Virtual simulation technology-driven production and teaching integrated teaching system

Abstract

The invention relates to the technical field of education informatization, in particular to a virtual simulation technology-driven production and education integrated teaching system, which is used for acquiring a behavior sequence and calculating association degree, extracting high-frequency behavior chain nodes, marking a failure stage and generating an identification result, optimizing semantic scene configuration, reconstructing a task execution time sequence and adjusting an access authority level. And generating a fusion scheduling path result. A high-frequency behavior chain is extracted by constructing a behavior sequence, a task path key node sequence is defined, and the behavior data recognition precision is enhanced; a node display strategy is adjusted based on the display frequency and the failure proportion, and the matching degree of task content and scene semantics is improved; rearranging a task node sequence according to the adaptation degree, and optimizing an execution path in combination with a role time period and a task priority; the access level is dynamically adjusted through the permission difference matrix, and permission distribution refinement is achieved.

Description

Technical Field

[0001] The present invention relates to the field of educational information technology, and in particular to an industry-education integration teaching system driven by virtual simulation technology. Background Art

[0002] The field of educational information technology includes a systematic method of using digital technology and educational scenarios to deeply integrate to achieve optimal allocation of teaching resources and intelligent management of the teaching process. The content involves building a virtual-real interactive teaching environment through virtual simulation, cloud computing, the Internet of Things and other technologies, integrating industrial practice data with educational theory models, solving problems such as lagging teaching content, insufficient practical resources, and disconnection between skill training and job requirements caused by the traditional separation of industry and education, covering digital modeling of teaching scenarios, real-time synchronization of multi-source data, cross-platform collaboration mechanism design, and the establishment of a dynamic feedback evaluation system.

[0003] Among them, a virtual simulation technology-driven industry-education integration teaching system refers to building a high-precision industry scenario model based on a virtual simulation engine, converting the enterprise production process, equipment operation specifications and job skill requirements into interactive teaching units, and realizing dynamic adaptation of teaching resources and industry standards through data-driven models, and relying on collaborative interactive interfaces to complete real-time operation feedback between teachers and students, enterprise personnel and simulation environments. The main technical means include: bidirectional mapping modeling of teaching scenarios and industrial environments, teaching behavior data collection based on standardized protocols, logical arrangement and conflict resolution mechanism of multi-role collaborative tasks, and real-time operation verification and path optimization based on rule engines.

[0004] In the existing technology, behavioral data processing remains at the level of static recording, lacks the extraction of the evolution law of task paths, and affects the effectiveness of dynamic task adjustment; task configuration relies on preset rules, and is unable to adjust the display strategy according to failed nodes, reducing the pertinence of teaching guidance; path scheduling does not consider adaptability and time resource matching, and often results in task accumulation or resource waste; authority allocation lacks a dynamic mechanism, and task authority configuration is disconnected from actual participation; node scheduling ignores the overlap of role time periods, and collaborative tasks are prone to conflicts, which limits the smooth execution of teaching tasks and the efficiency of resource allocation. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a production-education integration teaching system and method driven by virtual simulation technology.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a production-education integration teaching system driven by virtual simulation technology, the system comprising: The behavior recognition module obtains the interactive behavior data between students and industry roles, constructs behavior sequences according to task numbers, calculates the behavior correlation and the number of task successes, and classifies the behavior sequences to generate behavior pattern recognition results; Based on the behavior pattern recognition result, the semantic scenario optimization module maps teaching tasks and semantic tags, adjusts the display duration and order of nodes according to the mapping result, and generates an optimized result of semantic scenario configuration; The timing scheduling module calls the average execution duration and completion ratio of task nodes in the optimized result of semantic scenario configuration, sorts the node adaptation degrees, cross-compares the role time periods and task priorities, and generates a reconstructed result of task execution timing; The permission adjudication module reads the node role permissions in the reconstructed result of task execution timing, sorts the roles whose permission satisfaction times are lower than the median value of the total number of tasks according to the completion rate, and adjusts the role access permission levels to form a permission adjudication result; According to the available time periods and task priorities of the roles with adjusted permissions in the permission adjudication result, the industry-education integration task scheduling module extracts node combinations with an overlap degree of time periods exceeding the threshold, rearranges the time series to form an integrated scheduling path, and outputs an industry-education integration task scheduling result.

[0007] As a further solution of the present invention, the behavior pattern recognition result includes a behavior number index set, a task success frequency classification table, and a behavior sequence clustering label; the optimized result of semantic scenario configuration includes a task semantic mapping table, a node display duration adjustment item, and a node execution order rearrangement item; the reconstructed result of task execution timing includes a node adaptation degree sorting table, a task priority matching item, and a role time period cross matrix; the permission adjudication result includes a permission level adjustment item, a role access permission classification table, and a permission satisfaction frequency sorting table; the industry-education integration task scheduling result includes a node combination rearrangement sequence, a role available time period integration table, and an integrated task scheduling path set.

[0008] As a further solution of the present invention, the behavior recognition module includes: The behavior collection sub-module obtains the interaction data between students and industrial roles based on the operation logs, records the behaviors according to the task numbers and time sequence and extracts the operation time, calculates the time difference between behaviors and filters out the excessive behaviors, and obtains an interactive behavior interval data set; The sequence construction sub-module clusters the behavior records according to the interactive behavior interval data set, counts the successful times of each task and calculates the frequency, filters out the behavior sequences with a frequency exceeding the mean value, extracts the node order, and generates a high-frequency behavior node sorting sequence; The behavior chain recognition sub-module calls the high-frequency behavior node sorting sequence to identify the continuously failed task stages, calculates the node sorting deviation degree and combines it with the failure times ratio, and uses the formula: ; Obtain the deviation degree of each failure stage, establish the mapping between the stage and the node, and obtain the behavior pattern recognition result; Among them, represents the behavior chain order deviation degree value, Represents the position of the nth node in the high-frequency sorting, represents the average order of the nodes, represents the time difference between adjacent nodes, represents the duration of the failure stage, represents the node delay value.

[0009] As a further solution of the present invention, the semantic scenario optimization module includes: The node data extraction sub-module calculates the average display times based on the total display duration and times of the nodes in the behavior pattern recognition result, obtains the median value of the display times, determines whether the average display times of each node are lower than the median value, filters out the qualified nodes, and generates a low-frequency node set; The failure distribution measurement sub-module calls the failure times of each node in the low-frequency node set, calculates its failure ratio, obtains the average value of the failure ratios of all nodes, determines whether the node failure ratio is higher than this average value, and generates a list of node failure ratio exceeding the mean; The display duration adjustment sub-module calls the list of node failure ratio exceeding the mean and the original display duration, and according to the failure ratio, display times, skip delay time and time difference between nodes, uses the formula: ; Calculates the display duration adjustment value of the node through operation, rearranges the order of the task nodes according to the adjusted display duration, and generates an optimized result of the semantic scenario configuration; Among them, represents the adjusted display duration of the nth task node, is the original display duration of this node, is the failure rate of the nth node, is the average value of the failure rates of all nodes, is the number of node accesses, is the time interval between task nodes, is the delay time caused by the node being skipped.

[0010] As a further solution of the present invention, the timing scheduling module includes: The execution efficiency extraction sub-module calculates the total number of task completions and the total execution time based on the interaction records of each task node in the optimized result of the semantic scenario configuration, and respectively obtains the average execution duration and completion ratio to obtain the node average execution duration value and the node completion ratio value; The adaptability calculation sub-module calls the node average execution duration value and the node completion ratio value, combines the failure probability, execution fluctuation and task complexity indicators, and uses the formula: ; Obtain the task node adaptation degree value of the node through calculation, then compare the adaptation degree value with the set threshold, and screen and generate a high adaptation degree node set; Among them, represents the adaptation degree value of the th task node, is the node completion rate, is the failure probability, is the average execution duration of all nodes, is the average execution duration of node is the execution duration variance of node is the relative task complexity; The task time sequence generation sub-module calls the high adaptation degree node set, cross-compares according to the task priority and the available time period of the role, and rearranges the node order according to the time window to generate a task execution time sequence reconstruction result.

[0011] As a further solution of the present invention, the permission adjudication module includes: The role permission reading sub-module obtains the role access permission level and task node permission requirements of each task node according to the task execution time sequence reconstruction result, and generates a role access permission record; The permission difference matrix construction sub-module constructs a role permission difference matrix based on the role access permission record, and uses the formula: ; Calculate the permission difference of each role in different task nodes to generate a permission difference matrix; Among them, represents the permission difference between role and task node , represents the access permission level of role in task node , represents the permission requirement of task node in task , is the total number of tasks; The permission adjustment sub-module extracts the role set with the number of times of permission satisfaction lower than the median value of the total number of tasks in the permission satisfaction times matrix, obtains the completion rate of the role set in the task, adjusts the role access permission level, and generates a permission adjudication result.

[0012] As a further solution of the present invention, the industry-education integration task scheduling module includes: The role information reading sub-module reads the adjusted permission roles in the permission adjudication result again, reads the available time periods of the roles and the teaching task priorities again, obtains the available time period information of each role in a specific task, and sorts according to the priorities of the teaching tasks to generate a record of the available time periods of the roles; The task node adaptability sorting sub-module calculates the adaptability of the role nodes based on the role available time period record and the teaching task priorities, evaluates and sorts the matching degree between the roles and the task nodes to generate a task node adaptability sorting; The scheduling path construction sub-module extracts the task node combinations with the overlapping degree of available time periods between roles exceeding the overlapping degree threshold in the task node adaptability sorting, rearranges them according to the time sequence, and uses the formula: ; Performs operations to obtain the rescheduling of the scheduling path, forms an integration of production and education scheduling path, and generates a task scheduling result for the integration of production and education; Among them, represents the total length of the overlapping time period of the task node , represents the available time period length of the role in the task node , represents the task execution duration of the role in the task node , is the number of roles participating in this task.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by constructing a behavior sequence to extract high-frequency behavior chains, the order of key nodes of the task path is clarified, and the recognition accuracy of behavior data is enhanced; based on the display frequency and failure ratio, the node display strategy is adjusted to improve the semantic matching degree of task content and scenario; according to the adaptability, the order of task nodes is rearranged, and the execution path is optimized by combining the role time period and task priority; through the permission difference matrix, the access level is dynamically adjusted to achieve refined permission allocation; high-overlap node combinations are identified to rearrange the time sequence, improving the task collaboration efficiency and resource utilization rate, and constructing a task scheduling optimization mechanism with a closed structure loop. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is the system flow chart of the present invention; Figure 2 is the acquisition flow chart of the behavior recognition module of the present invention; Figure 3 is the acquisition flow chart of the semantic scenario optimization module of the present invention; Figure 4 is the acquisition flow chart of the time sequence scheduling module of the present invention; Figure 5The flowchart for obtaining the permission adjudication module of the present invention; Figure 6 The flowchart for obtaining the task scheduling module of the integration of production and education of the present invention. Specific implementation manner

[0015] The following combines the accompanying drawings to describe the technical solutions in the present invention.

[0016] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as more preferred or more advantageous than other embodiments or design solutions. Precisely, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0017] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same. "(of)", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same.

[0018] In the embodiments of the present invention, sometimes subscripts such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0019] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail in combination with the accompanying drawings and specific embodiments.

[0020] Please refer to Figure 1 , the present invention provides a technical solution: a production-education integration teaching system driven by virtual simulation technology. The system includes: The behavior recognition module obtains the interaction behavior data between students and industrial roles, constructs a behavior sequence according to the task number and time sequence, calculates the behavior correlation degree based on the operation time difference, counts the number of successful tasks according to the feedback information, classifies the behavior sequence and calculates the occurrence frequency, extracts the node order in the behavior chain with a frequency exceeding the sequence mean, and marks the task stage corresponding to the continuous failure of the behavior sequence to generate a behavior pattern recognition result; Based on the semantic content of task nodes in the behavior pattern recognition result, the semantic scenario optimization module maps teaching tasks to semantic labels, reads the total display duration and number of times of each node in the behavior sequence and calculates the average value, filters the set of nodes with the average display times lower than the median value of the nodes, counts the number of times students execute tasks unsuccessfully in the corresponding behavior paths and calculates the failure rate, redistributes the display duration of nodes with a failure rate exceeding the average failure rate and adjusts the order of task nodes to generate an optimized result of semantic scenario configuration; The timing scheduling module calls the total execution time and total completion times of task node interaction records in the optimized result of semantic scenario configuration, calculates the average execution duration and completion ratio according to node grouping, sorts the nodes based on the calculation results for adaptability, extracts the nodes with an adaptability exceeding the set adaptability threshold, and cross-compares the available time periods of roles and task priorities to generate a reconstructed result of task execution timing; The permission adjudication module reads the access permission levels of the corresponding roles of each task node and the task permission requirements of the nodes in the reconstructed result of task execution timing, constructs a role permission difference matrix and counts the number of times the roles meet the permissions in the tasks, extracts the set of roles with the number of times of permission satisfaction lower than the median value of the total number of tasks, obtains the task completion rate of the role set in all task records, sorts according to the completion rate and adjusts the role access permission levels to form a permission adjudication result; The industry-education integration task scheduling module re-reads the available time periods of roles and the priorities of teaching tasks according to the roles with adjusted permissions in the permission adjudication result, sorts the adaptability of role nodes in teaching tasks and constructs a task priority execution sequence, extracts the combination of task nodes with an overlap degree of available time periods between roles in the priority sequence exceeding the overlap degree threshold, rearranges according to the time series to form an integrated scheduling path, and outputs the industry-education integration task scheduling result.

[0021] The behavior pattern recognition result includes a behavior number index set, a task success frequency classification table, and a behavior sequence clustering label; the optimized result of semantic scenario configuration includes a task semantic mapping table, a node display duration adjustment item, and a node execution order rearrangement item; the reconstructed result of task execution timing includes a node adaptability sorting table, a task priority matching item, and a role time period cross matrix; the permission adjudication result includes a permission level adjustment item, a role access permission classification table, and a permission satisfaction frequency sorting table; the industry-education integration task scheduling result includes a node combination rearrangement sequence, a role available time period integration table, and an integrated task scheduling path set.

[0022] Please refer to Figure 2 , the behavior recognition module includes: The behavior collection sub-module obtains the interaction data between students and industrial roles based on the operation logs, records the behaviors according to the task numbers and time sequence and extracts the operation times, calculates the time differences between behaviors and filters out the excessive behaviors to obtain an interactive behavior interval data set; All the log records generated by the user on the interaction platform are called, and the record fields related to the task execution behavior are extracted, including user ID, task number, behavior event name, behavior event timestamp, behavior feedback flag, event duration, terminal device identifier, operation path and operation type. By traversing this type of log records, the behavior events are sorted in ascending order according to the timestamp field, and a behavior event sequence arranged in chronological order is constructed. For example, for a user with student ID S001, the behavior records generated during task phase A include a series of behavior events such as entering the task, clicking on resources, and submitting results, with timestamps of 10:03:12, 10:03:25, and 10:04:40 respectively. After the system sorts the behavior sequence in turn, it locates the operation order. Then, it is grouped by the task number field, and the start and end time points of each behavior sequence within each group of tasks are extracted and the duration is calculated to obtain the start and end times of each task behavior chain. Subsequently, the task feedback is classified according to "success" or "failure" in the feedback flag field. At the same time, a time difference operation is performed in the sequence, and the time difference threshold is set to 180 seconds. When the time interval between two consecutive behaviors exceeds this threshold, the system considers that an interrupted operation behavior has occurred, and this behavior segment is removed from the analysis sequence. For example, the behavior intervals of student S001 in task A are (13 seconds from 10:03:12 to 10:03:25, 75 seconds from 10:03:25 to 10:04:40), both of which do not exceed 180 seconds, so this behavior chain is retained. Conversely, if the interval exceeds 180 seconds, such as reaching 250 seconds, this segment of the behavior chain is removed, and the remaining behaviors are recombined into a new valid behavior segment. In addition, the time difference between each pair of retained behavior events is calculated to form a time interval list. For example, the time difference between clicking on resources and submitting results is 75 seconds, and this value is recorded as the time interval value of the behavior pair in the interaction behavior interval dataset. Each record in the interaction behavior interval dataset includes five fields: task number, pre-behavior time, post-behavior time, operation time difference, and behavior feedback flag. For example, task number A001, pre-behavior 10:03:25, post-behavior 10:04:40, time interval 75 seconds, feedback flag "success", and finally an interaction behavior interval dataset under multiple users and multiple task dimensions is formed.

[0023] The sequence construction sub-module clusters the behavior records according to the interaction behavior interval dataset, counts the number of successful times of each task and calculates the frequency, filters the behavior sequences with frequencies exceeding the mean value, extracts the node order, and generates a high-frequency behavior node sorting sequence; For each user task chain, the task behavior sequence is abstracted into a node sequence. The node identifier consists of the behavior event name plus the time difference, such as "Click on resource (13s) - Submit task (75s)". Subsequently, the system performs frequency statistics on each behavior node sequence. For each type of behavior path sequence, such as "Enter task - View resource - Submit task", the system records the number of times it appears in the dataset. By traversing all user task behavior chains and statistically calculating the frequency in the way of sequence combination matching, the statistical threshold is set as the frequency mean. Suppose the total number of behavior path types is 120, and the frequencies are as follows: Sequence A appears 32 times, B appears 26 times, C appears 18 times, D appears 7 times, etc., then the statistical mean is , and the system only retains the path sequences with frequencies greater than the mean. In this example, sequences A, B, and C are retained; after the screening is completed, it enters the node order extraction stage. For each high-frequency path, extract its node composition and the interval order according to the sequence of behavior occurrences. For example, sequence A is "Login - Browse - Ask questions - Submit", and its time intervals are (12s, 34s, 51s) in turn. The node order is N1→N2→N3→N4, which are respectively recorded as N1 = "Login", N2 = "Browse", etc. During the process of recording the node order, the same type of high-frequency sequences are standardized numbered and weight-converted. For example, if the frequency of the path "Browse - Submit - Answer questions" is 26 times, accounting for about 21.6% of all behavior paths, then the corresponding weight is recorded as 0.216. Each high-frequency behavior path calculates the weight value according to frequency normalization , where is the number of occurrences of the th high-frequency sequence, is the total number of high-frequency sequences. The finally formed high-frequency behavior node sorting sequence is a list of triples containing several "node order + behavior interval + weight value", such as: Sequence 1 = [N1 Login, N2 Browse, N3 Submit], the time intervals are (12s, 27s) respectively, and the weight is 0.268. This list will be used as the standard sorting reference data for subsequent behavior chain offset analysis and input into the system for behavior chain comparison processing in the failure stage.

[0024] The behavior chain recognition sub-module calls the high-frequency behavior node sorting sequence to identify the continuous failure task stage, calculates the node sorting offset degree and combines it with the failure times ratio, using the formula: ; Obtain the offset degree of each failure stage, establish the mapping between the stage and the node, and obtain the behavior pattern recognition result; Among them, represents the behavior chain order offset degree value, represents the position of the th node in the high-frequency sorting, represents the average order of the nodes, Represents the time difference between adjacent nodes, Represents the duration of the failure stage, Represents the node delay value; Read the behavior node chain of each task and its execution feedback label, screen the tasks with feedback of "failure" and segment them in chronological order. Assume that task numbers T001 to T004 fail continuously, and the system divides them into a failure stage. Subsequently, enter the offset calculation process. For all behavior sequences in this failure stage, the system calls the order of its behavior nodes and the high-frequency behavior node sorting sequence to compare node by node to calculate the offset of each node's position in the high-frequency sorting. Set the high-frequency sequence as [N1: Login, N2: Browse, N3: Submit, N4: Answer questions], and the actual execution sequence as [N2: Browse, N1: Login, N4: Answer questions, N3: Submit]. The system converts its position information into position indexes (the high-frequency sequence indexes are 1, 2, 3, 4, and the actual ones are 2, 1, 4, 3), and calculates the offset value of each node as |2−1| = 1, |1−2| = 1, |4−3| = 1, |3−4| = 1, and the total offset is 4; the average order of the nodes is , and the calculation of the difference part of the node order is . Subsequently, the system calls the list of time differences between adjacent behavior nodes for square sum processing. Assume the differences are 35 seconds, 28 seconds, and 47 seconds respectively, and get , and the square root is . Then obtain the total execution duration of this failure stage . Assume the start and end times of the behavior chain are from 10:05:12 to 10:09:52, and the total duration is 280 seconds. At the same time, obtain the response delay value of each node , for example, the node delays are 8 seconds, 10 seconds, and 6 seconds respectively, seconds, substitute into the formula: ; Obtain the order offset of the behavior chain in this failure stage . Subsequently, calculate the offset values for all failure stages, establish a mapping relationship, and record information such as the node sequence, offset value, and time span of each failure stage for behavior pattern structure comparison and risk identification; during this process, the system sets an offset warning threshold is 0.2. Judging from experience, the offset degree between 0.1 and 0.2 is the controllable stage, above 0.2 is the serious deviation, and below 0.05 is the standard operation stage. Combining the actual detection results, it is determined that this stage is the serious deviation stage. It is necessary to mark this behavior chain as an abnormal sequence for subsequent identification training processing, and finally obtain the behavior pattern recognition result in the form of "failure stage number T001 - T004, offset degree 0.227, high node partial order degree, action delay concentrated between node N3 submission and N4 answering questions, and it is necessary to focus on the extraction and correction of abnormal behavior characteristics".

[0025] Please refer to Figure 3 , the semantic scene optimization module includes: The node data extraction sub-module calculates the average display times based on the total display duration and times of the nodes in the behavior sequence, obtains the median value of the display times, judges whether the average display times of each node are lower than the median value, filters out the qualified nodes, and generates a low-frequency node set; Extract the duration data and access times data of the task nodes that appear in multiple user behavior data. Aggregate the behavior logs with "behavior type" as the field, filter out the display type nodes with complete start and end times in the behavior events, and record the duration data of each node activation. In the example, node A appears 3 times in the data of students S001, S002, and S003, with display durations of 12 seconds, 18 seconds, and 14 seconds each time, the total display duration is 44 seconds, and the access times is 3. Calculate its average display times as , and the average display duration is seconds; the system sorts all task nodes in ascending order according to the display times and obtains the median display times. If the total number of nodes in the system is 13 and its display times set is: 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 5, 6, 7, then the median display times is the 7th item, that is, 3. Then judge whether each node is less than this median value. For example, the display times of node A is 3, so it does not meet the conditions, while the display times of node B is 2, so it enters the screening set; the system performs this comparison operation node by node and outputs the set of all nodes with display times less than the median 3 as the subsequent processing items. In the actual dataset, the number of screened nodes is 6, and the node names are B, C, F, G, I, J, with access times of 2, 1, 2, 1, 1, 2 respectively, and the total display durations are 28 seconds, 12 seconds, 24 seconds, 15 seconds, 9 seconds, 18 seconds respectively; this judgment process is based on the median judgment logic and has a clear division interval. The display times interval can be set as: low-frequency display ≤ 2, normal display 3 - 5, high-frequency display > 5, to form a stable task execution frequency partition, and finally the system outputs the low-frequency node set.

[0026] The failure distribution calculation sub-module calls the failure times of each node in the low-frequency node set, calculates its failure proportion, obtains the average value of the failure proportions of all nodes, determines whether the failure proportion of a node is higher than this average value, and generates a list of the ratio of nodes with failure higher than the average value; Match the task feedback identification field from the behavior records, filter the records with the task result in the failure state, and count the number of times each low-frequency node appears in the failed tasks. Suppose node B appears 6 times in the failed task sequence and 12 times in all behavior paths in total, then its failure proportion is , and all low-frequency nodes are counted in this way in turn. The example calculation results are: node B (0.50), C (0.75), F (0.40), G (0.20), I (0.33), J (0.60). The system then calculates the average value of the failure proportions of these nodes as , this value is the "benchmark of the average failure proportion". Then, compare the failure proportion of each node with 0.46. If it is greater than this average value, it enters the high-risk display node set. For example, nodes C (0.75) and J (0.60) are higher than 0.46, and the system marks them as "nodes with relatively high failure proportion"; this judgment logic does not depend on the setting of abstract weights, but is determined by specific ratios, and the failure rate interval is delimited as follows: failure rate > 0.46 is a high-failure node, 0.30 - 0.46 is a medium-risk node, < 0.30 is a stable node. The division has a direct interval judgment logic and the values have engineering significance, and finally a list of the ratio of nodes with failure higher than the average value is generated.

[0027] The display duration adjustment sub-module calls the list of the ratio of nodes with failure higher than the average value and the original display duration, and according to the failure proportion, display times, skip delay time and time difference between nodes, uses the formula: ; Calculate through operations to obtain the display duration adjustment value of the node, rearrange the order of task nodes according to the adjusted display duration, and generate the optimized result of the semantic scene configuration; Among them, represents the adjusted display duration of the th task node, is the original display duration of this node, is the failure rate of the th node, is the average value of the failure rates of all nodes, is the number of times the node is accessed, is the time interval between task nodes, is the delay time caused by the node being skipped; Suppose the original display duration of node C is T k= 24 seconds, its failure rate FR1 = 0.75, the number of visits VC1 = 3, the time difference data between nodes is: TD1 = 6 seconds, TD2 = 5 seconds, TD3 = 8 seconds, the skip delay time is SD1 = 4 seconds, SD2 = 3 seconds, SD3 = 2 seconds, the calculated failure rate average is , substitute it into the formula, and the result is as follows: Each item is expanded as follows: The molecular part is: ; The denominator is: ; The overall incremental value is: ; Therefore, the display duration of node C is adjusted to: ; Finally, the node display duration adjustment value is output, and the nodes are rearranged according to the value to generate the semantic scene configuration optimization result. The result shows that the display time of node C needs to be increased due to its high failure rate and frequent skipping. The adjusted display duration is 24.07 seconds, which exceeds the original display duration by 0.07 seconds, and the corresponding adjustment ratio is 0.29%. This numerical result means that the visibility of the node in the semantic scene needs to be increased, and the display stage should be extended to align the density of its abnormal behavior characteristics. This value directly participates in the node order reordering as a display duration adjustment factor and affects the overall task chain structure, and is further used for the derivation and output of the semantic scene configuration optimization results.

[0028] See also Figure 4 , the timing scheduling module includes: The execution efficiency extraction submodule calculates the total number of task completions and the total execution time according to the node number based on the interaction records of each task node in the semantic scene configuration optimization result, and respectively obtains the average execution time and completion ratio of the node to obtain the average execution time value of the node and the node completion ratio value; The task nodes are numbered and grouped by node dimension. The total number of task completions and total execution time accumulated by each node in all interaction behaviors are extracted in turn. The number of task completions is obtained by counting the number of entries with the task status field as "completed" in the interaction records. The total execution time is obtained by summing the difference between the start and end time of the aggregation node. In the example, node T001 has 8 completed records out of 12 interactions, and the total execution time is 1020 seconds. The average execution time of the node is seconds, the node completion rate is , the system executes the same process for all nodes, obtains two key metric parameters corresponding to the node numbers, and establishes an index mapping table. In this mapping table, each node number is recorded along with its average execution duration and completion ratio. The system deletes abnormal records without duration or status fields based on field integrity verification. In the formal operation scenario of the system, students S001, S002, and S003 all execute task nodes T001, T002, and T003. There are a total of 72 behavior data records, involving 3 roles. The total number of task completion markers is 52, and the total number of activated node records is 96. Through the above statistics and calculations, the average execution duration value and node completion ratio value of all nodes are obtained.

[0029] The adaptability calculation sub-module calls the average execution duration value and node completion ratio value of the nodes, combines the failure probability, execution fluctuation, and task complexity metrics, and uses the formula: ; Performs calculations to obtain the task node adaptability value of the nodes, and then compares the adaptability value with the set threshold to screen and generate a set of high-adaptability nodes; Among them, represents the adaptability value of the th task node, is the node completion rate, is the failure probability, is the average execution duration of all nodes, is the average execution duration of node , is the variance of the execution duration of node , is the relative task complexity; The way to obtain the failure probability FP of a task node is the number of failures divided by the number of activations. Assuming that node T002 is activated 15 times and fails 5 times, we get FP2 = , the node completion rate PC is the number of successes divided by the number of activations. Assuming the number of successes is 10 times, then PC2 = , the average execution duration DT2 = 135 seconds. The average duration of all nodes is statistically seconds, the variance DV2 is set to 49 seconds², the complexity level code DC is set to 5, and the maximum level is 10. Therefore, RC2 = , substituting into the formula for calculation: ; Expanding and calculating the formula: The numerator = ; The denominator = ; The adaptability value = ; Referring to the analysis of the standard deviation of the stability interval and the mean completion rate of the reference nodes, if the adaptation threshold is set to 1.0, and derived from the distribution of the adapted nodes and abnormal nodes in the previous training set, the interval is set to 0.8 - 1.2. Therefore, those with an adaptation degree higher than 1.0 are determined to have sufficient adaptability. The node T002 meets the condition and is marked as a schedulable task; The result shows that the adaptation degree value of node T002 is 1.019, which has exceeded the 1.0 threshold, indicating that it can support the scheduling arrangement in terms of behavior stability, duration consistency, and complexity bearing capacity. This value directly constitutes the adaptation degree value of the task node and enters the subsequent node screening.

[0030] The task timing generation sub-module calls the set of nodes with high adaptation degrees, cross-compares according to the task priority and the available time periods of the roles, and rearranges the node order according to the time window to generate the result of the task execution timing reconstruction; Arrange the highly adapted nodes in descending order according to the task priority field, and then traverse the role to which each node belongs to query its available time period interval during the task execution cycle. For example, the role corresponding to node T002 is R03, and during the task cycle from June 1st to June 15th, 2024, the available time periods are 8:00 - 11:30 and 13:30 - 17:00 every day. The system cross-matches the estimated execution time of the task nodes with the idle periods of the roles. If the average execution time of the node is 135 seconds, the system inserts a time window in the role gap in minutes. After performing the same process for all nodes, sort all nodes according to the task priority and time sequence to generate a new node time schedule, which records the node number, estimated start time, estimated end time, bound role, and the task priority to which it belongs. The system checks whether there are conflicts and time overlaps between the nodes, and when an overlap is detected, eliminates the node with a lower priority or delays it according to the priority, and processes cyclically until a non-overlapping time plan for all tasks is generated, and finally obtains the result of the task execution timing reconstruction.

[0031] Please refer to Figure 5 , the permission adjudication module includes: The role permission reading sub-module obtains the role access permission level and the task node permission requirements of each task node according to the result of the task execution timing reconstruction, and generates a role access permission record; Suppose the access permission level of Role A is 2, indicating that the access permission of this role on the task node is at a medium level. The access permission level of Role B is 3, indicating that it has a higher access permission. The node task permission requirements define the minimum permission level required to complete the task. For example, if the permission requirement of a certain task node is 2, it means that the node requires the role to have at least medium permission to operate. During this process, all role access permissions and task node permission requirements are extracted from the database or records reconstructed according to the task execution time sequence and used as the data input for subsequent processing to generate a role access permission record. This record contains the permission information of each role on each task node and serves as the basis for constructing and adjusting the subsequent permission difference matrix.

[0032] Based on the role access permission record, the permission difference matrix construction sub-module constructs a role permission difference matrix using the formula: ; Calculate the permission difference of each role on different task nodes to generate a permission difference matrix; Among them, represents the permission difference between role and task node , represents the access permission level of role on task node , represents the permission requirement of task node in task , is the total number of tasks; Suppose there are 3 roles (A, B, C) and 3 task nodes, and the permission requirements of the task nodes are: [Task 1 permission requirement = 2, Task 2 permission requirement = 3, Task 3 permission requirement = 1]. The access permission levels of each role are as follows: Role A: Task 1 level = 2, Task 2 level = 3, Task 3 level = 1; Role B: Task 1 level = 1, Task 2 level = 2, Task 3 level = 1; Role C: Task 1 level = 3, Task 2 level = 2, Task 3 level = 3; Substitute into the formula for calculation: Step 1: Calculate the permission difference: For Role A: Task 1 difference: (Meet the permission requirement); Task 2 difference: (Meet the permission requirement); Task 3 difference: (Meet the permission requirement); For Role B: Difference in Task 1: (Permission requirements not met); Difference in Task 2: (Permission requirements not met); Difference in Task 3: (Permission requirements met); For Role C: Difference in Task 1: (Permission requirements not met); Difference in Task 2: (Permission requirements not met); Difference in Task 3: (Permission requirements not met).

[0033] Step 2: Count the number of times the permissions of each role are met in the tasks: For Role A: The number of tasks with permission requirements met = 3 (all tasks are met); For Role B: The number of tasks with permission requirements met = 1 (only Task 3 is met); For Role C: The number of tasks with permission requirements met = 0 (no task is met).

[0034] Step 3: Compare with the median of the total number of tasks: The total number of tasks is 3, so the median of the total number of tasks is 2. According to the median, the number of times Role A's permissions are met is higher than the median, while those of Role B and Role C are lower than the median.

[0035] Step 4: Generate a matrix of the number of times permissions are met: The generated matrix of the number of times permissions are met is as follows: For Role A: The number of times met = 3; For Role B: The number of times met = 1; For Role C: The number of times met = 0; This result indicates that Role A has met the permission requirements at all task nodes, meaning this role has sufficient permissions to execute all tasks. Therefore, the number of times its permissions are met is 3, exceeding the median of the total number of tasks, which is 2; Role B only meets the permission requirements in Task 3, with the number of times permissions are met being 1, lower than the median, so its access permissions need to be further adjusted; Role C fails to meet the permission requirements for any task, with the number of times permissions are met being 0, which means its permission settings do not meet the task requirements at all and also need permission adjustment.

[0036] The permission adjustment sub-module extracts the set of roles whose number of times permissions are met in the matrix of the number of times permissions are met is lower than the median of the total number of tasks, obtains the completion rate of the role set in the tasks, adjusts the role access permission levels, and generates a permission adjudication result; Extract the set of roles from the permission satisfaction frequency matrix whose permission satisfaction frequencies are lower than the median value of the total number of tasks. These roles may not meet the permission requirements on many task nodes due to insufficient or mismatched permissions. Calculate the task completion rate of these roles in all task nodes, that is, count the proportion of tasks completed by each role. Suppose role A completed tasks in task 1 and task 3, but did not complete task 2, then the completion rate of role A is 2 / 3, that is, 66.7%. By comparing the completion rates of all roles, sort the roles according to the completion rate. The roles with lower completion rates will have their access permission levels adjusted. Usually, the roles with lower completion rates will be given higher access permissions to ensure that they can complete tasks smoothly during task execution. Finally, adjust the access permission levels of roles according to the sorting results to obtain the permission adjudication result.

[0037] Please refer to Figure 6 , the industry-education integration task scheduling module includes: The role information reading sub-module re-reads the available time slots of roles and the teaching task priorities according to the roles with adjusted permissions in the permission adjudication result, obtains the available time slot information of each role in specific tasks, and sorts them according to the priorities of teaching tasks to generate role available time slot records; Obtain the permission adjustment situation of each role. The adjusted permissions determine the task nodes and time periods that the role can participate in. In the actual scenario, role A may be assigned to task node 1, role B to task node 2, and role C to task node 3. Obtaining the available time slot information of the role is based on the schedule of actual tasks, the idle time of the role, and the requirements of teaching tasks. Sort according to the available time slots of each role and the priorities of each task. The priority sorting is usually determined according to the importance, urgency of the task, and the key role of the role in the task. For example, the priority of task 1 is higher than that of task 2, and task 2 is higher than task 3. Finally, allocate suitable roles to each task node to generate role available time slot records.

[0038] The task node adaptability sorting sub-module calculates the adaptability of role nodes based on the role available time slot records and the teaching task priorities, evaluates and sorts the matching degree between the role and the task node to generate the task node adaptability sorting; The calculation of the fitness between a role and a task node is carried out by comparing the available time period of the role and the time requirement of the task node. For example, if task node 1 requires to be carried out from 8:00 to 10:00 and the available time period of role A is from 9:00 to 10:00, then the fitness of role A for this task node is 1 hour. The fitness is quantified according to the time overlap degree between the role and the task. For each role, the fitness value represents the actual available participation time of the role at the task node. Then, by evaluating the fitness of each role at the task node, the task nodes are sorted from high to low according to the role fitness. The combination of the role with high fitness and the task node will be scheduled first. The basis for sorting is mainly the priority of the task node and the available time of the role in the task, and the task node fitness sorting is generated.

[0039] The scheduling path construction sub-module extracts the combination of task nodes with the time overlap degree between roles exceeding the overlap degree threshold in the task node fitness sorting, rearranges them according to the time sequence, and uses the formula: ; Performs operations to obtain the rescheduling of the scheduling path, forms the scheduling path of the integration of production and education, and generates the scheduling result of the integration of production and education tasks; Among them, represents the total length of the overlapping time period of task node , represents the available time period length of role at task node , represents the task execution duration of role at task node , is the number of roles participating in this task; Suppose that for task node 1, role A has a task execution duration of 1 hour and the available time period of role A is 2 hours (9:00 - 11:00); for role B in task node 1, the task execution duration is 1.5 hours and the available time period of role B is 1 hour (9:00 - 10:00); for role A in task node 2, the task execution duration is 1 hour and the available time period of role A is 1.5 hours (9:00 - 10:30); for role B in task node 2, the task execution duration is 2 hours and the available time period of role B is 2 hours (9:00 - 11:00).

[0040] Overlapping time period of task node 1: For role A: ; For role B: ; ; Overlapping time period of task node 2: For role A: ; For character B: ; ; The calculated overlapping periods are 2 hours for task node 1 and 3 hours for task node 2. If the set overlapping degree threshold is 0.3 hours, it can be determined that task node 1 and task node 2 meet the combination conditions and are rearranged. In this case, since task node 2 has a longer overlapping period, it will be scheduled first, and finally a scheduling path is formed.

[0041] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claimed rights.

Claims

1. A production-education integration teaching system driven by virtual simulation technology, characterized in that, The system includes: The behavior recognition module obtains the interaction behavior data between students and industrial roles, constructs a behavior sequence according to the task number, calculates the behavior correlation degree and the number of task successes, classifies the behavior sequence, and generates a behavior pattern recognition result; The semantic scenario optimization module maps teaching tasks and semantic labels based on the behavior pattern recognition result, adjusts the display duration and order of nodes according to the mapping result, and generates a semantic scenario configuration optimization result; The timing scheduling module calls the average execution duration and completion ratio of task nodes in the semantic scenario configuration optimization result, sorts the node adaptability, cross-compares the role time period and task priority, and generates a task execution timing reconstruction result; The permission adjudication module reads the node role permissions in the task execution timing reconstruction result, sorts the roles whose permission satisfaction times are lower than the median value of the total number of tasks according to the completion rate, and adjusts the role access permission level to form a permission adjudication result; The industry-education integration task scheduling module extracts the node combination with the time period overlap degree exceeding the threshold according to the available time period and task priority of the roles with adjusted permissions in the permission adjudication result, rearranges the time series to form an integrated scheduling path, and outputs the industry-education integration task scheduling result.

2. The integrated production and education teaching system driven by virtual simulation technology according to claim 1, characterized in that: The behavior pattern recognition result includes a behavior number index set, a task success frequency classification table, and a behavior sequence clustering label; the semantic scenario configuration optimization result includes a task semantic mapping table, a node display duration adjustment item, and a node execution order rearrangement item; The task execution timing reconstruction result includes a node adaptability sorting table, a task priority matching item, and a role time period cross matrix; the permission adjudication result includes a permission level adjustment item, a role access permission classification table, and a permission satisfaction frequency sorting table; The industry-education integration task scheduling result includes a node combination rearrangement sequence, a role available time period integration table, and an integrated task scheduling path set.

3. The teaching system for integration of production and education driven by virtual simulation technology according to claim 1, characterized in that: The behavior recognition module includes: The behavior collection sub-module obtains the interaction data between students and industrial roles based on the operation log, records the behavior according to the task number and time sequence, extracts the operation time, calculates the time difference between behaviors, filters out the excessive behaviors, and obtains the interactive behavior interval data set; The sequence construction sub-module clusters the behavior records according to the interactive behavior interval data set, counts the number of successes of each task, calculates the frequency, filters out the behavior sequences with frequencies exceeding the mean value, extracts the node order, and generates a high-frequency behavior node sorting sequence; The behavior chain recognition sub-module calls the high-frequency behavior node sorting sequence to identify the continuously failed task stages, calculates the node sorting offset degree and combines it with the failure times ratio, and uses the formula: ； Obtain the offset degree of each failure stage, establish the mapping between the stage and the node, and obtain the behavior pattern recognition result; Among them, represents the offset value of the behavior chain order, represents the position of the th node in the high-frequency sorting, represents the average order of the nodes, represents the time difference between adjacent nodes, represents the duration of the failure stage, represents the node delay value.

4. The teaching system for integration of production and education driven by virtual simulation technology according to claim 1, characterized in that: The semantic scenario optimization module includes: The node data extraction sub-module calculates the average display times based on the total display duration and times of nodes in the behavior pattern recognition result, obtains the median value of the display times, judges whether the average display times of each node are lower than the median value, filters out the qualified nodes, and generates a low-frequency node set; The failure distribution measurement sub-module calls the failure times of each node in the low-frequency node set, calculates its failure proportion, obtains the average value of the failure proportions of all nodes, determines whether the node failure proportion is higher than this average value, and generates a list of node ratios with failure proportions above the average; The display duration adjustment sub-module calls the list of node ratios with failure proportions above the average and the original display duration, and based on the failure proportion, display times, skip delay time, and time difference between nodes, uses the formula: ； Performs calculations to obtain the display duration adjustment value for the node, rearranges the order of task nodes according to the adjusted display duration, and generates an optimized result for semantic scenario configuration; Among them, represents the display duration after adjustment of the th task node, is the original display duration of this node, is the failure rate of the th node, is the average value of the failure rates of all nodes, is the number of times the node is accessed, is the time interval between task nodes, is the delay time caused by the node being skipped.

5. The integrated production-education teaching system driven by virtual simulation technology according to claim 1, characterized in that: The timing scheduling module includes: The execution efficiency extraction sub-module, based on the interaction records of each task node in the optimized result of semantic scenario configuration, calculates the total number of task completions and the total execution time according to the node numbers, respectively obtains the average execution duration and completion ratio, and gets the node average execution duration value and the node completion ratio value; The adaptability calculation sub-module calls the node average execution duration value and the node completion ratio value, combines the failure probability, execution fluctuation, and task complexity indicators, and uses the formula: ； Performs calculations to obtain the task node adaptability value for the node, then compares the adaptability value with a set threshold, and filters and generates a set of nodes with high adaptability; Among them, represents the fitness value of the th task node, is the node completion rate, is the failure probability, is the average execution duration of all nodes, is the average execution duration of node , is the average execution duration of node is the variance of the execution duration of node is the relative task complexity; The task timing generation sub-module calls the set of nodes with high adaptability, cross-compares according to the task priority and the available time period of the role, rearranges the node order according to the time window, and generates a result of task execution timing reconstruction; 6. The teaching system for integration of production and education driven by virtual simulation technology according to claim 1, characterized in that: The permission adjudication module includes: The role permission reading sub-module, according to the result of task execution timing reconstruction, obtains the role access permission level and task node permission requirements of each task node, and generates a role access permission record; The permission difference matrix construction sub-module, based on the role access permission record, constructs a role permission difference matrix, and uses the formula: ； Calculates the permission differences of each role in different task nodes and generates a permission difference matrix; Among them, represents the role and the task node the permission difference between them, represents the role at the task node the access permission level, represents the task node in the task the permission requirement, is the total number of tasks; The permission adjustment sub-module extracts the set of roles whose permission satisfaction times in the permission satisfaction times matrix are lower than the median value of the total number of tasks, obtains the completion rate of the role set in the task, adjusts the role access permission level, and generates a permission adjudication result; 7. The integrated production-education teaching system driven by virtual simulation technology according to claim 1, wherein: The industry-education integration task scheduling module includes: The role information reading sub-module, according to the roles with adjusted permissions in the permission adjudication result, rereads the available time period of the role and the priority of teaching tasks, obtains the available time period information of each role in a specific task, and sorts according to the priority of teaching tasks, and generates a record of the available time period of the role; The task node adaptability sorting sub-module, based on the record of the available time period of the role and the priority of teaching tasks, calculates the adaptability of the role nodes, evaluates and sorts the matching degree between the role and the task nodes, and generates a task node adaptability sorting; The scheduling path construction sub-module extracts the task node combinations with an overlap degree of available time periods between roles in the task node adaptability sorting that exceeds the overlap degree threshold, rearranges them according to the time sequence, and uses the formula: ； Performs calculations to obtain the rescheduling of the scheduling path, forms an industry-education integration scheduling path, and generates an industry-education integration task scheduling result; Among them, represents the total length of the overlapping period of the task node , represents the role in the available period length of the task node , represents the role in the task execution duration of the task node , is the number of roles participating in this task.

Images