New energy intelligent training and examination system and method based on cloud rendering

Abstract

The present application relates to the technical field of new energy training, and particularly relates to a new energy intelligent training and examination system and method based on cloud rendering, which comprises a cloud rendering hub platform system that generates a computing resource allocation scheme according to virtualized and integrated graph computing resources based on course learning request data uploaded by a student terminal device and a student access terminal type; a post skill training and examination module that generates personalized learning evaluation data containing ability short board positioning according to course interaction data and student learning behavior data of the student in a virtual simulation environment; and a virtual-real fusion module that generates force feedback operation data that is synchronously updated with a real job training scene by fusing the virtual training environment and the real job training scene through a data coupling engine. The present application realizes a new energy intelligent training and examination system that seamlessly fuses virtual and real job scenes, and improves the utilization efficiency and precise allocation of cloud rendering platform resources.

Description

Technical Field

[0001] This invention relates to the field of new energy training technology, and in particular to a cloud-based intelligent training and assessment system and method for new energy. Background Technology

[0002] With the rapid development of the new energy industry, the importance of its training skills has become increasingly prominent. However, traditional training and assessment methods for new energy skills face many challenges. In terms of traditional training models, hands-on training based on physical equipment is limited by equipment costs and venue resources, making it difficult to achieve large-scale training. At the same time, offline theoretical training is constrained by both time and space. Although existing digital training systems have achieved large-scale coverage in theoretical teaching, these systems mostly adopt a combination of video courseware, graphic materials, and online exams. This two-dimensional information transmission method lacks the simulation of a real operating environment, making it difficult for trainees to combine theory with practice during the learning process.

[0003] While the introduction of virtual simulation technology has alleviated the problem of the disconnect between theory and practice to some extent, such software usually requires high-performance computer support, which not only increases training costs but also poses a risk of software resources being cracked and leaked. At the same time, existing training methods are struggling to keep up with the rapid development of the new energy industry. Existing training systems lack dynamic data integration capabilities and cannot obtain real-time data such as operating parameters and fault cases of production line equipment, creating a data gap between the training scenario and the actual working environment. As a result, training content is often out of touch with the industry's most advanced workflows and cannot meet the needs of the industry's rapid development.

[0004] In summary, existing methods for assessing and training new energy skills have many problems. There is an urgent need in the field of new energy skills training to provide a more efficient, flexible training and assessment method that can closely integrate theory and practice. Summary of the Invention

[0005] To address the above technical problems, this invention provides a cloud-based intelligent training and assessment system and method for new energy.

[0006] In a first aspect, the present invention provides a new energy intelligent training and assessment system based on cloud rendering, comprising:

[0007] The cloud rendering central platform system is used to virtualize and integrate graphics computing resources based on the course learning request data uploaded by the student terminal device and the student access terminal type, generate a computing resource allocation scheme, and transmit the virtual simulation environment rendering screen to the front-end interface of the student terminal device using different transmission strategies based on the computing resource allocation scheme and course type characteristics.

[0008] The job skills training and assessment module is used to quantitatively evaluate the standardization of new energy training parameters, the compliance rate of standard parameters, and the handling of abnormal working conditions based on the course interaction data and learning behavior data of trainees in the virtual simulation environment, and to generate personalized learning assessment data that includes the identification of competency gaps.

[0009] The virtual-real fusion module is used to merge the virtual training environment with the real-world training scenario based on the personalized learning assessment data and the real-time collected on-site training equipment operation data through a data coupling engine. This generates force feedback operation data that is updated synchronously with the real-world training scenario, and feeds the force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme.

[0010] In a further implementation, the cloud rendering central platform system includes a computing resource dynamic scheduling module, a cloud rendering course image transmission module, a human-environment-device data integration module, and a spatial computing course recommendation module;

[0011] The spatial computing course recommendation module is used to recommend a list of training programs with the shortest straight-line distance based on the student's geographical location information or spatial distribution information, and to generate course learning request data when the student selects a target training program based on all the training program course lists.

[0012] The computing resource dynamic scheduling module is used to identify the student access terminal type based on the course learning request data uploaded by the student terminal device, and dynamically schedule graphics computing resources from the server resource pool to run various basic courses based on the student access terminal type and the computing resource requirements of various basic courses, and generate a computing resource allocation scheme.

[0013] The cloud rendering course image transmission module is used to perform streaming media encoding processing on the virtual simulation environment rendering screen according to the characteristics of the course type to be learned, generate an encrypted rendering data stream, and transmit the encrypted rendering data stream to the front-end interface of the student terminal device through the computing resource allocation scheme.

[0014] The human-environment-equipment data integration module is used to update and integrate structured knowledge in real time the dynamic background data and student learning-related information involved in the job training and assessment courses.

[0015] In a further implementation, when the course type to be learned is a theoretical course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through a remote encrypted access mode.

[0016] When the course type to be learned is a practical course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through cloud rendering mode.

[0017] When the course type to be learned is an assessment course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through a front-end encrypted display mode.

[0018] In a further implementation, the spatial computing course recommendation module includes a spatial positioning unit, a distance calculation unit, and a course recommendation unit;

[0019] The spatial positioning unit is used to extract the instrument feature information, parameter identification information and operating environment information of the new energy training equipment from the current environment information uploaded by the student terminal device, generate a structured equipment positioning feature dataset, and associate and match the structured equipment positioning feature dataset with a preset new energy training equipment parameter database to generate positioning point data.

[0020] The distance calculation unit is used to determine the shortest straight-line distance between the student and the new energy training equipment based on the spatial coordinate information of the new energy training equipment and the student terminal IP address in the positioning point data.

[0021] The course recommendation unit is used to retrieve the job skills training and assessment course database based on the shortest straight-line distance in space, and sort the retrieval results according to the equipment relevance weight to obtain a list of training program courses.

[0022] In a further implementation, the computing resource dynamic scheduling module includes a computing cluster unit composed of multiple cloud rendering servers, a resource pool unit, and a dynamic scheduling unit;

[0023] The resource pool unit is used to virtualize and integrate the physical server cluster in the computing cluster unit according to the pre-acquired server performance parameters to form a dynamically allocable resource pool.

[0024] The dynamic scheduling unit is used to generate an initial resource allocation scheme based on the historical training scale and the resource demand matrix of a single course through an iterative optimization algorithm. It also monitors the number of people waiting in the course reservation queue and the current resource utilization rate in real time according to the initial resource allocation scheme. When it detects that the current resource utilization rate exceeds the resource pool capacity, it reallocates resources according to the course weight priority and sends an elastic expansion instruction to the administrator.

[0025] In a further implementation, the human-environment-equipment data integration module includes a data integration unit, a data mapping unit, a data update unit, and a large-scale intelligent model unit;

[0026] The data integration unit is used to collect basic information of trainees, job curriculum system, course learning status data, new energy training equipment parameters and training environment parameters to form a structured dataset.

[0027] The data mapping unit is used to update and integrate the dynamic background data and student learning-related information involved in the job training and assessment courses in real time according to the structured dataset, and generate standardized long text questions based on the new energy training scenario.

[0028] The large model intelligent unit is used to analyze the standardized long text problem and generate a new energy training solution.

[0029] The data update unit is used to update all information related to people, equipment, and environment involved in the job training and assessment courses according to the new energy training solution.

[0030] In a further implementation plan, the job skills training and assessment module includes a job course pool unit, a peripheral compatibility unit, and a learning evaluation unit;

[0031] The job course pool unit is connected to the cloud rendering central platform system and is used to store the new energy training basic course data set based on the distributed database architecture, add different job tags to the new energy training basic course data set, and set dependency unlocking relationships for new energy training basic courses with the same job tags.

[0032] The learning evaluation unit is used to calculate the student's learning activity, learning points and ranking based on the student's learning behavior data in each learning stage, and to generate personalized learning evaluation data including ability deficiency analysis based on the learning behavior data, learning activity, learning points and ranking.

[0033] The peripheral compatibility unit establishes a two-way data channel with the third-party training equipment and the cloud rendering central platform. The peripheral compatibility unit is used to automatically load the corresponding new energy training equipment driver according to the communication protocol type of the third-party training equipment, and communicate with the cloud rendering central platform system to transmit the data obtained from the third-party training equipment to the cloud rendering central platform system.

[0034] In a further implementation scheme, the new energy training basic course dataset includes structured course resources, virtual simulation programs, and peripheral driver components;

[0035] The structured course resources include digital files of new energy training principle courseware, new energy training equipment operation videos, and new energy training procedure test papers.

[0036] The virtual simulation program includes virtual simulation software for new energy training processes and simulators for non-standard training scenarios.

[0037] The peripheral driver components include dynamic link library files and automated deployment scripts required by the training software of third-party training equipment.

[0038] In a further implementation, the virtual-real fusion module includes a data coupling unit, a force feedback operation unit, and a virtual-real update unit;

[0039] The data coupling unit is used to establish a data communication link with the on-site training equipment through an industrial communication protocol, and to acquire the operating parameters of the new energy training equipment, training environment parameters, and new energy training indicators in real time, thereby forming on-site training equipment operating data.

[0040] The force feedback operation unit is used to generate force feedback operation data that is synchronously updated with the actual operation training scenario based on the operation data of the on-site training equipment.

[0041] The virtual-real update unit is used to dynamically update the basic course content of new energy training based on the operating data of the on-site training equipment and the current virtual scene parameters.

[0042] Secondly, the present invention provides a new energy intelligent training and assessment method based on cloud rendering, the method comprising the following steps:

[0043] Based on the course learning request data uploaded by the student's terminal device and the student's access terminal type, the graphics computing resources are virtualized and integrated to generate a computing resource allocation scheme.

[0044] Based on the aforementioned computing resource allocation scheme and course type characteristics, different transmission strategies are used to transmit the virtual simulation environment rendering screen to the front-end interface of the student's terminal device.

[0045] Based on the course interaction data and learning behavior data of trainees in the virtual simulation environment, the operation standardization of new energy training parameters, the standard parameter compliance rate, and abnormal working condition handling data are quantitatively evaluated to generate personalized learning evaluation data that includes the identification of capability gaps.

[0046] Based on the personalized learning assessment data and the real-time collected on-site training equipment operation data, the virtual training environment is integrated with the real-world operational training scenario through a data coupling engine to generate force feedback operation data that is updated synchronously with the real-world operational training scenario.

[0047] The force feedback operation data is fed back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme.

[0048] This invention provides a cloud-rendered intelligent training and assessment system and method for new energy. The system includes a cloud rendering central platform system for virtualizing and integrating graphics computing resources based on course learning request data uploaded by student terminal devices and the type of student access terminal, generating a computing resource allocation scheme, and transmitting the virtual simulation environment rendered screen to the front-end interface of the student terminal device using different transmission strategies based on the computing resource allocation scheme and course type characteristics; a job skills training and assessment module for quantitatively evaluating the standardization of new energy training parameter operation, standard parameter compliance rate, and abnormal working condition handling data based on student course interaction data and student learning behavior data in the virtual simulation environment, generating personalized learning assessment data including the identification of competency gaps; and a virtual-real fusion module for fusing the virtual training environment with real-world operational training scenarios through a data coupling engine based on the personalized learning assessment data and real-time collected on-site training equipment operation data, generating force feedback operation data that is updated synchronously with the real-world operational training scenarios, and feeding the force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme. Compared with existing technologies, this system achieves a new energy training system that seamlessly integrates virtual and real work scenarios through the collaborative work of the cloud rendering central platform system, the job skills training and assessment module, and the virtual-real fusion module, significantly improving the utilization efficiency and accurate allocation of cloud rendering platform resources. Attached Figure Description

[0049] Figure 1 This is a block diagram of a cloud-rendered intelligent training and assessment system for new energy provided in an embodiment of the present invention;

[0050] Figure 2 This is a specific embodiment of the new energy intelligent training and assessment system provided in this invention.

[0051] Figure 3 This is a schematic diagram of the student spatial positioning course recommendation process provided in an embodiment of the present invention;

[0052] Figure 4 This is a schematic diagram of the personalized learning assessment data generation process provided in an embodiment of the present invention;

[0053] Figure 5 This is a schematic diagram of the cloud-rendered intelligent training and assessment method for new energy provided in an embodiment of the present invention. Detailed Implementation

[0054] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0055] refer to Figure 1 This invention provides a cloud-rendered intelligent training and assessment system for new energy, such as... Figure 1 As shown, the system includes a cloud rendering central platform system 101, a job skills training and assessment module 102, and a virtual-real fusion module 103.

[0056] In some implementations, the cloud rendering hub platform system is used to virtualize and integrate graphics computing resources based on course learning request data uploaded by student terminal devices and the type of student access terminal, generate a computing resource allocation scheme, and, based on the computing resource allocation scheme and course type characteristics, transmit the virtual simulation environment rendering screen to the front-end interface of the student terminal device using different transmission strategies, such as... Figure 2 As shown, the cloud rendering central platform system includes a dynamic scheduling module for computing resources, a cloud rendering course image transmission module, a human-environment-device data integration module, and a spatial computing course recommendation module. Specific functional descriptions are as follows:

[0057] The spatial computing course recommendation module is used to recommend a list of training programs with the shortest straight-line distance based on the student's geographical location information or spatial distribution information, and to generate course learning request data when the student selects a target training program based on all the training program course lists.

[0058] The computing resource dynamic scheduling module is used to identify the student access terminal type based on the course learning request data uploaded by the student terminal device, and dynamically schedule graphics computing resources from the server resource pool to run various basic courses based on the student access terminal type and the computing resource requirements of various basic courses, and generate a computing resource allocation scheme.

[0059] The cloud rendering course image transmission module is used to perform streaming media encoding processing on the rendering screen of the virtual simulation environment according to the characteristics of the course type to be learned, generate an encrypted rendering data stream, and transmit the encrypted rendering data stream to the front-end interface of the student's terminal device through the computing resource allocation scheme.

[0060] The Human-Environment-Equipment Data Integration Module is used to update and integrate structured knowledge in real time the dynamic background data and learner learning-related information involved in job training and assessment courses.

[0061] Specifically, when the cloud rendering central platform system receives learning requests from students for different basic courses, this embodiment first intelligently identifies the type of access terminal used by the student. Then, the system flexibly allocates corresponding computing resources from the server resource pool to run the course. The courses can include diverse basic courses such as courseware presentations, video tutorials, online exams, virtual simulation training, and hands-on practice with third-party peripherals. Each course content is equipped with a specific instantiation script to ensure efficient operation. Real-time footage during course execution is transmitted over the network to the front-end interface of the student's terminal device. After the student completes various interactive operations, the system immediately collects relevant human-environment-device course learning data. This data is then synchronized, filtered, integrated, and analyzed to generate evaluation results, guiding students to strengthen their learning of courses they are weak in. The system can also quickly respond to student questions at any time, providing immediate Q&A services to comprehensively improve students' learning efficiency and experience. Based on the above embodiment, in a more specific embodiment, the spatial computing course recommendation module includes a spatial positioning unit, a distance calculation unit, and a course recommendation unit. The specific functions of each unit in the spatial computing course recommendation module are as follows:

[0062] The spatial positioning unit is used to extract the instrument feature information, parameter identification information and operating environment information of the new energy training equipment from the current environment information uploaded by the student terminal device, generate a structured equipment positioning feature dataset, and associate and match the structured equipment positioning feature dataset with the preset new energy training equipment parameter database to generate positioning point data;

[0063] The distance calculation unit is used to determine the shortest straight-line distance between the student and the new energy training equipment based on the spatial coordinate information of the new energy training equipment in the positioning point data and the student terminal IP address.

[0064] The course recommendation unit is used to retrieve job skills training and assessment course databases based on the shortest straight-line distance in space, and sort the search results according to the equipment relevance weight to obtain a list of training program courses.

[0065] Specifically, the spatial computing course recommendation module establishes bidirectional communication connections with the job skills training and assessment module and the human-environment-equipment data integration module through data interfaces. For example, when searching for relevant courses, the course recommendation unit can directly access the course database in the job skills training and assessment module. Figure 3As shown, when trainees access the cloud rendering hub platform system using terminal devices, they can upload photos, videos, or text information about their current location via their terminal devices (such as mobile phones, tablets, or computers). Alternatively, the system can obtain the trainee's geographical location information through the terminal device's built-in positioning function (such as GPS, Wi-Fi positioning, etc.) or IP address resolution. The spatial positioning unit performs image recognition and text analysis on the photos, videos, or text information of the current location uploaded by the trainee. It identifies the characteristic information, parameter identification information, and operating environment information of the new energy training equipment from the photos, videos, and text. The identified information is saved as a structured equipment positioning feature dataset. The structured equipment positioning feature dataset is then matched with a preset new energy training equipment parameter database to generate positioning point data. The positioning point data includes the spatial coordinate information of the new energy training equipment, the parameters of the new energy training equipment, and the human-environment-equipment association data of the new energy training equipment's associated courses. For example, after a successful match, it is determined that the new energy training equipment near the trainee's location is a specific model of new energy training performance testing instrument, and its spatial coordinates are (X, Y, Z).

[0066] Then, the distance calculation unit reads the positioning point data generated by the spatial positioning unit, and simultaneously reads the IP address and positioning information of the student's terminal device (which can be obtained through IP address resolution). Combining the positioning point data, the student's terminal IP address, and positioning information, it calculates the shortest straight-line distance between the new energy training equipment and the student's location. The course recommendation unit retrieves relevant courses from the job skills training and assessment module based on the shortest straight-line distance provided by the distance calculation unit. The search results are sorted according to the device relevance weight to form a training project course list. That is, the courses with higher relevance to the new energy training equipment near the student's current location are ranked higher. The sorted course list is displayed to the student in a way that highlights or sorts by priority. The student can view the recommended course list on the access page and select the target training project course according to their personal interests and needs. After selecting the target training project course, the system generates course learning request data containing information such as course ID, student ID, and student terminal type, and sends the course learning request data to the computing resource dynamic scheduling module.

[0067] Based on the above embodiments, in a more specific embodiment, the computing resource dynamic scheduling module includes a computing cluster unit composed of multiple cloud rendering servers, a resource pool unit, and a dynamic scheduling unit. The specific functions of each unit of the computing resource dynamic scheduling module are as follows:

[0068] The resource pool unit is used to virtualize and integrate the physical server cluster in the computing cluster unit according to the pre-acquired server performance parameters to form a dynamically allocable resource pool.

[0069] The dynamic scheduling unit generates an initial resource allocation scheme based on historical training scale and single-course resource demand matrix through an iterative optimization algorithm. It also monitors the number of people waiting in the course reservation queue and the current resource utilization rate in real time according to the initial resource allocation scheme. When it detects that the current resource utilization rate exceeds the resource pool capacity, it reallocates resources according to course weight priority and sends an elastic expansion command to the administrator.

[0070] Specifically, due to the fragmented, unpredictable, and random nature of computing resource scheduling demands and peripheral device access, this embodiment can pre-set the server computing resource capacity for various basic courses. After receiving learning requests submitted by students and their possible third-party peripheral device input data, computing resources are called from the pre-set resource pool. After a fixed interval, the student's learning progress is evaluated, and the server computing resource capacity is reallocated. The computing resource dynamic scheduling module parses the student terminal type information in the course learning request data to determine the type of device used by the student. Based on the type of the selected course (such as basic new energy training, advanced testing technology, etc.) and the predefined amount of graphics computing resources (such as CPU, GPU, memory, etc.) required for each course, the module allocates the computing resources accordingly. The system dynamically allocates corresponding graphics computing resources from the server resource pool to ensure the smooth operation of selected courses and generates a computing resource allocation scheme containing information such as resource ID, allocation time, and estimated usage duration. In this embodiment, the computing cluster unit is a server cluster composed of multiple cloud rendering servers. Each server is equipped with computing resources such as large-capacity CPU, high-performance GPU, sufficient memory, and high-speed network bandwidth. Based on the pre-obtained server performance parameters (such as CPU frequency, GPU computing power, memory size, and network bandwidth), the computing resources in the server cluster are virtualized, mapped, and merged through the resource pool unit to form a dynamically allocable resource pool, which facilitates unified allocation and scheduling by the dynamic scheduling unit.

[0071] The dynamic scheduling unit collects the computing resource requirements of existing basic courses. For example, a courseware presentation course requires relatively low GPU computing power and moderate network bandwidth, while a virtual simulation course requires high GPU computing power and large memory capacity. The dynamic scheduling unit constructs dynamic scheduling relationships based on these existing basic course computing resource requirements, that is, the correspondence between the types of computing resources virtualized by the resource pool unit and various basic courses. In this embodiment, it is assumed that the set of computing resources virtualized by the resource pool unit is X = {x1, x2, ..., x...}. p The number of basic courses that can be run online simultaneously is S = {s1, s2, ..., s}. r There are} types, and the requirements for each course are A = {a1, a2, ... a}. mIn this embodiment, an initial resource pool S1 = M0(X, S0, A, K) is preset, where S0 represents the number of basic courses of various types that can be run online simultaneously, as assessed by the administrator based on the number of trainees and the distribution of training courses; M0(*) is the initial planning algorithm of the resource scheduling unit, which iteratively optimizes the initial parameters to obtain the number of courses S1 that the computing cluster unit supports running online simultaneously, which is close to S0; K represents a preset course category weight table to ensure a lower limit on the number of basic courses of each type that can run simultaneously.

[0072] The dynamic scheduling unit constructs dynamic scheduling relationships based on the computing resource requirements of existing basic courses. This involves determining the types and quantities of virtualized computing resources in the resource pool unit, as well as the number of various basic courses that can run simultaneously online. Based on the administrator's assessment of the number of various basic courses running simultaneously online, and according to S0 and course requirements A, the initial planning algorithm M0(*) iteratively optimizes to obtain the number of courses S1 that the computing cluster unit can support running simultaneously online, which is close to S0. The initial planning algorithm can find the number of courses S1 that is closest to S0 under the existing computing cluster resources by simulating different resource allocation combinations. Simultaneously, a preset course category weight table K is introduced to ensure a lower limit on the number of simultaneous runs for each type of basic course. For example, important virtual simulation courses have higher weights to ensure that at least a certain number of instances can run simultaneously. In this embodiment, based on the optimized S1, an initial resource allocation scheme is generated. This initial resource allocation scheme includes the number of computing resources allocated to each course and the corresponding server resources. For example, a certain number of CPU cores and network bandwidth are allocated to courseware display courses, while more GPU computing power and memory are allocated to virtual simulation courses.

[0073] After the dynamic scheduling unit starts the course learning service according to the initial resource allocation plan, it monitors the number of people waiting in the course reservation queue in real time. During the real-time monitoring phase, when it detects that the current resource utilization exceeds the resource pool capacity (for example, CPU utilization reaches over 90% and memory utilization is close to saturation), it reallocates resources according to course weight priority. Based on the preset course category weight table K, it prioritizes the resource needs of important courses (such as virtual simulation courses) and appropriately reclaims some resources from other non-critical courses (such as courseware presentation courses). At the same time, between adjacent reservation time periods, it collects the reservation quantity W of various basic courses for the next reservation time period. n ={s1, s2, ..., s q}, based on the reservation demand for the next reservation period and the demand for each course, B = {b1, b2, ..., b} t} types of computing resources, using dynamic programming algorithms to calculate the computing cluster unit's support for W n The number of courses running online simultaneously (S) is similar. nIf S can be solved using dynamic programming algorithm n Then according to S n If the dynamic programming algorithm cannot solve for S when readjusting resource allocation... n This indicates that due to insufficient resources, all reservation requests cannot be met. Therefore, a resource pool with sufficient resources is set up according to the preset course category weight table for students to make reservations. For reservation requests exceeding the limit, the administrator will be prompted to urgently expand the computing cluster unit. If the expansion fails, the reservations will be canceled in the order of reservation and the corresponding students will be reminded to make reservations for a later time slot.

[0074] Based on the above embodiments, in a more specific embodiment, the cloud rendering course image transmission module is connected to the job skills training and assessment module. When a student initiates a learning request on the front-end interface of the cloud rendering central platform system, the cloud rendering course image transmission module receives the request and determines the course type selected by the student by analyzing the course identification information in the request. For example, if the student selects a courseware course on the structure and principle of new energy batteries, the module identifies it as a courseware type; if the student selects a virtual simulation course on troubleshooting new energy batteries, it identifies it as a virtual simulation course. Then, based on the identified course type, the corresponding image transmission method is determined. When the course type to be learned is a theoretical course, the encrypted rendering is transmitted through a remote encrypted access mode. The data stream is transmitted to the front-end interface of the student's terminal device. When the course type to be learned is a practical course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through cloud rendering mode. When the course type to be learned is an assessment course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through front-end encrypted display mode. For example, for courseware and video courses, the cloud rendering course image transmission module adopts a remote encrypted access method; for exam paper courses, the student's terminal device front-end encrypted display method is adopted to ensure that the exam paper content is not leaked; peripheral training courses and virtual simulation courses require real-time rendering and high-performance computing resources, therefore, cloud rendering is used to remotely transmit images.

[0075] The cloud rendering hub platform activates the corresponding virtual simulation environment based on the students' learning needs. The environment rendering screen is captured in real time using a specific capture tool. The captured rendering screen is processed by a streaming media encoder, which encodes continuous image frames into streaming media formats (such as H.264, H.265, etc.) to reduce transmission bandwidth and improve playback smoothness. After the streaming media encoding is completed, the generated rendering data stream is encrypted using an encryption algorithm (such as AES encryption algorithm) to generate an encrypted rendering data stream, ensuring data security during transmission. The encryption key is uniformly managed by the cloud rendering hub platform system and transmitted to the student's terminal device through a secure channel. In this embodiment, the priority and bandwidth allocation of data transmission are determined according to the computing resource allocation scheme generated by the computing resource dynamic scheduling module. For example, for virtual simulation courses with higher priority, it is ensured that their encrypted rendering data stream can obtain sufficient network bandwidth during transmission to ensure smooth transmission. The encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through network transmission protocols and data distribution strategies. Students can watch the rendering screen of the virtual simulation environment in real time and conduct interactive learning and training.

[0076] Based on the above embodiments, in a more specific embodiment, the human-environment-device data integration module includes a data integration unit, a data mapping unit, a data update unit, and a large-scale intelligent model unit. The specific functions of each unit of the human-environment-device data integration module are as follows:

[0077] The data integration unit is used to collect basic information of trainees, job curriculum system, course learning status data, new energy training equipment parameters and training environment parameters to form a structured dataset.

[0078] The data mapping unit is used to update and integrate the dynamic background data and student learning-related information involved in the job training and assessment courses in real time according to the structured dataset, and generate standardized long text questions based on the new energy training scenario.

[0079] The large model intelligent unit is used to analyze the standardized long text problem and generate a new energy training solution.

[0080] The data update unit is used to update all information related to people, equipment, and environment involved in the job training and assessment courses according to the new energy training solution.

[0081] Specifically, the human-environment-equipment data integration module is connected to the job skills training and assessment module. This module mainly comprises four core units: a data integration unit, a data mapping unit, a data update unit, and a large-scale intelligent model unit. The data update unit, connected to the job skills training and assessment module, receives real-time data on course updates, student learning progress, and changes in the work environment. Simultaneously, it checks and updates all background information on the people, equipment, and environment involved in the job training and assessment courses to ensure accuracy and timeliness. The data mapping unit pre-establishes mapping relationships between job training and assessment courses and related information, including associations between courses and equipment, courses and the work environment, and courses and student positions. When a student begins learning a job training and assessment course, the data mapping unit quickly retrieves all information associated with that course based on the course identifier, providing a foundation for subsequent data integration and analysis. The data integration unit, based on student basic information, job curriculum system, course learning progress, and work environment, organizes the information into several long text questions, for example, generating the following long text questions:

[0082] [Hints for constraining the large model's capabilities], Student H is currently studying the new energy battery repair course J, the student's learning status is Q, and the work background, equipment, and process knowledge involved is V. Please summarize the student's current learning status, [Hints for constraining the large model's capabilities].

[0083] [Hints for constraining large model capabilities], Student basic information H is currently learning the new energy battery repair course J, the student's learning status is Q, and the work background, equipment, and process knowledge involved is V. Please analyze the reasons for the student's errors and provide improvement suggestions. [Hints for constraining large model capabilities].

[0084] [Hints for constraining large model capabilities]: Student basic information H is currently studying the new energy battery repair course J. The student's learning progress is Q, and the knowledge of the work background, equipment, and processes involved is V. If 0 to 100 points represent the student's ability and qualities to perform the job, please accurately evaluate the student's current score and provide the evaluation basis. [Hints for constraining large model capabilities].

[0085] [Hints for constraining large model capabilities]: Student basic information H is currently studying the new energy battery repair course J. The student's learning status is Q. The work background, equipment, and process knowledge involved is V. The student is currently encountering problem T. Please help me analyze the possible causes of problem T, which courses need to be studied, and provide all possible methods to solve the problem. [Hints for constraining large model capabilities].

[0086] After receiving the long text question generated by the data integration unit, the large model intelligent unit first performs semantic analysis and extracts key information from the question. For example, for a question about summarizing students' learning progress, the large model intelligent unit extracts key information such as students' basic information, course learning progress, and relevant work background, and conducts in-depth analysis and answers to the long text question. The output of the large model intelligent unit will be directly fed back to students or instructors to help them better understand their learning progress, identify problems, and seek solutions.

[0087] In some implementations, the job skills training and assessment module is used to quantitatively evaluate the operational standardization of new energy training parameters, the compliance rate of standard parameters, and the handling data of abnormal operating conditions based on the course interaction data and student learning behavior data in the virtual simulation environment, and to generate personalized learning assessment data that includes the identification of competency gaps. In this embodiment, the job skills training and assessment module includes a job course pool unit, a peripheral compatibility unit, and a learning evaluation unit, and the specific functions are described below:

[0088] The job course pool unit is connected to the cloud rendering central platform system and is used to store the new energy training basic course data set based on the distributed database architecture, add different job tags to the new energy training basic course data set, and set dependency unlocking relationships for new energy training basic courses with the same job tags.

[0089] The learning evaluation unit is used to calculate the student's learning activity, learning points and ranking based on the student's learning behavior data in each learning stage, and to generate personalized learning evaluation data including ability deficiency analysis based on the learning behavior data, learning activity, learning points and ranking.

[0090] The peripheral compatibility unit establishes a two-way data channel with the third-party training equipment and the cloud rendering central platform. The peripheral compatibility unit is used to automatically load the corresponding new energy training equipment driver according to the communication protocol type of the third-party training equipment, and communicate with the cloud rendering central platform system to transmit the data obtained from the third-party training equipment to the cloud rendering central platform system.

[0091] Specifically, the job skills training and assessment module is connected to the cloud rendering course image transmission module, the computing resource dynamic scheduling module, and the human-environment-equipment data integration module. The job course pool unit uses a distributed database architecture to store the basic new energy training course data set. This architecture can efficiently process large amounts of course data and ensure high availability and scalability. The basic new energy training course data set includes structured course resources, virtual simulation programs, and peripheral driver components. The structured course resources include digital files of new energy training principle courseware (PDF format), new energy training equipment operation videos, and new energy training procedure test papers. The virtual simulation programs include new energy training process virtual simulation software and non-standard training scenario simulators. The peripheral driver components include dynamic link library files and automated deployment scripts required by third-party training equipment training software. Specifically, these basic new energy training course datasets include courseware, videos, test papers, process virtual simulation software, third-party peripheral training software, scenario simulator software, etc. Administrators will add different job tags to various basic courses. For example, for new energy battery repair positions, job tags such as "Battery Repair Basics," "Battery Performance Testing," and "Battery Troubleshooting" will be set. Dependency unlocking relationships will be set for courses with the same job tags, so that students must complete the previous stage of the course before they can unlock the next stage. For example, the system will only unlock the "New Energy Production Operation Process" course after the student completes the "Basic Knowledge of New Energy Production" course. The unlocking conditions for the next stage of basic courses are only met after the previous stage of basic courses is completed, thus avoiding knowledge gaps caused by skipping learning.

[0092] The trainee terminal device is equipped with a peripheral compatibility unit program, namely a data transceiver plugin. This plugin can establish an interactive input data communication relationship with third-party peripherals and the cloud rendering central platform system. When trainees use third-party training equipment for training, the peripheral compatibility unit automatically loads the corresponding new energy training equipment driver according to the communication protocol type of the third-party training equipment. After loading the driver, the peripheral compatibility unit communicates with the cloud rendering central platform system, so that the data obtained by the trainees during the training process can be transmitted to the cloud rendering central platform system in real time for subsequent analysis and evaluation.

[0093] like Figure 4As shown, the learning evaluation unit is a key component of the job skills training assessment module. Based on the learners' learning behavior data at each learning stage, it calculates their learning activity, learning points, and ranking. This data not only reflects the learners' learning progress and performance but also provides a foundation for subsequent personalized learning assessments. To generate personalized learning assessment data that includes competency gap analysis, the learning evaluation unit utilizes the data integration unit and the large-scale model intelligent unit to generate real-time learning reports for each learner. The data integration unit is responsible for integrating learners' course interaction data, learning behavior data, and data on the standardization of new energy training parameters, the compliance rate of standard parameters, and the handling of abnormal operating conditions in the virtual simulation environment. The large-scale model intelligent unit then processes this data... In-depth analysis is conducted to identify learners' skill gaps and generate personalized learning suggestions. For example, it may point out that learners have skill gaps in handling a specific type of battery fault diagnosis, such as improper operation or inaccurate parameter settings. In summary, when generating personalized learning assessment data, the learning evaluation unit will comprehensively consider learners' learning activity, learning points, performance ranking, and the analysis results of the large model intelligent unit to obtain a comprehensive learning report that shows learners' learning progress, performance, skill gaps, and improvement suggestions. Administrators can view all learners' learning reports through the system backend to understand the overall learning situation and existing problems. Learners can also view their own learning reports at any time to understand their learning situation and make adjustments and improvements based on the suggestions.

[0094] In some implementations, the virtual-real fusion module is used to fuse the virtual training environment with the real-world training scenario based on the personalized learning assessment data and real-time collected on-site training equipment operation data through a data coupling engine. This generates force feedback operation data that is synchronously updated with the real-world training scenario, and feeds this force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme. In this embodiment, the virtual-real fusion module includes a data coupling unit, a force feedback operation unit, and a virtual-real update unit, the specific functions of which are described below:

[0095] The data coupling unit is used to establish a data communication link with the on-site training equipment through an industrial communication protocol, and to acquire the operating parameters of the new energy training equipment, training environment parameters, and new energy training indicators in real time, thereby forming on-site training equipment operating data.

[0096] The force feedback operation unit is used to generate force feedback operation data that is synchronously updated with the actual operation training scenario based on the operation data of the on-site training equipment.

[0097] The virtual-real update unit is used to dynamically update the basic course content of new energy training based on the operating data of the on-site training equipment and the current virtual scene parameters.

[0098] Specifically, the data coupling unit is software that runs on the cloud rendering platform and can only be operated by the administrator. The data coupling unit establishes a stable communication link with the on-site new energy training equipment through industrial communication protocols. It can integrate the virtual training environment with the real-world training scenario through data synchronization and scene mapping technologies. The administrator can configure parameters such as device address and communication protocol on the data coupling interface to ensure smooth data transmission. Then, in this embodiment, the data coupling unit parses the data received from the on-site training equipment in real time, extracts the operating parameters of the new energy training equipment, training environment parameters, and new energy training indicators, and organizes these data into on-site training equipment operating data. At the same time, the data coupling unit can integrate the received personalized learning evaluation data with the on-site training equipment operating data. Through data cleaning, data transformation, and data association technologies, the data in the virtual training environment and the data in the real-world training scenario are matched and complemented.

[0099] The force feedback operation unit relies on force feedback virtual simulation software loaded on the cloud rendering central platform system and force feedback devices connected to the software. Trainee terminal devices interact with the cloud rendering platform through force feedback devices (such as force feedback handles or gloves). Based on the operational data of the on-site training equipment, the force feedback operation unit generates force feedback operation data that is synchronously updated with the actual training scenario. For example, when the on-site training equipment detects the reaction force generated by the battery during repair, the force feedback operation unit converts the magnitude and direction of this force into force feedback simulation data. The force feedback virtual simulation software then uses this data to control the force feedback device to simulate the corresponding operating feel. For instance, trainees use force feedback in the virtual simulation environment... When simulating battery repair using a handheld controller, the controller applies corresponding force to the trainee's hand based on force feedback data from the virtual scene, allowing the trainee to experience a realistic repair operation feel. This force feedback data is updated synchronously with the real-world training scenario, ensuring that the trainee experiences the same operational feel as in a real environment. In this embodiment, the generated force feedback data is transmitted over the network to the cloud rendering central platform system. After receiving this data, the cloud rendering central platform system can use it as new input to analyze factors such as the complexity of the trainee's operations in the virtual simulation environment and the real-time requirements of the force feedback data. It then re-evaluates the current computing resource allocation scheme, and the dynamic scheduling module for computing resources iteratively optimizes the allocation scheme based on the feedback force feedback data. By analyzing the trainee's operations in the virtual simulation environment and the real-time requirements of the force feedback data, the allocation of computing resources is adjusted to better meet training needs.

[0100] The Virtual-Real Update Unit is software that runs continuously on the cloud rendering platform and is only operable by the administrator. The administrator logs into the management interface of the cloud rendering platform through a secure authentication mechanism and starts the Virtual-Real Update Unit. The Virtual-Real Update Unit virtualizes the on-site work environment, constructing a virtual simulation scene consistent with the real environment. Each virtual simulation course is a partial work process of this virtual scene. When there is a discrepancy between the curriculum system of the cloud rendering central platform system and the on-site work environment, the Virtual-Real Update Unit initiates the curriculum system update process. For example, if a battery testing device is added to the on-site work environment, but the cloud rendering platform's curriculum system does not have corresponding training content, the Virtual-Real Update Unit analyzes the on-site training device's operating data and the current virtual scene parameters to determine the course content that needs to be updated. Thus, it uses the large model intelligent unit to automatically update the basic new energy training course content in the virtual simulation course, ensuring that the training system is updated synchronously with the on-site work environment.

[0101] In summary, this embodiment significantly increases the number of online courses on the cloud rendering platform by automatically allocating computing resources for various basic courses within each scheduled time slot using a dynamic scheduling method. Furthermore, students can now access the cloud rendering platform to learn relevant courses using any configured terminal device and third-party peripherals, breaking traditional limitations such as venue size, equipment performance, and student free time, making learning more flexible and convenient. Secondly, by introducing a human-environment-device data integration module, this embodiment can provide unified text descriptions of students' learning progress in different job-related course systems and generate personalized learning analysis, evaluation, solutions, and suggestion reports in real time. These reports enable students to promptly identify and address learning gaps during theoretical knowledge, virtual simulation, and practical training, allowing for targeted reinforcement of weak skills. Finally, the spatial computing course recommendation module in this embodiment can analyze information uploaded by students (photos, videos, or text descriptions) when they encounter problems. The human-environment-device data integration module uses spatial positioning and distance calculation units to determine the student's position relative to the problem, thereby recommending relevant job-related course content for on-site learning and problem-solving.

[0102] This invention provides a cloud-rendered intelligent training and assessment system for new energy. The system includes a cloud rendering central platform system for virtualizing and integrating graphics computing resources based on course learning request data uploaded by trainees' terminal devices and the type of trainees' access terminals, generating a computing resource allocation scheme. A job skills training and assessment module is used to quantitatively evaluate the operational standardization of new energy training parameters, the pass rate of standard parameters, and abnormal operating condition handling data based on trainees' course interaction data and learning behavior data in the virtual simulation environment, generating personalized learning assessment data that includes the identification of skill gaps. A virtual-real fusion module is used to fuse the virtual training environment with real-world operational training scenarios through a data coupling engine, based on personalized learning assessment data and real-time collected on-site training equipment operation data, generating force feedback operation data that is synchronously updated with the real-world operational training scenarios, and feeding the force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme. Compared with existing technologies, this system, through the collaborative work of the cloud rendering central platform system, the job skills training and assessment module, and the virtual-real fusion module, achieves a seamless integration of virtual and real-world operational scenarios in the intelligent training and assessment system for new energy, significantly improving the utilization efficiency and accurate allocation of cloud rendering platform resources.

[0103] In one embodiment, such as Figure 5 As shown in the figure, this invention provides a new energy intelligent training and assessment method based on cloud rendering, the method including the following steps:

[0104] S1. Based on the course learning request data uploaded by the student's terminal device and the student's access terminal type, the graphics computing resources are virtualized and integrated to generate a computing resource allocation scheme;

[0105] S2. Based on the aforementioned computing resource allocation scheme and course type characteristics, different transmission strategies are used to transmit the virtual simulation environment rendering screen to the front-end interface of the student's terminal device;

[0106] S3. Based on the course interaction data and learning behavior data of trainees in the virtual simulation environment, quantitatively evaluate the standardization of new energy training parameters, the compliance rate of standard parameters, and the data on handling abnormal working conditions, and generate personalized learning evaluation data that includes the identification of capability gaps.

[0107] S4. Based on the personalized learning assessment data and the real-time collected on-site training equipment operation data, the virtual training environment is integrated with the real-world operational training scenario through a data coupling engine to generate force feedback operation data that is updated synchronously with the real-world operational training scenario;

[0108] S5. Feed the force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme.

[0109] It should be noted that the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0110] For specific limitations regarding a cloud-rendered intelligent training and assessment method for new energy, please refer to the above-described limitations regarding a cloud-rendered intelligent training and assessment system for new energy, which will not be repeated here. Those skilled in the art will recognize that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0111] This invention provides a cloud-rendered intelligent training and assessment method for new energy. The method virtualizes and integrates graphics computing resources based on course learning request data uploaded by the student's terminal device and the student's access terminal type to generate a computing resource allocation scheme. Based on the computing resource allocation scheme and course type characteristics, different transmission strategies are used to transmit the rendered virtual simulation environment to the front-end interface of the student's terminal device. Based on the student's course interaction data and learning behavior data in the virtual simulation environment, the method quantitatively evaluates the operational standardization of new energy training parameters, the standard parameter compliance rate, and abnormal operating condition handling data, generating personalized learning assessment data that includes the identification of capability gaps. Based on the personalized learning assessment data and real-time collected on-site training equipment operation data, a data coupling engine merges the virtual training environment with the real-world training scenario, generating force feedback operation data that is updated synchronously with the real-world training scenario. The force feedback operation data is fed back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme. Compared with existing technologies, this method achieves seamless integration of virtual and real-world scenarios in the intelligent new energy training and assessment system through virtualized resource allocation, personalized learning positioning, and virtual-real fusion, significantly improving the utilization efficiency and accurate allocation of cloud rendering platform resources.

[0112] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this invention, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A cloud-rendered intelligent training and assessment system for new energy, characterized in that, include: The cloud rendering central platform system is used to virtualize and integrate graphics computing resources based on the course learning request data uploaded by the student terminal device and the student access terminal type, generate a computing resource allocation scheme, and transmit the virtual simulation environment rendering screen to the front-end interface of the student terminal device using different transmission strategies based on the computing resource allocation scheme and course type characteristics. The job skills training and assessment module is used to quantitatively evaluate the standardization of new energy training parameters, the compliance rate of standard parameters, and the handling of abnormal working conditions based on the course interaction data and learning behavior data of trainees in the virtual simulation environment, and to generate personalized learning assessment data that includes the identification of competency gaps. The virtual-real fusion module is used to fuse the virtual training environment with the real-world training scenario through a data coupling engine based on the personalized learning assessment data and the real-time collected on-site training equipment operation data. This generates force feedback operation data that is updated synchronously with the real-world training scenario and feeds the force feedback operation data back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme. The cloud rendering central platform system includes a computing resource dynamic scheduling module, a cloud rendering course image transmission module, a human-environment-device data integration module, and a spatial computing course recommendation module. The spatial computing course recommendation module is used to recommend a list of training projects with the shortest straight-line distance based on the student's geographical location information or spatial distribution information. When the student selects a target training project from all the training project course lists, course learning request data is generated. The shortest straight-line distance between the student and the new energy training equipment is determined based on the spatial coordinate information of the new energy training equipment in the positioning point data and the student's terminal IP address. The computing resource dynamic scheduling module is used to identify the student access terminal type based on the course learning request data uploaded by the student terminal device, and dynamically schedule graphics computing resources from the server resource pool to run various basic courses based on the student access terminal type and the computing resource requirements of various basic courses, and generate a computing resource allocation scheme. The cloud rendering course image transmission module is used to perform streaming media encoding processing on the virtual simulation environment rendering screen according to the characteristics of the course type to be learned, generate an encrypted rendering data stream, and transmit the encrypted rendering data stream to the front-end interface of the student terminal device through the computing resource allocation scheme. The human-environment-equipment data integration module is used to update and integrate structured knowledge in real time the dynamic background data and student learning-related information involved in the job training and assessment courses.

2. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: When the course type to be learned is a theoretical course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through a remote encrypted access mode. When the course type to be learned is a practical course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through cloud rendering mode. When the course type to be learned is an assessment course, the encrypted rendering data stream is transmitted to the front-end interface of the student's terminal device through a front-end encrypted display mode.

3. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: The spatial computing course recommendation module includes a spatial positioning unit and a course recommendation unit; The spatial positioning unit is used to extract the instrument feature information, parameter identification information and operating environment information of the new energy training equipment from the current environment information uploaded by the student terminal device, generate a structured equipment positioning feature dataset, and associate and match the structured equipment positioning feature dataset with a preset new energy training equipment parameter database to generate positioning point data. The course recommendation unit is used to retrieve the job skills training and assessment course database based on the shortest straight-line distance in space, and sort the retrieval results according to the equipment relevance weight to obtain a list of training program courses.

4. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: The computing resource dynamic scheduling module includes a computing cluster unit composed of multiple cloud rendering servers, a resource pool unit, and a dynamic scheduling unit. The resource pool unit is used to virtualize and integrate the physical server cluster in the computing cluster unit according to the pre-acquired server performance parameters to form a dynamically allocable resource pool. The dynamic scheduling unit is used to generate an initial resource allocation scheme based on the historical training scale and the resource demand matrix of a single course through an iterative optimization algorithm. It also monitors the number of people waiting in the course reservation queue and the current resource utilization rate in real time according to the initial resource allocation scheme. When it detects that the current resource utilization rate exceeds the resource pool capacity, it reallocates resources according to the course weight priority and sends an elastic expansion instruction to the administrator.

5. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: The human-environment-equipment data integration module includes a data integration unit, a data mapping unit, a data update unit, and a large-scale intelligent model unit; The data integration unit is used to collect basic information of trainees, job curriculum system, course learning status data, new energy training equipment parameters and training environment parameters to form a structured dataset. The data mapping unit is used to update and integrate the dynamic background data and student learning-related information involved in the job training and assessment courses in real time according to the structured dataset, and generate standardized long text questions based on the new energy training scenario. The large model intelligent unit is used to analyze the standardized long text problem and generate a new energy training solution. The data update unit is used to update all information related to people, equipment, and environment involved in the job training and assessment courses according to the new energy training solution.

6. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: The job skills training and assessment module includes a job course pool unit, a peripheral compatibility unit, and a learning evaluation unit. The job course pool unit is connected to the cloud rendering central platform system and is used to store the new energy training basic course data set based on the distributed database architecture, add different job tags to the new energy training basic course data set, and set dependency unlocking relationships for new energy training basic courses with the same job tags. The learning evaluation unit is used to calculate the student's learning activity, learning points and ranking based on the student's learning behavior data in each learning stage, and to generate personalized learning evaluation data including ability deficiency analysis based on the learning behavior data, learning activity, learning points and ranking. The peripheral compatibility unit establishes a two-way data channel with the third-party training equipment and the cloud rendering central platform. The peripheral compatibility unit is used to automatically load the corresponding new energy training equipment driver according to the communication protocol type of the third-party training equipment, and communicate with the cloud rendering central platform system to transmit the data obtained from the third-party training equipment to the cloud rendering central platform system.

7. The new energy intelligent training and assessment system based on cloud rendering as described in claim 6, characterized in that: The new energy training basic course dataset includes structured course resources, virtual simulation programs, and peripheral driver components. The structured course resources include digital files of new energy training principle courseware, new energy training equipment operation videos, and new energy training procedure test papers. The virtual simulation program includes virtual simulation software for new energy training processes and simulators for non-standard training scenarios. The peripheral driver components include dynamic link library files and automated deployment scripts required by the training software of third-party training equipment.

8. The new energy intelligent training and assessment system based on cloud rendering as described in claim 1, characterized in that: The virtual-real fusion module includes a data coupling unit, a force feedback operation unit, and a virtual-real update unit. The data coupling unit is used to establish a data communication link with the on-site training equipment through an industrial communication protocol, and to acquire the operating parameters of the new energy training equipment, training environment parameters, and new energy training indicators in real time, thereby forming on-site training equipment operating data. The force feedback operation unit is used to generate force feedback operation data that is synchronously updated with the actual operation training scenario based on the operation data of the on-site training equipment. The virtual-real update unit is used to dynamically update the basic course content of new energy training based on the operating data of the on-site training equipment and the current virtual scene parameters.

9. A new energy intelligent training and assessment method based on cloud rendering, characterized in that, Applied to any one of claims 1-8, in a new energy intelligent training and assessment system based on cloud rendering, the method includes the following steps: Based on the course learning request data uploaded by the student's terminal device and the student's access terminal type, the graphics computing resources are virtualized and integrated to generate a computing resource allocation scheme. Based on the aforementioned computing resource allocation scheme and course type characteristics, different transmission strategies are used to transmit the virtual simulation environment rendering screen to the front-end interface of the student's terminal device. Based on the course interaction data and learning behavior data of trainees in the virtual simulation environment, the operation standardization of new energy training parameters, the standard parameter compliance rate, and abnormal working condition handling data are quantitatively evaluated to generate personalized learning evaluation data that includes the identification of capability gaps. Based on the personalized learning assessment data and the real-time collected on-site training equipment operation data, the virtual training environment is integrated with the real-world operational training scenario through a data coupling engine to generate force feedback operation data that is updated synchronously with the real-world operational training scenario. The force feedback operation data is fed back to the cloud rendering central platform system to trigger iterative optimization of the computing resource allocation scheme.

Images