Smart campus Internet of Things multi-service collaborative scheduling control system based on all-optical network

Abstract

The invention discloses a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network, and relates to the technical field of Internet of Things, and the system comprises an Internet of Things digital platform which is in communication connection with the following modules: an all-optical network infrastructure module which is used for constructing an optical fiber backbone network in a campus, and each subsystem in the campus is accessed. The reliability and stability of the campus Internet of Things are enhanced by constructing all-optical network infrastructure, the subsystems are directly connected through optical fibers by adopting the passive optical network technology, and the use of active equipment is reduced, so that the equipment failure rate and the maintenance cost are reduced, optical wavelength resources are dynamically allocated by utilizing a software defined network controller, and the reliability and the stability of the campus Internet of Things are improved. The transmission path is adjusted according to the service priority, efficient operation of key services is ensured, network stability is improved, high speed and safety of data transmission are guaranteed, and solid network support is provided for various intelligent applications in the campus.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, and in particular to a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network. BACKGROUND

[0002] With the expansion of the scale of the school, the management difficulty is increasing, and it is necessary to build a smart campus through building intelligent building system. In the construction process of the smart campus, there are twenty intelligent subsystems including information network system, security monitoring system, intrusion alarm system, information publishing system, smart classroom system, entrance control system, video security monitoring system, etc. The smart campus Internet of Things can be used to uniformly manage and schedule various resources in the campus, so that each intelligent subsystem realizes interconnection through the Internet of Things digital platform.

[0003] In the prior art, the traditional Internet of Things construction adopts optical fiber backbone plus category 6 network cable to form a communication network, which has problems such as large use of non-renewable resource copper core cable, limited terminal expansion, and transmission rate limited by network cable. Moreover, when scheduling multi-service resources, it is easy to fall into a local optimal trap and it is difficult to dynamically balance the demand for multi-service resources. Therefore, how to design an all-optical network architecture to cover the entire campus, reduce the energy consumption of active devices, and use a particle swarm algorithm to introduce a chaotic sequence to generate an initial particle swarm to avoid falling into a local optimum, so as to optimize the scheduling of multi-service resources and ensure the collaborative scheduling efficiency of multi-service resources in the campus scene. For this reason, the present application proposes a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network. SUMMARY

[0004] The present application aims to provide a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network to solve the problems raised in the background art.

[0005] To solve the above technical problems, the technical solution adopted by the present application is as follows:

[0006] The smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network comprises an Internet of Things digital platform, and the Internet of Things digital platform is communicatively connected with the following modules, wherein:

[0007] An all-optical network infrastructure module is used to build a high-speed, low-latency, and low-energy campus optical fiber backbone network, and access each subsystem in the campus to cover the network globally.

[0008] A multi-service resource perception module is used to deploy multi-modal sensors, obtain a basic data set for service resource scheduling, mark the limit value of the service resource, and generate a resource mark sequence list.

[0009] The chaos sequence initialization engine introduces a chaos sequence to generate an initial particle swarm based on a particle swarm algorithm, replaces random initialization, makes particles uniformly distributed in a solution space, and avoids premature convergence caused by initial particle aggregation of a traditional particle swarm algorithm (PSO);

[0010] The business resource coordination optimization module is used for combining the resource marking sequence list and the initial particle swarm, optimizing a business resource allocation strategy, intelligently scheduling business demands of each subsystem in the campus, and ensuring efficient use of resources.

[0011] The intelligent decision control module is used for providing a visual cockpit, displaying a resource scheduling state, converting an optimization result into an executable instruction, and opening an API to call a scheduling result by each subsystem to perform cross-system linkage control.

[0012] Further improvement of the technical scheme of the application is that the all-optical network infrastructure module comprises a fiber backbone network unit and an optical network management and control unit.

[0013] The fiber backbone network unit adopts a passive optical network (PON) technology, directly connects the subsystems in the campus through optical fibers, and expands the number of subsystems through a splitter.

[0014] The optical network management and control unit is used for deploying a software-defined network (SDN) controller, allocating optical wavelength resources, adjusting a transmission path according to a business priority, and avoiding resource waste caused by traditional fixed-bandwidth allocation of a network cable.

[0015] Further improvement of the technical scheme of the application is that the fiber backbone network unit comprises:

[0016] The passive optical network (PON) device is deployed in a campus center machine room, each subsystem in the campus is connected to the passive optical network device through an optical fiber, and fiber direct connection is realized.

[0017] The splitter is used to split the optical fiber signal into multiple paths, and the splitter reduces the number of active devices, reduces energy consumption and device cost, improves the scalability and flexibility of the network, and meets the growing demand for subsystem access in the campus.

[0018] The passive optical network device manages and transmits the split signals, aggregates data signals generated by each subsystem to an Internet of Things digital platform in the campus center machine room for processing, and the PON technology uses passive devices to reduce the use of active devices and reduce device failure rate and maintenance cost.

[0019] Further improvement of the technical scheme of the application is that the optical network management and control unit comprises:

[0020] Deploy a software-defined network (SDN) controller in the campus center machine room, complete system initialization, and connect the controller to the passive optical network device to obtain network topology and device status information and build a network topology graph;

[0021] The software-defined network controller dynamically allocates optical wavelength resources according to the business needs of each subsystem, monitors network traffic in real time, adjusts wavelength allocation on demand, ensures that high-bandwidth services have sufficient resources, and avoids resource waste caused by fixed bandwidth allocation.

[0022] The software-defined network controller adjusts the transmission path according to the business priority, preferentially allocates low-latency paths for preset services with high real-time requirements, ensures fast data transmission, and guarantees efficient operation of critical services.

[0023] Further improvements of the technical solution of the application are that the multi-service resource perception module includes a resource monitoring unit and a resource marking unit.

[0024] The resource monitoring unit is used to deploy multi-modal sensors to cover security, energy consumption, classroom environment and other scenarios, and to monitor the use of business resources of each subsystem in the campus in real time, and to obtain the basic data set for business resource scheduling.

[0025] The resource marking unit, in combination with the basic data set for business resource scheduling and the pre-set business resource limit value of each subsystem, marks the subsystems and corresponding business resources with increased demand, and integrates to generate a resource marking sequence list.

[0026] Further improvements of the technical solution of the application are that the resource monitoring unit includes:

[0027] Multi-modal sensors are deployed in key scenarios on campus (security, energy consumption, classroom environment, etc.), including cameras, temperature and humidity sensors, and energy consumption monitors, etc., to achieve full coverage and real-time collection of business resource usage of each subsystem;

[0028] The multi-modal sensors collect security monitoring pictures, energy consumption data and classroom environment parameter business resource usage information in real time, and transmit the collected data to the resource monitoring unit through the Internet of Things network to ensure the real-time and accuracy of the data.

[0029] The resource monitoring unit pre-processes the received multi-modal data, including data cleaning, format conversion, etc., and stores the processed data in a database to form a basic data set for business resource scheduling.

[0030] Further improvements of the technical solution of the application are that the resource marking unit includes:

[0031] Load the latest 5 minutes of business resource scheduling basic data set from the database, classify by subsystem, combine the preset business resource limit value of each subsystem, analyze the current resource usage, evaluate the demand amount of each subsystem to the business resource, and identify the subsystem with increased demand amount and the corresponding resource;

[0032] Mark the subsystem with increased demand amount and the corresponding business resource, record the unique ID identification and the priority of resource demand, and generate marking information containing subsystem identification, resource type, demand amount and priority;

[0033] Integrate the marked resource information, sort according to the priority, and obtain the current available amount of the campus resource pool, and comprehensively generate a resource marking sequence table.

[0034] The further improvement of the technical scheme of the application is that the chaotic sequence initialization engine comprises:

[0035] A chaotic model based on Logistic chaotic mapping is selected, initial parameters and iteration times are set, and a series of chaotic values are generated by calculating through the iteration formula of chaotic mapping;

[0036] According to the values generated by the chaotic sequence, the dimension of the problem and the size of the particle swarm, the chaotic values are mapped into the solution space, each particle is given a corresponding position and speed, the particles are uniformly distributed in the solution space, and an initial particle swarm based on the chaotic sequence is formed to replace the traditional random initialization method, wherein the dimension of the problem represents the number of variables in the optimization problem;

[0037] The constructed initial particle swarm is substituted into the particle swarm algorithm, the position and speed of the particle are updated according to the updating rule of the traditional particle swarm algorithm (PSO) and the individual extreme value and global extreme value, in the iteration process, the uniform distribution characteristic of the chaotic sequence is utilized to avoid particle aggregation, effectively prevent premature convergence, and improve the optimization ability of the algorithm.

[0038] The further improvement of the technical scheme of the application is that the business resource collaborative optimization module comprises:

[0039] Extract the resource marking sequence table, analyze the resource type (CPU, bandwidth, storage) and real-time state (occupancy rate, remaining amount) of each subsystem, identify the resource demand of each subsystem, including the subsystem with increased demand amount and the corresponding resource type;

[0040] According to the resource dimension, the chaotic sequence is generated, is mapped to the initial position (resource allocation proportion) and speed (adjustment step) of the particle swarm, and the fitness function is defined as the weighted sum of resource utilization rate and service demand satisfaction rate, the particle position is updated according to the traditional particle swarm algorithm (PSO) rule, the global optimal solution is dynamically disturbed combined with the chaotic sequence, and the local optimum is avoided, until the termination condition is met, that is, the maximum iteration number is met, and the global optimal service resource allocation strategy is generated;

[0041] The service requests are classified according to the emergency degree and resource demand type, the request is routed to the subsystem with sufficient idle resources according to the resource marking sequence table and the global optimal service resource allocation strategy, the high-priority service is preferentially guaranteed, and the resource usage is monitored in real time, if resource conflict occurs, the particle swarm algorithm is triggered to re-optimize, the resource marking sequence table is updated, and then the service request-resource allocation mapping table and the dynamic adjustment log are output.

[0042] The further improvement of the technical scheme of the application is that the intelligent decision control module comprises:

[0043] The resource marking sequence table data including CPU / bandwidth / storage occupancy rate, remaining amount and service priority queue are pulled in real time through WebSocket, and multi-dimensional charts are rendered by using ECharts, alarm threshold configuration is integrated, CPU / bandwidth / storage occupancy rate is displayed in blocks according to subsystems, alarm threshold is marked by color gradient, and then an interactive cockpit interface is output.

[0044] The global optimal service resource allocation strategy of the service resource collaborative optimization module is analyzed, the resource adjustment amount of each subsystem is extracted, and the instruction template is encapsulated according to the subsystem protocol, the execution condition is added, the legality of the instruction is checked through the rule engine (Drools), the instruction sequence with timestamp is generated, the standardized instruction set is output, and the resource type, adjustment value, target subsystem ID and execution priority are included.

[0045] Based on the gRPC protocol, API endpoints are exposed, and input and output are defined, wherein the input is the standardized instruction set, the output is the execution status code, the supported resource type and instruction format are declared when the subsystem is registered, the route is automatically matched, at the same time, the asynchronous message queue (Kafka) is used to buffer the instruction, the execution result (success / failure reason) is returned after the subsystem pulls, the instruction state is updated and the compensation mechanism is triggered, that is, the failure is retried for 3 times, and then the system scheduling log is output, including the record instruction distribution time, subsystem response and final execution result, supporting audit and fault backtracking.

[0046] Due to the adoption of the above technical scheme, the application has the following technical progress compared with the prior art:

[0047] 1. The application provides a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network, which enhances the reliability and stability of the campus Internet of Things by constructing an all-optical network infrastructure, directly connects each subsystem through an optical fiber using passive optical network technology, reduces the use of active devices, thereby reducing the device failure rate and maintenance cost, and dynamically allocates optical wavelength resources using a software-defined network controller, adjusts the transmission path according to the service priority, and ensures the efficient operation of critical services, not only improving the stability of the network, but also ensuring the high speed and security of data transmission, providing a solid network support for various intelligent applications in the campus.

[0048] 2. The application provides a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network, which optimizes the scheduling of multi-service resources by introducing a chaotic sequence initialized particle swarm algorithm, and makes the particles uniformly distributed in the solution space by using a chaotic sequence to replace the traditional random initialization method when initializing the particle swarm, effectively avoiding the premature convergence problem caused by particle aggregation, and intelligently scheduling the service demand of each subsystem in the campus according to the resource marker sequence table and the initial particle swarm by defining the fitness function as the weighted sum of resource utilization rate and service demand satisfaction rate, ensuring efficient use of resources, not only improving the efficiency of resource scheduling, but also enhancing the optimization ability of the system, thereby optimizing the allocation of service resources. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0050] Figure 1 The workflow diagram of the present application;

[0051] Figure 2 The system function module diagram of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] Embodiment 1, as Figure 1 , Figure 2As shown, the present application provides a smart campus Internet of Things multi-service collaborative scheduling control system based on an all-optical network, comprising an Internet of Things digital platform, which is communicatively connected with the following modules, wherein:

[0054] An all-optical network infrastructure module is used to build a high-speed, low-latency, and low-energy campus fiber backbone network, and access various subsystems within the campus, a global coverage network, replacing traditional copper cables, supporting real-time transmission of multi-service data, and the all-optical network infrastructure module comprises a fiber backbone network unit and an optical network management and control unit.

[0055] The fiber backbone network unit uses passive optical network (PON) technology to directly connect the subsystems within the campus through optical fibers, reducing the level of active devices in each subsystem, and expanding the number of subsystems through an optical splitter. The passive optical network (PON) device is deployed in the campus central computer room, and each subsystem within the campus is connected to the passive optical network device through an optical fiber, realizing direct connection of optical fibers. The optical fiber has the characteristics of high bandwidth, low latency, and strong anti-interference ability, and can provide a high-speed and stable physical link for multi-service data transmission within the campus, ensuring the efficiency and reliability of data transmission. The subsystems include an information network system, a security monitoring system, an intrusion alarm system, an information publishing system, a smart classroom system, an entrance control system, a video security monitoring system, etc. The optical fiber signal is split by the optical splitter to expand the connection to each subsystem. The use of the optical splitter reduces the number of active devices, reduces energy consumption and equipment costs, and improves the scalability and flexibility of the network, meeting the growing demand for subsystem access within the campus. The passive optical network device manages and transmits the split signals, and the data signals generated by each subsystem are aggregated to the Internet of Things digital platform in the campus central computer room for processing. Since the PON technology uses passive devices, the use of active devices is reduced, reducing equipment failure rate and maintenance cost, and through optical fiber transmission, the high speed and stability of data transmission are ensured.

[0056] The specific process of the fiber backbone network unit is as follows:

[0057] The passive optical network (PON) equipment is deployed in the campus center room as the core node of the entire fiber backbone network. From the center room, optical fibers are laid to various subsystems (information network system, security monitoring system, intrusion alarm system, information publishing system, smart classroom system, access control system, video security monitoring system, etc.) in the campus. Optical fibers have the characteristics of high bandwidth, low delay and strong anti-interference ability, and can provide high-speed and stable physical links for multi-service data transmission. Optical splitters are deployed at key nodes of the fiber path to divide the signal of one optical fiber into multiple paths and expand the connection to various subsystems. The use of optical splitters reduces the number of active devices, reduces energy consumption and equipment cost, and improves the scalability and flexibility of the network. After the optical splitter divides the optical fiber signal into multiple paths, each path is connected to a subsystem to ensure that each subsystem can receive sufficient signal strength. The splitting ratio of the optical splitter is selected according to actual needs to meet the access requirements of subsystems in different areas of the campus. The data signals generated by various subsystems are transmitted to the passive optical network equipment in the center room through optical fibers. Due to the high speed and stability of optical fiber transmission, data signals can be transmitted under low delay and high bandwidth conditions to ensure the efficiency and reliability of data transmission. The passive optical network equipment manages and transmits the split signals, aggregates the data signals generated by various subsystems to the center room, and then transmits the aggregated data signals to the Internet of Things digital platform in the campus center room. The Internet of Things digital platform processes the aggregated data, including data analysis, storage, analysis, etc.

[0058] The optical network management unit is used to deploy a software-defined network (SDN) controller, allocate optical wavelength resources, adjust the transmission path according to the business priority, avoid resource waste caused by traditional fixed bandwidth allocation, ensure priority transmission of real-time requirements, deploy a software-defined network (SDN) controller in the campus center room, complete system initialization, and connect the controller to the passive optical network equipment to obtain network topology and device state information, build a network topology graph. The software-defined network controller dynamically allocates optical wavelength resources according to the business requirements of various subsystems, monitors network traffic in real time, adjusts wavelength allocation on demand, ensures that high-bandwidth businesses have enough resources, and avoids resource waste caused by fixed bandwidth allocation. The software-defined network controller adjusts the transmission path according to the business priority, and for pre-set businesses with high real-time requirements, it allocates low-delay paths first to ensure fast data transmission and efficient operation of critical businesses.

[0059] The specific process of the optical network management unit is as follows:

[0060] A software-defined network (SDN) controller is deployed in the campus center machine room as the core equipment of the optical network management and control. The hardware installation and software configuration of the SDN controller are completed to ensure its normal operation. The SDN controller is connected to the passive optical network (PON) equipment through a network interface to establish a communication link, ensuring that the SDN controller can interact with the PON equipment to obtain network topology and device state information of each subsystem. The SDN controller automatically discovers the network topology structure through a network protocol, including fiber links, optical splitters, and subsystem access points, to build a network topology graph and update the network connection relationship and topology changes in real time. The SDN controller periodically sends a state query request to the PON equipment to obtain device operating state information of each subsystem, including optical power, link status, and device temperature. The SDN controller dynamically analyzes traffic characteristics based on the business requirements (bandwidth requirements, real-time requirements, etc.) of each subsystem, collects network traffic data in real time through traffic monitoring, analyzes the traffic trend of each subsystem, and dynamically allocates optical wavelength resources based on business requirements and traffic monitoring results. For high-bandwidth services, sufficient wavelength resources are allocated to ensure efficient transmission of services. Network traffic is monitored in real time, and wavelength allocation is adjusted as needed to avoid resource waste caused by fixed bandwidth allocation. The SDN controller evaluates the priority of each subsystem's business based on pre-set business priority rules. For services with high real-time requirements (video monitoring, real-time teaching in smart classrooms, etc.), high priority is marked. Based on the business priority, the SDN controller adjusts the transmission path and allocates low-latency paths to high-priority services. The latency and bandwidth utilization of the path are monitored in real time, and the path is dynamically adjusted to ensure efficient operation of critical services.

[0061] A multi-service resource perception module is used to deploy multi-modal sensors to perceive personnel, equipment, and environmental status in the campus, obtain basic data sets for business resource scheduling, and perform limit value marking of business resources to generate a resource marking sequence list. The multi-service resource perception module includes a resource monitoring unit and a resource marking unit.

[0062] The resource monitoring unit is configured to deploy multi-modal sensors to cover security, energy consumption, classroom environment and other scenes, monitor the use of business resources of each subsystem in the campus in real time, and obtain a basic data set for business resource scheduling. Multi-modal sensors are deployed in each key scene (security, energy consumption, classroom environment, etc.) in the campus, including cameras, temperature and humidity sensors, energy consumption monitors and the like, to achieve full coverage and real-time collection of the use of business resources of each subsystem. The multi-modal sensors collect security monitoring pictures, energy consumption data and classroom environment parameter business resource use information in real time, and transmit the collected data to the resource monitoring unit through the Internet of Things network to ensure the real-time and accuracy of the data. The resource monitoring unit pre-processes the received multi-modal data, including data cleaning, format conversion and the like, and stores the processed data in a database to form a basic data set for business resource scheduling.

[0063] The specific process of the resource monitoring unit is as follows:

[0064] Multi-modal sensors are deployed in each key scene (security, energy consumption, classroom environment, etc.) in the campus, including cameras, temperature and humidity sensors, energy consumption monitors and the like, to ensure full coverage of the sensors and real-time collection of the use of business resources of each subsystem. The cameras collect security monitoring pictures in real time for the security monitoring system, the temperature and humidity sensors collect classroom environment parameters in real time for classroom environment monitoring, and the energy consumption monitors collect energy consumption data in real time for the energy consumption management system. The multi-modal sensors transmit the collected data to the resource monitoring unit through the Internet of Things network to ensure the stability and security of data transmission and avoid data loss or transmission errors. The resource monitoring unit receives real-time data streams from the multi-modal sensors, cleans the received multi-modal data to remove noise data, duplicate data and abnormal data, detects and processes missing values in the data to ensure the integrity and accuracy of the data, and then converts data of different formats into a unified format. The pre-processed data is stored in a database to form a basic data set for business resource scheduling, and an index is established for the stored data. The database is backed up and maintained regularly to ensure the security and reliability of the data.

[0065] The resource marking unit, in combination with the basic data set of service resource scheduling, the pre-set service resource limit value of each subsystem, marks the subsystems with increased demand and the corresponding service resources, integrates to generate a resource marking list, loads the latest 5-minute basic data set of service resource scheduling from the database, classifies according to the subsystems, analyzes the current resource usage in combination with the pre-set service resource limit value of each subsystem, evaluates the demand of each subsystem for service resources, identifies the subsystems with increased demand and their corresponding resources, marks the subsystems with increased demand and their corresponding service resources, records their unique ID identification and the priority of resource demand, generates marking information containing the subsystem identification, resource type, demand and priority, integrates the marked resource information, sorts according to the priority, and obtains the current available amount of the campus resource pool, and comprehensively generates a resource marking list;

[0066] The specific process of the resource marking unit is as follows:

[0067] The resource marking unit, in combination with the basic data set of service resource scheduling, the pre-set service resource limit value of each subsystem, marks the subsystems with increased demand and the corresponding service resources, integrates to generate a resource marking list, loads the latest 5-minute basic data set of service resource scheduling from the database, classifies according to the subsystems, analyzes the current resource usage in combination with the pre-set service resource limit value of each subsystem, evaluates the demand of each subsystem for service resources, identifies the subsystems with increased demand and their corresponding resources, marks the subsystems with increased demand and their corresponding service resources, records their unique ID identification and the priority of resource demand, generates marking information containing the subsystem identification, resource type, demand and priority, integrates the marked resource information, sorts according to the priority, and obtains the current available amount of the campus resource pool, and comprehensively generates a resource marking list;

[0068] The chaos sequence initialization engine introduces chaos sequence to generate initial particle swarm based on particle swarm algorithm, replaces random initialization, makes particles uniformly distributed in solution space, and avoids premature convergence caused by initial particle aggregation in traditional particle swarm algorithm (PSO);

[0069] The business resource collaborative optimization module is used to combine the resource marking list and the initial particle swarm, optimize the business resource allocation strategy, and intelligently schedule the business demand of each subsystem in the campus, to ensure efficient use of resources;

[0070] The intelligent decision-making control module is used to provide a visual cockpit, display the resource scheduling status, convert the optimization results into executable instructions, and open the API for each subsystem to call the scheduling results for cross-system linkage control.

[0071] Example 2, as Figure 1 、 Figure 2 As shown, based on Example 1, the present invention provides a technical solution: preferably, the chaotic sequence initialization engine includes:

[0072] A chaotic model based on the Logistic chaotic map is selected, and the initial parameters and number of iterations are set. The chaotic map is used for calculation through the iterative formula to generate a series of chaotic values. Based on the values ​​generated by the chaotic sequence, the chaotic values ​​are mapped to the solution space in combination with the dimension of the problem and the size of the particle swarm. Each particle is assigned a corresponding position and velocity, so that the particles are evenly distributed in the solution space, forming an initial particle swarm based on the chaotic sequence, replacing the traditional random initialization method. The dimension of the problem represents the number of variables in the optimization problem. The constructed initial particle swarm is substituted into the particle swarm algorithm. According to the update rules of the traditional particle swarm algorithm (PSO), the position and velocity of the particles are updated according to the individual extreme value and the global extreme value. During the iteration process, the uniform distribution characteristics of the chaotic sequence are used to avoid particle aggregation, effectively prevent premature convergence, and improve the algorithm's optimization ability.

[0073] The specific process of the chaos sequence initialization engine is as follows:

[0074] The chaos model based on Logistic chaos map is selected. Logistic chaos map is a classic chaotic system, and its iterative formula is:

[0075] ;

[0076] in, Is the current iteration value, a value between 0 and 1, indicating the The result of the iterations, is the parameter of the chaotic mapping, which is the parameter that controls the intensity of chaotic behavior. Its value range is , to ensure that the system is in a chaotic state, is the value of the next iteration, i.e. The result of the iteration;

[0077] Set the initial parameters, including the initial value, the number of iterations and the parameters of the chaos map. The number of iterations is determined according to the complexity of the problem and the required particle swarm size. The parameters of the chaos map ensure that the system is in a chaotic state. Then, the iterative formula of the Logistic chaos map is used for calculation to generate a series of chaotic values. , each iteration is updated according to the iterative formula until the set number of iterations is reached The generated chaotic values ​​have the characteristics of randomness and uniform distribution, which can replace the traditional random initialization method; according to the dimension of the optimization problem and particle swarm size , determine the number of particles to be generated and the dimensions of each particle, for a dimensional optimization problem, the position and velocity of each particle are dimensional vector, and generate the chaotic value Mapped into the solution space, for each particle , each dimension of its position and velocity Initialized to chaotic values, the formulas for position and velocity are as follows:

[0078] ;

[0079] ;

[0080] in, It is The particle in Position in dimension, It is The particle in The speed in dimension, and is the first in the chaotic numerical sequence The value and The values ​​represent the continuous values ​​in the chaotic numerical sequence, which are used to initialize the position and velocity of the particles. It is the dimension of the problem, that is, the number of variables in the optimization problem. It is the size of the particle swarm, that is, the number of particles. is the index of the chaotic numerical sequence, It is a function that maps chaotic values ​​to the solution space and is scaled according to the upper and lower bounds of the problem; assigns corresponding positions and velocities to each particle so that the particles are evenly distributed in the solution space, ensuring that the distribution of the initial particle swarm has good diversity and uniformity, avoiding particle aggregation, and combining all initialized particles into an initial particle swarm as the input of the particle swarm algorithm (PSO). Then, the particle swarm algorithm is initialized and the constructed initial particle swarm is substituted into the particle swarm algorithm (PSO). The individual extreme value and the global extreme value are initialized. The individual extreme value is the initial position of each particle, and the global extreme value is the optimal position in the initial particle swarm. According to the update rule of the traditional particle swarm algorithm (PSO), the position and velocity of the particles are updated according to the individual extreme value and the global extreme value. In the iterative process, the uniform distribution characteristics of the chaotic sequence are used to avoid particle aggregation, effectively prevent premature convergence, improve the global search ability and convergence speed of the algorithm, ensure that the algorithm can find the global optimal solution, and update the formula of the particle position and velocity as follows:

[0081]

[0082] in, It is The particle in In the dimension The speed of iterations, It is The particle in In the dimension The speed of iterations, It is The particle in In the dimension The position of the iteration, It is The particle in In the dimension The position of the iteration, is the inertia weight, which controls the influence of the particle's previous velocity, and is a learning factor used to adjust the tendency of particles to move toward individual extreme values ​​and global extreme values. and Is a random number, randomly generated from the interval [0,1], to increase the randomness of the algorithm, It is The particle in The individual extreme value in dimension (i.e. the optimal position found by the particle in history), It is in The global extremum in dimension (i.e. the optimal position found historically by all particles), is the current iteration number;

[0083] The business resource coordination optimization module comprises:

[0084] The resource marker sequence list is extracted, the resource types (CPU, bandwidth, storage) and real-time states (occupancy, remaining amount) of each subsystem are analyzed, the resource requirements of each subsystem are identified, including the subsystems with increased demand and the corresponding resource types, a chaotic sequence is generated according to the resource dimension, which is mapped to the initial position (resource allocation ratio) and speed (adjustment step) of the particle swarm, and the fitness function is defined as the weighted sum of resource utilization rate and business demand satisfaction rate, the particle position is updated according to the traditional particle swarm optimization (PSO) rule, the global optimal solution is dynamically disturbed by combining the chaotic sequence, and the local optimum is avoided, until the termination condition is met, that is, the maximum iteration number is met, the global optimal business resource allocation strategy is generated, the business requests are classified according to the emergency degree and resource demand type, the requests are routed to the subsystem with sufficient idle resources according to the resource marker sequence list and the global optimal business resource allocation strategy, the high-priority business is preferentially guaranteed, and the resource usage is monitored in real time, if a resource conflict occurs, the particle swarm optimization is triggered to re-optimize, the resource marker sequence list is updated, and then the business request-resource allocation mapping table and dynamic adjustment log are output;

[0085] The intelligent decision control module comprises:

[0086] The resource marker sequence list data including CPU / bandwidth / storage occupancy, remaining amount and business priority queue is pulled in real time through WebSocket, and ECharts is used to render multi-dimensional charts, integrate alarm threshold configuration, display CPU / bandwidth / storage occupancy by sub-system block, and color gradient mark alarm threshold, such as red for CPU>90%, and then output an interactive cockpit interface for operation and maintenance personnel to master the global resource distribution and bottleneck in real time, analyze the global optimal business resource allocation strategy of the business resource coordination optimization module, extract the resource adjustment amount of each subsystem, and encapsulate the instruction template according to the subsystem protocol, add execution conditions, check the legality of the instructions through the rule engine (Drools), generate a timestamped instruction sequence, output a standardized instruction set including resource type, adjustment value, target subsystem ID and execution priority, expose API endpoints based on gRPC protocol, define input and output, wherein the input is the standardized instruction set and the output is the execution status code, and use the resource types and instruction formats supported by the subsystem declared when registering to automatically match the routing, at the same time, use the asynchronous message queue (Kafka) to buffer the instructions, and the subsystem pulls the execution results (success / failure reason) and updates the instruction state and triggers the compensation mechanism, i.e. 3 times of failure retry, and then output the system scheduling log, including the record instruction distribution time, subsystem response and final execution result, supporting audit and fault backtracking;

[0087] The specific process of the intelligent decision control module is:

[0088] A long connection is established through a WebSocket, and resource marker sequence table data is continuously pulled from each subsystem, including CPU / bandwidth / storage occupancy rate, remaining amount, and service priority queue information. An interactive cockpit interface is generated using ECharts, and resource occupancy rates such as CPU, bandwidth, and storage are displayed in blocks according to the subsystems. Warning thresholds are marked using color gradients, such as red for CPU>90%, yellow for 70%-90%, and green for <70%. Dynamic refreshing and interactive operations are supported. A visual configuration panel is provided, allowing operation and maintenance personnel to customize the warning thresholds for each resource type. After configuration, the thresholds take effect in real time and update the chart color markings. Based on the algorithm output of the business resource collaborative optimization module, the global optimal resource allocation strategy is analyzed, and the resource types (CPU / bandwidth / storage) and adjustment amounts that each subsystem needs to adjust are determined. According to the subsystem protocol specification, the resource adjustment amounts are packaged into standardized instruction templates, including resource type, adjustment value, target subsystem ID, execution priority, and execution conditions. The Drools rule engine is used to check the legality of the instructions, including whether the instructions meet the subsystem resource constraints and whether the execution conditions are met. Illegal instructions are filtered out, and a verification report is generated. Timestamps are added to the verified instructions, and a standardized instruction set is generated. The instructions are sorted by execution priority and marked as "to be distributed". The standardized instruction set includes unique instruction ID, resource type, adjustment value, target subsystem ID, execution priority, timestamp, execution condition, and other metadata, ensuring that the instructions are traceable and auditable. Service interfaces are defined based on the gRPC protocol, with input being the standardized instruction set and output being the execution status code (200 for success, 400 for parameter error, and 500 for system exception). The supported resource types and instruction formats declared by the subsystem during registration are used to automatically match the resource types in the instructions to the target subsystem, implement instruction routing and distribution, and avoid subsystem instantaneous overload by buffering instructions through a Kafka message queue. The subsystems pull instructions as needed and return execution results (success / failure reasons). The control module updates the instruction status to "completed" or "failed". For instructions that fail to execute, automatic retries are performed three times, with an increasing interval between each retry (1s / 2s / 4s). If the execution still fails, the instruction is marked as "terminated" and the failure reason is recorded. System scheduling logs are generated to record the entire life cycle information of the instructions, including instruction distribution time, subsystem response time, final execution result (success / failure times and reasons), compensation mechanism trigger record, and other information. A log query interface is provided to support retrieval based on instruction ID, subsystem ID, and time range, assisting operation and maintenance personnel in analyzing resource bottlenecks and locating fault sources. In addition, log data is integrated into the cockpit interface to display instruction execution success rate, failure rate, and high-frequency fault subsystems in the form of a timeline or heat map.

[0089] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A multi-service collaborative dispatching and control system for the smart campus Internet of Things based on an all-optical network, including an Internet of Things digital platform, characterized by: The IoT digital platform is connected to the following modules: All-optical network infrastructure module, used to build the campus fiber backbone network and access various subsystems on campus, providing full coverage network; The multi-service resource perception module is used to deploy multimodal sensors, obtain the basic data set for service resource scheduling, mark the limit of service resources, and generate a resource marking sequence table; Chaotic sequence initialization engine, based on the particle swarm algorithm, introduces a chaotic sequence to generate the initial particle swarm, replacing random initialization, so that particles are evenly distributed in the solution space; The business resource collaborative optimization module is used to combine the resource tag sequence table and the initial particle swarm to optimize the business resource allocation strategy and intelligently schedule the business needs of each subsystem on campus; The intelligent decision-making control module is used to provide a visual cockpit, display resource scheduling status, convert optimization results into executable instructions, and open API for cross-system linkage control.

2. The all-optical network-based smart campus IoT multi-service collaborative dispatching and control system according to claim 1 is characterized by: The all-optical network infrastructure module includes an optical fiber backbone network unit and an optical network management and control unit; The fiber optic backbone network unit uses passive optical network technology to directly connect to the subsystems on campus through optical fibers, while expanding the number of subsystems through optical splitters; The optical network management and control unit is used to deploy a software-defined network controller, allocate optical wavelength resources, and adjust transmission paths according to service priorities.

3. The all-optical network-based smart campus IoT multi-service collaborative dispatching and control system according to claim 2 is characterized by: The optical fiber backbone network unit comprises: Deploy passive optical network equipment in the campus central computer room and connect various subsystems on campus to the passive optical network equipment through optical fiber to achieve direct optical fiber connection; Use optical splitters to split optical fiber signals, dividing the signal of one optical fiber into multiple paths and extending the connection to each subsystem; The split signals are managed and transmitted through passive optical network equipment, and the data signals generated by each subsystem are aggregated to the Internet of Things digital platform in the campus central computer room for processing.

4. The all-optical network-based smart campus IoT multi-service collaborative dispatching and control system according to claim 2 is characterized by: The optical network management and control unit includes: Deploy a software-defined network controller in the campus central computer room, complete system initialization, and connect the controller to passive optical network devices to obtain network topology and device status information and build a network topology map; The software-defined network controller dynamically allocates optical wavelength resources based on the business needs of each subsystem, monitors network traffic in real time, and adjusts wavelength allocation as needed; The software-defined network controller adjusts the transmission path based on business priorities, and prioritizes low-latency paths for services with high preset real-time requirements.

5. The multi-service collaborative dispatching and control system for the smart campus Internet of Things based on the all-optical network according to claim 2 is characterized by: The multi-service resource perception module includes a resource monitoring unit and a resource marking unit; The resource monitoring unit is used to deploy multimodal sensors to monitor the usage of business resources of various subsystems on campus in real time, thereby obtaining a basic data set for business resource scheduling; The resource marking unit combines the basic data set of business resource scheduling and the pre-set business resource limits of each subsystem, marks the subsystems with increased demand and the corresponding business resources, and integrates them to generate a resource marking sequence table.

6. The multi-service collaborative dispatching and control system for the smart campus Internet of Things based on an all-optical network according to claim 5 is characterized by: The resource monitoring unit includes: Deploy multimodal sensors in key scenarios across the campus, including cameras, temperature and humidity sensors, and energy consumption monitors, to collect real-time data on the service resource usage of each subsystem. Multimodal sensors collect security monitoring images, energy consumption data, and business resource usage information of classroom environmental parameters in real time, and transmit the collected data to the resource monitoring unit through the Internet of Things network; The resource monitoring unit pre-processes the received multimodal data and stores the processed data in the database to form a basic data set for business resource scheduling.

7. The all-optical network-based smart campus IoT multi-service collaborative dispatching and control system according to claim 5 is characterized by: The resource marking unit includes: Load the latest 5-minute basic data set for business resource scheduling from the database, categorize it by subsystem, and analyze the current resource usage based on the pre-set business resource limits for each subsystem. This assesses the demand for business resources by each subsystem and identifies subsystems with increased demand and their corresponding resources. Mark subsystems with increased demand and their corresponding business resources, record their unique IDs and resource demand priorities, and generate marking information including subsystem ID, resource type, demand, and priority. Integrate the marked resource information, sort it by priority, obtain the current available amount of the campus resource pool, and comprehensively generate a resource marking sequence list.

8. The multi-service collaborative dispatching and control system for the smart campus Internet of Things based on an all-optical network according to claim 5 is characterized by: The chaotic sequence initialization engine includes: Select a chaotic model based on Logistic chaotic mapping, set the initial parameters and the number of iterations, calculate through the iterative formula of chaotic mapping, and generate a series of chaotic values; Based on the values ​​generated by the chaotic sequence, combined with the dimension of the problem and the size of the particle swarm, the chaotic values ​​are mapped to the solution space, and each particle is assigned a corresponding position and velocity, so that the particles are evenly distributed in the solution space, forming an initial particle swarm based on the chaotic sequence. The dimension of the problem represents the number of variables in the optimization problem. The constructed initial particle swarm is substituted into the particle swarm algorithm, and the position and velocity of the particles are updated according to the individual extreme value and the global extreme value according to the update rules of the traditional particle swarm algorithm.

9. The all-optical network-based smart campus Internet of Things multi-service collaborative dispatching and control system according to claim 8 is characterized by: The business resource collaborative optimization module includes: Extract the resource tag sequence table, analyze the resource type and real-time status of each subsystem, and identify the resource requirements of each subsystem, including subsystems with increased resource requirements and their corresponding resource types; A chaotic sequence is generated based on the resource dimension and mapped to the initial position and velocity of the particle swarm. The fitness function is defined as the weighted sum of resource utilization and business demand satisfaction. The particle positions are updated according to the traditional particle swarm algorithm rules. The global optimal solution is dynamically perturbed by the chaotic sequence until the termination condition is met, that is, the maximum number of iterations is met, and the global optimal business resource allocation strategy is generated. Business requests are classified according to urgency and resource requirement type. Based on the resource tag sequence table and the global optimal business resource allocation strategy, the requests are routed to subsystems with sufficient idle resources. Resource usage is monitored in real time. If a resource conflict occurs, the particle swarm algorithm is triggered to re-optimize and update the resource tag sequence table, thereby outputting a business request-resource allocation mapping table and a dynamic adjustment log.

10. The all-optical network-based smart campus Internet of Things multi-service collaborative dispatching and control system according to claim 9 is characterized by: The intelligent decision-making control module includes: Resource tag sequence table data, including CPU / bandwidth / storage utilization, remaining capacity, and service priority queues, is pulled in real time via WebSocket. ECharts is used to render multi-dimensional charts, integrating alarm threshold configuration. CPU / bandwidth / storage utilization is displayed in blocks by subsystem, and color gradients are used to mark alarm thresholds, thereby outputting an interactive cockpit interface. Analyze the global optimal business resource allocation strategy of the business resource collaborative optimization module, extract the resource adjustment amount of each subsystem, and encapsulate the instruction template according to the subsystem protocol, add execution conditions, verify the legitimacy of the instruction through the rule engine, generate a timestamp instruction sequence, and output a standardized instruction set including resource type, adjustment value, target subsystem ID, and execution priority; Based on the gRPC protocol, API endpoints are exposed and input and output are defined. The input is a standardized instruction set and the output is an execution status code. The resource types and instruction formats supported by the subsystem are declared when registering the subsystem to automatically match the route. At the same time, an asynchronous message queue is used to buffer instructions. After the subsystem pulls the instructions, it returns the execution results, updates the instruction status and triggers the compensation mechanism, and then outputs the system scheduling log, including recording the instruction distribution time, subsystem response and final execution results.

Images