[0249] Through the implementation of this VR training method, the following effects can be achieved:

[0250] Full Process Coverage: Cover the whole process from equipment inspection to operation completion.

[0251] Intelligent Assessment and Feedback: Provide accurate

[0252] instantaneous scoring and improvement suggestions.

[0253] Implementation Method 3. Combined with Figure 1 To illustrate this implementation method, this implementation method expands the above-provided technical solution through specific embodiments, including:

[0254] This implementation method is achieved through the following five core modules:

[0255] 1. Full Immersive Scene Construction Based on VR Technology

[0256] 2. Multimodal Interaction Design of Standardized Switching Operation Process

[0257] 3. Dynamic Action Scoring and Intelligent Behavior Assessment

[0258] 4. Real-time Optimization and Learning Transfer of Virtual Character Models

[0259] 5. Data-driven Personalized Training and Operation Improvement

[0260] 1. Full Immersive Scene Construction Based on VR Technology

[0261] 1.1 High-precision Scene Modeling

[0262] Specific Implementation Method:

[0263] 1. Data Acquisition:

[0264] Use a Lidar device to scan the actual environment of a 500 kV substation and collect data on equipment layout, line routing, and ground texture.

[0265] A high-resolution camera captures the material characteristics of the device surface for texture mapping.

[0266] 2. Model construction and optimization:

[0267] Import the collected point cloud data into 3D modeling software (such as Blender) to generate a basic geometric model.

[0268] Apply texture mapping and material optimization to refine key details such as equipment edges, nameplates, and buttons.

[0269] 3. Dynamic environment settings:

[0270] Design a real-time adjustable weather system, including scene changes such as sunny days, rainy days, and foggy days.

[0271] Simulate the day-night switch to test the trainees' operation ability under insufficient light conditions.

[0272] 1.2 Scene multi-task adaptation

[0273] Specific implementation methods:

[0274] 1. Task classification and modular design:

[0275] Divide the operation tasks into three categories: daily operations, fault troubleshooting, and emergency handling.

[0276] Create independent modules for each type of task. For example, the fault troubleshooting task includes subtasks such as circuit breaker inspection and equipment status assessment.

[0277] 2. Task logic arrangement:

[0278] Use script programming (such as Python or C) to define the execution order and trigger conditions of the tasks.

[0279] Set the dependency relationships between tasks. For example, the opening operation can only be performed after the equipment status is confirmed.

[0280] 3. Abnormal scene simulation:

[0281] Design the trigger conditions for equipment failures, such as high-temperature alarms and abnormal currents.

[0282] Configure the equipment feedback behavior for sudden scenarios (such as abnormal switch switching).

[0283] 1.3 Physical feedback and immersion enhancement

[0284] Specific implementation methods:

[0285] 1. Haptic feedback:

[0286] The force feedback glove realizes a realistic simulation of device operation and combines a vibration module to reproduce the mechanical resistance of the closing and opening operations.

[0287] 2. Sound effect simulation:

[0288] Add a sound effect library corresponding to the operating device, such as the current cut-off sound when the circuit breaker opens.

[0289] Introduce ambient background sounds, such as the current humming sound and wind noise in the substation.

[0290] 3. Rendering optimization:

[0291] Use real-time ray tracing technology to optimize the light and shadow effects and ensure visual authenticity.

[0292] 2. Multimodal interaction design for standardizing the switching operation process

[0293] 2.1 Visual operation process guidance

[0294] Specific implementation methods:

[0295] 1. Dynamic device annotation:

[0296] Augmented reality (AR) technology superimposes and displays the current state of the device (such as closed / open) in real time.

[0297] Highlight the device to be operated and display the function description of the device on the side of the screen.

[0298] 2. Process navigation:

[0299] The task execution path is marked by virtual arrows or flashing indicator lights.

[0300] According to the operation progress of the trainee, the navigation prompt is updated in real time.

[0301] 3. Step-by-step operation prompt:

[0302] After each step of the operation is completed, the operation requirements for the next step are automatically displayed.

[0303] An incorrect operation will trigger a specific correction prompt.

[0304] 2.2 Multimodal interaction interface

[0305] Specific implementation methods:

[0306] 1. Voice interaction:

[0307] Build in a standard operation instruction set to allow trainees to issue commands by voice, such as "Close the disconnecting switch".

[0308] The system recognizes the voice in real time and executes the corresponding operation.

[0309] 2. Gesture capture:

[0310] Use LeapMotion to capture hand movements and verify whether the movement path conforms to the standard process. Incorrect movements will automatically generate correction suggestions and display animation examples.

[0311] 3. Smart handle operation:

[0312] The handle button area is defined for quick operations, such as switching scenes, pausing training, etc.

[0313] Supports the action recording function for trainees to review and optimize training.

[0314] 2.3 Intelligent perception of the operating environment

[0315] Specific implementation method:

[0316] 1. Behavior analysis module:

[0317] Use sensors and motion capture devices to detect the operation range of trainees and the device interaction status. Record the movement trajectories of trainees and issue warnings when they do not conform to the standard actions.

[0318] 2. Error prompt and dynamic adjustment:

[0319] Incorrect operations are prompted by sound and visual means, such as red flashing or alarm sounds. Automatically pause the task and provide operation guides to guide trainees to complete the correct operations.

[0320] 3. Dynamic action scoring and intelligent behavior evaluation

[0321] 3.1 Real-time motion capture and standardized comparison

[0322] Specific implementation method:

[0323] 1. High-precision capture:

[0324] The OptiTrack device captures the starting point, movement path, and ending point of each action.

[0325] The motion data is stored as a time series for subsequent comparison.

[0326] 2. Standardized comparison and feedback:

[0327] Compare the trainee's actions with the standard actions frame by frame, and calculate the trajectory error and time difference.

[0328] Generate an instant score according to the error size.

[0329] 3.2 Comprehensive scoring model

[0330] Specific implementation method:

[0331] 1. Improvement of Scoring Formula:

[0332] Consider the action complexity and completion time, and dynamically adjust the weights.

[0333] Scoring Formula:

[0334]

[0335] Formula Definition:

[0336] S: Comprehensive Score.

[0337] α, β, δ: Weight parameters, used to balance the proportion of accuracy, correction ability, and fluency in scoring.

[0338] n: Total number of tasks.

[0339] ω i : Single-task accuracy rate, indicating the matching degree between the operation and the standard action.

[0340] c i : Task complexity coefficient, used to highlight the importance of high-difficulty tasks.

[0341] t i : Single-task completion time.

[0342] ε: Time compensation coefficient, to prevent unreasonable magnification of scores when the time is close to zero.

[0343] e: Cumulative number of errors made by the trainee in the whole task.

[0344] γ: Exponential coefficient of correction ability, reflecting the learning curve of error reduction of the trainee.

[0345] e -γ·e : Non-linear function, used to reward trainees with strong error correction ability.

[0346] F l : Action fluency score, calculated based on the smoothness of joint movement curves and action continuity.

[0347] δ: Weight coefficient of action fluency, used to evaluate the overall operation quality of the trainee.

[0348] 2. Graded Evaluation:

[0349] Excellent (above 90 points): Smooth actions and no errors.

[0350] Good (70 - 89 points): There are a small number of trajectory deviations.

[0351] Needs Improvement (below 70 points): Key steps that need to be practiced again.

[0352] 3.3 Dynamic Feedback and Adjustment

[0353] Specific implementation methods:

[0354] 1. Real-time feedback:

[0355] Display optimization suggestions immediately when an error occurs, such as adjusting the standing position or gestures.

[0356] 2. Step-by-step adjustment mechanism:

[0357] Recommend corresponding training modules according to the weak links of the trainees.

[0358] Adjust in real time according to the operation performance of the trainees, error correction ability, and task execution fluency.

[0359] Dynamic feedback adjustment formula:

[0360] F d = ρ·(1 - e -ξ·Δt )·(α·E c + β·F l ) + γ·ΔA

[0361] Formula definition:

[0362] F d : Dynamic feedback adjustment coefficient, which represents the dynamic coefficient for adjusting task difficulty and score according to the current operation performance and feedback behavior of the trainees.

[0363] ρ: Sensitivity coefficient of trainee feedback response, which represents the response intensity of the trainee to feedback. The value range is [0, 1], and the closer to 1, the more sensitive the trainee is to feedback.

[0364] ξ: Time decay coefficient, which reflects the time window effect of feedback. As time goes by, the feedback effect gradually weakens, ensuring that trainees are more sensitive to immediate feedback.

[0365] Δt: Time difference between trainee operation and feedback. The smaller the time difference, the stronger the feedback effect.

[0366] α, β: Weight coefficients, which respectively correspond to the influence of error correction ability (E_c) and action fluency (F_l) on feedback adjustment.

[0367] E c : Trainee error correction ability, evaluated based on the speed and frequency of trainees correcting errors. For example, the number of error corrections and the success rate of corrections.

[0368] F l : Action fluency score, calculated based on the action fluency and precision of the trainees, such as the smoothness of hand movements and the coherence of operations.

[0369] γ: The feedback correction coefficient, representing the correction intensity of the feedback on the task adjustment. A higher γ value indicates a greater impact of the feedback on the task difficulty.

[0370] ΔA: The task adjustment amount, representing the change in the task difficulty. For example, the number of steps required for the task, the complexity of the equipment, etc.

[0371] 4. Real - time Optimization and Learning Transfer of Virtual Character Models

[0372] 4.1 Action - driven Skeleton Optimization

[0373] Specific implementation methods:

[0374] 1. Joint parameter adjustment:

[0375] Capture the operation habits of the trainees and dynamically optimize the skeleton binding of the virtual character.

[0376] 2. Action curve optimization:

[0377] Optimize the interpolation algorithm for joint movement to reduce stuttering.

[0378] The interpolation algorithm is:

[0379] P(t)=(1 - λ(t))·P p rev(t)+λ(t)·P n ext(t)+ω(t)·f(t)

[0380] Formula definitions:

[0381] P(t): The interpolation position of the optimized action curve, representing the target position at time point t.

[0382] P p rev(t), P n ext(t): Respectively represent the key - frame positions before and after time point t. Generate the action trajectory through smooth transitions between key frames.

[0383] λ(t): The time - weighted factor, which controls the interpolation ratio over time. The value range of λ(t) is [0,1], representing the weighting coefficient between the previous frame and the next frame. Its specific calculation formula is as follows:

[0384]

[0385] -κ: The rate factor, controlling the speed of the interpolation transition.

[0386] t 0 : The reference time of the key frame, identifying the occurrence time of the key action.

[0387] ω(t): Smoothness optimization weight, which is gradually adjusted over time to increase or decrease the smoothness of actions. Its value can be dynamically adjusted based on training feedback, such as increasing or decreasing the adjustment of actions based on the proficiency of the trainee.

[0388] ω(t) = 1 / (1 + α·(t - t 0 ) 2 )

[0389] α: Smoothness factor, which controls the intensity of action smoothness. A larger value of α will enhance the effect of action smoothness and reduce unnecessary rapid action switches.

[0390] f(t): Fluency enhancement function, which is used to further optimize the naturalness and fluency of actions. This function is weighted based on the trainee's hand movement trajectory, joint rotation smoothness, and overall operation fluency:

[0391] f(t) = 1 / (1 + β·cos(2π·t / T))

[0392] β: Fluency adjustment factor, which is used to fine-tune action fluency. A higher value of β will result in higher fluency feedback.

[0393] T: Action period, which represents the time length of actions during the entire operation process.

[0394] 4.2 Multi-scenario Migration and Adaptation

[0395] Specific implementation method:

[0396] 1. Action mapping and adjustment:

[0397] Reuse training data in different scenarios and adjust the action amplitude through an adaptive algorithm.

[0398] 2. New task generation:

[0399] Expand new task modules based on existing tasks, such as emergency operation training.

[0400] 5. Data-driven Personalized Training and Operation Improvement

[0401] 5.1 Data Analysis and Optimization Suggestions

[0402] Specific implementation method:

[0403] 1. Data integration and display:

[0404] Visualize the trainee's operation history, including key metrics such as completion time and error frequency.

[0405] 2. Optimization plan generation:

[0406] The system automatically generates a targeted intensive training plan based on the data.

[0407] 5.2 Teamwork Evaluation

[0408] Specific implementation methods:

[0409] 1. Collaborative task simulation:

[0410] Multiple people participate in tasks online, and the collaboration efficiency and the rationality of task division are recorded in real time.

[0411] 2. Team feedback:

[0412] Provide team scores and individual contribution rate reports to optimize the role assignment strategy.

[0413] Technical implementation effects

[0414] Through the 500 kV standardized switching operation method based on VR virtual reality technology, the following effects can be achieved:

[0415] 1. Full-process coverage:

[0416] Full-scene coverage from equipment inspection to operation completion, improving the systematicness and integrity of training.

[0417] 2. Intelligent assessment and feedback:

[0418] Action scores and assessment results are generated in real time, providing accurate improvement suggestions.

[0419] 3. Dynamic optimization and adaptation:

[0420] The model and scene are dynamically optimized according to the performance of trainees to improve the training effect.

[0421] 4. Personalized learning:

[0422] Customize training programs according to trainee data to achieve efficient improvement of skills.

[0423] 5. Team operation collaboration:

[0424] Support multi-person collaborative tasks and improve the teamwork ability of complex grid operations.

[0425] The above further describes the technical solutions provided by the present invention through several specific implementation manners to highlight the advantages and beneficial effects of the technical solutions provided by the present invention. However, the above-mentioned several specific implementation manners are not used as limitations to the present invention. Any reasonable modifications and improvements to the present invention, combinations of implementation manners, and equivalent replacements within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A 500kV standardized switching operation method based on VR virtual reality technology, characterized in that: include: The steps to build a fully immersive virtual reality scene include: Collect actual environmental data of 500kV substation, including equipment layout, line direction, ground texture and equipment material characteristics; Based on the collected data, a high-precision scene model is built in a virtual reality environment, and texture mapping and material optimization are applied to simulate the real substation environment; Design dynamic environment settings; The multimodal interaction steps for designing a standardized switching operation process include: In the virtual reality scene, design visual operation process guidance, and use augmented reality technology to display the current status and operation path of the device; Provide a multimodal interactive interface; Introducing intelligent perception of the operating environment to detect the user's operating range and device interaction status in real time, and record the user's action trajectory; The steps to achieve dynamic action scoring and intelligent behavior assessment include: Use high-precision motion capture equipment to capture the user's switching operation actions in real time; Compare the captured action with the standard action frame by frame, calculate the trajectory error and time difference, and generate an instant score; Provide dynamic feedback based on the scoring results, including error prompts, optimization suggestions, and task difficulty adjustments; The steps of real-time optimization and learning transfer of the virtual character model include: Capture the user's operating habits and dynamically optimize the virtual character's skeleton binding and action curves to make the virtual character's actions consistent with the user's actual operations; Realize multi-scenario migration and adaptation.

2. The method for 500kV standardized switching operation based on VR virtual reality technology according to claim 1 is characterized in that: Also includes Steps for data-driven personalized training and operational improvement include: Collect and analyze user operation data, including key indicators such as completion time and error frequency; Generate optimization suggestions based on data analysis results; Provide team collaboration evaluation function, record team collaboration efficiency and rationality of task division, and optimize role allocation strategy.

3. The method for 500kV standardized switching operation based on VR virtual reality technology according to claim 1 is characterized in that: The instant scoring is achieved by the following formula: Where S is the comprehensive score, α, β, δ are weight parameters used to balance the proportion of accuracy, correction ability and fluency in the score, n is the total number of tasks, and w is i : Single-task accuracy, indicating the degree of matching between the operation and the standard action, c i : Task complexity coefficient, used to highlight the importance of high-difficulty tasks, t i : Single task completion time, ε: Time compensation coefficient, to prevent unreasonable amplification of scores when the time is close to zero, e: The number of errors accumulated by the user in the entire task, γ: The exponential coefficient of correction ability, reflecting the learning curve of user error reduction, e -γ·e : Nonlinear function used to reward users with strong error correction ability, F l : Action fluency score, calculated based on the smoothness of the joint motion curve and the action continuity, δ: The weight coefficient of action fluency, used to evaluate the overall operation quality of the user.

4. The method for 500kV standardized switching operation based on VR virtual reality technology according to claim 1 is characterized in that: The optimization of the action curve is achieved through the following interpolation algorithm: P(t)=(1-λ(t))·P p rev(t)+λ(t)·P n ext(t)+ω(t)·f(t) P(t): The interpolation position of the optimized action curve, indicating the target position at time point t, P p rev(t),P n ext(t): represents the key frame positions before and after time point t, respectively. The motion trajectory is generated through the smooth transition between key frames. λ(t): time weighting factor. As time goes by, it controls the proportion of interpolation. The value range of λ(t) is [0,1], which represents the weighting coefficient from the previous frame to the next frame. The specific calculation formula is as follows: -k: rate factor, controlling the speed of interpolation transition, t0: base time of key frame, marking the moment of occurrence of key action, ω(t): smoothness optimization weight, gradually adjusted over time to increase or decrease the smoothness of the action, its value can be dynamically adjusted based on training feedback, for example, to increase or decrease the adjustment of the action based on the user's proficiency, ω(t) = 1 / (1+α·(t-t0) 2 ) α: Smoothing factor, which controls the intensity of action smoothing. A larger α value will enhance the effect of action smoothing and reduce unnecessary rapid action switching. f(t): Fluency enhancement function, which is used to further optimize the naturalness and fluency of the action. This function is weighted based on the user's hand motion trajectory, joint rotation smoothness, and overall operation fluency: f(t)=1 / (1+β·cos(2π·t / T)) β: Fluency adjustment factor, used to fine-tune the action fluency. A higher β value will result in higher fluency feedback. T: Action cycle, which indicates the duration of the action during the entire operation process.

5. The method for 500kV standardized switching operation based on VR virtual reality technology according to claim 1 is characterized in that: The step of constructing a fully immersive virtual reality scene also includes: Use 3D modeling software to model substation equipment according to the size, shape and material of the actual equipment; Design interactive hotspots in the virtual environment to provide users with device information.

6. The method for 500kV standardized switching operation based on VR virtual reality technology according to claim 1 is characterized in that: The multimodal interaction step of designing a standardized switching operation process also includes: Provide a voice command library, and users can control device operations in the virtual environment through preset voice commands; Provide gesture recognition algorithm to identify the user's hand movements and convert them into operation instructions in the virtual environment.

7. A 500kV standardized switching operation device based on VR virtual reality technology, characterized in that: include: Modules for building fully immersive virtual reality scenes, including: Collect actual environmental data of 500kV substation, including equipment layout, line direction, ground texture and equipment material characteristics; Based on the collected data, a high-precision scene model is built in a virtual reality environment, and texture mapping and material optimization are applied to simulate the real substation environment; Design dynamic environment settings; Design a multi-modal interaction module for standardized switching operation procedures, including: In the virtual reality scene, design visual operation process guidance, and use augmented reality technology to display the current status and operation path of the device; Provide a multimodal interactive interface; Introducing intelligent perception of the operating environment to detect the user's operating range and device interaction status in real time, and record the user's action trajectory; Modules for implementing dynamic action scoring and intelligent behavior evaluation include: Use high-precision motion capture equipment to capture the user's switching operation actions in real time; Compare the captured action with the standard action frame by frame, calculate the trajectory error and time difference, and generate an instant score; Provide dynamic feedback based on the scoring results, including error prompts, optimization suggestions, and task difficulty adjustments; The modules for real-time optimization and learning transfer of virtual character models include: Capture the user's operating habits and dynamically optimize the virtual character's skeleton binding and action curves to make the virtual character's actions consistent with the user's actual operations; Realize multi-scenario migration and adaptation.

8. A computer storage medium for storing a computing program, characterized in that: When the computer program is read by a computer, the computer executes the method of claim 1 .

9. A computer, comprising a processor and a storage medium, characterized in that: When the processor reads the computer program stored in the storage medium, the computer executes the method of claim 1 .

10. A computer program product, being a computer program, characterized in that When the computer program is executed, the method of claim 1 is implemented.

Images