Training system constructed based on operation and maintenance platform
By using a data acquisition module, a personalized learning engine, and a containerized virtual experimental environment, the problem of fixed teaching content in existing system operation and maintenance training has been solved, enabling personalized teaching and efficient experimental environment deployment, thereby improving training quality and efficiency.
Patent Information
- Application Number
- CN202511100643.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-07
- Publication Date
- 2025-11-07
AI Technical Summary
Existing system operation and maintenance training lacks a personalized adjustment mechanism, has low efficiency in deploying experimental environments, low resource utilization, and cannot meet the needs of large-scale concurrent training. Furthermore, the learning feedback mechanism is insufficient, affecting the quality and efficiency of training.
The system employs a data acquisition module to collect student learning behavior data, a personalized learning engine to calculate knowledge mastery, and recommends teaching content and experimental tasks; the virtual experimental environment is built using a containerization approach and provides interactive operation; and a feedback optimization mechanism dynamically adjusts learning strategies.
It enables personalized adaptation of teaching content and experimental tasks, improves training quality and student learning experience, enhances the deployment efficiency and resource utilization of the experimental environment, and supports large-scale concurrent training.
Smart Images
Figure CN120913469A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the technical field of training, and particularly relates to a training system based on an operation and maintenance platform. BACKGROUND
[0002] In the existing system operation and maintenance training scheme, fixed course content and unified experimental environment are usually adopted for teaching. Learners learn theoretical knowledge by watching videos, reading documents, etc., and complete operation practice combined with standardized experimental platforms. However, such methods have obvious limitations: on the one hand, the teaching content lacks individualized adjustment mechanism, and it is difficult to adapt to the learning rhythm and mastery level of different learners; on the other hand, the experimental environment is mostly based on physical equipment or full virtualization architecture, which has high resource occupation, long deployment period, and is not easy to expand, and cannot meet the needs of large-scale concurrent training.
[0003] In addition, the existing training system has a weak feedback mechanism for the learning process of learners, and it is difficult to evaluate their knowledge mastery in real time and dynamically optimize the teaching path. This leads to some learners losing confidence because the tasks are too difficult, and some learners lack challenges because the tasks are too simple, and the overall training effect is limited.
[0004] In summary, one of the main technical problems in the prior art is the lack of ability to dynamically adjust the teaching content and experimental difficulty according to individual differences of learners, while the experimental environment deployment efficiency is low and the resource utilization rate is not high, which affects the quality and efficiency of operation and maintenance training. SUMMARY
[0005] The purpose of the present application is to provide a training system based on an operation and maintenance platform, which solves the problems of fixed teaching content and low experimental environment deployment efficiency in the prior art, realizes the individualized adaptation and efficient execution of teaching content and experimental tasks in the operation and maintenance training process, and significantly improves the training quality and learner learning experience.
[0006] To achieve the above purpose, the present application adopts the following technical solution: a training system based on an operation and maintenance platform, comprising a data acquisition module, a personalized learning engine, a virtual experimental environment and a feedback optimization mechanism; the data acquisition module is used to collect learning behavior data of learners, including access frequency, correct answer rate and experimental completion time; the personalized learning engine calculates the knowledge mastery degree of learners based on the data of the data acquisition module, and recommends teaching content and experimental tasks accordingly; the virtual experimental environment is containerized, and is used to deploy operation and maintenance scenarios and provide interactive operation; the feedback optimization mechanism adjusts the personalized learning strategy according to the learning progress and historical data of learners.
[0007] Preferably, the data collection module generates learning behavior data by recording the correct rate of the answers of the trainee and the experiment completion time; the personalized learning engine calculates the knowledge mastery degree of the trainee according to the learning behavior data, and recommends matched teaching content based on the knowledge mastery degree.
[0008] Preferably, the personalized learning engine calculates the knowledge mastery degree of the trainee by a weighted scoring method; the knowledge mastery degree calculation formula is: Wherein, s j represents the mastery degree of the jth knowledge point, alpha j is the weight coefficient corresponding to the knowledge point, and m is the total number of knowledge points.
[0009] Preferably, the personalized learning engine adjusts the difficulty of the experiment task based on a difficulty adjustment factor; the difficulty adjustment factor calculation formula is: D = 1 + beta. (mu-s); wherein, mu is the average mastery rate of the knowledge point, s is the actual mastery rate of the trainee, and beta is an experience parameter.
[0010] Preferably, the virtual experiment environment adopts a lightweight containerization architecture, and realizes the isolation between containers through a namespace and a control group; the containerization architecture supports rapid deployment and operation and maintenance scenarios, and limits the influence of malicious operations on the host.
[0011] Preferably, the virtual experiment environment realizes automatic snapshot and rollback through copy-on-write; the snapshot allows the trainee to restore to the previous state when an operation is failed.
[0012] Preferably, the resource management module of the virtual experiment environment dynamically allocates available resources according to the resource demand of the container; the resource management module ensures that the total demand of the container resources does not exceed the total amount of the available resource pool.
[0013] Preferably, the feedback optimization mechanism calculates a reward value by a reinforcement learning method to optimize the learning path; the reward value is based on the change of the knowledge mastery degree before and after the trainee completes the experiment task, and is adjusted by a decay coefficient.
[0014] Preferably, further comprising a multi-dimensional evaluation system for measuring the learning achievements of the trainee; the multi-dimensional evaluation system comprises a theoretical knowledge mastery degree, an experimental operation proficiency and a comprehensive application ability.
[0015] Preferably, the multi-dimensional evaluation system generates learning achievement data through regular tests, experimental operation records and project-based learning tasks; the learning achievement data is used to adjust the teaching content and the experiment task.
[0016] The technical effects and advantages of the present application are as follows:
[0017] The present application realizes comprehensive recording of the learning behavior of students by introducing a data acquisition module, and calculates the knowledge mastery degree of the students according to the collected data by a personalized learning engine, thereby recommending matched teaching content and experimental tasks. This method can dynamically adjust the teaching strategy according to the learning progress and understanding ability of different students, and improve the pertinence and effectiveness of learning. The virtual experimental environment is constructed by using containerization technology, which supports rapid deployment of various operation and maintenance scenarios, and has good resource isolation and scalability, effectively reducing the maintenance cost and resource consumption of the experimental environment, and improving the response speed and concurrent processing capacity of the system. Through the above improvements, the present application solves the problems of fixed teaching content and low deployment efficiency of experimental environment in the prior art, realizes personalized adaptation and efficient execution of teaching content and experimental tasks in the operation and maintenance training process, and significantly improves the training quality and student learning experience. BRIEF DESCRIPTION OF DRAWINGS
[0018] Figure 1 The figure is a block diagram of a training system constructed based on an operation and maintenance platform. DETAILED DESCRIPTION
[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The specific embodiments described herein are only used to explain the present application, and are not used to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.
[0020] The present application provides a training system constructed based on an operation and maintenance platform as shown in Figure 1 The system uses artificial intelligence technology to drive the teaching process, combines personalized learning algorithms and virtual experimental operations, and realizes efficient practical training of intelligent system operation and maintenance management. The core goal of the system is to improve the practical ability of students in the field of system operation and maintenance, dynamically adjusts the teaching content and experimental difficulty by intelligently analyzing the behavior characteristics and knowledge mastery of learners, thereby providing highly customized learning experience. In addition, the system uses virtualization technology to simulate a real operation and maintenance environment, so that students can perform practical training in a safe and controllable environment, and improve their ability to deal with complex operation and maintenance problems.
[0021] Specifically, the training system constructed based on the operation and maintenance platform includes a data acquisition module, a personalized learning engine, a virtual experimental environment, and a feedback optimization mechanism.
[0022] The data collection module is configured to collect learning behavior data of the student, including access frequency, correct answer rate and experiment completion time; the data collection module is configured to record the correct answer rate and experiment completion time of the student to generate learning behavior data; the personalized learning engine is configured to calculate the knowledge mastery degree of the student based on the learning behavior data, and recommend matched teaching content based on the knowledge mastery degree.
[0023] The personalized learning engine is configured to calculate the knowledge mastery degree of the student based on data of the data collection module, and recommend teaching content and experiment tasks accordingly; the personalized learning engine is configured to calculate the knowledge mastery degree of the student by using a weighted scoring method; a knowledge mastery degree calculation formula is as follows: wherein, s j represents a mastery degree of the jth knowledge point, a j is a weight coefficient corresponding to the knowledge point, and m is a total number of knowledge points.
[0024] Further, the personalized learning engine is configured to adjust the difficulty of the experiment task based on a difficulty adjustment factor; a difficulty adjustment factor calculation formula is as follows: D = 1 + β · (μ - s); wherein, μ is an average mastery rate of the knowledge point, s is an actual mastery rate of the student, and β is an experience parameter.
[0025] The virtual experiment environment is configured to be constructed by using containerization, to be used for deployment and operation scenarios and to provide interactive operation; the virtual experiment environment is configured to use a lightweight containerization architecture, to realize isolation between containers by using a namespace and a control group; the containerization architecture supports rapid deployment of operation scenarios, and limits influence of malicious operation on a host computer.
[0026] In addition, the virtual experiment environment is configured to realize automatic snapshot and rollback by using copy-on-write; the snapshot allows the student to restore to a previous state when an operation is failed. The resource management module of the virtual experiment environment is configured to dynamically allocate available resources according to resource requirements of the container; the resource management module is configured to ensure that a total resource requirement of the container does not exceed a total amount of an available resource pool.
[0027] The feedback optimization mechanism is configured to adjust the personalized learning strategy according to learning progress and historical data of the student; the feedback optimization mechanism is configured to calculate a reward value by using a reinforcement learning method, to optimize a learning path; the reward value is based on a change in the knowledge mastery degree of the student before and after completing the experiment task, and is adjusted by a decay coefficient.
[0028] Preferably, the system further comprises a multi-dimensional evaluation system configured to measure learning achievements of the student; the multi-dimensional evaluation system comprises a theoretical knowledge mastery degree, an experiment operation proficiency and a comprehensive application ability. The multi-dimensional evaluation system is configured to generate learning achievement data by using regular testing, experiment operation records and project-based learning tasks; the learning achievement data is configured to adjust the teaching content and the experiment task.
[0029] In addition, each of the above modules is further configured to implement the following functions when executed:
[0030] The data collection module is responsible for collecting the learning behavior data of students, such as access frequency, correct answer rate, experiment completion time, etc., and inputting these data into the personalized learning engine. The personalized learning engine calculates the current knowledge mastery level K(t) of the learner according to the past performance, and accordingly recommends the teaching content and experimental tasks of corresponding difficulty. The knowledge mastery level of the learner can be represented by the following formula:
[0031] where x i (t) represents the mastery degree of the ith knowledge point at time t, w i is the weight of the corresponding knowledge point, reflecting its importance in the overall knowledge system.
[0032] The virtual experiment environment is built based on containerization technology, which can quickly deploy various operation and maintenance scenarios such as server configuration, network troubleshooting, and log analysis. Students can interactively operate in this environment and obtain immediate feedback.
[0033] In addition, the feedback optimization mechanism continuously monitors the learning progress of students and adjusts the personalized learning strategy combined with historical data. For example, if a student performs poorly in a specific type of experimental task, the system will automatically reduce the difficulty of subsequent tasks and increase the review of related knowledge points.
[0034] The personalized learning algorithm in this embodiment is based on the historical learning data and real-time behavior feedback of students, dynamically adjusting the presentation mode and difficulty level of teaching content to ensure that each student can learn at the pace most suitable for their cognitive level. The core idea of this algorithm is to establish an adaptive knowledge mastery model that can continuously adjust the teaching strategy according to the learning progress and understanding ability of students, thereby improving learning efficiency.
[0035] In the process of personalized learning, the system first obtains the learning trajectory of students through the data collection module, including knowledge point mastery, exercise correct rate, and experimental operation time, etc. key indicators. Then, the system uses the weighted scoring method to process these data to measure the knowledge mastery level of students. Suppose a student has completed the learning of m knowledge points in a certain stage, the mastery degree of each knowledge point is represented by s j (where j = 1, 2,..., m), and the corresponding weight coefficient is α j , then the overall knowledge mastery level S of this student is represented as:
[0036] where the weight coefficient α j reflects the importance of different knowledge points, which is usually evaluated by experts or based on historical data analysis. This formula can be used to measure the comprehensive ability of students in a certain learning stage, and serves as the basis for adjusting subsequent teaching content.
[0037] After determining the knowledge mastery of the student, the system uses a dynamic adjustment mechanism to optimize the learning path. Specifically, the system calculates a difficulty adjustment factor D based on the student's learning performance, which determines the difficulty level of the next stage of teaching content. Assuming that the average mastery rate of the current knowledge point is μ, and the actual mastery rate of the student is s, the difficulty adjustment factor can be defined as: D = 1 + β · (μ - s);
[0038] where β is an empirical parameter used to control the adjustment amplitude. When the student's mastery rate is lower than the average level, the difficulty adjustment factor is less than 1, indicating that the system should appropriately reduce the difficulty of subsequent content; conversely, if the mastery rate is higher than the average level, the difficulty adjustment factor is greater than 1, meaning that the challenge can be appropriately increased.
[0039] In addition, the system also introduces an adaptive feedback mechanism to further optimize the individualized learning process. This mechanism adjusts the teaching strategy continuously, so that the student's learning curve gradually approaches the optimal state. Specifically, the system records the results of each learning task and calculates a reward value R to measure the effectiveness of the learning. The reward value is calculated as follows: R = γ · (s new -s old );
[0040] where s new and s old represent the student's knowledge mastery before and after completing the new task, respectively, and γ is a decay coefficient used to balance short-term gains and long-term learning effectiveness. By maximizing the reward value, the system can dynamically adjust the teaching strategy to enable the student to achieve the best learning effect within a limited time.
[0041] The virtual experiment operation environment in this embodiment is built based on containerization technology, aiming to provide a highly simulated system operation practice platform for students. This environment allows students to perform various operation and maintenance tasks, such as server configuration, network debugging, log analysis, and automation script writing, in an isolated and controllable virtual space. In this way, students can perform high-risk operations without interfering with the actual production environment, while gaining practical experience close to the real scene.
[0042] The core technology of the virtual experiment environment is a lightweight containerization architecture that uses operating system-level virtualization technology to create multiple independent user space instances on the host machine, each of which can run a complete operation and maintenance tool chain. Compared to traditional full virtualization solutions, this architecture has lower resource consumption and higher execution efficiency, making large-scale concurrent experiments possible. The isolation between containers is guaranteed by namespaces (Namespace) and control groups (Cgroup), ensuring that each experiment environment does not affect each other, while preventing malicious operations from causing system crashes.
[0043] To achieve efficient task scheduling and resource allocation, the system adopts a dynamic resource management mechanism. Assume that there are N running experiment containers in the system, and the resource requirement of each container is represented by the vector r i =(CPU i ,RAM i ,DISK i ), where CPU i , RAM i , and DISK i represent the computing resources, memory size, and storage capacity required by the i-th container, respectively. The system maintains an available resource pool R available =(CPU total ,RAM total ,DISK total ), and calculates the feasibility of resource allocation according to the following formula:
[0044] If the condition is met, the system can accept new experiment requests; otherwise, it needs to wait for some experiments to end and release resources, or dynamically adjust the resource configuration of existing experiments.
[0045] In addition, the virtual experiment environment also integrates automatic snapshot and rollback functions to quickly restore the previous state when the student makes a mistake. This function relies on the Copy-on-Write technology, which creates a temporary branch based on the original image, and all modifications only affect this branch without affecting the base image. When rollback is needed, the system only needs to discard the current branch and reload the original image.
[0046] To ensure the security of the experiment environment, the system adopts a multi-level protection strategy. First, all experiment containers run under restricted user permissions, prohibiting direct access to critical files and system interfaces of the host machine. Second, the system has a built-in traffic monitoring module that detects abnormal network activity in real time and automatically blocks connections when potential attack behavior is detected. Finally, all experiment operations are recorded in audit logs, facilitating subsequent tracking and analysis of the student's operation patterns.
[0047] In this embodiment, the personalized learning algorithm and the virtual experiment environment work closely together to build a dynamically adjusted learning process. The personalized learning algorithm determines the difficulty and type of the experiment task by analyzing the student's learning behavior and knowledge mastery, while the virtual experiment environment provides the corresponding practice platform, allowing the student to operate under appropriate challenges.
[0048] In the specific implementation process, the personalized learning algorithm first calculates the student's current knowledge mastery S based on their historical learning data, as described earlier:
[0049] This value reflects the learner's understanding of the learned knowledge points. Subsequently, the system determines the difficulty level of the next experimental task using the difficulty adjustment factor D: D = 1 + β · (μ - s);
[0050] where μ is the average mastery rate of the knowledge point, and s is the actual mastery rate of the learner. According to the value of D, the system will select experimental tasks of different complexity. For example, if D < 1, it indicates that the learner has not mastered the knowledge point, and the system will recommend basic experimental tasks, such as simple server configuration or basic log analysis; if D > 1, it indicates that the learner has mastered the knowledge point well, and the system will provide more complex tasks, such as automated operation script writing or advanced network troubleshooting.
[0051] Once the experimental task is determined, the virtual experimental environment begins to prepare the corresponding experimental scene. This environment is based on containerization technology and can quickly deploy experimental environments that meet task requirements. Assuming that there are N experimental containers in the system, the resource requirements of each container are r i =(CPU i ,RAM i ,DISK i ), and the available resource pool is R available =(CPU total ,RAM total ,DISK total ), then the system determines whether it can start a new experiment according to the following conditions:
[0052] If the conditions are met, the system will start the corresponding experimental container and open the experimental interface to the learner. In this process, the personalized learning algorithm continues to monitor the learner's operation behavior and adjusts the subsequent learning path according to their experimental performance. For example, if the learner repeatedly attempts to configure incorrectly during the experiment, but eventually successfully solves the problem, the system will record this experience and recommend similar experimental tasks in the future to strengthen their relevant skills.
[0053] In addition, the system also introduces a feedback mechanism to optimize the learning path. Whenever the learner completes an experimental task, the system will calculate a reward value R to measure the learning effect: R = γ · (s new -s old );
[0054] where s new and s oldrespectively represent the degree of mastery of relevant knowledge points before and after the completion of the experiment, γ is the attenuation coefficient, used to balance short-term benefits and long-term learning effectiveness. If the reward value is high, it means that the experimental task has significant help for the learner's knowledge improvement, and the system will increase the proportion of similar tasks in the future learning path; on the contrary, if the reward value is low, it will reduce the frequency of related experimental tasks and provide more basic training opportunities.
[0055] Through the above mechanism, the personalized learning algorithm and the virtual experimental environment form a closed loop, constantly adjusting the difficulty and type of experimental tasks to match the learning progress of the learners. This dynamic adjustment not only improves learning efficiency, but also enhances the learners' autonomous exploration ability, enabling them to practice in an environment most suitable for their own level, thereby improving overall learning effectiveness.
[0056] The training system in this embodiment adopts a multi-dimensional evaluation system to accurately measure the learning achievements of learners and dynamically adjusts teaching strategies in combination with feedback optimization mechanisms to improve learning effectiveness. The evaluation system covers three core aspects: theoretical knowledge mastery, experimental operation proficiency, and comprehensive application ability, ensuring comprehensive reflection of learners' learning progress.
[0057] In terms of theoretical knowledge evaluation, the system collects learners' answer data through regular tests and in-class exercises, and calculates their knowledge mastery probability P(K i |θ) based on Bayesian inference method, where K i represents the mastery of the i-th knowledge point, and θ is the overall ability parameter of the learner. This probability can be calculated by the following formula:
[0058] where P(θ|K i ) is the probability of the learner having the ability θ under the condition of mastering the knowledge point K i , P(K i ) is the prior probability of the knowledge point K i , and P(θ) is the probability distribution of the overall ability of the learner. Through this method, the system can dynamically adjust the test difficulty and provide additional learning resources for weak knowledge points.
[0059] In terms of experimental operation evaluation, the system records learners' operation steps, error rate and completion time in the virtual experimental environment, and calculates the experimental score E s , whose expression is as follows:
[0060] where T is the total number of steps, w t is the weight of the t-th step, e t is the number of errors in that step, and m tThis represents the maximum allowed number of errors. This formula is used to quantify trainees' experimental performance and serves as an important basis for adjusting the difficulty of subsequent experimental tasks.
[0061] The comprehensive application ability assessment is based on project-based learning tasks, requiring trainees to complete a series of closely related operation and maintenance tasks within a limited time. The system calculates a comprehensive score (C) based on task completion quality, code standardization, and problem-solving efficiency. s Its formula is as follows: C s =α·Q c +β·C p +γ·T e ;
[0062] Among them, Q c To rate the quality of task completion, C p For code standardization scoring, T e The task completion efficiency is scored, with α, β, and γ representing the weighting coefficients of each indicator. This score measures the learner's comprehensive practical ability and influences their subsequent learning path planning.
[0063] The feedback optimization mechanism dynamically adjusts teaching content and experimental tasks based on the aforementioned evaluation results. For example, if a student frequently makes the same type of error during an experiment, the system will reduce the difficulty of subsequent experiments and add review sessions on relevant knowledge points. Furthermore, the system will incorporate reinforcement learning methods, adjusting recommendation strategies based on student feedback to optimize their learning path. Through this mechanism, the training system can continuously improve its teaching plans, ensuring students learn in the best possible condition.
[0064] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A training system based on an operation and maintenance platform, characterized in that, The system comprises a data collection module, a personalized learning engine, a virtual experiment environment, and a feedback optimization mechanism. The data collection module is used to collect learning behavior data of students, including access frequency, correct answer rate, and experiment completion time. The personalized learning engine calculates the knowledge mastery degree of students based on the data collected by the data collection module, and accordingly recommends teaching content and experiment tasks. The virtual experiment environment is built using containerization, which is used to deploy and operate scenarios and provide interactive operations. The feedback optimization mechanism adjusts the personalized learning strategy according to the learning progress and historical data of students.
2. The training system based on the operation and maintenance platform according to claim 1, wherein, The data collection module generates learning behavior data by recording the correct answer rate and experiment completion time of students. The personalized learning engine calculates the knowledge mastery degree of students according to the learning behavior data, and recommends matching teaching content based on the knowledge mastery degree. 3.The training system based on the operation and maintenance platform according to claim 1, characterized in that, The personalized learning engine calculates the knowledge mastery degree of the student by a weighted scoring method; a knowledge mastery degree calculation formula is: Wherein, s j represents the mastery degree of the jth knowledge point, α j is the weight coefficient of the corresponding knowledge point, and m is the total number of knowledge points.
4. The training system based on the operation and maintenance platform according to claim 1, characterized in that, The personalized learning engine adjusts the difficulty of experiment tasks based on a difficulty adjustment factor; the formula for calculating the difficulty adjustment factor is: D = 1 + β·(μ-s); where μ is the average mastery rate of knowledge points, s is the actual mastery rate of students, and β is an experience parameter.
5. The training system based on the operation and maintenance platform according to claim 1, wherein, The virtual experiment environment uses a lightweight containerization architecture to achieve isolation between containers through namespaces and control groups. The containerization architecture supports rapid deployment and operation of scenarios, and limits the impact of malicious operations on the host.
6. The training system based on the operation and maintenance platform according to claim 1, wherein, The virtual experiment environment uses copy-on-write to achieve automated snapshots and rollback; snapshots allow students to restore to a previous state in case of operation errors.
7. The training system based on the operation and maintenance platform according to claim 1, wherein, The resource management module of the virtual experiment environment dynamically allocates available resources based on container resource requirements; the resource management module ensures that the total demand for container resources does not exceed the total amount of the available resource pool. 8.The training system based on an operation and maintenance platform according to claim 1, wherein, The feedback optimization mechanism calculates a reward value using reinforcement learning methods to optimize the learning path; the reward value is based on the change in knowledge mastery degree before and after students complete experiment tasks, and is adjusted by a decay coefficient. 9.The training system based on an operation and maintenance platform according to claim 1, wherein, It also includes a multi-dimensional evaluation system for measuring the learning achievements of students; the multi-dimensional evaluation system includes theoretical knowledge mastery, experiment operation proficiency, and comprehensive application ability.
10. The training system based on the operation and maintenance platform according to claim 9, characterized in that, The multi-dimensional evaluation system generates learning achievement data through regular tests, experiment operation records, and project-based learning tasks. The learning achievement data is used to adjust teaching content and experiment tasks.