A smart campus teaching space overall management system

Abstract

The present invention belongs to the field of intelligent management technology. The present invention discloses a smart campus teaching space overall management system, including a data acquisition module for acquiring teaching space data, teaching plan data, resource usage data and reservation request data; a data processing module for preprocessing the acquired data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data; weighted fusion of the teaching space parameter data and the resource usage parameter data to obtain space feature data; a resource management module for constructing a teaching space resource library based on the space feature data; an intelligent scheduling module for performing initial allocation of teaching space based on the teaching plan parameter data and obtaining free space data; a reservation management module for performing secondary allocation of teaching space; and a communication module for sending reservation messages to users. The present invention provides an efficient and flexible solution for the management of smart campuses.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management technology, and more specifically, to a smart campus teaching space overall management system. Background Art

[0002] Amidst the rapid development of information technology, the education industry is constantly seeking change and innovation. The construction of smart campuses has become a crucial component of modern education management. By integrating and optimizing various campus information and resources, it provides convenient and efficient educational services, creating a brand-new learning and living environment for teachers and students. As a crucial component of smart campus construction, the smart campus teaching space management system has gradually developed and improved, playing a significant role in improving teaching efficiency and quality, and promoting educational equity and sustainable development.

[0003] Patent application publication number CN117216928A discloses a full-process intelligent teaching space construction management system, which relates to the technical field of intelligent education. It includes a first target subsystem, a second target subsystem, and a third target subsystem, and the first target subsystem, the second target subsystem, and the third target subsystem are interconnected through interfaces; the first target subsystem is used to store, search, consult, and download standard documents related to teaching space construction models; the second target subsystem is used to present the teaching space construction model and provide guidance on the teaching space construction model; the third target subsystem is used to store excellent case space construction models and provide a panoramic display of the excellent case space construction models. This application has the effect of facilitating the management of teaching space construction.

[0004] Traditional campus teaching space management systems mainly rely on manual operations and simple electronic management, and have the following major problems:

[0005] It is unable to seamlessly integrate with other systems to achieve data sharing and business collaboration, and its data collection and system integration capabilities are weak; not only is the time and labor costs high, but it is also unable to meet the ever-changing teaching space management needs in a timely manner; it is unable to automatically adjust the allocation strategy according to different teaching plans, and has poor ability to handle complex allocation problems, and cannot guarantee the flexibility and accuracy of allocation; the allocation process is complicated and cumbersome, not flexible and efficient enough, and cannot effectively improve the utilization efficiency of teaching space; it relies too much on historical experience and neglects to explore new and potentially more effective allocation methods; the lack of effective resource optimization means may lead to some teaching spaces not being fully utilized, resulting in waste of resources; when the number of teaching spaces is large or the needs are complex, traditional methods are difficult to carry out effective large-scale management.

[0006] In view of this, the present invention proposes a smart campus teaching space overall management system to solve the above problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention provides the following technical solutions: a smart campus teaching space overall management system, comprising:

[0008] Data acquisition module, used to obtain teaching space data, teaching plan data, resource usage data and reservation request data;

[0009] The data processing module is used to pre-process the acquired teaching space data, teaching plan data, resource usage data and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data; and perform weighted fusion on the teaching space parameter data and resource usage parameter data to obtain spatial feature data;

[0010] Resource management module, used to build a teaching space resource library based on spatial feature data;

[0011] The intelligent scheduling module makes an initial allocation of teaching space based on the teaching plan parameter data; based on the initial allocation results, it updates the teaching space status in the teaching space resource library in real time and obtains free space data;

[0012] The reservation management module re-allocates teaching space based on reservation parameter data and free space data;

[0013] The communication module sends the reservation message to the user according to the result of the secondary allocation; each module is connected through wired and / or wireless means.

[0014] Furthermore, the teaching space data includes basic space information, space equipment information and opening period information; basic space information includes the type, capacity, number, location and status data of the teaching space; space equipment information includes the type, number and status of the equipment in the teaching space;

[0015] Teaching plan data includes course data and teaching resource demand data; course data includes course title, course category, instructor, class time, and participating student information; teaching resource demand data includes equipment requirements and space requirements; resource usage data includes space usage frequency and space utilization rate; and reservation request data includes reservation user information, reservation time period, reservation purpose, and required equipment.

[0016] Teaching space data and teaching plan data are collected through the educational affairs system; resource usage data are collected through visual terminals; and reservation request data are collected through user terminals.

[0017] Furthermore, the method of preprocessing the acquired teaching space data, teaching plan data, resource usage data and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data includes:

[0018] Use the K-means clustering algorithm to identify outliers in the teaching space data, teaching plan data, resource usage data, and reservation request data, and remove the identified outliers to obtain processed teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data;

[0019] The processed teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data are normalized by standard deviation and converted into standard normal distribution to obtain normalized teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data.

[0020] Furthermore, the method of weightedly fusing the teaching space parameter data and the resource usage parameter data to obtain the spatial feature data includes:

[0021] The normalized teaching space parameter data and resource usage parameter data are fused through a weighted formula to obtain spatial feature data, which is recorded as , resource usage parameter data is recorded as ;

[0022] The weighted formula is: ;in, The weight coefficient of the parameter data used for the resource; The weight coefficient for using parameter data for the resource.

[0023] Furthermore, the method for constructing a teaching space resource library includes:

[0024] The teaching space resource library is a relational database MySQL. Based on the acquired spatial feature data, the database table structure is defined, the database is created using the SQL language, and an API interface is written to implement the data addition, deletion, modification, and query functions to obtain the established teaching space resource library. The specific method is as follows:

[0025] S41. Define a database table structure based on the acquired spatial feature data. The database table structure consists of a teaching space table, a device table, and a timetable. The teaching space table is used to store basic spatial information for each teaching space. The device table is used to store spatial device information for each teaching space. The timetable is used to store available time period information for each teaching space.

[0026] S42. Use SQL to create a teaching space table, an equipment table, and a schedule table in a relational database; define the field types, primary keys, and foreign keys of each table; the teaching space table uses SpaceID as the primary key; the equipment table uses EquipmentID as the primary key and SpaceID as the foreign key; the schedule uses ScheduleID as the primary key and SpaceID as the foreign key; a one-to-many relationship is established between the teaching space table and the equipment table using the unique identifier of the SpaceID space; and a one-to-many relationship is established between the teaching space table and the schedule using SpaceID as the foreign key.

[0027] S43. Create an API framework using Flask and set up a server; write processing functions at each API endpoint to implement data addition, deletion, modification, and query functions; connect the API framework to a relational database through a database connection library and perform database operations required by the API endpoint; use API testing tools to verify the functionality of the API framework;

[0028] After the above steps, a well-constructed teaching space resource library is finally obtained.

[0029] Furthermore, the method for initially allocating teaching space according to teaching plan parameter data includes:

[0030] The allocation of teaching space is modeled as a Markov decision process, and the Q-learning reinforcement learning algorithm is used to learn the Q-value function. When the Q-value reaches convergence, the optimal allocation strategy is obtained. The specific steps are as follows:

[0031] S61. Build a reinforcement learning framework and define 、 、 and ;in, For teaching space Usage status at the moment; For The action taken at each moment is to allocate teaching space for a certain teaching plan; To select a strategy, that is, according to the current state Select the best action ; is the index of the moment; For Always perform actions Immediate rewards after

[0032] It is obtained through the spatial adaptability function. The specific mathematical formula is: ;in, The difference between teaching resource requirements and teaching space. The smaller the difference, the higher the adaptability.

[0033] S62, using Q-learning reinforcement learning algorithm, initialize the Q value function, set the learning rate and discount factor ;

[0034] S63. Get the current status based on the teaching space resource library At every moment , according to the current Select the best action ; Execute the optimal action , get the new state and instant rewards ;Optimal action use Greedy strategy is used for selection;

[0035] S64. According to the new state and instant rewards Update the Q value function. The specific mathematical formula is: ;in, For the current state Next action Q value; In the new state All possible moves In the example, choose the action that maximizes the Q value and obtains the maximum expected reward; In the new state All possible moves;

[0036] S65, will Set it as the new current state, repeat steps S53-S54 until the Q value reaches convergence, and obtain the optimal allocation strategy; perform the initial allocation of the teaching space according to the optimal allocation strategy, and obtain the initial allocation result.

[0037] Furthermore, the setting learning rate and discount factor The methods include:

[0038] Using the learning rate adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial learning rate; is the updated learning rate; is the update step size, the value range is (0,1]; is the time step;

[0039] Using the factor adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial discount factor; is the updated discount factor; To reward importance;

[0040] Furthermore, the method for secondary allocation of teaching space using the reservation parameter data and the free space data includes:

[0041] The multi-armed bandit model is used to secondary allocate the idle teaching space. The specific steps are as follows:

[0042] S81. Assume that Different arms, each arm corresponds to a different free teaching space; set the initial number of times each arm is pulled , the initial average reward value of each arm is ;in, is the index of the arm, ; Set parameters , controls the width of the confidence interval;

[0043] S82. For each reservation request, calculate the upper confidence limit of each arm, specifically: ;in, is the current total number of pulls; select the arm with the largest upper confidence limit and get the reward ; According to the reward , determine the appointment result;

[0044] S83, updating the average reward value and number of pulls of the selected arm;

[0045] S84, repeat steps S82-S83 until a predetermined number of iterations is reached;

[0046] Furthermore, the reward , the methods for determining the appointment results include:

[0047] Design a reward mechanism, taking successful appointments and conflict failures as feedback results; When it is a positive number, the reservation is considered successful; when the reward If it is 0 or a negative number, the reservation is considered failed;

[0048] When the reservation is successful, the teaching space represented by the selected arm is the result of secondary allocation of the teaching space based on the reservation request and the free space data.

[0049] Furthermore, the method of sending the reservation message to the user according to the result of the secondary allocation includes:

[0050] The user who sent the reservation request will receive the reservation result through the user terminal; if the reservation is successful, a reservation success notification will be sent; if the reservation fails, a reservation failure notification and the reason for failure will be sent.

[0051] The present invention is based on the technical effects and advantages of a smart campus teaching space overall management system:

[0052] The present invention can seamlessly integrate with other systems, such as the teaching space resource library based on the MySQL database, such as the teaching system and the campus card system, to achieve data sharing and business collaboration, and has powerful data collection and system integration capabilities. The database table structure can be easily expanded according to actual needs to meet the ever-changing teaching space management requirements.

[0053] The reinforcement learning algorithm not only automatically adjusts the allocation strategy according to different teaching plans to adapt to changing teaching needs, but also handles complex allocation problems and quickly responds to changes in teaching plans. The learning rate adjustment formula dynamically adjusts the learning rate as the time step increases, allowing the algorithm to converge quickly in the early learning phase and adjust the strategy more finely in the later stages. The discount factor adjustment formula balances immediate and long-term rewards based on reward importance, allowing the algorithm to consider current interests while also focusing on future gains.

[0054] Using the multi-armed bandit model for secondary allocation of teaching space can effectively improve the utilization efficiency of teaching space, enhance the system's adaptability and response speed, while simplifying the complexity of the allocation process, providing an efficient and flexible solution for the management of smart campuses. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a schematic diagram of the structure of a smart campus teaching space overall management system of the present invention;

[0056] Figure 2 This is a flow chart of a method for overall management of smart campus teaching space according to the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] Example 1

[0059] See also Figure 1As shown, this embodiment provides a smart campus teaching space coordination and management system, including:

[0060] Data acquisition module, used to obtain teaching space data, teaching plan data, resource usage data and reservation request data;

[0061] The data processing module is used to pre-process the acquired teaching space data, teaching plan data, resource usage data and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data; and perform weighted fusion on the teaching space parameter data and resource usage parameter data to obtain spatial feature data;

[0062] Resource management module, used to build a teaching space resource library based on spatial feature data;

[0063] The intelligent scheduling module makes an initial allocation of teaching space based on the teaching plan parameter data; based on the initial allocation results, it updates the teaching space status in the teaching space resource library in real time and obtains free space data;

[0064] The reservation management module re-allocates teaching space based on reservation parameter data and free space data;

[0065] The communication module sends the reservation message to the user according to the result of the secondary allocation; each module is connected through wired and / or wireless means.

[0066] The teaching space data includes basic space information, space equipment information and opening period information; basic space information includes the type, capacity, number, location and status data of the teaching space; space equipment information includes the type, number and status of the equipment in the teaching space;

[0067] Teaching plan data includes course data and teaching resource demand data; course data includes course title, course category, instructor, class time, and participating student information; teaching resource demand data includes equipment requirements and space requirements; resource usage data includes space usage frequency and space utilization rate; and reservation request data includes reservation user information, reservation time period, reservation purpose, and required equipment.

[0068] Teaching space data and teaching plan data are collected through the educational affairs system; resource usage data are collected through visual terminals; and reservation request data are collected through user terminals.

[0069] The method of preprocessing the acquired teaching space data, teaching plan data, resource usage data, and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data includes:

[0070] Use the K-means clustering algorithm to identify outliers in the teaching space data, teaching plan data, resource usage data, and reservation request data, and remove the identified outliers to obtain processed teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data;

[0071] The processed teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data are normalized by standard deviation and converted into standard normal distribution to obtain normalized teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data.

[0072] The method of weightedly fusing the teaching space parameter data and the resource usage parameter data to obtain the spatial feature data includes:

[0073] The normalized teaching space parameter data and resource usage parameter data are fused through a weighted formula to obtain spatial feature data, which is recorded as , resource usage parameter data is recorded as ;

[0074] The weighted formula is: ;in, The weight coefficient of the parameter data used for the resource; The weight coefficient for using parameter data for the resource.

[0075] The method for constructing a teaching space resource library includes:

[0076] The teaching space resource library is a relational database MySQL. Based on the acquired spatial feature data, the database table structure is defined, the database is created using the SQL language, and an API interface is written to implement the data addition, deletion, modification, and query functions to obtain the established teaching space resource library. The specific method is as follows:

[0077] S41. Define a database table structure based on the acquired spatial feature data. The database table structure consists of a teaching space table, a device table, and a timetable. The teaching space table is used to store basic spatial information for each teaching space. The device table is used to store spatial device information for each teaching space. The timetable is used to store available time period information for each teaching space.

[0078] S42. Use SQL to create a teaching space table, an equipment table, and a schedule table in a relational database; define the field types, primary keys, and foreign keys of each table; the teaching space table uses SpaceID as the primary key; the equipment table uses EquipmentID as the primary key and SpaceID as the foreign key; the schedule uses ScheduleID as the primary key and SpaceID as the foreign key; a one-to-many relationship is established between the teaching space table and the equipment table using the unique identifier of the SpaceID space; and a one-to-many relationship is established between the teaching space table and the schedule using SpaceID as the foreign key.

[0079] S43. Create an API framework using Flask and set up a server; write processing functions at each API endpoint to implement data addition, deletion, modification, and query functions; connect the API framework to a relational database through a database connection library and perform database operations required by the API endpoint; use API testing tools to verify the functionality of the API framework;

[0080] After the above steps, a well-constructed teaching space resource library is finally obtained.

[0081] The method for initially allocating teaching space according to teaching plan parameter data includes:

[0082] The allocation of teaching space is modeled as a Markov decision process, and the Q-learning reinforcement learning algorithm is used to learn the Q-value function. When the Q-value reaches convergence, the optimal allocation strategy is obtained. The specific steps are as follows:

[0083] S61. Build a reinforcement learning framework and define 、 、 and ;in, For teaching space Usage status at the moment; For The action taken at each moment is to allocate teaching space for a certain teaching plan; To select a strategy, that is, according to the current state Select the best action ; is the index of the moment; For Always perform actions Immediate rewards after

[0084] It is obtained through the spatial adaptability function. The specific mathematical formula is: ;in, The difference between teaching resource requirements and teaching space. The smaller the difference, the higher the adaptability.

[0085] S62, using Q-learning reinforcement learning algorithm, initialize the Q value function, set the learning rate and discount factor ;

[0086] S63. Get the current status based on the teaching space resource library At every moment , according to the current Select the best action ; Execute the optimal action , get the new state and instant rewards ;Optimal action use Greedy strategy is used for selection;

[0087] For example: Assume that in a certain state, there are three optional actions b1, b2, and b3, and their Q values ​​are: Q(b1) = 10; Q(b2) = 15; Q(b3) = 8;

[0088] According to the greedy strategy, the optimal action we choose is ; Next, execute action b2 and update the Q value;

[0089] S64. According to the new state and instant rewards Update the Q value function. The specific mathematical formula is: ;in, For the current state Next action Q value; In the new state All possible moves In the example, choose the action that maximizes the Q value and obtains the maximum expected reward; In the new state All possible moves;

[0090] S65, will Set it as the new current state, repeat steps S53-S54 until the Q value reaches convergence, and obtain the optimal allocation strategy; perform the initial allocation of the teaching space according to the optimal allocation strategy, and obtain the initial allocation result.

[0091] The setting learning rate and discount factor The methods include:

[0092] Using the learning rate adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial learning rate; is the updated learning rate; is the update step size, the value range is (0,1]; is the time step;

[0093] Using the factor adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial discount factor; is the updated discount factor; To reward importance, ;

[0094] For example: Assuming the initial learning rate is 0.2, the update step is 0.5, and the time step is 0.2, according to the learning rate adjustment formula, the updated learning rate is ;

[0095] Assuming the initial discount factor is 0.9, the reward importance is 0.7, and the update step is 0.2, according to the factor adjustment formula, the updated discount factor is ;

[0096] The method for secondary allocation of teaching space using reservation parameter data and free space data includes:

[0097] The multi-armed bandit model is used to secondary allocate the idle teaching space. The specific steps are as follows:

[0098] S81. Assume that Different arms, each arm corresponds to a different free teaching space; set the initial number of times each arm is pulled , the initial average reward value of each arm is ;in, is the index of the arm, ; Set parameters , controls the width of the confidence interval;

[0099] S82. For each reservation request, calculate the upper confidence limit of each arm, specifically: ;in, is the current total number of pulls; select the arm with the largest upper confidence limit and get the reward ; According to the reward , determine the appointment result;

[0100] S83, updating the average reward value and number of pulls of the selected arm;

[0101] S84, repeat steps S82-S83 until a predetermined number of iterations is reached;

[0102] For example, suppose there are four arms, representing the free teaching spaces f1, f2, f3, and f4; initially, the number of times each space is selected and the average reward value are both 0, and the parameter is 0.2; after several iterations, the current state is:

[0103] Space f1: average reward value = 3.8, number of choices = 30; Space f2: average reward value = 3.2, number of choices = 25; Space f3: average reward value = 3.0, number of choices = 20; Space f4: average reward value = 2.5, number of choices = 15;

[0104] Compute the upper confidence bound for each arm for a reservation request:

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] According to the UCB value, space f1 has the highest UCB. Space f1 is selected as the teaching space for secondary distribution. The reward for selecting space f1 is calculated. If the reward If it is a positive number, the reservation is considered successful. If it is a negative number, the reservation is considered failed;

[0110] Assuming the reward in the selection space f1 Is a positive number, update the average reward value of space f1: ;The number of times it was pulled is 31;

[0111] The reward , the methods for determining the appointment results include:

[0112] Design a reward mechanism, taking successful appointments and conflict failures as feedback results; When it is a positive number, the reservation is considered successful; when the reward If it is 0 or a negative number, the reservation is considered failed;

[0113] When the reservation is successful, the teaching space represented by the selected arm is the result of secondary allocation of the teaching space based on the reservation request and the free space data.

[0114] The method of sending the reservation message to the user according to the result of the secondary allocation includes:

[0115] Send the reservation result to the user who made the reservation request through the user terminal; if the reservation is successful, send a reservation success notification; if the reservation fails, send a reservation failure notification and the reason for failure;

[0116] In this embodiment, the teaching space resource library based on the MySQL database can be seamlessly integrated with other systems, such as the educational administration system and the campus card system, to achieve data sharing and business collaboration, and has powerful data collection and system integration capabilities. The database table structure can be easily expanded according to actual needs to meet the ever-changing teaching space management requirements.

[0117] The reinforcement learning algorithm not only automatically adjusts the allocation strategy according to different teaching plans to adapt to changing teaching needs, but also handles complex allocation problems and quickly responds to changes in teaching plans. The learning rate adjustment formula dynamically adjusts the learning rate as the time step increases, allowing the algorithm to converge quickly in the early learning phase and adjust the strategy more finely in the later stages. The discount factor adjustment formula balances immediate and long-term rewards based on reward importance, allowing the algorithm to consider current interests while also focusing on future gains.

[0118] Using the multi-armed bandit model for secondary allocation of teaching space can effectively improve the utilization efficiency of teaching space, enhance the system's adaptability and response speed, while simplifying the complexity of the allocation process, providing an efficient and flexible solution for the management of smart campuses.

[0119] Example 2

[0120] See also Figure 2 As shown, this embodiment provides a method for overall management of smart campus teaching space, including:

[0121] S1. Obtain teaching space data, teaching plan data, resource usage data, and reservation request data;

[0122] S2. Preprocess the acquired teaching space data, teaching plan data, resource usage data, and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data; perform weighted fusion of the teaching space parameter data and resource usage parameter data to obtain spatial feature data;

[0123] S3. Construct a teaching space resource library based on spatial feature data;

[0124] S4. Initial allocation of teaching space according to teaching plan parameter data; based on the initial allocation results, the teaching space resource library in real time updates the teaching space status and obtains free space data;

[0125] S5. Perform secondary allocation of teaching space based on reservation parameter data and free space data;

[0126] S6. According to the result of the secondary allocation, the reservation message is sent to the user; each module is connected via wired and / or wireless means.

[0127] Since the electronic device introduced in this embodiment is an electronic device used to implement the embodiment of this application based on a smart campus teaching space overall management system, based on the smart campus teaching space overall management system introduced in the embodiment of this application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as technical personnel in this field implement the electronic device used in the embodiment of this application based on a smart campus teaching space overall management system, it falls within the scope of protection of this application.

[0128] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0129] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for users of ordinary skill in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A smart campus teaching space management system, characterized by: include: Data acquisition module, used to obtain teaching space data, teaching plan data, resource usage data and reservation request data; A data processing module is used to pre-process the acquired teaching space data, teaching plan data, resource usage data and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data; Perform weighted fusion of teaching space parameter data and resource usage parameter data to obtain spatial feature data; Resource management module, used to build a teaching space resource library based on spatial feature data; The intelligent scheduling module is used to make an initial allocation of teaching space according to the teaching plan parameter data; based on the initial allocation results, the teaching space status is updated in real time in the teaching space resource library and the free space data is obtained; The method for initially allocating teaching space according to teaching plan parameter data includes: The allocation of teaching space is modeled as a Markov decision process, and the Q-learning reinforcement learning algorithm is used to learn the Q-value function. When the Q-value reaches convergence, the optimal allocation strategy is obtained. The specific steps are as follows: S61. Build a reinforcement learning framework and define 、 、 and ;in, For teaching space Usage status at the moment; For the actions taken at every moment; To choose a strategy; Index of the moment; For Always perform actions Immediate rewards after It is obtained through the spatial adaptability function. The specific mathematical formula is: ;in, The difference between teaching resource requirements and teaching space. The smaller the difference, the higher the adaptability. S62, using Q-learning reinforcement learning algorithm, initialize the Q value function, set the learning rate and discount factor ; S63. Get the current status based on the teaching space resource library At every moment , according to the current Select the best action ; Execute the optimal action , get the new state and instant rewards ;Optimal action use Greedy strategy is used for selection; S64. According to the new state and instant rewards Update the Q-value function; S65, will Set it as the new current state, repeat steps S63-S64 until the Q value reaches convergence, and obtain the optimal allocation strategy; perform the initial allocation of the teaching space according to the optimal allocation strategy, and obtain the initial allocation result; The reservation management module is used to perform secondary allocation of teaching space based on reservation parameter data and free space data; The method for secondary allocation of teaching space according to reservation parameter data and free space data includes: The multi-armed bandit model is used to secondary allocate the idle teaching space. The specific steps are as follows: S81. Assume that Different arms, each arm corresponds to a different free teaching space; set the initial number of times each arm is pulled , the initial average reward value of each arm is ;in, is the index of the arm, ; Set parameters , controls the width of the confidence interval; S82. For each reservation request, calculate the upper confidence limit of each arm, specifically: ;in, is the current total number of pulls; select the arm with the largest upper confidence limit and get the reward ; According to the reward , determine the appointment result; S83, updating the average reward value and number of pulls of the selected arm; S84, repeat steps S82-S83 until a predetermined number of iterations is reached; The communication module is used to send the reservation message to the user according to the result of the secondary allocation; each module is connected by wired and / or wireless means.

2. A smart campus teaching space overall management system according to claim 1, characterized in that: The teaching space data includes basic space information, space equipment information and opening period information; basic space information includes the type, capacity, number, location and status data of the teaching space; space equipment information includes the type, number and status of the equipment in the teaching space; Teaching plan data includes course data and teaching resource demand data; course data includes course title, course category, instructor, class time, and participating student information; teaching resource demand data includes equipment requirements and space requirements; resource usage data includes space usage frequency and space utilization rate; and reservation request data includes reservation user information, reservation time period, reservation purpose, and required equipment. Teaching space data and teaching plan data are collected through the educational system; resource usage data are collected through visual terminals; The reservation request data is collected through the user terminal.

3. The smart campus teaching space overall management system according to claim 2 is characterized in that: The method of preprocessing the acquired teaching space data, teaching plan data, resource usage data, and reservation request data to obtain teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data includes: Use the K-means clustering algorithm to identify outliers in the teaching space data, teaching plan data, resource usage data, and reservation request data, and remove the identified outliers to obtain processed teaching space parameter data, teaching plan parameter data, resource usage parameter data, and reservation parameter data; The processed teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data are normalized by standard deviation and converted into standard normal distribution to obtain normalized teaching space parameter data, teaching plan parameter data, resource usage parameter data and reservation parameter data.

4. The smart campus teaching space overall management system according to claim 3 is characterized in that: The method of weightedly fusing the teaching space parameter data and the resource usage parameter data to obtain the spatial feature data includes: The normalized teaching space parameter data and resource usage parameter data are fused through a weighted formula to obtain spatial feature data, which is recorded as , resource usage parameter data is recorded as ; The weighted formula is: ;in, The weight coefficient of the parameter data used for the resource; The weight coefficient for using parameter data for the resource.

5. The smart campus teaching space overall management system according to claim 4 is characterized in that: The method for constructing a teaching space resource library includes: The teaching space resource library is a relational database MySQL. The database table structure is defined according to the acquired spatial feature data, the database is created using the SQL language, and the API interface is written to implement the data addition, deletion, modification and query functions to obtain the established teaching space resource library.

6. The smart campus teaching space overall management system according to claim 5 is characterized in that: The setting learning rate and discount factor The methods include: Using the learning rate adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial learning rate; is the updated learning rate; is the update step size; is the time step; Using the factor adjustment formula To limit it, the specific mathematical formula is: ;in, is the initial discount factor; is the updated discount factor; To reward importance.

7. The smart campus teaching space overall management system according to claim 6 is characterized in that: The reward , the methods for determining the appointment results include: Design a reward mechanism, taking successful appointments and conflict failures as feedback results; When it is a positive number, the reservation is considered successful; when the reward If it is 0 or a negative number, the reservation is considered failed; When the reservation is successful, the teaching space represented by the selected arm is the result of secondary allocation of the teaching space based on the reservation request and the free space data.

8. The smart campus teaching space overall management system according to claim 7 is characterized in that: The method of sending the reservation message to the user according to the result of the secondary allocation includes: The user who sent the reservation request will receive the reservation result through the user terminal; if the reservation is successful, a reservation success notification will be sent; if the reservation fails, a reservation failure notification and the reason for failure will be sent.

Images