Multi-screen centralized synchronous operation system and method and electronic equipment
An adaptive task scheduling mechanism combining deep learning and particle swarm optimization algorithms solves the problems of response lag and uneven resource allocation in high-load scenarios for multi-screen operating systems, enabling intelligent and collaborative operation of multi-screen systems and improving task scheduling accuracy and remote management efficiency.
Patent Information
- Application Number
- CN202511662004.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-13
- Publication Date
- 2026-02-10
AI Technical Summary
Existing multi-screen operating systems lack the ability to dynamically perceive and intelligently respond to the running status in high-concurrency, highly interactive real-time applications, resulting in uneven resource utilization, increased response latency, and insufficient remote management and anomaly response capabilities, making it difficult to adapt to complex application environments with frequent changes in task load and constant fluctuations in terminal status.
A deep learning module is used to predict screen states through convolutional neural networks, and a particle swarm optimization algorithm is combined to optimize task execution order and resource allocation. An adaptive task scheduling module is introduced to dynamically adjust priorities and allocation strategies, and a system management module enables remote control and monitoring, forming a multi-screen centralized synchronous operating system.
It enables sensitive response to real-time changes and dynamic optimization of task priorities, improves the flexibility and stability of task scheduling, enhances the controllability of remote management and the convenience of operation and maintenance, improves the ability to automatically adjust to abnormal states, and promotes the development of multi-screen systems towards intelligence and collaboration.
Smart Images
Figure CN121501232A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of multi-screen scheduling technology, and in particular to a multi-screen centralized synchronization operating system, method and electronic device. Background Technology
[0002] With the widespread adoption of smart terminal devices and the rapid growth in demand for multimedia information processing, multi-screen collaborative display systems have been widely deployed in application scenarios such as command and dispatch, security monitoring, smart conferencing, and commercial presentations. Especially in high-concurrency, highly interactive real-time applications, achieving information synchronization, task collaboration, and resource sharing among multiple screens has become a crucial design objective. However, existing multi-screen operating systems largely rely on static configurations and preset rules, lacking dynamic perception and intelligent response capabilities regarding operational status, making it difficult to adapt to complex application environments with frequently changing task loads and fluctuating terminal states.
[0003] In existing technologies, multi-screen management solutions typically perform initial configuration and status presentation through a centralized control platform or software. However, they lack a mechanism for deep integration with the real-time status of the screens at the task scheduling level. Most systems use fixed priorities or simple rules for task allocation, making it difficult to automatically adjust task strategies based on real-time changes in terminal status. This leads to uneven resource utilization, increased response latency, and even partial task processing failures. Especially in scenarios involving multiple screens executing complex tasks in parallel, traditional scheduling solutions have significant limitations in task distribution and load balancing.
[0004] Meanwhile, while existing systems possess some centralized control capabilities, they largely rely on manual intervention, lacking sufficient remote management and anomaly response capabilities. When a screen exhibits an abnormal state (such as abnormal brightness or load spikes), the system lacks state-triggered intelligent control logic, often requiring manual intervention from administrators, resulting in delayed and inefficient processing. Furthermore, the perception of multi-terminal operating status is mostly achieved through periodic polling, lacking a real-time response mechanism for sudden state changes, thus impacting the overall system's collaborative operating efficiency.
[0005] In terms of task execution visualization, traditional systems focus more on presenting task results, lacking full-process monitoring and feedback of the execution process, making it difficult to evaluate and optimize task scheduling behavior. The correspondence between task execution status and screen physical status has not been effectively modeled and utilized, further limiting the development space of intelligent scheduling strategies for the system. Summary of the Invention
[0006] The purpose of this invention is to provide a multi-screen centralized synchronization operating system, method, and electronic device, which solves the problems of unintelligent multi-screen task scheduling, untimely status response, and insufficient remote centralized management capabilities in the prior art.
[0007] To achieve the above objectives, the present invention provides the following technical solution: a multi-screen centralized synchronous operating system, comprising: The deep learning module is used to predict screen status based on historical operation data and real-time feedback from multiple screens, and generate task scheduling suggestions. The optimization algorithm module is used to optimize the task execution order and resource allocation based on task scheduling suggestions, screen status, and task requirements, and generate scheduling results. The real-time data stream module is used to collect screen status data in real time and obtain the current screen status; The adaptive task scheduling module is used to dynamically adjust task priorities and allocation strategies based on real-time collected screen status data and scheduling results. The system management module is used to uniformly configure, remotely control and monitor the screen status based on real-time collected screen status data, and display the task execution status.
[0008] Preferably, the deep learning module predicts the screen state using a convolutional neural network, and the prediction model of the convolutional neural network satisfies: ; in, Indicates time The set of status parameters of the screen at that time, including brightness Resolution ,load ; The input feature set includes user operation instructions. Screen resource utilization Environmental parameters ; This is the prediction function output by the convolutional neural network.
[0009] Preferably, the optimization algorithm module uses particle swarm optimization to optimize the task execution order. The optimization objective of the particle swarm optimization algorithm is to minimize the total task execution time while satisfying preset constraints. ; in, This represents the total execution time of all tasks; Indicates the first The execution time of each task; This represents the total number of tasks.
[0010] Preferably, the matching constraint between the task resource requirements and screen resource capacity under the preset constraints satisfies: ; in, Indicates the first The task in the first The amount of resources used per screen; Indicates the first Maximum resource capacity per screen; Total number of screens; This represents the total number of tasks.
[0011] Preferably, the step of dynamically adjusting task priority and allocation strategy includes: Receive the scheduling results generated by the optimization algorithm module and the real-time screen status data provided by the real-time data stream module. ,in Indicates time The actual screen state; Calculate the screen state change based on measured data from adjacent time points. Based on the scheduling results and state change Through formula ; Update task priority, where: Set the initial task priority; The adjusted task priority; The adaptive weighting coefficients satisfy the following conditions: .
[0012] Preferably, the adaptive weighting coefficient The calculation formula is: ; in, Indicates the first Real-time load of each screen; This represents the average load across all screens; This represents the total number of connected screens in the system.
[0013] Preferably, the remote control command generation rule of the system management module is as follows: ; in, For remote control commands; For the screen In time Real-time brightness; This is the preset brightness threshold.
[0014] This invention also provides a method for centralized synchronization of multiple screens, comprising the following steps: Based on historical operation data and real-time feedback from multiple screens, a deep learning module is used to predict screen states and generate task scheduling suggestions. Based on the task scheduling suggestions and considering the screen status and task requirements, the optimization algorithm module is used to optimize the task execution order and resource allocation, and a scheduling result is generated. Real-time acquisition of screen status data; the current screen status is obtained through the real-time data stream module. Based on real-time collected screen status data and combined with scheduling results, the adaptive task scheduling module dynamically adjusts task priorities and allocation strategies. Based on real-time collected screen status data, the system can uniformly configure, remotely control and monitor screen status, display task execution status, and use the system management module for remote control.
[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, is capable of implementing the methods described above.
[0016] The present invention also provides a storage medium having a computer program stored thereon, which, when executed by a processor, can implement the above-described method.
[0017] In summary, the present invention has at least one of the following beneficial technical effects: 1. This invention employs an adaptive task scheduling mechanism and a screen state change-linked adjustment strategy, achieving the technical effect of sensitive response and dynamic optimization of task priorities based on real-time changes. Compared to existing technologies with fixed priority configurations or simple rule-driven scheduling strategies, it solves the problems of response lag, task congestion, and uneven resource allocation in multi-screen, high-load scenarios, significantly enhancing the system's task scheduling flexibility and stability.
[0018] 2. This invention employs an integrated architecture design with a system management module for unified monitoring and remote control of the operating status of multiple screens, achieving centralized management and unified configuration of distributed terminals. Compared to existing solutions that rely on independent control terminals for each screen, this invention solves the problems of difficulty in unifying command responses and low efficiency in status feedback across multiple terminals, thereby improving system controllability and ease of operation and maintenance.
[0019] 3. This invention introduces a brightness threshold-driven remote control triggering mechanism to achieve automatic adjustment and control of the screen under abnormal conditions, thus achieving the technical effect of intelligent resource response. Compared with the control methods in the prior art that rely solely on manual intervention or timed strategies, this invention solves the problems of untimely remote adjustment response and difficulty in identifying sudden states, thereby improving the system's intelligent management and control capabilities in dynamic environments.
[0020] 4. This invention combines real-time screen status data with scheduling results in a coupled modeling approach, achieving a deep fusion of status awareness and task allocation behavior, thus improving the accuracy of task scheduling. Compared to existing technologies that rely solely on scheduling models or static state parameters for control logic, this invention solves the problems of low data utilization and poor system adaptability, promoting the development of multi-screen systems towards intelligence and collaboration. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a system architecture diagram of the present invention; Figure 3 This is a schematic diagram of the computer device structure of the present invention.
[0022] Among them, 10 is computer equipment; 11 is processor; 12 is memory; and 13 is storage medium. Detailed Implementation
[0023] The following is in conjunction with the appendix Figure 1 -Appendix Figure 3 The present invention will be further described in detail below.
[0024] This invention provides a multi-screen centralized synchronization operating system, method, and electronic device. like Figure 1 As shown, this multi-screen centralized synchronization operating system includes: The deep learning module is used to predict screen status based on historical operation data and real-time feedback from multiple screens, and generate task scheduling suggestions. The deep learning module learns from historical operation data and real-time feedback from multiple screens to predict screen states and generate task scheduling suggestions. Its core function is to model screen states using convolutional neural networks (CNNs) to accurately predict future screen states and provide valuable predictive information for subsequent task scheduling. Beyond predicting screen states, this module also provides a basis for task scheduling to the optimization algorithm module, further optimizing task execution order and resource allocation.
[0025] In this embodiment, the deep learning module predicts the screen state using a convolutional neural network. Specifically, the prediction model of this network can be expressed as the following formula: ; in, Indicates time The set of status parameters of the screen at that time, including brightness Resolution ,load These parameters reflect the current working status of the screen and are important bases for subsequent task scheduling and resource optimization.
[0026] The input feature set contains the following important components: User operation instructions : Indicates the user's time On-screen commands, such as user-initiated tasks and interactions.
[0027] Screen resource utilization This indicates the current resource usage of the screen, including the utilization of computing resources, display resources, etc.
[0028] Environmental parameters This includes external environmental factors that affect the screen's status, such as temperature, humidity, and network bandwidth.
[0029] function This is the prediction function output by the convolutional neural network. Through this function, the network predicts the next time step based on the current state and historical operation information. The function trains the convolutional layer based on historical data and real-time feedback, gradually adjusting the weights and learning the patterns of screen state changes to obtain more accurate prediction results.
[0030] Typically, deep learning modules are trained using supervised learning. Training data includes a large amount of historical screen state data and its corresponding user operation commands, screen resource utilization, and environmental parameters. Using this data, the convolutional neural network can learn the relationship between various input features and the screen state, thereby predicting the screen state at future moments.
[0031] Alternatively, a convolutional neural network (CNN) architecture can include multiple convolutional and pooling layers, enabling the extraction and dimensionality reduction of spatial features of the screen state. Through repeated training and optimization, the network can adaptively learn the optimal combination of parameters, thereby improving prediction accuracy.
[0032] Specifically, in this embodiment, the training process of the convolutional neural network can be optimized using the backpropagation algorithm. During each training process, the network adjusts its weights based on the error between the predicted and actual results, ultimately improving the prediction function. Accurate modeling of screen state changes is achieved. The trained convolutional neural network can not only predict the current screen state, but also provide high-precision predictions of the state at future moments, thereby supporting subsequent task scheduling decisions.
[0033] In one possible implementation, to further improve prediction accuracy, the deep learning module can also employ transfer learning or online learning techniques, depending on the needs of the actual application. Transfer learning can quickly adapt to different screen state characteristics and user operation methods by transferring pre-trained network models to new multi-screen systems, thereby reducing training time and improving prediction efficiency.
[0034] Based on the output of the convolutional neural network, the prediction model provides... This will serve as one of the input data for subsequent task scheduling. This will be achieved by predicting the screen state. Combined with other modules (such as the optimization algorithm module), the system can rationally arrange the task execution order according to the future screen state and optimize resource allocation, thereby ensuring the efficient execution of tasks on multiple screens.
[0035] Furthermore, the predictions from deep learning modules can help the system identify potential resource bottlenecks or task conflicts in advance. For example, if the load on a certain screen increases at a future time... If the forecast is too high, the system can adjust the priority of the task in advance or allocate other resources to avoid performance degradation or task delays caused by excessive load.
[0036] In summary, the deep learning module utilizes convolutional neural networks to predict screen states and generate task scheduling suggestions, providing support for subsequent optimization algorithm modules and adaptive task scheduling modules. By accurately predicting screen states and learning the relationship between screen states and task execution, the deep learning module plays a crucial role in multi-screen centralized synchronous operating systems, providing strong technical support for the system's efficient operation and flexible scheduling.
[0037] The optimization algorithm module is used to optimize the task execution order and resource allocation based on task scheduling suggestions, screen status, and task requirements, and generate scheduling results. The main task of the optimization algorithm module is to optimize the execution order and resource allocation of tasks based on the task scheduling suggestions generated by the aforementioned deep learning module, combined with the current state of the screen and the resource requirements of the tasks, thereby generating the final task scheduling result. By employing the Particle Swarm Optimization (PSO) algorithm, the optimization algorithm module can minimize the total execution time of tasks while considering task execution time and resource constraints, ensuring the efficiency and rational resource utilization of the multi-screen system when executing tasks.
[0038] In this embodiment, the optimization algorithm module uses the particle swarm optimization algorithm to optimize the task execution order. Specifically, the optimization objective of the particle swarm optimization algorithm is to minimize the total execution time of all tasks, i.e.: ; in, This represents the total execution time of all tasks; Indicates the first The execution time of each task; Let be the total number of tasks. The optimization algorithm adjusts the execution order of tasks to minimize the execution time of all tasks while satisfying screen resource constraints, thereby improving the overall efficiency of the system.
[0039] Specifically, the Particle Swarm Optimization (PSO) algorithm simulates the process of particles in a swarm searching for the optimal solution in the search space. Each particle represents a solution for a task execution order, and the particle's fitness function is the total task execution time. Through multiple iterations, the PSO algorithm gradually updates the particle's velocity and position, eventually converging to the optimal solution. The particle update formula is as follows: ; ; in, It is a particle At any moment speed; It is a particle At any moment The position (i.e., the task execution order or the state of the solution); It is the inertial weight, which controls the influence of the particle's own velocity; and It is the acceleration constant, which controls the particle's ability to learn from its own historical experience and the global best experience, respectively. and It is a random number between 0 and 1, which increases the randomness of particle search and ensures a balance between global and local search; It is a particle The individual's best position, that is, the best solution that the particle has found so far; It is the globally optimal position, that is, the optimal solution found among all particles.
[0040] In this optimization process, each iteration considers the resource requirements of the task and the resource capacity of the screen to ensure that the maximum capacity of the screen resources is not exceeded when optimizing the task execution order.
[0041] As an alternative, the particle swarm optimization algorithm can further optimize task scheduling results by introducing constraints. Specifically, the matching constraint between task resource requirements and screen resource capacity needs to meet the following conditions: ; in, Indicates the first The task in the first The amount of resources used per screen; Indicates the first Maximum resource capacity per screen; Total number of screens; This represents the total number of tasks.
[0042] Generally, by using such constraints, optimization algorithms can not only minimize the total execution time of tasks, but also ensure the rational use of resources and avoid resource overload or idleness, thereby achieving balance and stability in task scheduling.
[0043] In one possible implementation, the particle swarm optimization algorithm can also be combined with other optimization algorithms (such as genetic algorithms and simulated annealing) to further improve the solution performance. By combining with other algorithms, the system can adapt to more complex task scheduling scenarios and provide more accurate scheduling results.
[0044] Through the particle swarm optimization algorithm, the optimization module can further optimize the task execution order and resource allocation scheme based on the task scheduling suggestions generated by the deep learning module, combined with the current state of each screen and the resource requirements of each task. In this process, the module aims to minimize the total task execution time, adjusting the task allocation method while meeting resource constraints, thereby obtaining a more reasonable scheduling result.
[0045] The real-time data stream module is used to collect screen status data in real time and obtain the current screen status; To ensure the real-time performance and accuracy of task scheduling, it is necessary to acquire the screen's state information during operation. This state information, as input features, is not only directly used for state prediction in the deep learning module but also forms a crucial foundation for the optimization algorithm module to formulate resource allocation strategies. Therefore, to achieve dynamic perception of the system state, a real-time data stream module needs to be set up to collect and process various operating state parameters in the multi-screen environment in real time, thereby obtaining the current screen state.
[0046] In this embodiment, the real-time data stream module includes a data acquisition unit, a status parsing unit, and a data synchronization unit, which are used to realize real-time monitoring and transmission of the operating parameters of each screen.
[0047] Specifically, the data acquisition unit of the real-time data stream module is deployed on each physical screen node, and is mainly responsible for collecting screen operation data including but not limited to the following parameters: Screen brightness : Indicates the current luminous intensity of the screen, which can be expressed in nits, and is used to reflect the visual output status of the screen; resolution This includes the pixel matrix size of the screen, typically expressed as... ,in The width in pixels. The height is the number of pixels; Load level : Indicates the computational load of the tasks currently being handled by the screen, expressed in units such as CPU percentage, memory usage, or image rendering frame rate; Input signal status Used to indicate whether the current screen is receiving user commands or control signals; Network bandwidth This reflects the current communication bandwidth status of the node where the screen is located. Temperature parameters Used to detect the operating temperature of equipment and reflect changes in the external environment or hardware status.
[0048] The above parameters together form the screen's state vector. It can be represented as: ; Typically, the data acquisition unit acquires data by calling underlying operating system APIs or connecting to hardware sensor interfaces. Alternatively, to improve the accuracy and consistency of data acquisition, the module can also incorporate edge computing processing mechanisms, performing preprocessing and standardization on local nodes before uploading the data to a centralized management server for fusion and analysis.
[0049] The state resolution unit is used to convert raw collected data into structured state information and sort it by timestamp. The data is labeled for use by subsequent modules. This unit can cache state data in a temporary data buffer and extract near-temporal state change sequences using a sliding window method for subsequent convolutional neural network modeling and prediction.
[0050] In one possible implementation, the data synchronization unit is responsible for centrally aggregating and aligning the data collected from multiple screen nodes. Due to geographical or communication latency differences between screens, this unit uses a global timestamp synchronization mechanism. This unifies data from different screens onto a standard timeline to meet the timing consistency requirements of deep learning modules.
[0051] In addition, to ensure the integrity and real-time nature of the status information, the module may also be equipped with an anomaly monitoring mechanism. When collected data is missing, delays exceed a threshold, or parameters fluctuate abnormally, the system can automatically issue an anomaly flag signal. This is used to prompt the optimization module to adjust the scheduling strategy appropriately.
[0052] In some embodiments, to further reduce latency and improve system response speed, the real-time data stream module can implement parallel acquisition and transmission based on a multi-threaded mechanism and adopt zero-copy transfer technology to reduce communication overhead.
[0053] The real-time data stream module collects and processes screen state data in real time to generate a vector of current screen state parameters. This provides accurate input for the adaptive task scheduling module and the system management module.
[0054] The adaptive task scheduling module is used to dynamically adjust task priorities and allocation strategies based on real-time collected screen status data and scheduling results. The module dynamically adjusts task priorities and resource allocation strategies. It operates based on the predictions from the deep learning module and the initial scheduling scheme from the optimization algorithm module, and incorporates real-time screen status information collected by the real-time data stream module. This further refines and corrects task scheduling behavior to adapt to changes in the multi-screen operating environment.
[0055] In this embodiment, the adaptive task scheduling module receives the scheduling result generated by the optimization algorithm module. It also acquires screen status data from the real-time data stream module in real time. ,in Indicates the current time.
[0056] Under normal circumstances, the system operates at a fixed interval. Perform a state update. Calculate the state change between the current time step and the previous time step: ; The system can accurately capture fluctuations in screen status. (Status change amount) It can be in vector form, including, for example, load variations. Brightness variation Bandwidth changes Multiple dimensions are used to reflect real-time changes in screen usage status.
[0057] After obtaining the aforementioned state changes, the system bases its decisions on the scheduling results. and state change Combined with the previously set initial task priorities Update the task priority using the following formula: ; in: Set the initial task priority; The adjusted task priority; This item indicates the impact of screen state changes on priority. The adaptive weighting coefficients satisfy the following conditions: .
[0058] Adaptive weight coefficients The selection of weighting coefficients is crucial, directly affecting the sensitivity and stability of priority adjustments. To achieve a reasonable response to system load, the weighting coefficients... Calculated dynamically using the following formula: ; in, Indicates the first Real-time load of each screen; This represents the average load across all screens; This represents the total number of connected screens in the system.
[0059] The above formula indicates that the system determines the aggressiveness of scheduling adjustments based on the variance of screen load. When the load fluctuations between screens are large, As the load increases, the adjustment range for priority also increases accordingly to quickly adapt to sudden changes in local states. When the load distribution tends to be balanced, If the value is reduced, the system will maintain the current strategy to enhance scheduling stability.
[0060] Specifically, during task execution, if the load on a screen node assigned to a task increases significantly, the system can automatically increase the processing priority of the task using the above formula, or reschedule it to a node with a lower load, thereby achieving task migration and load balancing.
[0061] As an option, the adaptive task scheduling module can also set a task migration threshold. This is used to control the sensitivity of scheduling behavior. For example, only when the relative change in screen load exceeds [a certain threshold]. Only when necessary will the system trigger a reassignment or priority adjustment, avoiding unnecessary system jitter caused by frequent adjustments.
[0062] In one possible implementation, the module can also introduce a feedback learning mechanism to record the execution effect after each priority adjustment, continuously optimize the weight coefficient calculation method or adjustment strategy, and achieve a certain degree of task scheduling self-learning and evolution.
[0063] In summary, the adaptive task scheduling module described in this embodiment dynamically adjusts task priorities and allocation strategies during system operation. Based on the fusion analysis of real-time status feedback and predetermined scheduling results, combined with an adaptive weight control mechanism, it achieves dynamic balance and stable operation of the multi-screen scheduling system, and is a key component in the system's intelligent scheduling logic.
[0064] The system management module is used to uniformly configure, remotely control and monitor the screen status based on real-time collected screen status data, and display the task execution status.
[0065] The system management module is responsible for uniformly configuring, remotely controlling, and monitoring the working status of each screen based on real-time collected screen status data. Furthermore, the system management module is used to display task execution status, ensuring information transparency and real-time feedback during task scheduling. This module works closely with the real-time data stream module, optimization algorithm module, and adaptive task scheduling module to ensure more efficient and flexible task scheduling and resource allocation in multi-screen environments.
[0066] In this embodiment, the system management module collects screen status data in real time. And combined with the task execution status The system management platform displays the current working status and task execution progress of each screen. Through a centralized monitoring interface, operators can view the real-time operating status of each screen, including information on multiple dimensions such as brightness, load, resolution, and network status. To enable remote control of the screens, the system management module sets rules for generating remote control commands based on screen status.
[0067] Specifically, the rules for generating remote control commands are as follows: ; in, This is a remote control command. A value of 1 indicates that remote control is triggered, and a value of 0 indicates that remote control is not triggered. For the screen In time The real-time brightness reflects the display status of the screen; A preset brightness threshold is set. When the screen brightness exceeds this threshold, a remote control command is triggered.
[0068] The basic idea behind this control rule is: when the brightness of a certain screen... Reaching or exceeding a preset threshold At that time, the system will automatically trigger remote control commands. The system allows for remote control adjustments to the screen's state. For example, it might adjust brightness, turn off the screen, or change the screen's operating mode to optimize resource utilization or save energy.
[0069] In some embodiments, the system management module can also dynamically adjust the generation rules of remote control commands based on other real-time status information of the screen (such as load, temperature, etc.). For example, when the screen load exceeds a certain preset threshold, the system can reduce the burden on the screen through corresponding control commands to avoid overload. Through real-time monitoring and dynamic control, the system can effectively improve the stability and operating efficiency of multi-screen systems.
[0070] As an alternative, the system management module can also combine task execution progress, screen resource utilization, and environmental changes (such as temperature changes) to comprehensively determine whether to trigger a remote control command. Specifically, if the brightness of a screen exceeds a threshold for an extended period... Furthermore, given its high load, the system can comprehensively consider factors such as resource allocation and task priority to decide whether to remotely migrate tasks to other screens or adjust the display mode of the screen, thereby optimizing resource allocation.
[0071] In one possible implementation, the system management module can further optimize the remote control strategy using artificial intelligence algorithms. By analyzing historical data and patterns in screen state changes, the system can gradually learn and predict which screens are more prone to overload or failure under specific conditions, thereby taking preventative measures in advance to avoid system instability.
[0072] In summary, the system management module achieves dynamic monitoring and management of the screen through a remote control mechanism based on real-time screen status data, combined with preset brightness threshold rules. This module not only displays task execution status in real time but also responds quickly to real-time status changes, ensuring efficient utilization of system resources and smooth task execution.
[0073] Please see the appendix Figure 2 The present invention also provides a multi-screen centralized synchronization operation method including the following steps: S1. Based on historical operation data and real-time feedback from multiple screens, use a deep learning module to predict screen states and generate task scheduling suggestions; S2. Based on the task scheduling suggestions and the screen status and task requirements, use the optimization algorithm module to optimize the task execution order and resource allocation, and generate scheduling results; S3. Real-time acquisition of screen status data, obtaining the current screen status through the real-time data stream module; S4. Based on the real-time collected screen status data and combined with the scheduling results, use the adaptive task scheduling module to dynamically adjust task priority and allocation strategy. S5. Based on real-time collected screen status data, uniformly configure, remotely control and monitor screen status, display task execution status, and use the system management module for remote control.
[0074] The method in this embodiment can be used to execute the above system embodiment, and its principle and technical effect are similar, so it will not be described again here.
[0075] Please see the appendix Figure 3The present invention also provides a computer device 10, including a processor 11 and a memory 12, wherein the memory 12 stores a computer program executable by the processor, and the computer program performs the method described above when executed by the processor.
[0076] The present invention also provides a storage medium 13 on which a computer program is stored, and the computer program is executed by a processor 11 to perform the method described above.
[0077] The storage medium 13 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, disk, or optical.
[0078] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A multi-screen centralized synchronous operating system, characterized in that, include: The deep learning module is used to predict screen states based on historical operation data and real-time feedback from multiple screens, and to generate task scheduling suggestions. The optimization algorithm module is used to optimize the task execution order and resource allocation based on task scheduling suggestions, screen status, and task requirements, and generate scheduling results. The real-time data stream module is used to collect screen status data in real time and obtain the current screen status; The adaptive task scheduling module is used to dynamically adjust task priorities and allocation strategies based on real-time collected screen status data and scheduling results. The system management module is used to uniformly configure, remotely control and monitor the screen status based on real-time collected screen status data, and display the task execution status.
2. The multi-screen centralized synchronization operating system according to claim 1, characterized in that, The deep learning module predicts the screen state using a convolutional neural network, and the prediction model of the convolutional neural network satisfies: ； in, Indicates time The set of status parameters of the screen at that time, including brightness Resolution ,load ; The input feature set includes user operation instructions. Screen resource utilization Environmental parameters ; This is the prediction function output by the convolutional neural network.
3. The multi-screen centralized synchronization operating system according to claim 1, characterized in that, The optimization algorithm module uses particle swarm optimization to optimize the task execution order. The optimization objective of the particle swarm optimization algorithm is to minimize the total task execution time while satisfying preset constraints. ； in, This represents the total execution time of all tasks; Indicates the first The execution time of each task; This represents the total number of tasks.
4. The multi-screen centralized synchronization operating system according to claim 3, characterized in that, The matching constraints between task resource requirements and screen resource capacity under the preset constraints are satisfied as follows: ； in, Indicates the first The task in the first The amount of resources used per screen; Indicates the first Maximum resource capacity per screen; Total number of screens; This represents the total number of tasks.
5. The multi-screen centralized synchronization operating system according to claim 1, characterized in that, The steps for dynamically adjusting task priorities and allocation strategies include: Receive the scheduling results generated by the optimization algorithm module and the real-time screen status data provided by the real-time data stream module. ,in Indicates time The actual screen state; Calculate the screen state change based on measured data from adjacent time points. Based on the scheduling results and state change Through formula ； Update task priority, where: Set the initial task priority; The adjusted task priority; The adaptive weight coefficients satisfy the following conditions: .
6. The multi-screen centralized synchronization operating system according to claim 5, characterized in that, The adaptive weighting coefficient The calculation formula is: ； in, Indicates the first Real-time load of each screen; This represents the average load across all screens; This represents the total number of connected screens in the system.
7. The multi-screen centralized synchronization operating system according to claim 1, characterized in that, The remote control command generation rules of the system management module are as follows: ； in, For remote control commands; For the screen In time Real-time brightness; This is the preset brightness threshold.
8. A multi-screen centralized synchronization operation method, applied to the multi-screen centralized synchronization operating system as described in any one of claims 1-7, characterized in that, Includes the following steps: Based on historical operation data and real-time feedback from multiple screens, a deep learning module is used to predict screen states and generate task scheduling suggestions. Based on the task scheduling suggestions and considering the screen status and task requirements, the optimization algorithm module is used to optimize the task execution order and resource allocation, and a scheduling result is generated. Real-time acquisition of screen status data; the current screen status is obtained through the real-time data stream module. Based on real-time collected screen status data and combined with scheduling results, the adaptive task scheduling module dynamically adjusts task priorities and allocation strategies. Based on real-time collected screen status data, the system can uniformly configure, remotely control and monitor screen status, display task execution status, and use the system management module for remote control.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.
10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.