An intelligent course arrangement method and system based on an improved five-dimensional digital twin model

Abstract

This invention discloses an intelligent scheduling method and system based on an improved five-dimensional digital twin model. The method includes: collecting campus physical entity data to construct a campus physical entity set; constructing a scheduling virtual entity set based on the campus physical entity set; solving the scheduling virtual entity set using a multi-lion swarm optimization algorithm to generate an optimal scheduling scheme; establishing a scheduling twin database based on the campus physical entity set, the scheduling virtual entity set, and the optimal scheduling scheme; and reading from the scheduling twin database to generate a three-dimensional scene interface of the scheduling scheme using a three-dimensional visualization construction method. The system includes a campus physical entity construction module, a scheduling virtual entity construction module, a scheduling twin database module, a multi-lion swarm algorithm scheduling module, a three-dimensional visualization module, and a data connection module. This invention combines digital twins with a multi-lion swarm optimization algorithm, effectively improving the rationality, visualization, and solution efficiency of the scheduling scheme.

Description

Technical Field

[0001] This invention relates to the field of smart education and educational informatization, and in particular to an intelligent scheduling method and system based on an improved five-dimensional digital twin model. Background Technology

[0002] Course scheduling has always been a top priority in academic affairs at universities, affecting the organization and scheduling of the entire university's teaching order. However, traditional manual scheduling is inefficient and difficult to optimize. Therefore, automatic scheduling has received widespread attention from the industry for many years.

[0003] Existing automatic time scheduling algorithms can be categorized into mathematical methods, heuristic methods, metaheuristic algorithms, and hybrid algorithms. Mathematical methods, centered on precise mathematical models, possess strong theoretical rigor and ensure optimal results. However, their computational complexity increases dramatically as the scale of time scheduling expands and constraints become more complex, often making it difficult to solve large-scale practical problems within a reasonable timeframe. Heuristic methods, relying on empirical strategies and specific search patterns, can quickly output relatively optimal solutions that meet basic requirements. However, they are highly dependent on empirical rules, struggle to break through local optima, and lack global exploration capabilities. Metaheuristic methods achieve global search by simulating natural evolution or intelligent behavior, exhibiting excellent versatility and flexibility, and can handle multi-constraint, large-scale time scheduling problems. However, their global search characteristics also lead to slower convergence speeds, potentially requiring even longer to approach the optimal solution when dealing with ultra-large-scale time scheduling tasks. Hybrid methods combine the above methods, leveraging their advantages at different stages to address complex time scheduling problems. However, their design and implementation are relatively complex, requiring a deep understanding of the characteristics of different algorithms. The challenge lies in designing reasonable collaboration methods between different approaches.

[0004] Existing course scheduling results are often based on information provided in multiple two-dimensional tables (teaching tasks - week, day of the week - period, teaching building - classroom), which makes it difficult to reflect information such as the location, distance, and available classrooms of teaching buildings from a three-dimensional perspective and macroscopically in a virtual space, thus lacking decision support for administrators. Summary of the Invention

[0005] This invention provides an intelligent scheduling method and system based on an improved five-dimensional digital twin model to overcome the problems of excessive time spent processing ultra-large-scale scheduling tasks, reasonable collaboration between different methods, and difficulty in providing decision support for managers from a three-dimensional perspective and macro level in existing technologies.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: A smart course scheduling method based on an improved five-dimensional digital twin model includes: S1. Collect campus physical entity data and construct a campus physical entity set; S2. Construct a set of virtual entities for scheduling based on the set of physical entities on campus; S3. Based on the set of virtual entities for scheduling, generate the optimal scheduling scheme through the multi-lion group optimization algorithm; S4. Establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme; S5. Read the data from the course scheduling twin database, and generate a 3D scene interface of the course scheduling plan through a 3D visualization construction method to realize the campus course scheduling visualization service.

[0007] Furthermore, the set of campus physical entities includes spatial entities and personnel entities; The spatial entities include a set of teaching buildings, a set of floors, and a set of classrooms; The set of teaching buildings can be represented as ,in, For the first One teaching building, For the teaching building index, The total number of teaching buildings; the set of floors can be represented as ,in, For the teaching building The One floor, For the teaching building Floor index, For the teaching building The total number of floors; the set of classrooms can be represented as ,in, For the first One classroom, For classroom indexing, Total number of classrooms; The personnel entities include a set of students and a set of teachers; The student set can be represented as ,in, For the first One student, Index for students, Let the total number of students be ; the set of teachers can be represented as ,in, For the first One teacher, For teachers' index, This represents the total number of teachers.

[0008] Furthermore, the set of virtual entities for scheduling includes a geometric model, business-related entities, a rule model, and a behavior model; The geometric model is used to construct a spatial layout set mapping covering the campus level, teaching building level and classroom level based on spatial entities through a virtual mapping method; The business-related entities include teaching tasks and scheduling results; the teaching task can be represented as... ,in, For the first One teaching task As an index for teaching tasks, The total number of teaching tasks; the scheduling result can be expressed as ,in, For the first The scheduling results for each teaching task; The rule model aims to minimize the fluctuation of teacher and student class time, minimize classroom vacancy rate, and minimize the average distance for teachers and students to change teaching buildings between consecutive time slots. It is constructed with constraints such as student time non-conflict constraint, teacher time non-conflict constraint, classroom time non-conflict constraint, classroom capacity constraint, weekly class hour constraint, total number of classrooms constraint, and class week constraint. The behavioral model includes classroom status update behavior and floor congestion warning behavior.

[0009] Furthermore, the expression for the student time non-conflict constraint is:

[0010] In the formula, A set of time periods for the teaching week; For the first The time period of a teaching week; For students The selected set of teaching tasks; This is the set of all scheduling events; The expression for the teacher time non-conflict constraint is:

[0011] In the formula, For teachers The selected set of teaching tasks; The expression for the classroom time non-conflict constraint is:

[0012] In the formula, For the course Are you in the classroom? and the time period of the teaching week Attend class; The expression for the classroom capacity constraint is:

[0013] In the formula, For classroom Number of seats; To select teaching tasks The number of students; For teaching tasks Are you in the classroom? Attend class; The expression for the weekly study hour constraint is:

[0014] In the formula, For the teaching week collection; For the first One teaching week; For teaching week Teaching tasks The collection of class time slots; Teaching tasks The target number of time periods in odd-numbered weeks; Teaching tasks The target number of time periods in even-numbered weeks; The expression for the total number of classrooms constraint is:

[0015] In the formula, A collection of data for a one-week period; The first of the week A time period; In the teaching week time period A collection of teaching tasks for the class; The expression for the class week constraint is:

[0016]

[0017] In the formula, For teaching tasks Class week; For teaching tasks The beginning of the week; For teaching tasks The end of the week; For teaching tasks Is it an odd or even week? and All of these are time periods during the week when classes are held.

[0018] Furthermore, the expression for the fluctuation of teacher-student class time is:

[0019] In the formula, To mitigate the fluctuations in class times for teachers and students; Weighting for students; Weighting for teachers; Gather for the day of class; For the first One class day; For students During the teaching week Secondary school days Number of classes; For teachers During the teaching week Secondary school days Number of classes; This represents the average number of times a student attends class per day. This represents the average number of classes taught by teachers per day. The expression for the classroom vacancy rate is:

[0020] In the formula, Classroom vacancy rate; For classroom In time period The number of empty seats inside; For classroom In time period The actual number of seats used inside; The expression for the average distance that teachers and students travel between teaching buildings continuously over a period of time is:

[0021] In the formula, The average distance between teaching buildings for teachers and students to change shifts over a continuous period of time; A collection of time periods within a single day; For the first time in a day A time period; For students During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period; For teachers During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period.

[0022] Furthermore, the floor congestion warning behavior is based on floor congestion and a set threshold; the expression for floor congestion is:

[0023] In the formula, For the time period middle school building floor The degree of congestion; For the time period middle school building floor Total number of students attending classes in all classrooms; For the time period middle school building floor The total number of seats in all classrooms.

[0024] Furthermore, the specific execution steps of the multi-lion swarm optimization algorithm include: S31. Based on the number of objective functions, set the number of optimized lion prides and set the parameters for each optimized lion pride; S32. Initialize all lions and set the lion with the lowest initial penalty value as the Lion King. S33. Set up a shared pride and copy the lion kings from each optimized pride to the shared pride to obtain the initial shared pride. S34. Optimize and update the positions of individual lions in each optimized pride to obtain updated individual lions; S35. Calculate the penalty value of each updated lion in each optimized pride; select the lion with the lowest penalty value as the new lion king, and add the new lion king of each optimized pride to the shared pride to obtain the updated shared pride. S36. Perform Pareto sorting on the alpha lions in the shared pride to obtain and retain the first-level alpha lions after sorting. S37. When the maximum number of iterations is reached, output the lion king with the highest crowding degree calculated by Pareto sort in the shared pride as the optimal scheduling scheme; otherwise, return to S34 to continue iterating.

[0025] Furthermore, the specific rules for optimizing and updating the positions of individual lions within each optimized pride are as follows: The expression for updating the Lion King's individual position is:

[0026] In the formula, The updated position of the Lion King individual; For random selection The perturbed position matrix of each teaching task; For the remainder The unperturbed position matrix of each teaching task; in, and The expression is:

[0027]

[0028] In the formula, For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; The expression for updating the lioness's position is:

[0029] In the formula, The updated position of the individual lionesses; For random selection The disturbed position matrix after collaboration between a teaching task and a shared pride; For random selection The perturbed position matrix after the teaching task and the cooperation of the lionesses in this optimized pride; For the remainder The unperturbed position matrix of each teaching task; The probability of the lioness referencing the shared pride for iteration; This is a probabilistic random number randomly generated for the current lioness individual in each iteration; in, , and The expression is:

[0030]

[0031]

[0032] In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task The globally optimal position; To optimize the first Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; The expression for updating the position of individual lion cubs is:

[0033] In the formula, The updated position of the individual lion cub; For random selection The perturbed position matrix after collaboration between a teaching task and lions in a shared pride; For random selection The perturbed position matrix after the teaching task and the cubs of this optimized pride collaborate; For the remainder An undisturbed position matrix consisting of teaching tasks; This is a probabilistic random number randomly generated for the current lion cub in each iteration; The probability of a lion cub iterating by referencing a shared pride; in, , and The expression is:

[0034]

[0035]

[0036] In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task A position far from the global optimum; To optimize the first Only Lion Second iteration teaching task The historical best position; For one in Random numbers within an interval; and These represent the number of updated and unupdated teaching tasks for lion cubs, respectively. , To dynamically adjust and update the number of teaching tasks.

[0037] An intelligent scheduling system based on an improved five-dimensional digital twin model is characterized by comprising a campus physical entity construction module, a scheduling virtual entity construction module, a scheduling twin database module, a multi-lion group algorithm scheduling module, a three-dimensional visualization module, and a data connection module. The campus physical entity construction module is used to collect campus physical entity data and construct a set of campus physical entities; The virtual entity construction module for scheduling is used to construct a set of virtual entities for scheduling based on the set of physical entities on campus. The multi-lion group algorithm scheduling module is used to generate the optimal scheduling scheme based on the set of virtual entities for scheduling using the multi-lion group optimization algorithm. The scheduling twin database module is used to establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme. The 3D visualization module is used to read data from the scheduling twin database and generate a 3D scene interface of the scheduling scheme through a 3D visualization construction method. The data connection module is used to connect the campus physical entity construction module, the course scheduling virtual entity construction module, the course scheduling twin database module, the multi-lion group algorithm course scheduling module, and the 3D visualization module through a full-dimensional connection network.

[0038] Beneficial Effects: This invention provides an intelligent scheduling method and system based on an improved five-dimensional digital twin model. By collecting physical entity data of the campus, a set of physical entities is constructed, and a set of virtual entities for scheduling is built accordingly. This achieves accurate mapping and real-time synchronization between the physical campus and the virtual campus. This mechanism can completely map physical constraints such as classroom seating capacity, multimedia equipment status, and floor distribution to the virtual space, ensuring that the scheduling plan fully matches the actual resource situation of the campus from the initial generation. This effectively avoids the problem of scheduling plans being infeasible due to factors such as mismatched classroom capacity and conflicting equipment resources, and significantly improves the actual executability of the scheduling plan. By employing a multi-lion swarm optimization algorithm to generate the optimal course scheduling scheme, the limitations of traditional course scheduling methods in handling large-scale, multi-constraint course scheduling problems are effectively overcome. The multi-lion swarm optimization algorithm introduces a collaborative search mechanism among multiple swarms, enabling it to quickly lock the global optimum in a vast solution space. At the same time, the parallel structure of independent evolution of each swarm significantly enhances the diversity of the population, effectively avoiding the technical defects of single-swarm algorithms that are prone to getting trapped in local optima. Compared with traditional optimization methods such as genetic algorithms and particle swarm optimization, it has significant advantages in terms of convergence speed and solution quality, and is particularly suitable for smart campus course scheduling scenarios with a large number of courses and complex constraints. By establishing a scheduling twin database, campus physical entity data, scheduling virtual entity data, and optimal scheduling scheme data are stored and managed in a unified manner. This database not only provides real-time and accurate data support for the 3D visualization module, but also supports historical retrospection, scheme comparison, and iterative optimization of scheduling schemes, enabling scheduling management to have continuous evolution capabilities and providing a solid data foundation for the dynamic allocation of teaching resources and optimization of scheduling strategies. By using a 3D visualization construction method, a 3D scene interface for the scheduling plan is generated, which innovates the traditional way of presenting scheduling results in a 2D table format to a 3D visualization presentation. Users can intuitively view the course arrangements, teacher allocation, and student distribution in each classroom at each time period from an immersive perspective. It supports multi-level penetrating viewing from the overall campus to specific classrooms, which greatly reduces the cognitive burden on scheduling managers in the process of understanding and reviewing the plan, and significantly improves the intuitiveness of the scheduling plan display and user experience. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating the intelligent course scheduling method of the present invention; Figure 2 This is a schematic diagram of the intelligent course scheduling system of the present invention; Figure 3 This is a flowchart of the multi-lion flock optimization algorithm in an embodiment of the present invention; Figure 4 This is a diagram showing the connection architecture of each module in an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] This embodiment provides an intelligent course scheduling method based on an improved five-dimensional digital twin model, such as... Figure 1 As shown, it includes: S1. Collect campus physical entity data and construct a campus physical entity set; Preferably, the set of campus physical entities includes spatial entities and personnel entities; The spatial entities include a set of teaching buildings, a set of floors, and a set of classrooms; The set of teaching buildings can be represented as ,in, For the first One teaching building, For the teaching building index, The total number of teaching buildings; the set of floors can be represented as ,in, For the teaching building The One floor, For the teaching building Floor index, For the teaching building The total number of floors; the set of classrooms can be represented as ,in, For the first One classroom, For classroom indexing, Total number of classrooms; The personnel entities include a set of students and a set of teachers; The student set can be represented as ,in, For the first One student, Index for students, Let the total number of students be ; the set of teachers can be represented as ,in, For the first One teacher, For teachers' index, This represents the total number of teachers.

[0043] S2. Construct a set of virtual entities for scheduling based on the set of physical entities on campus; Preferably, the set of virtual entities for scheduling includes a geometric model, business-related entities, a rule model, and a behavior model; The geometric model is used to construct a spatial layout set mapping covering the campus level, teaching building level and classroom level based on spatial entities through a virtual mapping method; In this embodiment, the geometric model is based on map data to collect the actual location, geometric dimensions, and shape information of each teaching building to complete the virtual mapping of the teaching buildings; a lightweight color block presentation method is used to associate the basic attributes such as the teaching building and floor to which the classroom belongs to display the usage status of the classroom; and a spatial layout set mapping covering the campus level, teaching building level, and classroom level is constructed. The virtual mapping method includes: collecting the three-dimensional coordinates, geometric dimensions, and shape data of the teaching building; establishing a floor space model based on the teaching building's floor plan and floor information; extracting attributes such as the location, area, and number of seats for each classroom; constructing a three-dimensional space model in the Unity engine according to the actual scale; and mapping the physical entity attributes to the corresponding nodes of the virtual entities. The business-related entities include teaching tasks and scheduling results; the teaching task can be represented as... ,in, For the first One teaching task As an index for teaching tasks, The total number of teaching tasks; the scheduling result can be expressed as ,in, For the first The scheduling results for each teaching task; The rule model aims to minimize the fluctuation of teacher and student class time, minimize classroom vacancy rate, and minimize the average distance for teachers and students to change teaching buildings between consecutive time slots. It is constructed with constraints such as student time non-conflict constraint, teacher time non-conflict constraint, classroom time non-conflict constraint, classroom capacity constraint, weekly class hour constraint, total number of classrooms constraint, and class week constraint. Preferably, the student time non-conflict constraint means that each student can only take one course at a time. The expression for the student time non-conflict constraint is:

[0044] In the formula, It is a set of time periods within a teaching week, and ; For the first The time period of a teaching week; For students The selected set of teaching tasks; Let be the set of all scheduling events, and ,in, The total number of scheduling events. For scheduling events, each scheduling event is a triple (role, course, time period), indicating that a certain role participates in a certain course during a certain time period; The teacher time non-conflict constraint means that each teacher can only teach one course at a time. The expression for the teacher time non-conflict constraint is:

[0045] In the formula, For teachers The selected set of teaching tasks; The classroom time non-conflict constraint means that only one course can be scheduled in each classroom at any given time. The expression for the classroom time non-conflict constraint is:

[0046] In the formula, For the course Are you in the classroom? and the time period of the teaching week Attend class; The classroom capacity constraint means that the number of seats in a classroom must be greater than or equal to the number of students attending the class. The expression for the classroom capacity constraint is:

[0047] In the formula, For classroom Number of seats; To select teaching tasks The number of students; For teaching tasks Are you in the classroom? Attend class; The expression for the weekly study hour constraint is:

[0048] In the formula, For the teaching week collection; For the first One teaching week; For teaching week Teaching tasks The collection of class time slots; Teaching tasks The target number of time periods in odd-numbered weeks; Teaching tasks The target number of time periods in even-numbered weeks; The total number of classrooms constraint means that the total number of classrooms is greater than the total number of courses scheduled at the same time. The expression for the total number of classrooms constraint is:

[0049] In the formula, A collection of data for a one-week period; The first of the week A time period; In the teaching week time period A collection of teaching tasks for the class; The aforementioned class week constraint means that the class week must be within the start and end weeks specified in the teaching task, and the class week should conform to the odd-even week arrangement requirement. The expression for the class week constraint is:

[0050]

[0051] In the formula, For teaching tasks Class week; For teaching tasks The beginning of the week; For teaching tasks The end of the week; For teaching tasks Is it an odd or even week? and All of these are time periods during the school week; The teaching tasks Whether it is an odd or even week can be expressed as: .

[0052] Preferably, the fluctuation of class time for teachers and students represents the variance of the number of classes per day for all students (or teachers) related to the teaching tasks of this semester within the teaching start and end week interval; the smaller the variance of the scheduling scheme, the smaller the time fluctuation of the students (or teachers), the more uniform the distribution, and the higher the satisfaction; the overall fluctuation of the two is weighted. The expression for the fluctuation of teacher and student class time is:

[0053] In the formula, To mitigate the fluctuations in class times for teachers and students; Weighting for students; Weighting for teachers; Gather for the day of class; For the first One class day; For students During the teaching week Secondary school days Number of classes; For teachers During the teaching week Secondary school days Number of classes; This represents the average number of times a student attends class per day. This represents the average number of classes taught by teachers per day. The expressions for the average number of classes per day for students and the average number of classes per day for teachers are as follows:

[0054] ; The classroom vacancy rate refers to the ratio of the number of empty seats in a classroom to the number of seats in the classroom in use. The expression for the classroom vacancy rate is:

[0055] In the formula, Classroom vacancy rate; For classroom In time period The number of empty seats inside; For classroom In time period The actual number of seats used inside; The classroom vacancy rate reflects the utilization of classrooms over a period of time. If the scheduling is not reasonable, some classrooms may have a high vacancy rate during a specific period of time. A good scheduling plan should minimize the classroom vacancy rate as much as possible, thereby improving the full utilization of teaching resources. The average distance that teachers and students travel between teaching buildings in consecutive time slots represents the average distance they walk when changing teaching buildings due to course schedules between two consecutive time slots of a day. The expression for the average distance that teachers and students travel between teaching buildings continuously over a period of time is:

[0056] In the formula, The average distance between teaching buildings for teachers and students to change shifts over a continuous period of time; A collection of time periods within a single day; For the first time in a day A time period; For students During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period; For teachers During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period.

[0057] In this embodiment, the volatility of teacher and student class times is used to assess the rationality of the course scheduling for teachers or students, avoiding situations where teachers or students have overly concentrated class schedules while having relatively free time at other times. Therefore, in this embodiment, the lower the volatility of teacher and student class times, the better. A classroom vacancy rate model is used to assess classroom vacancy. The lower the vacancy rate of a classroom, the higher its utilization efficiency. Therefore, the lower the average vacancy rate of the scheduled classrooms, the better. The average distance for teachers and students to change teaching buildings between consecutive time periods is used to assess the rationality of the classroom scheduling for courses, avoiding situations where teachers or students have to travel too far between two classes. Therefore, in this embodiment, the lower the average distance for teachers and students to change teaching buildings between consecutive time periods, the better.

[0058] The behavioral model includes classroom status update behavior and floor congestion warning behavior.

[0059] Specifically, the classroom status update behavior enables dynamic updates of classroom and floor status in a scheduling scenario; Classrooms are defined as follows In time period Status:

[0060] As time slots change, classrooms will automatically change their status, and the status of floors will change according to their crowding level. The crowding level of a floor is determined by the total number of students attending class on that floor in the current time slot and the total capacity of the floor.

[0061] Preferably, the floor congestion warning behavior is based on the floor congestion level and a set threshold. When the congestion level exceeds the set threshold (0.8), the floor will enter a congestion warning state, and a corresponding warning behavior will be issued in the visualization service. The expression for the floor congestion level is:

[0062] In the formula, For the time period middle school building floor The degree of congestion; For the time period middle school building floor Total number of students attending classes in all classrooms; For the time period middle school building floor Total number of seats in all classrooms; Among them, in the time period middle school building floor The total number of students attending classes in all classrooms can be expressed as:

[0063] In the formula, For the teaching building All classrooms gathered together; This is a function indicating classroom occupancy status, and .

[0064] S3. Based on the set of virtual entities for scheduling, generate the optimal scheduling scheme through the multi-lion group optimization algorithm; Preferred, such as Figure 3 As shown, the specific execution steps of the multi-lion swarm optimization algorithm include: S31. Based on the number of objective functions, set the number of optimized lion prides and set the parameters for each optimized lion pride; Specifically, the optimal lion pride size is set to 3 to correspond to the three objective functions in the scheduling model; the total number of lions in each optimal pride is set to... The proportion of adult lions in the total number of lions is The maximum number of iterations is Each lion represents a scheduling method; the maximum capacity of the shared lion pride is set to [value missing]. The probability of a female lion in a pride referencing and sharing the pride iteration is optimized as follows: The probability that a lion cub will refer to the shared pride during iteration is: ; S32. Initialize all lions and set the lion with the lowest initial penalty value as the Lion King. Specifically, for each optimized pride, the number of mother lions and cubs is determined based on the proportion of adult lions to the total number of lions, and all lions are initialized, with the lion with the lowest initial penalty value being set as the lion king. The penalty value is calculated based on the fluctuation of class time for teachers and students, classroom vacancy rate, and the average distance for teachers and students to change teaching buildings during continuous time. The expression is:

[0065] In the formula, This is the penalty value; , and This is the penalty weighting coefficient; S33. Set up a shared pride and copy the lion kings from each optimized pride to the shared pride to obtain the initial shared pride. Specifically, the expression for defining and initializing the shared pride is as follows:

[0066] In the formula, For the shared pride of lions Only Lion Second iteration teaching task Location; The shared pride of lions is a group that includes all lions. A complete plan for scheduling results of each teaching task; S34. Optimize and update the positions of individual lions in each optimized pride to obtain updated individual lions; S35. Calculate the penalty value of each updated lion in each optimized pride; select the lion with the lowest penalty value as the new lion king, and add the new lion king of each optimized pride to the shared pride to obtain the updated shared pride. S36. Perform Pareto sorting on the alpha lions in the shared pride to obtain and retain the first-level alpha lions after sorting. Specifically, the update expression for the shared pride is:

[0067] In the formula, For the first The shared pride in the next iteration; For the first The set of lion kings for each optimized pride at the next iteration; This is the maximum number of lions that can be shared in a pride. Use the target filtering function to filter out the set. Center front The detailed steps for identifying individuals with better objective functions are as follows: First, based on the rule model in the set of virtual entities for scheduling, calculate the penalty values ​​of all lions on all three objective functions; Secondly, all lions are sorted using a non-dominated sorting method to obtain the Pareto solution set. Only the Pareto front is retained, and the solution sets of other layers are deleted. Then, determine the number of lions in the Pareto front; if it exceeds the shared pride capacity... Then, the lions in the front are sorted by crowding degree, and the lions with higher crowding degree are retained; The crowding degree is calculated by defining the boundary points in the same front as infinity. When applied to the lion pack scheduling algorithm, in order to avoid the optimal solution always being at the boundary point, the boundary crowding degree is defined as the ratio of twice the distance to its adjacent point to the difference between the maximum and minimum values ​​in the front. The crowding degree of the other solutions is defined as the ratio of the distance between the two adjacent nodes before and after the node to the difference between the maximum and minimum values ​​in the front. The expression for calculating the congestion level is:

[0068] In the formula, For the same frontier, the first The crowding degree of each solution; The number of objective functions; The number of solutions in the same frontier; For the Pareto frontier, the first The solution is at the th solution. The values ​​of each objective function; and To solve the problem in the first place The maximum and minimum values ​​of each objective function; Index for the objective function; Index of the solution; S37. When the maximum number of iterations is reached, output the lion king with the highest crowding degree calculated by Pareto sort in the shared pride as the optimal scheduling scheme; otherwise, return to S34 to continue iterating.

[0069] The preferred specific rules for optimizing and updating the positions of individual lions in each optimized pride are as follows: The update of the lion king's individual position involves a small-range movement from the globally optimal position within the pride. The expression for updating the Lion King's individual position is:

[0070] In the formula, The updated position of the Lion King individual; For random selection The perturbed position matrix of each teaching task; For the remainder The unperturbed position matrix of each teaching task; in, and The expression is:

[0071]

[0072] In the formula, For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; in, , , To follow the normal distribution The generated random numbers are truncated and restricted to a range. Inside; The update of the individual lion's position refers to the process by which a lion needs to cooperate with another lion or a lion in a shared pride to adjust its position and make a positive update during the optimization of the schedule. First, a wide range of teaching tasks are selected for position updates to increase the diversity of solutions; as the number of iterations increases, the lioness will gradually select a smaller range of teaching tasks for position updates, and gradually find a relatively better scheduling scheme. The expression for updating the lioness's position is:

[0073] In the formula, The updated position of the individual lionesses; For random selection The disturbed position matrix after collaboration between a teaching task and a shared pride; For random selection The perturbed position matrix after the teaching task and the cooperation of the lionesses in this optimized pride; For the remainder The unperturbed position matrix of each teaching task; The probability of the lioness referencing the shared pride for iteration; This is a probabilistic random number randomly generated for the current lioness individual in each iteration; in, , and The expression is:

[0074]

[0075]

[0076] In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task The globally optimal position; To optimize the first Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; in, , The definition is as follows:

[0077] In the formula, for One of the numbers represents the step size factor, used to accelerate convergence. The value is usually greater than 0.5; For iterative indexing; This represents the maximum number of iterations to be performed throughout the optimization process; the number of updated teaching tasks will be dynamically adjusted during the optimization process, initially (when...). When the process is relatively short, allow the lioness to exchange teaching task positions across a wide range within the optimization space. As the iteration progresses (i.e., ... Gradually approaching ), By gradually reducing the number of teaching tasks and moving them to less important positions, a relatively optimal scheduling scheme can be found. The updating of the individual lion cub's location reflects three different stages in its growth process, and the mechanism by which it dynamically adjusts its survival location based on its own experience and environmental information. The expression for updating the position of individual lion cubs is:

[0078] In the formula, The updated position of the individual lion cub; For random selection The perturbed position matrix after collaboration between a teaching task and lions in a shared pride; For random selection The perturbed position matrix after the teaching task and the cubs of this optimized pride collaborate; For the remainder An undisturbed position matrix consisting of teaching tasks; This is a probabilistic random number randomly generated for the current lion cub in each iteration; The probability of a lion cub iterating by referencing a shared pride; in, , and The expression is:

[0079]

[0080]

[0081] In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task A position far from the global optimum; To optimize the first Only Lion Second iteration teaching task The historical best position; For one in Random numbers within an interval; and These represent the number of updated and unupdated teaching tasks for lion cubs, respectively. , To dynamically adjust and update the number of teaching tasks; Initially, the perturbation factor is relatively large, which can select a better position for a large number of teaching tasks. As the number of iterations increases, the perturbation factor gradually decreases and gradually converges to find a better scheduling scheme. This search strategy is a linear transition from extensive exploration to precise positioning. and The expression is: , .

[0082] S4. Establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme; The scheduling twin database includes geometric foundation data, scheduling business data, and scheduling result data: The aforementioned geometric data includes the location coordinates, geometric dimensions, and shape information of the teaching building, as well as the spatial layout of the floors and the specific locations of the classrooms. The course scheduling data includes basic student information, basic teacher information, basic classroom information, basic building information, distance information between building and teaching building, teaching task information, and course selection information. The basic student information includes student ID, student name, major, and class. The basic information of the teachers includes teacher ID, teacher name, and major. The basic information of the classroom includes classroom number, classroom name, classroom type, number of students, and the teaching building number to which it belongs; The basic information of the teaching building includes the teaching building number and the teaching building name; The distance information between the teaching buildings includes the starting teaching building number, the ending teaching building number, and the distance between the two teaching buildings; The teaching task information includes teaching task number, course number, course name, total class hours, assessment type, academic year, semester, instructor number, teaching class number, academic year / semester, start week, end week, whether it is an odd or even week, and weekly class hours; The course selection information includes the teaching task number and student ID.

[0083] S5. Read the data from the course scheduling twin database, and generate a 3D scene interface of the course scheduling plan through a 3D visualization construction method to realize the campus course scheduling visualization service.

[0084] In a specific embodiment, the 3D visualization construction method includes a 3D scene construction module, a classroom status visualization module, a floor congestion warning module, a data integration module, and an interactive control module: The aforementioned 3D scene construction module builds a campus model in the Unity engine. The teaching building is rendered with a semi-transparent material. By adjusting the Alpha channel parameters, a balance is achieved between the visibility of the building's exterior and the perspective effect of the internal classrooms, ensuring that users can observe the usage of the internal classrooms through the building. The classroom status visualization module abstracts the classroom into color block elements and establishes a three-color coding rule: red color blocks correspond to the status of having classes, green color blocks correspond to the status of being available, and gray color blocks correspond to the status of being unavailable, such as equipment maintenance. The positions of the color blocks are mapped one-to-one with the actual physical positions of the classrooms. The aforementioned floor congestion warning module calculates the congestion level based on the total number of students and total capacity on each floor using a formula. When the congestion level exceeds a preset threshold, an early warning is triggered. The module provides visual prompts and differentiated visual indicators to help managers identify high-load floors, providing a basis for resource allocation and class scheduling optimization. The data integration module dynamically updates the color block display status of all classrooms at the corresponding time node through time slice parameter settings, realizing a global presentation of the campus classroom utilization at a specific time point; it also supports the import of scheduling result data. The interactive control module allows for free adjustment of the viewing angle through mouse dragging and scroll wheel zooming; by clicking on the name of the target teaching building on the UI interface, a close-up view of the teaching building can be quickly switched, and the camera angle and zoom ratio are automatically optimized in close-up mode to highlight the details of the classrooms inside the target building; The specific steps of the 3D visualization construction method include: a 3D scene construction module reads the geometric basic data from the scheduling twin database and builds a 3D campus model in the Unity engine; a data integration module extracts classroom status data for the specified time period from the scheduling twin database based on the user-selected time slice parameters; a classroom status visualization module renders classroom color blocks to the corresponding colors (red / green / gray) based on the classroom status data; a floor congestion warning module calculates the congestion level of each floor and triggers a warning sign for floors exceeding the threshold; an interactive control module responds to user operations such as perspective adjustment, zooming, and clicking, updating the scene display; and the 3D scene interface is updated in real time to complete the visualization service.

[0085] The second embodiment provides an intelligent course scheduling system based on an improved five-dimensional digital twin model, such as... Figure 2 As shown, it includes a campus physical entity construction module, a course scheduling virtual entity construction module, a course scheduling twin database module, a multi-lion group algorithm course scheduling module, a 3D visualization module, and a data connection module; The campus physical entity construction module is used to collect campus physical entity data and construct a set of campus physical entities; The virtual entity construction module for scheduling is used to construct a set of virtual entities for scheduling based on the set of physical entities on campus. The multi-lion group algorithm scheduling module is used to generate the optimal scheduling scheme based on the set of virtual entities for scheduling using the multi-lion group optimization algorithm. The scheduling twin database module is used to establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme. The 3D visualization module is used to read data from the scheduling twin database and generate a 3D scene interface of the scheduling scheme through a 3D visualization construction method. The data connection module is used to connect the campus physical entity construction module, the course scheduling virtual entity construction module, the course scheduling twin database module, the multi-lion group algorithm course scheduling module, and the 3D visualization module through a full-dimensional connection network.

[0086] Specifically, such as Figure 4As shown, the data connection module constructs a full-dimensional connection network covering all data flow links in the entire scheduling process based on standardized interfaces and a data bus. This full-dimensional connection network at least implements the following four types of specific data connections: Physical Entity-Data Storage Connection (CN_PD): The CN_PD is a dedicated data connection between a physical entity (PE) and a data storage (DD). By calling the API interface of map data and campus information system, a two-way data interaction link is established between the physical entity and the data storage, which is used to complete the real-time update of the basic information of the physical entity in the data storage (DD), the supplementation of missing information, and the verification of data consistency. Virtual Entity-Data Storage Connection (CN_VD): The CN_VD is a dedicated data connection between the virtual entity (VE) and the data storage (DD). It uses a standardized database interface to build a stable data transmission channel between the virtual entity and the data storage, realizes bidirectional data synchronization between the virtual entity (VE) and the data storage (DD), and ensures the accuracy of the parameters, the timeliness of the state and the integrity of the data in the virtual entity (VE) model. Virtual Entity-Service Module Connection (CN_VS): CN_VS is a dedicated data connection between the virtual entity (VE) and the service module (Ss). The virtual entity (VE) provides full-process data and model support for two core services of the service module (Ss): First, the scheduling rule model built into the virtual entity (VE) generates service output scheduling constraints (including but not limited to teacher configuration constraints, classroom resource constraints, course time constraints, and student course selection constraints) and objective functions (including but not limited to minimizing scheduling conflicts, maximizing resource utilization, and optimizing teaching effectiveness) for the algorithm scheduling scheme of the service module (Ss), ensuring that the algorithm optimization direction is accurately matched with the actual needs of the scheduling business; Second, the virtual entity (VE) provides geometric model data, real-time status data, and behavioral pattern data to the visualization service of the service module (Ss), providing a model foundation and data support for visualization display; Service Module - Data Storage Connection (CN_SD): CN_SD is a dedicated data connection between the service module (Ss) and the data storage (DD). Through this connection, the service module (Ss) reads basic scheduling data such as student information, teacher information, classroom information, and teaching task information from the data storage (DD), calls the multi-lion swarm algorithm to process the basic data, generates an optimized scheduling plan, and encrypts and stores the scheduling plan in the data storage (DD). At the same time, the service module (Ss) can also read historical scheduling results data, real-time scheduling status data, and related statistical data from the data storage (DD) through this connection, providing data support for visualization, teaching management decisions, and scheduling plan adjustments.

[0087] In the third embodiment, to verify the effectiveness of the intelligent course scheduling method of this embodiment, based on real data from the second semester of the 2023-2024 academic year of a certain university, totaling 296 teaching tasks, a comparative experiment was conducted using four algorithms—MOPSO, NSGA-II, MOLSO, and MOGRO—and the Multi-Lion Group Algorithm (CMLSO). The evaluation index was the weighted total penalty value; the smaller the total penalty value, the better the course scheduling scheme obtained by the algorithm. This experiment increased the scale of the scheduling problem by continuously increasing the number of teaching tasks, in order to evaluate the performance of the multi-lion swarm algorithm under different problem scales. Each experiment was repeated 10 times and the average penalty value was taken to reduce the error caused by random factors and ensure the reliability of the experimental results. The experimental results are shown in Table 1. Table 1 Experimental Results

[0088] This embodiment also uses the ITC-2019 scheduling information dataset, employing the multi-lion herd algorithm and classic multi-objective optimization methods (MOPSO, NSGA-II) to solve the dataset, and comparing it with three recent representative solutions: the simulated annealing algorithm (SA) proposed by Sylejmani et al., the MIP modeling method by Rappos et al., and the graph-based reformulation method (PGB-MIP) by Holm et al. The algorithm performance is evaluated using a penalty value, the formula for which is:

[0089] In the formula, The total penalty for the course scheduling plan; The time penalty value for the course scheduling plan; Classroom penalty value for the class scheduling plan; The penalty value for violating the regulations regarding the course scheduling plan; The number of student conflicts in the course scheduling plan; the penalty values ​​are shown in Table 2. Table 2 Penalty Value Results

[0090] The method proposed in this embodiment can flexibly cope with various constraints and effectively manage resource allocation in the scheduling problem, reduce the total penalty value of the scheduling scheme, and achieve efficient and stable multi-objective automatic scheduling.

[0091] The present invention has the following beneficial effects: This invention discloses an intelligent scheduling method and system based on an improved five-dimensional digital twin model. By collecting physical entity data of the campus, a set of physical entities of the campus is constructed, and a set of virtual entities for scheduling is constructed accordingly. This achieves accurate mapping and real-time synchronization between the physical campus and the virtual campus. This mechanism can completely map physical constraints such as classroom seating capacity, multimedia equipment status, and floor distribution to the virtual space, so that the scheduling plan fully matches the actual resource situation of the campus from the beginning. It effectively avoids the problem of scheduling plan infeasibility caused by factors such as classroom capacity mismatch and equipment resource conflict, and significantly improves the actual executability of the scheduling plan. By employing a multi-lion swarm optimization algorithm to generate the optimal course scheduling scheme, the limitations of traditional course scheduling methods in handling large-scale, multi-constraint course scheduling problems are effectively overcome. The multi-lion swarm optimization algorithm introduces a collaborative search mechanism among multiple swarms, enabling it to quickly lock the global optimum in a vast solution space. At the same time, the parallel structure of independent evolution of each swarm significantly enhances the diversity of the population, effectively avoiding the technical defects of single-swarm algorithms that are prone to getting trapped in local optima. Compared with traditional optimization methods such as genetic algorithms and particle swarm optimization, it has significant advantages in terms of convergence speed and solution quality, and is particularly suitable for smart campus course scheduling scenarios with a large number of courses and complex constraints. By establishing a scheduling twin database, campus physical entity data, scheduling virtual entity data, and optimal scheduling scheme data are stored and managed in a unified manner. This database not only provides real-time and accurate data support for the 3D visualization module, but also supports historical retrospection, scheme comparison, and iterative optimization of scheduling schemes, enabling scheduling management to have continuous evolution capabilities and providing a solid data foundation for the dynamic allocation of teaching resources and optimization of scheduling strategies. By using a 3D visualization construction method, a 3D scene interface for the scheduling plan is generated, which innovates the traditional way of presenting scheduling results in a 2D table format to a 3D visualization presentation. Users can intuitively view the course arrangements, teacher allocation, and student distribution in each classroom at each time period from an immersive perspective. It supports multi-level penetrating viewing from the overall campus to specific classrooms, which greatly reduces the cognitive burden on scheduling managers in the process of understanding and reviewing the plan, and significantly improves the intuitiveness of the scheduling plan display and user experience.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent course scheduling method based on an improved five-dimensional digital twin model, characterized in that, include: S1. Collect campus physical entity data and construct a campus physical entity set; S2. Construct a set of virtual entities for scheduling based on the set of physical entities on campus; S3. Based on the set of virtual entities for scheduling, generate the optimal scheduling scheme through the multi-lion group optimization algorithm; S4. Establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme; S5. Read the data from the course scheduling twin database, and generate a 3D scene interface of the course scheduling plan through a 3D visualization construction method to realize the campus course scheduling visualization service.

2. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 1, characterized in that, The set of physical entities on campus includes spatial entities and personnel entities; The spatial entities include a set of teaching buildings, a set of floors, and a set of classrooms; The set of teaching buildings can be represented as ,in, For the first One teaching building, For the teaching building index, The total number of teaching buildings; the set of floors can be represented as ,in, For the teaching building The One floor, For the teaching building Floor index, For the teaching building The total number of floors; the set of classrooms can be represented as ,in, For the first One classroom, For classroom indexing, Total number of classrooms; The personnel entities include a set of students and a set of teachers; The student set can be represented as ,in, For the first One student, Index for students, The total number of students; the set of teachers can be represented as ,in, For the first One teacher, For teachers' index, This represents the total number of teachers.

3. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 2, characterized in that, The set of virtual entities for scheduling includes a geometric model, business-related entities, a rule model, and a behavior model; The geometric model is used to construct a set mapping of spatial layouts covering the campus, teaching building, and classroom levels based on spatial entities using a virtual mapping method. The business-related entities include teaching tasks and scheduling results; the teaching task can be represented as... ,in, For the first One teaching task As an index for teaching tasks, The total number of teaching tasks; the scheduling result can be expressed as ,in, For the first The scheduling results for each teaching task; The rule model aims to minimize the fluctuation of teacher and student class time, minimize classroom vacancy rate, and minimize the average distance for teachers and students to change teaching buildings between consecutive time slots. It is constructed with constraints such as student time non-conflict constraint, teacher time non-conflict constraint, classroom time non-conflict constraint, classroom capacity constraint, weekly class hour constraint, total number of classrooms constraint, and class week constraint. The behavioral model includes classroom status update behavior and floor congestion warning behavior.

4. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 3, characterized in that, The expression for the student time non-conflict constraint is: In the formula, A set of time periods for the teaching week; For the first The time period of a teaching week; For students The selected set of teaching tasks; This is the set of all scheduling events; The expression for the teacher time non-conflict constraint is: In the formula, For teachers The selected set of teaching tasks; The expression for the classroom time non-conflict constraint is: In the formula, For the course Are you in the classroom? and the time period of the teaching week Attend class; The expression for the classroom capacity constraint is: In the formula, For classroom Number of seats; To select teaching tasks The number of students; For teaching tasks Are you in the classroom? Attend class; The expression for the weekly study hour constraint is: In the formula, For the teaching week collection; For the first One teaching week; For teaching week Teaching tasks The collection of class time slots; Teaching tasks The target number of time periods in odd-numbered weeks; Teaching tasks The target number of time periods in even-numbered weeks; The expression for the total number of classrooms constraint is: In the formula, A collection of data for a one-week period; The first of the week A time period; In the teaching week time period A collection of teaching tasks for the class; The expression for the class week constraint is: In the formula, For teaching tasks Class week; For teaching tasks The beginning of the week; For teaching tasks The end of the week; For teaching tasks Is it an odd or even week? and All of these are time periods during the week when classes are held.

5. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 4, characterized in that, The expression for the fluctuation of teacher and student class time is: In the formula, To mitigate the fluctuations in class times for teachers and students; Weighting for students; Weighting for teachers; Gather for the day of class; For the first One class day; For students During the teaching week Secondary school days Number of classes; For teachers During the teaching week Secondary school days Number of classes; This represents the average number of times a student attends class per day. This represents the average number of classes taught by teachers per day. The expression for the classroom vacancy rate is: In the formula, Classroom vacancy rate; For classroom In time period The number of empty seats inside; For classroom In time period The actual number of seats used inside; The expression for the average distance that teachers and students travel between teaching buildings continuously over a period of time is: In the formula, The average distance between teaching buildings for teachers and students to change shifts over a continuous period of time; A collection of time periods within a single day; For the first time in a day A time period; For students During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period; For teachers During the teaching week Secondary school days The The time period and the first The distance between different teaching buildings within a certain time period.

6. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 5, characterized in that, The floor congestion warning behavior is based on the floor congestion level and a set threshold; the expression for the floor congestion level is: In the formula, For the time period middle school building floor The degree of congestion; For the time period middle school building floor Total number of students attending classes in all classrooms; For the time period middle school building floor The total number of seats in all classrooms.

7. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 6, characterized in that, The specific execution steps of the multi-lion group optimization algorithm include: S31. Based on the number of objective functions, set the number of optimized lion prides and set the parameters for each optimized lion pride; S32. Initialize all lions and set the lion with the lowest initial penalty value as the Lion King. S33. Set up a shared pride and copy the lion kings from each optimized pride to the shared pride to obtain the initial shared pride. S34. Optimize and update the positions of individual lions in each optimized pride to obtain updated individual lions; S35. Calculate the penalty value of each updated lion in each optimized pride; select the lion with the lowest penalty value as the new lion king, and add the new lion king of each optimized pride to the shared pride to obtain the updated shared pride. S36. Perform Pareto sorting on the alpha lions in the shared pride to obtain and retain the first-level alpha lions after sorting. S37. When the maximum number of iterations is reached, output the lion king with the highest crowding degree calculated by Pareto sort in the shared pride as the optimal scheduling scheme; otherwise, return to S34 to continue iterating.

8. The intelligent scheduling method based on an improved five-dimensional digital twin model according to claim 7, characterized in that, The specific rules for optimizing and updating the positions of individual lions within each optimized pride are as follows: The expression for updating the Lion King's individual position is: In the formula, The updated position of the Lion King individual; For random selection The perturbation position matrix of each teaching task; For the remainder The unperturbed position matrix of each teaching task; in, and The expression is: In the formula, For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; For the first Second iteration teaching task The globally optimal position; For the first Only Lion Second iteration teaching task The historical best position; The expression for updating the lioness's position is: In the formula, The updated position of the individual lionesses; For random selection The disturbed position matrix after collaboration between a teaching task and a shared pride; For random selection The perturbed position matrix after the teaching task and the cooperation of the lionesses in this optimized pride; For the remainder The unperturbed position matrix of each teaching task; The probability of the lioness referencing the shared pride for iteration; This is a probabilistic random number randomly generated for the current lioness individual in each iteration; in, , and The expression is: In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task The globally optimal position; To optimize the first in the lion pride Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; For the first Only Lion Second iteration teaching task The historical best position; The expression for updating the position of individual lion cubs is: In the formula, The updated position of the individual lion cub; For random selection The perturbed position matrix after collaboration between a teaching task and lions in a shared pride; For random selection The perturbed position matrix after the teaching task and the cubs of this optimized pride collaborate; For the remainder An undisturbed position matrix consisting of teaching tasks; This is a probabilistic random number randomly generated for the current lion cub in each iteration; The probability of a lion cub iterating by referencing a shared pride; in, , and The expression is: In the formula, Randomly select the first from the shared pride Only Lion Second iteration teaching task A position far from the global optimum; To optimize the first in the lion pride Only Lion Second iteration teaching task The historical best position; For one in Random numbers within an interval; and These represent the number of updated and unupdated teaching tasks for lion cubs, respectively. , To dynamically adjust and update the number of teaching tasks.

9. An intelligent scheduling system based on an improved five-dimensional digital twin model, according to the method described in claims 1 to 8, characterized in that, It includes a campus physical entity construction module, a course scheduling virtual entity construction module, a course scheduling twin database module, a multi-lion group algorithm course scheduling module, a 3D visualization module, and a data connection module; The campus physical entity construction module is used to collect campus physical entity data and construct a set of campus physical entities; The virtual entity construction module for scheduling is used to construct a set of virtual entities for scheduling based on the set of physical entities on campus. The multi-lion group algorithm scheduling module is used to generate the optimal scheduling scheme based on the set of virtual entities for scheduling using the multi-lion group optimization algorithm. The scheduling twin database module is used to establish a scheduling twin database based on the set of physical entities on campus, the set of virtual entities for scheduling, and the optimal scheduling scheme. The 3D visualization module is used to read data from the scheduling twin database and generate a 3D scene interface of the scheduling scheme through a 3D visualization construction method. The data connection module is used to connect the campus physical entity construction module, the course scheduling virtual entity construction module, the course scheduling twin database module, the multi-lion group algorithm course scheduling module, and the 3D visualization module through a full-dimensional connection network.

Images