New energy full-scene practical training platform based on interactive simulation and simulation method thereof

Abstract

The invention provides a new energy full-scene practical training platform based on interactive simulation and a simulation method thereof, and the platform comprises a three-dimensional data obtaining module which is used for obtaining three-dimensional data of new energy full-scene practical training, and the new energy full-scene practical training comprises fan scene practical training, photovoltaic scene practical training and booster station scene practical training; the modeling module is used for performing 3D digital twinning according to the three-dimensional data to obtain a virtual model; the synchronization module is used for performing data synchronization on the virtual model of the operation data input by the user and the physical equipment of the new energy full-scene practical training; and the training module is used for performing corresponding training based on the synchronization result. According to the new energy full-scene practical training platform based on interactive simulation and the simulation method thereof, the training efficiency is improved, and the training scene is more comprehensive.

Description

Technical Field

[0001] This invention relates to the field of simulation training technology, and in particular to a new energy full-scenario practical training platform based on interactive simulation and its simulation method. Background Technology

[0002] As the global energy structure shifts towards cleaner and lower-carbon energy, the new energy industry is experiencing explosive growth, with a surge in demand for highly skilled personnel in emerging fields such as wind power, photovoltaics, and energy storage. However, the traditional training system has lagged far behind the industry's development needs, revealing four major structural defects: First, theoretical teaching and on-site practice operate in parallel, requiring trainees an average of 6-8 months to adapt, with an operation and maintenance error rate as high as 23%. Second, existing simulation systems mostly use static mathematical models, resulting in simulation errors of up to 18% for dynamic operating conditions such as photovoltaic array shading and wind turbine yaw errors, leading to a significant deviation between training effectiveness and actual operating conditions. Third, training coverage for extreme scenarios such as grid frequency oscillations and battery thermal runaway is less than 5%, resulting in a lack of emergency response capabilities for major accidents. More importantly, current training resources are characterized by "island-like" features, with practical training for subsystems such as wind power, photovoltaics, and energy storage being fragmented and unable to simulate new operating scenarios such as multi-energy complementarity and virtual power plants.

[0003] In view of this, there is an urgent need for a new energy full-scenario practical training platform and its simulation methods based on interactive simulation, in order to at least solve the above-mentioned shortcomings. Summary of the Invention

[0004] One objective of this invention is to provide a new energy full-scenario practical training platform and its simulation method based on interactive simulation. It introduces 3D data from wind turbines, photovoltaics, and booster stations for full-scenario practical training and establishes virtual models. The virtual model, containing user-input operation data, is synchronized with the physical equipment used in the new energy full-scenario practical training. When the user's practical operation corresponding to the operation data leads to a fault, the virtual model automatically switches to a fault state, highlights the abnormal component, and triggers error operation prompts and maintenance guidance animations. Simultaneously, the simulation model verifies the legality of the operation to ensure that virtual erroneous operations do not actually damage the equipment. Otherwise, both the virtual model and the physical equipment respond to the user's practical operation, intuitively displaying the equipment status to the user. After the entire practical training operation is completed, the system automatically generates a scoring report including operation steps and time consumption, completing the training, improving training efficiency, and providing a more comprehensive training scenario.

[0005] The interactive simulation-based new energy full-scenario practical training platform provided in this embodiment of the invention includes:

[0006] The 3D data acquisition module is used to acquire 3D data for new energy full-scenario training, which includes wind turbine scenario training, photovoltaic scenario training and booster station scenario training.

[0007] The modeling module is used to create a 3D digital twin based on 3D data to obtain a virtual model.

[0008] The synchronization module is used to synchronize the virtual model of user input operation data with the physical equipment of new energy full-scenario training.

[0009] The training module is used to conduct relevant training based on the synchronization results.

[0010] Preferably, the physical equipment for wind turbine scenario training in the 3D data acquisition module includes a 1.5MW doubly-fed wind turbine.

[0011] Preferably, the physical equipment for photovoltaic scene training in the 3D data acquisition module includes: 30kW photovoltaic modules, photovoltaic brackets, combiner boxes, inverters, 50kW / 100kWh energy storage systems, connecting cables, 3.6kW photovoltaic modules, photovoltaic brackets, inverters, distribution boxes, cables, and auxiliary materials and tools for making MC4 plugs.

[0012] Preferably, the physical equipment for the substation scenario training in the 3D data acquisition module includes: 10kV switchgear, distribution cabinet, 400V motor and supporting control cabinet, 10kV cross-linked polyethylene cable head making materials and tools, relay protection cabinet, DC power supply, substation simulation panel, UPS, workstation and cabinet.

[0013] Preferably, the modeling module performs 3D digital twin generation based on 3D data to obtain a virtual model, including:

[0014] The virtual model is obtained by using Unity3D and BladeGen to jointly model the 3D data.

[0015] Preferably, the synchronization module synchronizes the virtual model containing user-input operation data with the physical equipment used in the new energy full-scenario training, including:

[0016] When the physical equipment is a 10kV switchgear for a substation, bidirectional data interaction between the SCADA system and the virtual model is achieved through the IEC 61850 and / or MODBUS protocols.

[0017] Preferably, the training module conducts corresponding training based on the synchronization results, including:

[0018] If the synchronization result indicates an operation failure, obtain the first operation score of the corresponding user's input operation data;

[0019] The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score; the remedial performance includes: the immediate handling of the failure after the operation failure and the re-entry of operation data after the failure handling is completed.

[0020] If the second operation score is less than or equal to the preset operation score threshold, the abnormal component is highlighted and an error operation prompt and maintenance guidance animation are triggered.

[0021] If the synchronization result is successful, obtain the action matching between the user's viewed action and the standard viewed action after successful synchronization;

[0022] If the action matching fails, or if the user's confused expression is extracted when the action matching is successful, the running component will be highlighted and an operation prompt will be triggered.

[0023] Preferably, the standard steps for obtaining the viewing action include:

[0024] The training knowledge is obtained by retrieving the training knowledge base based on the semantics of the running components.

[0025] Train a knowledge understanding structure generation model based on training records labeled with the semantics of historical training knowledge and manually drawn knowledge understanding structures.

[0026] Input the semantics of practical training knowledge into the knowledge understanding structure generation model to generate the knowledge understanding structure of the running components;

[0027] Traverse the structural nodes in the knowledge understanding structure of the running components from beginning to end, and extract the descriptive bytes and importance keywords of the running components based on the descriptive component content in the structural node information;

[0028] The viewing time for running components is quantified based on description bytes and keywords of importance.

[0029] Standard viewing actions are generated based on the viewing duration of the sequentially quantified operating components.

[0030] Preferably, before triggering the maintenance guidance animation, if the corresponding user input operation data matches the data in the historical high-frequency error operation database, the maintenance guidance animation is triggered; otherwise, the system is reset directly after triggering the error operation prompt. The reset state is the virtual model state before the corresponding user input operation data of the operation fault is entered, with the synchronization result being the virtual model state before the operation fault is entered.

[0031] Preferably, the first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score, including:

[0032] Based on the processing characteristics of fault handling immediately after an operational failure, the user's fault understanding is obtained;

[0033] Based on the re-entered operation data and the standard operation data corresponding to the operation steps of operation failures, obtain the user's operation understanding verification degree;

[0034] The operational understanding verification score corresponds to the third operational score based on the initial input of operational data.

[0035] The second operation score is calculated based on the first operation score, fault understanding, and the third operation score.

[0036] The interactive simulation-based training method for new energy full-scenario practical training provided in this embodiment of the invention includes:

[0037] Step 1: Obtain 3D data for the new energy full-scenario training, which includes: wind turbine scenario training, photovoltaic scenario training, and booster station scenario training;

[0038] Step 2: Create a 3D digital twin based on the 3D data to obtain a virtual model;

[0039] Step 3: Synchronize the data between the virtual model containing the user's input operation data and the physical equipment used in the new energy full-scenario training.

[0040] Step 4: Based on the synchronization results, conduct corresponding training.

[0041] The beneficial effects of this invention are as follows:

[0042] This invention introduces 3D data from full-scenario training involving wind turbines, photovoltaics, and booster stations, and establishes virtual models. The virtual model, containing user-inputted operation data, is synchronized with the physical equipment used in the full-scenario training. When a user's training operation leads to a malfunction, the virtual model automatically switches to a fault state, highlighting the abnormal component and triggering error operation prompts and maintenance guidance animations. Simultaneously, the simulation model verifies the legality of the operation to ensure that virtual malfunctions do not actually damage the equipment. Otherwise, both the virtual model and the physical equipment respond to the user's training operation, intuitively displaying the equipment status to the user. After the entire training operation is completed, the system automatically generates a scoring report including operation steps and time consumption, completing the training, improving training efficiency, and providing a more comprehensive training scenario.

[0043] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0044] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0045] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0046] Figure 1This is a schematic diagram of a new energy full-scenario practical training platform based on interactive simulation in an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of a simulation method for new energy full-scenario practical training based on interactive simulation in an embodiment of the present invention. Detailed Implementation

[0048] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0049] This invention provides a new energy full-scenario practical training platform based on interactive simulation, such as... Figure 1 As shown, it includes:

[0050] 3D data acquisition module 1 is used to acquire 3D data for new energy full-scenario training, which includes wind turbine scenario training, photovoltaic scenario training and booster station scenario training.

[0051] The New Energy Full-Scenario Training refers to a comprehensive training environment covering wind power generation, photovoltaic power generation, and substation operation and maintenance. It includes the following sub-scenarios: Wind Turbine Scenario Training: Simulating the operation of doubly-fed induction generator (DFIG) wind turbines, covering core subsystem operations such as yaw, pitch, and drivetrain; Photovoltaic Scenario Training: Replicating the joint scheduling and fault handling of photovoltaic arrays, inverters, and energy storage systems; Substation Scenario Training: Training on the operational procedures of grid-connected equipment such as 10kV switchgear and relay protection devices. The 3D data includes the geometric dimensions (such as wind turbine tower height and photovoltaic panel tilt angle) and material textures of the physical equipment in the New Energy Full-Scenario Training.

[0052] Among them, the physical equipment for wind turbine scenario training in the 3D data acquisition module includes a 1.5MW doubly-fed wind turbine;

[0053] Among them, the physical equipment for photovoltaic scene training in the 3D data acquisition module includes: 30kW photovoltaic modules, photovoltaic brackets, combiner boxes, inverters, 50kW / 100kWh energy storage systems, connecting cables, 3.6kW photovoltaic modules, photovoltaic brackets, inverters, distribution boxes, cables, and auxiliary materials and tools for making MC4 plugs;

[0054] Among them, the physical equipment for the substation scenario training in the 3D data acquisition module includes: 10kV switch cabinet, distribution cabinet, 400V motor and supporting control cabinet, 10kV cross-linked polyethylene cable head making materials and tools, relay protection cabinet, DC screen, substation simulation screen, UPS, workstation and cabinet.

[0055] Modeling module 2 is used to create a 3D digital twin based on 3D data to obtain a virtual model, and performs the following operations:

[0056] The 3D data is modeled using Unity3D and BladeGen to obtain a virtual model;

[0057] Taking the physical equipment as an example of building a virtual model of a doubly fed wind turbine, the Unity3D+BladeGen joint modeling is used to realize the 1:1 fine modeling of the doubly fed wind turbine, including multiple components such as the aerodynamic shape of the blades and the internal gear meshing of the gearbox, and supports dynamic simulation of pitch angle from 0 to 91°.

[0058] Synchronization module 3 is used to synchronize the virtual model of user input operation data with the physical equipment of new energy full-scenario training.

[0059] During data synchronization, IoT, edge computing, and digital twin technologies are integrated to build a two-way real-time interactive system. During synchronization, if a user's training operation causes a fault, the virtual model will automatically switch to the fault state. At the same time, the simulation model will verify the legality of the operation to ensure that the virtual misoperation will not actually damage the equipment. Otherwise, both the virtual model and the physical equipment will respond to the user's training operation.

[0060] Training Module 4 is used to conduct corresponding training based on the synchronization results.

[0061] When training is conducted based on synchronized results, if a user's training operation leads to a malfunction, the virtual model will highlight the abnormal component and trigger error operation prompts and maintenance guidance animations. After the entire training operation is completed, the system automatically generates a scoring report including the operation steps and time consumed, thus completing the training.

[0062] The working principle and beneficial effects of the above technical solution are as follows:

[0063] This invention introduces 3D data from full-scenario training involving wind turbines, photovoltaics, and booster stations, and establishes virtual models. The virtual model, containing user-inputted operation data, is synchronized with the physical equipment used in the full-scenario training. When a user's training operation leads to a malfunction, the virtual model automatically switches to a fault state, highlighting the abnormal component and triggering error operation prompts and maintenance guidance animations. Simultaneously, the simulation model verifies the legality of the operation to ensure that virtual malfunctions do not actually damage the equipment. Otherwise, both the virtual model and the physical equipment respond to the user's training operation, intuitively displaying the equipment status to the user. After the entire training operation is completed, the system automatically generates a scoring report including operation steps and time consumption, completing the training, improving training efficiency, and providing a more comprehensive training scenario.

[0064] In one embodiment, the training module performs corresponding training based on the synchronization results, including:

[0065] If the synchronization result indicates an operation failure, obtain the first operation score of the corresponding user's input operation data;

[0066] The first operation score is determined based on the severity of the operation failure; the higher the severity, the lower the corresponding first operation score.

[0067] The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score; the remedial performance includes: the immediate handling of the failure after the operation failure and the re-entry of operation data after the failure handling is completed.

[0068] The second performance score is the final score adjusted by combining the first performance score and the remedial performance.

[0069] If the second operation score is less than or equal to the preset operation score threshold, the abnormal component is highlighted and an error operation prompt and maintenance guidance animation are triggered.

[0070] The preset operation rating threshold is manually set to determine whether the user needs assistance; highlighting abnormal parts refers to displaying faulty parts in the virtual model; error operation prompts are text or voice prompts provided by the system to explain the cause of the operation error; maintenance guidance animation is a dynamic 3D animation demonstrating how to repair the operation fault.

[0071] If the synchronization result is successful, obtain the action matching between the user's viewed action and the standard viewed action after successful synchronization;

[0072] User viewing actions refer to the actions taken by the user to view the data panel and virtual components after successful synchronization; standard viewing actions refer to the expected and correct viewing behavior; action matching status refers to whether the user viewing actions and standard viewing actions are consistent after comparison.

[0073] If the action matching fails, or if the user's confused expression is extracted when the action matching is successful, the running component will be highlighted and an operation prompt will be triggered.

[0074] The confused expression feature is a user's facial expression (such as frowning, tilting head, or staring for a long time) detected by a camera or sensor. When a confused expression feature is detected, it indicates that the user is confused. The running component is a virtual component triggered by the user's input operation data that has been successfully synchronized. It is highlighted to guide the user's attention. The operation prompt is a status explanation generated by the system status of the corresponding virtual component triggered by the user's operation data that has been successfully synchronized. It is dynamically displayed in the blank virtual space next to the highlighted running component.

[0075] The working principle and beneficial effects of the above technical solution are as follows:

[0076] During simulation training, a scenario arises after a simulation failure: a user enters incorrect operation data due to misoperation. The user actually knows the correct operation, but is simply unfamiliar with it. It's unreasonable to directly treat such users as completely unaware of the cause of the operation failure, and it wastes subsequent guidance resources. Therefore, based on the severity of the operation failure, a first operation score is determined, and then the final score (second operation score) is obtained by combining the user's immediate remedial performance after the operation failure. If the second operation score is still less than or equal to the preset operation score threshold, it indicates that the user is indeed unfamiliar with the operation step. The abnormal component is highlighted, and an error operation prompt and maintenance guidance animation are triggered for precise guidance. Furthermore, even if no operation failure occurs, the user may not fully understand the operation and maintenance principles. For example, they may only know how to do it (i.e., what operation data to input) but are unaware of the internal operating mechanism of the physical equipment. Therefore, after successful synchronization, the system obtains the action matching between the user's viewing action and the standard viewing action. If the action matching fails, or if the user's confused expression is extracted when the action matching succeeds, it indicates that the user has missed viewing the data of the standard viewing action whose action matching failed, and has not understood the data corresponding to the standard viewing action whose user's confused expression is extracted when the action matching succeeds. Further understanding is needed. Therefore, the running component is highlighted and an operation prompt is triggered, which greatly improves the learning efficiency of simulation training and makes the training more targeted.

[0077] In one embodiment, the steps for obtaining a standard viewing action include:

[0078] The training knowledge is obtained by retrieving the training knowledge base based on the semantics of the running components.

[0079] Among them, component semantics is a semantic description of the name of the running component;

[0080] Among them, the practical training knowledge base is a database composed of practical training knowledge related to wind turbine scenario training, photovoltaic scenario training and booster station scenario training;

[0081] Among them, the practical training knowledge is the knowledge data (component function, component operation, and component working principle) of the virtual component triggered by the input operation data retrieved from the practical training knowledge base, which corresponds to the actual component.

[0082] Train a knowledge understanding structure generation model based on training records labeled with the semantics of historical training knowledge and manually drawn knowledge understanding structures.

[0083] Among them, the manually drawn knowledge understanding structure is a structured flowchart drawn by humans based on the semantics of historical training knowledge to help understand the corresponding knowledge;

[0084] In training the knowledge understanding structure generation model, the CNN model is trained using historical training knowledge semantics as input and corresponding manually drawn knowledge understanding structures as output.

[0085] Input the semantics of practical training knowledge into the knowledge understanding structure generation model to generate the knowledge understanding structure of the running components;

[0086] Among them, the knowledge understanding structure of the running components is a flowchart drawn based on the practical training knowledge of the running components that actually trigger virtual operation, which helps to understand the corresponding knowledge.

[0087] Traverse the structural nodes in the knowledge understanding structure of the running components from beginning to end, and extract the descriptive bytes and importance keywords of the running components based on the descriptive component content in the structural node information;

[0088] Here, structural nodes are process nodes in the flowchart, structural node information is node information in the process nodes, and the content of the descriptive component is the semantic result of the structural node information.

[0089] The viewing time for running components is quantified based on description bytes and keywords of importance.

[0090] When quantifying the viewing time of a running component, the more bytes described and the more important the keywords representing the key, the longer the quantified viewing time. For example, the preset viewing time for 150 bytes is 2 minutes, and the preset viewing time for "key components" is 3 minutes, determined by the byte count-time table. The quantified viewing time of the running component is the average of these determined viewing times (2.5 minutes).

[0091] Standard viewing actions are generated based on the viewing duration of the sequentially quantified operating components.

[0092] The working principle and beneficial effects of the above technical solution are as follows:

[0093] Traditional training relies on verbal instruction from coaches, and trainees spend arbitrarily time examining components (e.g., a novice might only spend 3 seconds at a high-voltage node, far less than the 10 seconds required for safety), leading to a high rate of missed detections. Therefore, this invention retrieves training knowledge based on component semantics and generates a corresponding operational component knowledge understanding structure based on a trained knowledge understanding structure generation model. It iterates through the structural nodes in the operational component knowledge understanding structure from beginning to end, extracting and quantifying the descriptive bytes and importance keywords of the operational components in each node's component description content to determine the viewing duration. Based on the quantified viewing duration of each operational component, standard viewing actions are generated. This invention achieves automatic quantification of viewing actions, establishing calculable operational standards in full-scenario training, breaking through the experience-driven model of traditional training, and becoming more intelligent. Furthermore, it can dynamically generate matching standards (standard viewing actions) without the need for a pre-built database of massive standard viewing actions, greatly improving matching efficiency.

[0094] In one embodiment, if the corresponding user input operation data matches the data in the historical high-frequency error operation database before triggering the maintenance guidance animation, the maintenance guidance animation is triggered; otherwise, the system is reset directly after triggering the error operation prompt. The reset state is the virtual model state before the corresponding user input operation data of the synchronization result is the operation fault.

[0095] The historical high-frequency error operation database is constructed by staff based on the proportion of historical error operation data for the same operation step to all historical operation data. If the proportion of a certain historical error operation data is greater than a preset proportion threshold (e.g., 10%), then the staff will enter the historical high-frequency error operation data into the historical high-frequency error operation database.

[0096] The working principle and beneficial effects of the above technical solution are as follows:

[0097] Generally, not all virtual faults have maintenance guidance value. When beginners are training, their erroneous operation records may indeed be frequently encountered errors in subsequent practice. These errors are likely to be reproduced in future practice and thus have maintenance guidance value. Therefore, if the user's input operation data matches data in the historical high-frequency error operation database, a maintenance guidance animation is triggered. Otherwise, after triggering the error operation prompt, the simulation characteristics of the virtual model are used to directly reset the system, reducing worthless guidance and improving guidance efficiency.

[0098] In one embodiment, a second operation score is obtained by adjusting the first operation score based on the user's immediate remedial performance after the operation failure, including:

[0099] Based on the processing characteristics of fault handling immediately after an operational failure, the user's fault understanding γ is obtained;

[0100] Processing characteristics include: processing speed and processing results; fault understanding is the user's level of understanding of the fault; when obtaining the user's fault understanding, based on the historical processing characteristics of the fault, the processing speed and processing results of the immediate fault processing are scored separately, the average score is taken and normalized to obtain the fault understanding. For example, if the score based on processing speed is 86 and the score based on processing results is 90, then the normalized fault understanding is 0.88.

[0101] Based on the re-entered operation data and the standard operation data corresponding to the operation steps of operation failures, obtain the user's operation understanding verification degree;

[0102] The operation understanding verification score is the data similarity between the re-input operation data and the standard operation data; obtaining the operation understanding verification score corresponds to the third operation score of the first input operation data.

[0103] Among them, the third operation score corresponding to the first input operation data refers to the operation score of the first input operation when the data similarity between the first input operation data and the standard operation data corresponding to the operation failure is the operation understanding verification score. The conversion relationship between the third operation score of the first input operation and the operation understanding verification score corresponding to the first input operation is obtained according to the manually preset scoring table. For the same operation understanding verification score, the score of the first input operation is greater than the score of the re-input operation.

[0104] Based on the first operation score Fault understanding γ and third operation score Calculate the score for the second operation The calculation formula is as follows:

[0105]

[0106] The working principle and beneficial effects of the above technical solution are as follows:

[0107] This invention scores and normalizes the processing speed and results of immediate fault handling based on the processing features (processing speed and results) extracted immediately after an operational failure and the historical processing features of the failure, thus determining the fault understanding degree. The similarity between re-input operation data and standard operation data is used as the operation understanding verification degree. Simultaneously, the operation understanding degree of the initial input operation data is obtained as the third operation score at the operation understanding verification degree. The fault understanding degree is used as a correction weight to adjust the difference between the third operation score and the first operation score to obtain a corrected result. The corrected result is added to the first operation score to obtain an adjusted second operation score. This precisely quantifies the remedial performance, upgrading the system from "judging right or wrong" to "understanding the user's cognitive state," achieving accurate operation evaluation, and making subsequent intervention and assistance more appropriate.

[0108] This invention provides a simulation method for real-world training in new energy scenarios based on interactive simulation, such as... Figure 2 As shown, it includes:

[0109] Step 1: Obtain 3D data for the new energy full-scenario training, which includes: wind turbine scenario training, photovoltaic scenario training, and booster station scenario training;

[0110] Step 2: Create a 3D digital twin based on the 3D data to obtain a virtual model;

[0111] Step 3: Synchronize the data between the virtual model containing the user's input operation data and the physical equipment used in the new energy full-scenario training.

[0112] Step 4: Based on the synchronization results, conduct corresponding training, including:

[0113] If the synchronization result indicates an operation failure, obtain the first operation score of the corresponding user's input operation data;

[0114] The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score; the remedial performance includes: the immediate handling of the failure after the operation failure and the re-entry of operation data after the failure handling is completed.

[0115] If the second operation score is less than or equal to the preset operation score threshold, the abnormal component is highlighted and an error operation prompt and maintenance guidance animation are triggered.

[0116] If the synchronization result is successful, obtain the action matching between the user's viewed action and the standard viewed action after successful synchronization;

[0117] If the action matching fails, or if the user's confused expression is extracted when the action matching is successful, the running component will be highlighted and an operation prompt will be triggered.

[0118] The standard steps for obtaining a viewing action include:

[0119] The training knowledge is obtained by retrieving the training knowledge base based on the semantics of the running components.

[0120] Train a knowledge understanding structure generation model based on training records labeled with the semantics of historical training knowledge and manually drawn knowledge understanding structures.

[0121] Input the semantics of practical training knowledge into the knowledge understanding structure generation model to generate the knowledge understanding structure of the running components;

[0122] Traverse the structural nodes in the knowledge understanding structure of the running components from beginning to end, and extract the descriptive bytes and importance keywords of the running components based on the descriptive component content in the structural node information;

[0123] The viewing time for running components is quantified based on description bytes and keywords of importance.

[0124] Based on the sequentially quantified viewing duration of the running components, standard viewing actions are generated;

[0125] Before triggering the maintenance guidance animation, if the corresponding user input operation data matches the data in the historical high-frequency error operation database, the maintenance guidance animation will be triggered; otherwise, the system will be reset directly after triggering the error operation prompt. The reset state is the virtual model state before the corresponding user input operation data of the operation failure was entered, which is the synchronization result.

[0126] The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score, which includes:

[0127] Based on the processing characteristics of fault handling immediately after an operational failure, the user's fault understanding is obtained;

[0128] Based on the re-entered operation data and the standard operation data corresponding to the operation steps of operation failures, obtain the user's operation understanding verification degree;

[0129] The operational understanding verification score corresponds to the third operational score based on the initial input of operational data.

[0130] The second operation score is calculated based on the first operation score, fault understanding, and the third operation score.

[0131] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A new energy full-scenario practical training platform based on interactive simulation, characterized in that: include: The 3D data acquisition module is used to acquire 3D data for new energy full-scenario training, which includes wind turbine scenario training, photovoltaic scenario training and booster station scenario training. The modeling module is used to create a 3D digital twin based on 3D data to obtain a virtual model. The synchronization module is used to synchronize the virtual model of user input operation data with the physical equipment of new energy full-scenario training. The training module is used to conduct relevant training based on the synchronization results.

2. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The physical equipment for wind turbine scenario training in the 3D data acquisition module includes a 1.5MW doubly-fed wind turbine.

3. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The physical equipment for photovoltaic scene training in the 3D data acquisition module includes: 30kW photovoltaic modules, photovoltaic brackets, combiner boxes, inverters, 50kW / 100kWh energy storage systems, connecting cables, 3.6kW photovoltaic modules, photovoltaic brackets, inverters, distribution boxes, cables, and auxiliary materials and tools for making MC4 plugs.

4. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The physical equipment for the substation scenario training in the 3D data acquisition module includes: 10kV switchgear, distribution cabinet, 400V motor and supporting control cabinet, 10kV cross-linked polyethylene cable head making materials and tools, relay protection cabinet, DC power supply, substation simulation panel, UPS, workstation and cabinet.

5. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The modeling module generates a 3D digital twin based on the 3D data to obtain a virtual model, including: The virtual model is obtained by using Unity3D and BladeGen to jointly model the 3D data.

6. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The synchronization module synchronizes user-input operation data between the virtual model and the physical equipment used in the new energy full-scenario training, including: When the physical equipment is a 10kV switchgear for a substation, bidirectional data interaction between the SCADA system and the virtual model is achieved through the IEC 61850 and / or MODBUS protocols.

7. The new energy full-scenario practical training platform based on interactive simulation as described in claim 1, characterized in that, The training module, based on the synchronization results, provides corresponding training, including: If the synchronization result indicates an operation failure, obtain the first operation score of the corresponding user's input operation data; The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score; the remedial performance includes: the immediate handling of the failure after the operation failure and the re-entry of operation data after the failure handling is completed. If the second operation score is less than or equal to the preset operation score threshold, the abnormal component is highlighted and an error operation prompt and maintenance guidance animation are triggered. If the synchronization result is successful, obtain the action matching between the user's viewed action and the standard viewed action after successful synchronization; If the action matching fails, or if the user's confused expression is extracted when the action matching is successful, the running component will be highlighted and an operation prompt will be triggered.

8. The new energy full-scenario practical training platform based on interactive simulation as described in claim 7, characterized in that, Before triggering the maintenance guidance animation, if the corresponding user input operation data matches the data in the historical high-frequency error operation database, the maintenance guidance animation will be triggered; otherwise, the system will be reset directly after triggering the error operation prompt. The reset state is the virtual model state before the corresponding user input operation data of the operation fault was entered, with the synchronization result being the virtual model state before the operation fault was entered.

9. The new energy full-scenario practical training platform based on interactive simulation as described in claim 7, characterized in that, The first operation score is adjusted based on the user's immediate remedial performance after the operation failure, resulting in a second operation score, which includes: Based on the processing characteristics of fault handling immediately after an operational failure, the user's fault understanding is obtained; Based on the re-entered operation data and the standard operation data corresponding to the operation steps of operation failures, obtain the user's operation understanding verification degree; The operational understanding verification score corresponds to the third operational score based on the initial input of operational data. The second operation score is calculated based on the first operation score, fault understanding, and the third operation score.

10. A simulation method for new energy full-scenario practical training based on interactive simulation, characterized in that, include: Step 1: Obtain 3D data for the new energy full-scenario training, which includes: wind turbine scenario training, photovoltaic scenario training, and booster station scenario training; Step 2: Create a 3D digital twin based on the 3D data to obtain a virtual model; Step 3: Synchronize the data between the virtual model containing the user's input operation data and the physical equipment used in the new energy full-scenario training. Step 4: Based on the synchronization results, conduct corresponding training.

Images