Real-time modeling training quality management method and system for civil aviation

Abstract

The invention relates to the technical field of civil aviation training, in particular to a real-time modeling training quality management method and system for civil aviation, and the method comprises the steps: obtaining the historical learning data and historical examination data of a student, and evaluating the training level and training state of the student according to the historical examination data and the historical learning data; formulating a corresponding training strategy for the trainee according to the training level and the training state, determining a corresponding scene training state in response to the operation fluency and the training score in the training process, and determining a training characterization state of the trainee at the corresponding training level according to each scene training state in the training level so as to adjust the training strategy. Comprising the steps of adjusting the trueness of a training model according to the average operation fluency and the average training score of each scene, and adjusting the training duration after training according to the average training frequency of each scene. According to the invention, different training strategies are formulated according to individual differences of trained students, so that waste of training resources is reduced, and training pertinence and training efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of civil aviation training, and particularly to a real-time modeling training quality management method and system for civil aviation. Background Art

[0002] The civil aviation industry has extremely high requirements for the professional qualities and skills of personnel. Training is a key link to ensure the smooth progress of various aspects of work such as flight safety and ground service quality. There are common problems in the quality management of existing civil aviation training, such as inaccurate and untimely evaluation of training effects, difficulty in dynamically adjusting according to the individual differences of trainees, and unreasonable allocation of training resources; traditional training quality management methods mostly rely on manual experience judgment and post-event summary, lacking real-time and systematic modeling analysis means, resulting in the inability to timely discover problems and carry out effective intervention during the training process, thus affecting the improvement of training quality and efficiency, and it is difficult to meet the growing demand for high-quality talent training in the civil aviation industry.

[0003] Chinese Patent Publication No. CN113112385B discloses an air traffic control training platform system for civil aviation, including a PC online learning module, a mobile online learning module, and a business management module. The air traffic control training platform system of the present invention, by establishing a PC online learning module, a mobile online learning module, and a business management module, electronicizes course resources and integrates examination and training, and establishes a flexible and efficient learning and training system, which can meet the needs of digital training for air traffic control in civil aviation. It can be seen that the present invention has the following problems: It does not formulate different training strategies according to the individual differences of training trainees, resulting in waste of training resources and poor training pertinence, and thus low training efficiency. Summary of the Invention

[0004] For this reason, the present invention provides a real-time modeling training quality management method and system for civil aviation to overcome the problems of waste of training resources, poor training pertinence, and low training efficiency caused by not formulating different training strategies according to the individual differences of training trainees in the prior art.

[0005] To achieve the above object, on the one hand, the present invention provides a real-time modeling training quality management method for civil aviation, including: Obtaining the historical learning data and historical assessment data of trainees, where the historical learning data includes the historical learning times and the learning duration of each time; Evaluating the training level of trainees according to the historical assessment data, and evaluating the training status of trainees according to the historical learning data, where the training status includes frequent training and intermittent training; Formulating corresponding training strategies for trainees according to the training level and the training status, where the training strategies include the authenticity of the training model and the training duration after training; Determine the corresponding scenario training status in response to the operation fluency and training performance during the training process, where the scenario training status includes a passing status and a practice status; Determine the training representation status of the trainee at the corresponding training level according to the scenario training statuses in the training level to adjust the training strategy, including, Adjust the authenticity of the training model according to the average operation fluency and average training performance of each scenario; And, adjust the post-training duration according to the average number of training times of each scenario; Wherein, the training level includes at least four levels.

[0006] Further, the method for evaluating the training level of the trainee according to the historical assessment data includes: Determine the highest assessment score, the lowest assessment score, and the average assessment score according to the historical assessment data; Preliminarily determine the training level according to the average assessment score; Determine the accuracy of the training level according to the score difference between the highest assessment score and the lowest assessment score to re-determine the training level, where, If the score difference is less than or equal to the preset score difference, it is determined that the training level is accurate and there is no need to re-determine the training level; If the score difference is greater than the preset score difference, it is determined that the training level is inaccurate and the training level is re-determined according to the ratio of the score difference to the preset score difference.

[0007] Further, the method for evaluating the training status according to the historical learning data includes: Filter out invalid historical learning times according to each learning duration to obtain the effective historical learning times; Determine the effective learning frequency according to the effective historical learning times; Determine the training status according to the effective learning frequency, where, If the effective learning frequency is greater than the preset frequency, it is determined that the training status is frequent training; If the effective learning frequency is less than or equal to the preset frequency, it is determined that the training status is intermittent training.

[0008] Further, the method for formulating a corresponding training strategy for the trainee according to the training level and the training status includes: Determine the authenticity of the training model according to the determination results of the training level and the training status; And, determine the post-training duration according to the training status and the effective learning frequency.

[0009] Further, the method for determining the authenticity of the training model according to the determination results of the training level and the training status includes: Determine the point cloud density and its distribution mode according to the determination result of the training level; And determine whether to add texture rendering according to the determination result of the training status.

[0010] Furthermore, determine the post-training duration according to the training status and the effective learning frequency, including, If the training status is frequent training, determine that the post-training duration is the standard duration; If the training status is intermittent training, determine to increase the post-training duration and determine the post-training duration according to the ratio of the preset frequency to the effective learning frequency.

[0011] Furthermore, determine the corresponding scenario training status in response to the operation fluency and training performance during the training process, including, If the operation fluency is greater than or equal to the preset fluency and the training performance is greater than or equal to the passing score, determine that the corresponding scenario training status is the passed status; If the operation fluency is less than the preset fluency and / or the training performance is less than the passing score, determine that the corresponding scenario training status is the practice status.

[0012] Furthermore, determine the training representation status of the trainee at the corresponding training level according to the scenario training status in the training level, including, If the ratio of the number of scenarios with the passed scenario training status to the total number of scenarios is greater than or equal to the preset ratio, determine that the training representation status is to complete the current level of training; If the ratio of the number of scenarios with the passed scenario training status to the total number of scenarios is less than the preset ratio, determine that the training representation status is to continue the current level of training.

[0013] Furthermore, the method for adjusting the training strategy according to the training representation status includes: Obtain the training representation status of the trainee at the training level to determine whether to adjust the training strategy, where, If the training representation status is to continue the current level of training, determine to adjust the training strategy; In response to the determination result of adjusting the training strategy, obtain the number of training times of the trainee in each scenario training and the operation fluency and training performance of each training; Determine the average operation fluency and the average training performance respectively according to the operation fluency and the training performance of each training; Determine to reduce the authenticity of the training model according to the average operation fluency and the average training performance; Increase the post-training duration according to the average number of training times in each scenario.

[0014] On the other hand, the present invention also provides a civil aviation real-time modeling training quality management system, including: A data acquisition module for obtaining the historical learning data and historical assessment data of students; A training evaluation module connected to the data acquisition module for evaluating the training level of students according to the historical assessment data, and evaluating the training status of students according to the historical learning data to formulate corresponding training strategies; A practice evaluation module for determining the corresponding scenario training status according to the operation fluency and training performance during the training process; A training adjustment module connected to the training evaluation module and the practice evaluation module respectively for determining the training representation status of students at the corresponding training level according to the scenario training status in each training level to adjust the training strategy.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The civil aviation real-time modeling training quality management method provided by the present invention can evaluate the corresponding training level and status of students through the analysis of the historical learning data and assessment data of students, and can formulate targeted training strategies for different students to meet the personalized learning needs of students and improve the training efficiency and effect. In implementation, for students with good foundations and in frequent training, a training model with higher authenticity can be provided to accelerate their skill improvement; while for students with weak foundations or passive training, a relatively simple and lower authenticity model can be used for basic consolidation training first to avoid discouraging students due to excessive difficulty; secondly, the present invention determines the scenario training status of the scenario according to the operation fluency and training performance of students during the scenario training process, and determines the training representation status according to the scenario training status of each scenario to determine whether to further adjust the training strategy so that the training can be dynamically optimized according to the real-time learning situation of students; thirdly, the present invention comprehensively evaluates students with data from multiple aspects, not only paying attention to the assessment results but also considering the data during the learning process (such as the number of learning times, duration, and operation conditions during the training process, etc.) to provide more comprehensive learning feedback for students, which helps to understand the real learning situation of students, discover the advantages and disadvantages of students, and thus improve and enhance targeted; Furthermore, the training level is determined by comprehensively considering the highest score, the lowest score, and the average score in the assessment, which avoids the one-sidedness brought about by relying solely on a single piece of data (such as the average score) for evaluation. The comprehensive evaluation method can more accurately locate the true level of the trainee among the overall trainee group, so as to allocate more appropriate training resources and course content with appropriate difficulty levels for them. The difference between the highest score and the lowest score is used to verify the accuracy of the preliminarily determined assessment level and make a secondary determination when necessary, which can effectively adapt to the characteristics of different assessment scenarios and trainee groups. For example, in some small training batches, due to the limited sample size, the contingency of a single assessment result may be relatively large, resulting in a relatively large score difference. At this time, the secondary determination of the assessment level can correct the preliminary level deviation caused by accidental factors, making the finally determined training level more in line with the true ability of the trainee and ensuring the effectiveness and pertinence of subsequent training strategies. This assessment method provides a fairer and more just level assessment mechanism for each trainee. Regardless of how the overall difficulty of the assessment fluctuates or whether individual trainees perform exceptionally well or poorly in a certain assessment, by comprehensively considering the highest score, the lowest score, the average score, and the score difference, these factors can be incorporated into the consideration range of the level assessment, thus avoiding unreasonable level assessment phenomena caused by special circumstances and ensuring the scientificity and objectivity of the level assessment in the entire training system, enabling trainees to receive training and evaluation in a fair environment; Furthermore, judging the training status through the effective learning frequency can more objectively measure the enthusiasm of trainees for training. The effective learning frequency directly reflects the frequency of trainees actively and effectively participating in learning within a certain period of time, avoiding the one-sidedness of judging trainees' enthusiasm solely based on subjective feelings or a single learning indicator (such as the total learning duration). Some trainees may have a long total learning duration but a small number of effective learning times, indicating that they may be easily distracted or have low learning efficiency during the learning process. The effective learning frequency can accurately capture this situation, thus making a more realistic assessment of the training status of trainees; Furthermore, determining the point cloud density, distribution pattern, and texture rendering according to the actual situation of the trainees can avoid adopting the highest-level visual effects in the training models of all trainees, thus saving computing resources and storage resources. For trainees with a low training level or intermittent training, reducing unnecessary high-resolution texture rendering and high-density point clouds can enable the system to occupy less memory and processor resources when processing these models, improve the overall operation efficiency of the system, and at the same time reduce the time for data transmission and loading, allowing trainees to enter the learning scenario faster. This targeted visual effect setting helps trainees learn and understand faster. Appropriate point cloud density and texture rendering can highlight the key learning content and reduce irrelevant visual interference. In flight training, by appropriately reducing the point cloud density and simplifying the texture of the training scenario for junior trainees, the main instruments and control components of the aircraft become more prominent, and trainees can more efficiently familiarize themselves with the positions and functions of these key parts, rather than being confused by the complex external details of the aircraft, thus accelerating the learning progress and improving the learning efficiency. Furthermore, by considering both the operation fluency and training performance indicators simultaneously, it is possible to more comprehensively and accurately judge the learning effectiveness of trainees in a specific scenario. Operation fluency reflects the coherence, proficiency of trainees when performing operations, and the naturalness of their interaction with the virtual environment, while training performance reflects the degree of trainees' mastery of the knowledge and skills related to the scenario and the completion quality in the simulation task. In the takeoff scenario of simulated flight, only looking at the training performance may only show whether trainees know the takeoff process and related parameter settings, but combining the operation fluency can further clarify the proficiency and naturalness of trainees when actually operating actions such as pulling the control column and adjusting the throttle of the aircraft, thus more accurately grasping the real learning situation of trainees in this scenario. Furthermore, more effective allocation of training resources can be obtained according to the training representation status. For trainees who are judged to have completed the current level of training, more advanced training resources (such as more complex simulation equipment, professional senior instructor guidance, etc.) can be allocated to them to meet their further improvement needs. For trainees who need to continue the current level of training, resources can be concentrated on strengthening the training of their weak scenarios, such as providing additional practice time, targeted learning materials, etc., improving the resource utilization efficiency and enhancing the overall training quality. Furthermore, adjusting the training strategy according to the training representation status can determine the learning progress of the trainees at the current training level, timely adjust the strategy for trainees continuing with the current level of training, focus on weak links, and adjust the authenticity of the training model based on operation fluency and performance data to adapt to the trainees' abilities. At the same time, reasonably increase the post-training duration according to the average number of training sessions and set compulsory training to meet personalized learning needs. For trainees who have completed the current level of training, the next-level strategy can be planned in a timely manner, which not only efficiently utilizes training resources, avoids wasting resources on already mastered content, but also optimizes the overall training progress, ensures that trainees progress steadily, makes the training process smoother and more efficient, and comprehensively improves the quality and effect of training. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of the steps of the civil aviation real-time modeling training quality management method according to the embodiment of the present invention; Figure 2 It is a flowchart of evaluating the training level and training status of trainees according to the embodiment of the present invention; Figure 3 It is a flowchart of determining the training representation status of trainees at the corresponding training level to adjust the training strategy according to the embodiment of the present invention; Figure 4 It is a connection diagram of the civil aviation real-time modeling training quality management system according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only for explaining the present invention and are not used to limit the present invention.

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0019] It should be noted that in the description of the present invention, the terms indicating directions or positional relationships such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings. This is only for convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention.

[0020] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and defined, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0021] Please refer to Figure 1 as shown in the figure, which is the step diagram of the real-time modeling training quality management method for civil aviation in the embodiment of the present invention. The present invention provides a real-time modeling training quality management method for civil aviation, including: Step S1, obtaining the historical learning data and historical assessment data of the trainees, where the historical learning data includes the historical learning times and the learning durations of each time; Step S2, evaluating the training level of the trainees according to the historical assessment data, and evaluating the training status of the trainees according to the historical learning data, where the training status includes frequent training and intermittent training; Step S3, formulating corresponding training strategies for the trainees according to the training level and the training status, where the training strategies include the authenticity of the training model and the post-training duration; Step S4, determining the corresponding scenario training status in response to the operation fluency and training performance during the training, where the scenario training status includes the passed status and the practice status; Step S5, determining the training representation status of the trainees at the corresponding training level according to the scenario training status in the training level to adjust the training strategy, including, adjusting the authenticity of the training model according to the average operation fluency and average training performance of each scenario; and, adjusting the post-training duration according to the average training times of each scenario; wherein, the training level includes at least four levels.

[0022] It can be understood that the real-time modeling training quality management method for civil aviation provided by the present invention can formulate targeted training strategies for different students to meet their personalized learning needs and improve training efficiency and effect by analyzing and evaluating the corresponding training level and status of the students' historical learning data and assessment data. In implementation, for students with good foundation and frequent training, a training model with higher authenticity can be provided to accelerate their skill improvement; for students with weak foundation or passive training, a relatively simple model with lower authenticity can be used for basic consolidation training to avoid hitting the enthusiasm of students due to excessive difficulty; secondly, the present invention determines the scene training state of the scene according to the student's operation fluency and training results during the scene training process, and determines the training representation state according to the scene training state of each scene to determine whether to further adjust the training strategy so that the training can be dynamically optimized according to the real-time learning situation of the students; thirdly, the present invention evaluates the students by integrating various data, not only focusing on the assessment results but also considering the data in the learning process (such as the number of learning times, duration and operation conditions during training, etc.) to provide students with more comprehensive learning feedback, which is helpful to understand the real learning situation of the students, discover the advantages and disadvantages of the students, and thus improve and improve them in a targeted manner.

[0023] See also Figure 2 As shown, it is a flow chart of evaluating the training level and training status of trainees according to an embodiment of the present invention. Figure 2 , a flowchart of first performing step S21 and then performing step S22 is shown. In implementation, step S2 includes step S21 and step S22, wherein: Step S21, evaluating the trainee's training level according to the historical assessment data; Step S22, evaluating the training status according to the historical learning data; It can be understood that step S21 and step S22 can be performed simultaneously, or step S21 can be performed first and then step S22, or step S22 can be performed first and then step S21.

[0024] Specifically, in step S21, the method for evaluating the trainee's training level according to the historical assessment data includes: Step S211, determining the highest assessment score, the lowest assessment score and the average assessment score according to the historical assessment data; Step S212, preliminarily determining the training level according to the average assessment score; Step S213, determining the accuracy of the training level according to the difference between the highest assessment score and the lowest assessment score to determine the training level twice, wherein: If the score difference is less than or equal to the preset score difference, it is determined that the student's assessment score is stable, so the training level is accurate and there is no need to determine the training level again; If the score difference is greater than the preset score difference, it is determined that the training level is inaccurate and the training level is determined again based on the ratio of the score difference to the preset score difference; if only the average score is relied on, the fluctuations in the students' scores may be ignored. When the average scores are the same but the difference between the highest score and the lowest score is large, it indicates that the students' score stability is poor, and their actual ability level may deviate from the level determined only based on the average score.

[0025] In implementation, a comparison table of assessment average scores, training levels, and training times is established in advance. The training levels include at least four levels, usually including the entry level, the primary level, the intermediate level, and the advanced level; among them, the assessment average score corresponding to the entry level is less than 60 points, and the training time is 15 days. The assessment average score corresponding to the primary level is 60 - 70 points (excluding 70 points), and the training time is 25 days. The assessment average score corresponding to the intermediate level is 70 - 85 points (including 70 points but excluding 85 points), and the training time is 40 days. The assessment average score corresponding to the advanced level is greater than or equal to 85 points, and the training time is 60 days; in implementation, more assessment levels can also be added to make the score segments corresponding to each level more refined.

[0026] It can be understood that the score fluctuations of a student can be determined by the highest and lowest assessment scores of the student. If the score difference of a certain student is too large (greater than the preset score difference), it means that the score fluctuations of the student are very large (that is, the scores are unstable), and the training level determined by their assessment average score is considered inaccurate. Therefore, it is necessary to determine their training level again.

[0027] In implementation, the training level determined again = the initially determined training level - score difference ÷ preset score difference; among them, if the ratio of the score difference to the preset score difference is a decimal, the value of its quotient is determined according to the principle of "rounding off"; in one implementation, if the quotient of the calculated score difference ÷ preset score difference is 1, the training level determined again is 1 level lower than the initially determined training level. If the quotient of the calculated score difference ÷ preset score difference is 2, the training level determined again is 2 levels lower than the initially determined training level.

[0028] In implementation, the preset score difference for each assessment level should be less than the score difference corresponding to the score range of that level. Preferably, it is set to half of the score difference corresponding to the score range of that level. In one implementation, the score range corresponding to the intermediate level is 70 - 85 points, and the score difference of this score range is 15 points. Then the preset score difference for this range is 8 (15÷2 = 7.5, rounded up to 8) points. The score range corresponding to the primary level is 60 - 70 points, and the score difference of this score range is 10 points. Then the preset score difference for this range is 5 (10÷2) points. Therefore, the preset score differences for each level may vary. In implementation, the assessment level should be initially determined first, then the preset score difference should be determined according to the score difference of the score range of the initially determined assessment level. Finally, compare the size of the score difference and the preset score difference to determine whether to re - determine the assessment level for the second time.

[0029] Specifically, in step S22, the method for evaluating the training status according to the historical learning data includes: Step S221, filtering out invalid historical learning times according to each learning duration to obtain valid historical learning times. It can be understood that too short learning duration means that the learner only simply opens the learning software or enters the learning environment but quickly exits. Such short - term learning behaviors cannot really enable the learner to absorb knowledge or improve skills. After filtering out these invalid learning times, the remaining valid historical learning times can more accurately reflect the actual situation of the learner's active participation in learning, providing more reliable data for evaluating the learner's learning attitude and effort level. It can be understood that according to each learning duration, it is determined whether this learning is a valid learning or an invalid learning. Among them, if the learning duration of a single learning is less than the minimum learning duration, then this learning is an invalid learning. In implementation, learning theoretical knowledge usually requires a certain amount of time to understand concepts and memorize key points. Therefore, the minimum learning duration is usually determined by the average course duration and is usually set to one - third to one - half of the average course duration. The average course duration is the average value of each course duration. In one implementation, an average course duration is 60 minutes, then the minimum learning duration can be set between 20 minutes and 30 minutes.

[0030] Step S222, determining the effective learning frequency according to the valid historical learning times. In implementation, the effective semester period is determined according to the time of the first effective learning and the time of the last effective learning, and the effective learning frequency is determined according to the ratio of the valid historical learning times to the effective learning period. Step S223, determining the training status according to the effective learning frequency, where if the effective learning frequency is greater than the preset frequency, it is determined that the training status is frequent training; If the effective learning frequency is less than or equal to the preset frequency, it is determined that the training status is intermittent training. It can be understood that considering the skill requirements and training expectations of the civil aviation industry, the preset frequency is determined according to the reasonable learning progress generally recognized in the industry: Generally speaking, in order to reach the skill level recognized by the industry, the average effective learning frequency of trainees during the entire training period should not be less than 3 times per week. This is because civil aviation work involves many key areas such as flight safety and passenger service, and trainees need to achieve corresponding professional qualities and skill standards through continuous effective learning.

[0031] Specifically, in step S3, the method of formulating a corresponding training strategy for trainees according to the training level and the training status includes: Step S31, determining the authenticity of the training model according to the determination results of the training level and the training status. It can be understood that determining the authenticity of the training model according to the training level and the training status can provide appropriate learning challenges for trainees with different levels and learning attitudes. For trainees with a high training level and in frequent training, providing a training model with high authenticity can meet their needs for in-depth expansion of knowledge and skills and accelerate their growth into professional civil aviation talents; while for trainees with a relatively low training level or passive training, appropriately reducing the authenticity of the training model and first focusing on the consolidation of their basic content can avoid discouraging learning enthusiasm or causing learning obstacles due to excessive difficulty. And step S32, determining the post-training duration according to the training status and the effective learning frequency. It can be understood that determining the post-training duration according to the training status and the effective learning frequency can further optimize the allocation of training resources. Trainees with active training represent that they can independently learn knowledge and skills, so the post-training duration can be appropriately shortened; while trainees with passive training represent that their autonomy in learning knowledge is relatively poor, so mandatory post-training needs to be arranged and a longer post-training duration is set to strengthen the learning and consolidation of the learned content and ensure that they can keep up with the training progress. Step S32 enables each trainee to improve their abilities at the most suitable training rhythm for themselves.

[0032] It can be understood that by accurately matching the authenticity of the training model and the post-training duration in step S3, the learning enthusiasm and participation of trainees can be improved. When trainees face training content and duration that are suitable for their own abilities and status, they are more likely to understand and master knowledge and skills, reduce learning frustration, enhance self-confidence, and thus be more actively involved in training; in simulated flight training, appropriate model authenticity and post-training duration can enable trainees to better apply theoretical knowledge to practical operations, improve operation proficiency and the ability to handle emergencies, achieve better training effects within a limited training time, and improve the overall training efficiency.

[0033] Specifically, in step S31, the method for determining the authenticity of the training model according to the determination results of the training level and the training status includes: Determine the point cloud density and its distribution mode according to the determination result of the training level; it can be understood that in 3D modeling, the number of points (i.e., point cloud density) can affect the accuracy of the space. A point cloud is a set of discrete points in 3D space, and the shape of an object is described by a large number of points. A higher point cloud density can capture the shape details of the object more precisely; in addition to the point cloud density, the distribution mode of the points can also affect the accuracy of the space. The uniform distribution of points on the surface and in the space of the object can ensure the consistency of space accuracy. At the same time, the distribution of points should be adaptively adjusted according to the shape and detail level of the object; And, determine whether to add texture rendering according to the determination result of the training status; it can be understood that texture mapping is the process of covering a 2D image (texture) on the surface of a 3D object, and its accuracy affects the realism and accuracy of the space. The texture must match the geometric shape of the object; high-resolution textures can provide more details to improve the accuracy of the space. The texture should also support the level of detail (LOD) technology and dynamically adjust the texture details according to the distance between the observer and the object to maintain appropriate accuracy and visual effects from different perspectives.

[0034] In implementation, a comparison relationship table of the training level, point cloud density, and point distribution mode is established in advance. Usually, there is a positive correlation between the training level and the point cloud density (i.e., the higher the training level, the greater the point cloud density). Usually, the distribution of points at the highest training level and the second-highest training level is adaptively adjusted according to the shape and detail level of the object, and the distribution of points at the remaining training levels is evenly distributed; In implementation, adding texture rendering to frequently trained students can significantly enhance their learning experience. Not adding texture rendering to intermittently trained students can prevent them from being distracted by too many complex visual details; adding texture rendering is also a kind of incentive measure for students. When students see that they can enjoy a more advanced and realistic training environment because of their positive learning status, it will further strengthen their positive behavior. This positive feedback can encourage students to maintain a good learning state and strive to improve their training level to obtain more high-quality learning resources and a more realistic training experience.

[0035] It is understandable that determining the point cloud density and its distribution pattern based on the training level can provide appropriate visual details for trainees at different levels: for trainees with a high training level, a higher point cloud density and a refined distribution pattern can present more complex and realistic object shapes and scene details; for example, in scenarios simulating the internal structure of an aircraft or complex airport facilities, advanced trainees can see more precise shapes of mechanical components and details of building structures, which helps them deeply understand the actual situation of civil aviation equipment and the environment and meet their need for in-depth exploration of knowledge; while for trainees with a low training level, appropriately reducing the point cloud density can avoid the cognitive burden caused by excessive detail information, enabling them to first focus on the key structural and functional parts and gradually improve their understanding ability. It is understandable that deciding whether to add texture rendering according to the training status can further optimize the learning experience; for trainees in frequent training, adding texture rendering can enhance the realism and immersion of the scene. For example, in a scenario simulating cabin service, seats, decorative panels, etc. with texture rendering can allow trainees to more vividly feel the cabin environment, thereby improving their learning enthusiasm and participation; while for trainees in intermittent training, temporarily not adding texture rendering or using simple textures can first simplify the visual information, enabling them to focus on the learning content itself rather than being distracted by overly complex visual effects.

[0036] Specifically, in step S32, determining the post-training duration according to the training status and the effective learning frequency includes If the training status is frequent training, then determine the post-training duration as the standard duration; it is understandable that frequent training can indicate a high level of self-learning ability of the trainee. Therefore, setting a standard post-training duration for trainees in frequent training can avoid fatigue or aversion caused by overtraining. A stable post-training duration also helps trainees reasonably arrange their learning plans and improve the efficiency and coherence of learning. If the training status is intermittent training, then determine to increase the post-training duration and determine the post-training duration according to the ratio of the preset frequency to the effective learning frequency; it is understandable that when the training status is intermittent training, it can indicate that the trainee is not very active in learning. Therefore, increasing the post-training duration for trainees in intermittent training is an effective intervention measure. These trainees may encounter difficulties or lack sufficient learning motivation during the learning process. By extending the post-training time, more learning support and guidance can be provided for them; in implementation, determining the post-training duration according to the ratio of the preset frequency to the effective learning frequency can make the increase in the training duration more targeted. If the effective learning frequency of the trainee is much lower than the preset frequency, then the post-training duration will be increased significantly accordingly, giving them enough time to make up for the learning deficiencies and gradually catch up with the progress of other trainees.

[0037] In practice, the duration of post-training under intermittent training = preset frequency ÷ effective learning frequency × standard duration; it is understandable that the standard duration is related to the training content: (1) The standard duration of post-training for theoretical knowledge is usually 30 minutes to 45 minutes, which is enough to review, summarize and answer questions about the key knowledge of a theoretical class; (2) The standard duration of post-training for practical operations is usually 20 minutes to 40 minutes, which can meet the needs of students' operational feedback, skill explanation and repeated practice; therefore, it is preferred to set the standard duration to 30 minutes.

[0038] In implementation, the preset frequency is no less than 3 times per week, then the training status of each trainee should be re-evaluated every 2 to 4 weeks to timely grasp the training situation of each trainee, usually re-evaluated every 2 weeks; it can be understood that the post-training training duration is dynamically adjusted according to the changes in the trainees' training status and learning frequency. If a trainee who originally trained passively gradually turns to frequent training, then the post-training training duration can be adjusted back to the standard duration; conversely, if the trainees who originally trained actively enter intermittent training for some reason (such as excessive learning pressure, encountering learning bottlenecks, etc.), the post-training training duration can be increased accordingly to help them overcome difficulties; this dynamic adjustment mechanism can better adapt to various changes in trainees during the training process and ensure the effectiveness and pertinence of the training.

[0039] Specifically, in step S4, the corresponding scene training state is determined in response to the operation fluency and training results during the training process, including: If the operation fluency is greater than or equal to the preset fluency and the training score is greater than or equal to the qualified score, the corresponding scene training status is determined to be a passed status, indicating that the trainee has reached a certain level of ability in this scene; If the operation fluency is less than the preset fluency and / or the training score is less than the qualified score, the corresponding scene training status is determined to be a practice status; this indicates that the trainee's training in this scene is insufficient. Whether the operation is not smooth enough or the training score is not up to standard, the training system can identify the problem and arrange special exercises / post-training content for these weak links, such as designing a special operation skill training course for the problem of unsmooth operation, or conducting intensive review of knowledge points for poor training results.

[0040] In implementation, the passing score is ≥80 points (out of 100).

[0041] See also Figure 3As shown, it is a flowchart for the embodiment of the present invention to determine the training representation state of a trainee at the corresponding training level and adjust the training strategy. Specifically, step S5 includes step S51, which determines the training representation state of the trainee at the corresponding training level according to the training status of each scenario in the training time corresponding to the training level, including, If the ratio of the number of scenarios with a passed training status to the total number of scenarios is greater than or equal to a preset ratio, it is determined that the training representation state is to complete the current level of training; indicating that they have the conditions to enter the next training level, and at this time, more advanced training content can be timely enabled for the trainee, so that the training can proceed orderly according to the actual ability level of the trainee; If the ratio of the number of scenarios with a passed training status to the total number of scenarios is less than the preset ratio, it is determined that the training representation state is to continue the current level of training; the training system can then focus on strengthening the training for the scenarios that the trainee has not passed, ensuring that the trainee can firmly master all necessary knowledge and skills at the current level, avoiding the training progress being too fast or too slow, and optimizing the arrangement of the training process.

[0042] It can be understood that by considering the training status of each scenario and their proportion in the total scenarios to determine the training representation state of the trainee at their training level, the overall level of the trainee at the current training level can be comprehensively and synthetically measured; it avoids judging whether the trainee has completed the current level of training only based on the performance of a single scenario or a few scenarios, because civil aviation training usually includes multiple complex and interrelated scenarios, such as takeoff, cruise, and landing scenarios in flight training, and boarding service, in-flight service, emergency handling, etc. scenarios in cabin service; only when the trainee reaches the passed status in most key scenarios can it be shown that they have truly mastered the knowledge and skills required for this level, and this evaluation method more accurately reflects the actual training progress of the trainee.

[0043] It can be understood that the preset ratio ≤ 1. Using the preset ratio as the judgment standard provides an objective and quantitative measurement scale, which enables both training instructors and trainees themselves to clearly understand whether they have reached a sufficient ability level at the current training level; in implementation, the preset ratio is usually greater than 0.7. The larger the preset ratio, the higher the degree of mastery of the trainee at the current training level. When the number of scenarios with a passed status accounts for the total number of scenarios reaching or exceeding the preset ratio, it can be objectively considered that the trainee has completed the current level of training, rather than judging by subjective feelings or vague impressions, increasing the fairness and accuracy of the evaluation; preferably, the preset ratio is 0.8.

[0044] It can be understood that each trainee will determine the training time when determining the training level, determine their training representation state at that level when reaching their training time, and then adjust the training strategy according to the training representation state; Specifically, step S5 includes step S52, which adjusts the training strategy according to the training representation state. The method for adjusting the training strategy according to the training representation state includes: Step S521, obtaining the training representation state of the trainee at the current training level to determine whether to adjust the training strategy, where if the training representation state is to continue training at this level, it is determined to adjust the training strategy (the training strategy here is for this training level); if the training representation state is to complete training at this level, it is determined not to adjust the training strategy (the training strategy here is for this training level), but a training strategy corresponding to the next training level will be re - formulated in the state of completing training at this level; Step S522, in response to the determination result of adjusting the training strategy, obtaining the number of training times of the trainee in each scenario training, the operation fluency and training performance of each training; Step S523, respectively determining the average operation fluency and average training performance according to the operation fluency and training performance of each training; Step S524, determining the reduction of the authenticity of the training model according to the average operation fluency and the average training performance; in implementation, first determine the ratio of the average operation fluency to the preset fluency and the ratio of the average training performance to the qualified performance as the first ratio and the second ratio respectively, compare the magnitude relationship between the first ratio and the second ratio, and select the smaller ratio to determine its adjusted authenticity (at this time, the authenticity is adjusted by adjusting the point cloud density. If the first ratio < the second ratio, the adjusted point cloud density = the first ratio × the pre - adjusted point cloud density; if the second ratio < the first ratio, the adjusted point cloud density = the second ratio × the pre - adjusted point cloud density); Step S525, increasing the post - training duration according to the average number of training times in each scenario; the average number of training times can reflect the learning difficulty and mastery degree of the trainee in each scenario. When the average number of training times is relatively high, it indicates that the trainee may have encountered more difficulties in each scenario and needs to spend more time learning and consolidating; therefore, there is a positive correlation between the average number of training times and the post - training duration, and the adjusted post - training duration = the average number of training times ÷ the target number of times × the pre - adjusted post - training duration; where the target number of times is mostly set to 5 times. If the average number of training times ≤ the target number of times, it means that the trainee's number of training times is insufficient, resulting in continuing training at this level. At this time, n compulsory trainings are set, and n = (the target number of times - the average number of training times) × the number of scenarios at this training level.

[0045] Specifically, the operation fluency of the trainee's single - training is determined according to the operation duration of the trainee's single - training, the average reaction time to emergencies, and the smoothness of the change in flight attitude.

[0046] In implementation, weights are assigned to the operation duration, the average response time to emergencies, and the smoothness of the change in flight attitude respectively. Among them, the operation duration and the average response time to emergencies have a greater impact on the operation fluency. Therefore, the weight of the operation duration ≥ the weight of the average response time to emergencies > the weight of the smoothness of the change in flight attitude. Generally, the weight of the operation duration ∈ [0.35, 0.45], the weight of the average response time to emergencies ∈ [0.35, 0.45], and the weight of the smoothness of the change in flight attitude ∈ [0.1, 0.3]. The weighted operation duration, the average response time to emergencies, and the smoothness of the change in flight attitude are added together to determine the operation fluency. It can be understood that the smaller the calculated value of the operation fluency, the smoother the operation of the trainee. In implementation, by judging the smoothness of the change in the aircraft attitude (such as pitch, roll, yaw), usually the change in the roll angle of the aircraft is gradual without sudden large swings, and the pitch attitude can be well controlled during the turning process to maintain the stability of the flight altitude and speed, making the entire flight attitude change process look natural and smooth, indicating that its change smoothness is high. The smoothness of the change in flight attitude is automatically generated by the operating system after each training. In implementation, the smoother the change in flight attitude, the smaller the value (0 - 1) generated by the system.

[0047] In implementation, the preset fluency is determined when the weights are 0.45, 0.4, 0.15 respectively, the average response time to emergencies is 5s (generally speaking, for key emergencies, the preset average response time can be around 3s - 7s), the smoothness of the change in flight attitude is 0.5, and the operation time is 1.2 times the standard operation time. It can be understood that for each training, there is a specified standard operation time.

[0048] In implementation, the training system also includes a machine learning model for generating the smoothness of the change in flight attitude.

[0049] Please refer to Figure 4 as shown, which is the connection diagram of the civil aviation real - time modeling training quality management system according to the embodiment of the present invention. The present invention also provides a civil aviation real - time modeling training quality management system, including, A data acquisition module for obtaining the historical learning data and historical assessment data of trainees; A training evaluation module, connected to the data acquisition module, for evaluating the training level of trainees according to the historical assessment data and evaluating the training status of trainees according to the historical learning data to formulate corresponding training strategies; A practice evaluation module for determining the corresponding scenario training status according to the operation fluency and training performance during the training process; A training adjustment module, which is respectively connected to the training evaluation module and the practice evaluation module, is used to determine the training representation state of the trainee at the corresponding training level according to the training state of each scenario in the training level so as to adjust the training strategy.

[0050] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0051] The above are only the preferred embodiments of the present invention and are not used to limit the present invention; for those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for quality management of real-time modeling training for civil aviation, characterized in that: include: Obtain the student's historical learning data and historical assessment data, wherein the historical learning data includes the number of historical learning times and the duration of each learning time; Evaluate the trainee's training level based on the historical assessment data, and evaluate the trainee's training status based on the historical learning data, wherein the training status includes frequent training and intermittent training; Formulate a corresponding training strategy for the trainee according to the training level and the training status, wherein the training strategy includes the authenticity of the training model and the duration of post-training training; Determine a corresponding scenario training state in response to the operation fluency and training results during the training process, wherein the scenario training state includes a pass state and a practice state; Determining the training representation state of the trainee at the corresponding training level according to the training state of each scene in the training level to adjust the training strategy includes: Adjust the realism of the training model based on the average operation fluency and average training performance of each scenario; and,adjusting the duration of post-training training based on the average number of training sessions for each scenario; Wherein, the training levels include at least four levels.

2. The method for quality management of civil aviation real-time modeling training according to claim 1, characterized in that: The method for evaluating the training level of the trainee according to the historical assessment data includes: Determine the highest assessment score, the lowest assessment score and the average assessment score based on historical assessment data; The training level is initially determined based on the average score of the assessment; The accuracy of the training level is determined according to the difference between the highest score and the lowest score of the assessment to determine the training level twice, wherein: If the score difference is less than or equal to the preset score difference, it is determined that the training level is accurate and there is no need to determine the training level again; If the score difference is greater than a preset score difference, it is determined that the training level is inaccurate and the training level is re-determined based on the ratio of the score difference to the preset score difference.

3. The method for quality management of civil aviation real-time modeling training according to claim 1, characterized in that: The method for evaluating the training status according to the historical learning data includes: Filter invalid historical learning times according to the duration of each learning time to obtain valid historical learning times; Determine the effective learning frequency according to the effective historical learning times; The training state is determined according to the effective learning frequency, wherein: If the effective learning frequency is greater than the preset frequency, the training state is determined to be frequent training; If the effective learning frequency is less than or equal to the preset frequency, the training state is determined to be intermittent training.

4. The method for quality management of civil aviation real-time modeling training according to claim 3 is characterized in that: The method for formulating a corresponding training strategy for a trainee according to the training level and the training status includes: Determining the authenticity of the training model according to the determination results of the training level and the training state; and, determining a post-training training duration based on the training status and the effective learning frequency.

5. The method for quality management of civil aviation real-time modeling training according to claim 4 is characterized in that: The method for determining the authenticity of the training model according to the determination results of the training level and the training state includes: Determining the point cloud density and its distribution mode according to the determination result of the training level; and, determining whether to add texture rendering according to a determination result of the training state.

6. The method for quality management of civil aviation real-time modeling training according to claim 4, characterized in that: Determine the duration of post-training training according to the training status and the effective learning frequency, include, If the training status is frequent training, the post-training training duration is determined to be the standard duration; If the training state is intermittent training, it is determined to increase the post-training training duration and the post-training training duration is determined according to the ratio of the preset frequency to the effective learning frequency.

7. The method for quality management of civil aviation real-time modeling training according to claim 1, characterized in that: Determining the corresponding scene training state in response to the operation fluency and training results during the training process, including: If the operation fluency is greater than or equal to the preset fluency and the training score is greater than or equal to the qualified score, the corresponding scene training status is determined to be a passed status; If the operation fluency is less than a preset fluency and / or the training score is less than a qualified score, the corresponding scene training state is determined to be a practice state.

8. The method for quality management of civil aviation real-time modeling training according to claim 7, characterized in that: Determine the training representation state of the trainee at the corresponding training level according to the training state of each scene in the training level, including: If the ratio of the number of scenes in which the scene training status is a passing state to the total number of scenes is greater than or equal to a preset ratio, then the training representation state is determined to be the completion of the current level of training; If the ratio of the number of scenes in which the scene training status is the passed state to the total number of scenes is less than a preset ratio, it is determined that the training representation state is to continue the current level of training.

9. The method for quality management of civil aviation real-time modeling training according to claim 8, characterized in that: The method for adjusting the training strategy according to the training representation state includes: Obtaining the training representation status of the trainee at the training level to determine whether to adjust the training strategy, wherein: If the training representation state is to continue the current level training, then determining to adjust the training strategy; In response to the determination result of adjusting the training strategy, obtaining the number of training times of the trainee in each training scenario and the operation fluency and training results of each training; Determine the average operation fluency and the average training result according to the operation fluency and the training result of each training respectively; Determining, according to the average operation fluency and the average training performance, to reduce the authenticity of the training model; Increase the length of post-training training based on the average number of training sessions for each scenario.

10. A civil aviation real-time modeling training quality management system using the civil aviation real-time modeling training quality management method according to any one of claims 1 to 9, characterized in that: include, Data collection module, used to obtain students' historical learning data and historical assessment data; A training evaluation module, which is connected to the data collection module, is used to evaluate the training level of the trainees according to the historical assessment data, and to evaluate the training status of the trainees according to the historical learning data, so as to formulate corresponding training strategies; The practice evaluation module is used to determine the corresponding scenario training status according to the operation fluency and training results during the training process; The training adjustment module is connected to the training assessment module and the practice evaluation module respectively, and is used to determine the training representation state of the trainee at the corresponding training level according to the training state of each scene in the training level to adjust the training strategy.

Images