Intelligent Scheduling Method, System, Device and Medium Based on Device Usage Frequency

Abstract

The present invention relates to the technical field of equipment scheduling, and discloses an intelligent scheduling method, system, device and medium based on equipment usage frequency, including: collecting the basic information of tasks and the relevant information of all devices to be used; respectively establishing a scheduling model for multiple tasks belonging to the same type and all devices to be used corresponding to tasks of this type, and optimizing the scheduling model to obtain an optimal scheduling plan; performing scheduling allocation of devices and tasks according to the optimal scheduling plan. The present invention utilizes intelligent algorithms to achieve reasonable scheduling of equipment resources, reduce equipment idle time and unnecessary waiting.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment scheduling, and particularly relates to an intelligent scheduling method, system, device and medium based on equipment usage frequency. Background Art

[0002] In equipment-intensive industries, medical and logistics scenarios, etc., the equipment scheduling efficiency directly affects the effect of task completion and economic costs. Traditional equipment scheduling methods rely on factors such as equipment availability and static priorities of tasks for scheduling. Such methods are designed based on static rules, lack real-time response capabilities, are difficult to adapt to frequently changing task requirements, and are also difficult to balance the actual workload of equipment.

[0003] In a complex equipment management environment, factors such as the usage frequency of different equipment, the urgency of tasks, and the maintenance status of equipment affect the overall utilization efficiency of equipment. Existing scheduling systems, due to relying on static preset rules and fixed priorities, often cannot be dynamically adjusted to adapt to real-time changing requirements. This has led to problems such as overuse of high-frequency equipment, idleness of low-frequency equipment, and delay of urgent tasks, directly affecting the service life of equipment and the task completion efficiency.

[0004] Existing equipment scheduling systems usually include the following methods:

[0005] Fixed priority scheduling: Scheduling equipment by setting static priorities, and giving priority to scheduling important equipment. However, static priorities cannot adapt to frequent task changes.

[0006] Round-robin scheduling: Adopting a fixed polling strategy to allocate equipment tasks. This scheme is not sensitive to task changes and cannot optimize high-frequency equipment.

[0007] Static load balancing: Based on the static load capacity of equipment for balanced allocation, but cannot dynamically schedule according to real-time usage conditions, resulting in resource waste.

[0008] In the above-mentioned prior art, there is a lack of effective combination of equipment frequency, priority and urgency, resulting in unreasonable utilization of equipment resources, and especially unable to achieve an optimized effect in frequently changing scenarios.

[0009] However, the above methods also have the following problems:

[0010] Insufficient dynamic response: Existing scheduling methods cannot respond to real-time task requirements in a timely manner, resulting in scheduling delays.

[0011] Waste of equipment resources: Lack of analysis of equipment usage frequency, resulting in idleness of low-frequency equipment and overloading of high-frequency equipment.

[0012] Insufficient handling of task priorities: Do not consider the urgency of tasks and cannot give priority to scheduling urgent tasks in a short time. Summary of the Invention

[0013] The present invention provides an intelligent scheduling method and system based on device usage frequency. Through real-time analysis of device usage frequency, task urgency, and device health status and intelligent scheduling algorithms, a rational allocation model of devices is provided to optimize device utilization rate and achieve efficient task allocation.

[0014] The present invention is realized through the following technical solutions:

[0015] An intelligent scheduling method based on device usage frequency includes:

[0016] Collect the basic information of the task and the relevant information of all devices to be used.

[0017] Establish a scheduling model for multiple tasks of the same type and all devices to be used corresponding to this type of task, and optimize the scheduling model to obtain an optimal scheduling plan.

[0018] Perform scheduling allocation of devices and tasks according to the optimal scheduling plan.

[0019] As an optimization, the basic information of the task includes the type of the task, the start time of the task, the remaining time, and the preset priority.

[0020] As an optimization, the relevant information of the devices to be used includes the health status of the devices, the task types corresponding to the devices, and the usage frequency of the devices.

[0021] As an optimization, the method for optimizing the scheduling model is to use a reinforcement learning algorithm to optimize the scheduling model. Among them, in the reinforcement learning algorithm, the state is the scheduling plan, the action is the corresponding switch of devices and tasks, and the reward is a reward value considering the health status, usage frequency, and preset priority.

[0022] As an optimization, the method for optimizing the scheduling model is to use a genetic algorithm to optimize the scheduling model. The specific process is as follows:

[0023] A1. Set the number of iterations of the genetic algorithm and the number of individuals to be solved in each generation of the population. Among them, one individual to be solved is a chromosome, and one chromosome includes a combination of different devices and different tasks of corresponding types. Among them, one gene locus of the chromosome is a combination of one device and one task, and the devices at different gene loci are different, and the tasks at different gene loci are different.

[0024] A2. Establish a scheduling model. The scheduling model includes an objective function and total constraint conditions. Among them, the objective function is:

[0025] ;

[0026] Among them, F is the objective function, is the number of tasks to be scheduled, , , are the weight factors of task urgency, equipment health status, and equipment usage frequency respectively, , , are the normalized task urgency, the health status of the equipment and the usage frequency of the equipment respectively;

[0027] A3. Encode the combination of the equipment and tasks to obtain the encoding value of the individual to be solved, and randomly generate an initial population composed of several individuals to be solved based on the total constraint conditions. Let this initial population be the parental population, and let the individuals to be solved in the parental population be parental individuals. The encoding value of each parental individual includes different equipment and different tasks belonging to the same type;

[0028] A4. Perform scheduling simulation according to the encoding values of the parental individuals in the parental population to obtain the simulation scheduling result;

[0029] A5. Calculate the fitness function according to the simulation scheduling result, and calculate and sort the fitness values of each parental individual in the parental population according to the simulation scheduling result;

[0030] A6. Save the first H parental individuals with the largest fitness values in the parental population. Select parental individuals by roulette wheel selection from all parental individuals other than the first H parental individuals with the largest fitness values for crossover and mutation operations to obtain offspring individuals. Then calculate the fitness values of the offspring individuals after crossover and mutation and sort them. Re-insert the offspring individuals into the parental population according to the fitness values, select a set number of individuals to be solved to form a new parental population, and then return to A4;

[0031] A7. Repeat A4 - A6 until the iteration number is reached or the objective function value converges. The parental individual with the largest fitness value in the finally obtained parental population is the optimal scheduling plan.

[0032] As an optimization, the total constraint conditions include general constraints, the first constraint when the number of equipment is greater than or equal to the number of tasks, and the second constraint when the number of equipment is less than the number of tasks. Among them,

[0033] The general constraint is:

[0034] Each task must be assigned to a device, and the execution time of the tasks assigned to the device does not exceed the remaining execution time of the device;

[0035] The specific first constraint is:

[0036] Each task must be assigned a device, and ;

[0037] Among them, represents the status value for judging whether the device is assigned, represents the device is assigned to the task , represents the device is not assigned, is the number of devices, represents all the devices corresponding to the tasks of this type;

[0038] The specific second constraint is:

[0039] Sort the task urgencies from high to low, and obtain the top m tasks as the tasks for the genetic algorithm.

[0040] As an optimization, during the device scheduling process, when sudden tasks or device failures occur, the genetic algorithm is used to dynamically adjust the scheduling plan again, and a new scheduling plan that meets the set conditions is taken for device scheduling. The set conditions are:

[0041] ;

[0042] Among them, represents the optimal scheduling plan after readjustment through the genetic algorithm after sudden tasks or device failures occur, represents the optimal scheduling plan before sudden tasks or device failures occur.

[0043] The present invention also discloses an intelligent scheduling system based on device usage frequency for executing the foregoing intelligent scheduling method based on device usage frequency, including:

[0044] A data collection module for the basic information of tasks and the relevant information of all devices to be used;

[0045] A device usage frequency analysis module for calculating the usage frequency of devices through historical data analysis, identifying high-frequency devices with a usage frequency higher than the set threshold and low-frequency devices with a usage frequency lower than the set threshold, and providing data support for scheduling decisions;

[0046] A task urgency evaluation module, which is used to evaluate the task urgency in real time according to the remaining time and priority of the task, so as to ensure that high-priority tasks are processed first;

[0047] An intelligent scheduling optimization module, which is used to combine the equipment usage frequency and task urgency, and use an intelligent optimization algorithm to realize the dynamic allocation of equipment to obtain an optimal scheduling plan;

[0048] A scheduling execution module, which is used to transmit the optimal scheduling plan to the equipment control system to ensure that the equipment can execute tasks according to the optimal scheduling plan.

[0049] The present invention also discloses an electronic device, including at least one processor and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute an intelligent scheduling method based on equipment usage frequency as described above.

[0050] The present invention also discloses a storage medium storing a computer program, and when the computer program is executed by a processor, it implements the intelligent scheduling method based on equipment usage frequency as described above.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] The present invention dynamically and real-time responds to the equipment usage frequency and task urgency, and significantly improves the equipment utilization rate on the premise of meeting the task requirements. At the same time, it improves the flexibility of equipment scheduling.

[0053] The present invention uses an intelligent algorithm to realize the reasonable scheduling of equipment resources and reduce equipment idle time and unnecessary waiting. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not limit the embodiments of the present invention. In the drawings:

[0055] Figure 1 is a flowchart of an intelligent scheduling method based on equipment usage frequency according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and the drawings. The illustrative embodiments and descriptions thereof of the present invention are only used to explain the present invention and do not limit the present invention.

[0057] In order to solve the problems in the background art, the present invention mainly realizes the following functions:

[0058] 1. Dynamically monitor the device usage frequency, flexibly schedule according to real-time task requirements, and be able to dynamically optimize device allocation based on multi-dimensional factors such as device usage frequency, task urgency, and device status.

[0059] 2. Adjust the scheduling priority according to the urgency of the task, reduce the delay of urgent tasks, improve device utilization rate, reduce device idle rate and delay, and achieve intelligent allocation of device resources.

[0060] 3. Adopt machine learning and optimization algorithms to improve the intelligence level of scheduling decisions, and improve device utilization rate and task completion efficiency. By optimizing the scheduling algorithm, ensure that the task scheduling of the device can respond to demand changes in real time and meet the task priorities in different scenarios.

[0061] Combine the historical usage frequency of the device, task type and priority to provide a more intelligent scheduling model, improve the device utilization efficiency, and extend the service life of the device. Embodiment 1

[0062] Provide an intelligent scheduling method based on device usage frequency, as Figure 1 shown, including:

[0063] S1. Collect the basic information of the task and the relevant information of all devices to be used.

[0064] S2. Establish a scheduling model for multiple tasks of the same type and all devices to be used corresponding to this type of task respectively, and optimize the scheduling model to obtain the optimal scheduling plan.

[0065] S3. Perform scheduling allocation of devices and tasks according to the optimal scheduling plan.

[0066] In this embodiment, the basic information of the task includes the type of the task, the start time of the task, the remaining time, and the preset priority.

[0067] In this embodiment, the relevant information of the device to be used includes the health status of the device, the task type corresponding to the device, and the usage frequency of the device.

[0068] In this embodiment, the method for optimizing the scheduling model is to optimize the scheduling model using a reinforcement learning algorithm. Among them, in the reinforcement learning algorithm, the state is the scheduling plan, the action is the corresponding switch of the device and the task, and the reward is the reward value considering the health status, usage frequency, and preset priority.

[0069] It should be noted that there may be multiple types of tasks, and each task will have multiple devices. Therefore, in the case of multiple tasks of each type and corresponding multiple devices, the present invention separately sets up a scheduling model for each task type and conducts optimized training.

[0070] In this embodiment, a reinforcement learning algorithm is used to train the scheduling model.

[0071] In reinforcement learning, ;

[0072] wherein, represents the value of executing action a in state s, is the learning rate, is the current immediate reward, which can be , , , are the weight factors of task urgency, device health status, and device usage frequency respectively, , , are the urgency of the normalized task , the health status of the device , and the usage frequency of the device respectively; is the discount factor, is the new state after executing action a, is the optimal action in the new state . The state s represents the current situation of the environment or the current configuration of the system, that is, the scheduling scheme. Embodiment 2

[0073] Different from Embodiment 1, the optimized scheduling model is implemented using a genetic algorithm.

[0074] The specific process is as follows:

[0075] A1. Set the number of iterations of the genetic algorithm and the number of individuals to be solved in each generation of the population. Among them, one individual to be solved is a chromosome, and one chromosome includes a combination of different devices and different tasks of corresponding types. Among them, one gene position of the chromosome is a combination of one device and one task, and the devices of different gene positions are different, and the tasks of different gene positions are different;

[0076] That is to say, for 5 tasks of the same type, assuming there are 5 devices to be scheduled, which are , , , and , and the 5 tasks are respectively , , , and , a chromosome can be: , and another chromosome can be: .

[0077] A2. Establish a scheduling model, where the scheduling model includes an objective function and total constraint conditions. Among them, the objective function is:

[0078] ;

[0079] Among them, F is the objective function, is the number of tasks to be scheduled, , , are the weight factors of task urgency, equipment health status, and equipment usage frequency respectively, which are used to balance different influencing factors (task urgency, equipment health status, equipment usage frequency), , , are the urgency of the normalized task , the health status of the equipment , and the usage frequency of the equipment respectively;

[0080] The core goal of equipment scheduling is to allocate limited equipment resources to task requirements, and optimize the equipment utilization efficiency, task response time, and balance of resource allocation as much as possible. Therefore, the above objective function is designed.

[0081] A3. Encode the combination of the equipment and tasks to obtain the encoded value of the individual to be solved, and randomly generate an initial population formed by several individuals to be solved based on the total constraint conditions. Let this initial population be the parental population, and let the individuals to be solved in the parental population be parental individuals. The encoded value of each parental individual includes different equipment and different tasks belonging to the same type;

[0082] A4. Perform scheduling simulation according to the encoded values of the parental individuals in the parental population to obtain the simulated scheduling result;

[0083] A5. Calculate the fitness function according to the simulated scheduling result, and calculate and sort the fitness values of each parental individual in the parental population according to the simulated scheduling result;

[0084] A6. Save the top H parental individuals with the largest fitness values in the parental population. Select parental individuals from all parental individuals except the top H parental individuals with the largest fitness values through roulette wheel selection for crossover and mutation operations to obtain offspring individuals. Then calculate the fitness values of the offspring individuals after crossover and mutation and sort them, re-insert the offspring individuals into the parental population according to the fitness values, select a set number of individuals to be solved to form a new parental population, and then return to A4; H is a positive integer and H is less than the number of individuals in the population.

[0085] A7. Repeat A4 - A6 until the iteration number is reached or the objective function value converges. The parental individual with the largest fitness value in the finally obtained parental population is the optimal scheduling scheme.

[0086] When the fitness value of the genetic algorithm reaches the convergence condition, output the optimal scheduling scheme, including: the device allocation for each task. At the same time, output the queuing order of unassigned tasks and the status of unused devices.

[0087] In the intelligent scheduling system based on device usage frequency, it is a common practical scenario that the number of devices is inconsistent with the number of tasks. In response to this situation, the present invention needs to set appropriate constraint conditions and strategies in the scheduling optimization algorithm to ensure the rationality and effectiveness of the scheduling scheme. The following analyzes possible scenarios in detail and gives solutions.

[0088] Scenario analysis:

[0089] When the number of devices > the number of tasks:

[0090] Scenario description: The number of devices to be scheduled is more than the number of tasks to be processed currently. For example, there are 10 available devices, but only 5 new tasks need to be assigned.

[0091] Potential problem: Some devices may be idle, resulting in waste of resources. How to select the most suitable devices to execute tasks is the key to optimization.

[0092] When the number of devices < the number of tasks:

[0093] Scenario description: The current number of tasks is more than the number of available devices. For example, there are 10 devices, but 15 tasks need to be processed.

[0094] Potential problem: Tasks may queue up or be executed with delay, which may affect the timeliness of tasks. How to efficiently allocate devices and reasonably arrange task priorities is the focus of optimization.

[0095] When the number of devices is equal to the number of tasks:

[0096] Scenario description: The number of devices is equal to the number of tasks. Direct allocation is usually the optimal solution.

[0097] Potential problem: If the device status is uneven (such as differences in health), it is necessary to optimize the matching relationship between devices and tasks.

[0098] Solution strategy:

[0099] Scheduling constraint conditions:

[0100] According to different scenarios (the first scenario and the second scenario), the following scheduling constraint conditions are designed according to the scenario characteristics:

[0101] The number of devices ≥ the number of tasks:

[0102] The optimization goal is to select the device with the optimal status (based on usage frequency, health status, etc.) to assign tasks, and the unused devices enter the standby state.

[0103] The number of devices < the number of tasks:

[0104] The optimization goal is to give priority to assigning tasks with a high degree of urgency, and non-urgent tasks enter the waiting queue or are rescheduled.

[0105] Scheduling optimization algorithm adjustment: To cope with the scenario where the number of devices and tasks is inconsistent, an adaptive adjustment mechanism is introduced into the genetic algorithm:

[0106] Expression of constraint conditions: Let the device set be E = {e1, e2, …, e m}, and the task set be T = {t1, t2, …, t n}, where m and n are the number of devices and tasks respectively, then:

[0107] When m ≥ n: Each task must be assigned a device.

[0108] When m < n: Priority is given to satisfying the assignment of high-priority tasks, and unassigned tasks enter the queue.

[0109] In summary, the total constraint conditions include general constraints, the first constraint when the number of devices is greater than the number of tasks, and the second constraint when the number of devices is less than the number of tasks, where,

[0110] The general constraint is:

[0111] Each task must be assigned to a device, and the execution time of the task assigned to the device does not exceed the remaining execution time of the device, and the total scheduling time is minimized;

[0112] The specific content of the first constraint is:

[0113] Each task must be assigned a device, ;

[0114] Among them, represents the judgment device The status value that has been assigned, indicating the device has been assigned to the task , indicating the device has not been assigned to the task , is the number of devices, indicating all the devices corresponding to the tasks of this type;

[0115] indicating the task is only assigned to one of the m devices. The unassigned devices are set to the standby state to improve the utilization rate of system resources.

[0116] The second constraint is specifically:

[0117] Sort the task urgency from high to low, and obtain the top m tasks as the tasks for the genetic algorithm.

[0118] Sorted by task urgency, if , and , divide the tasks into k batches, each batch containing at most m tasks. First, assign the m tasks with the highest urgency, and then, after these m tasks are completed, sequentially assign the next m tasks with high urgency, and so on.

[0119] To cope with sudden tasks or equipment failures, the system introduces a dynamic adjustment mechanism:

[0120] Reorder the tasks and add sudden tasks;

[0121] Use mutation operations to randomly adjust device allocation to improve adaptability to sudden situations.

[0122] That is to say, during the process of device scheduling, when a sudden task or equipment failure occurs, re - dynamically adjust the scheduling plan through the genetic algorithm, and take a new scheduling plan that meets the set conditions for device scheduling. The set conditions are:

[0123] ;

[0124] Among them, represents the optimal scheduling plan after re - adjustment through the genetic algorithm after a sudden task or equipment failure, represents the optimal scheduling plan before a sudden task or equipment failure.

[0125] ; <(

[0126] and respectively represent the change in task urgency and the change in equipment usage frequency before and after readjustment.

[0127] In summary, in this embodiment, by introducing a genetic algorithm, dynamic constraint adjustment, and real-time optimization mechanism, an intelligent and efficient equipment scheduling system is realized, which can adapt to complex scenarios where the number of equipment and tasks is inconsistent, and has significant technological innovation and practical application value. The specific advantages are as follows:

[0128] Dynamic adaptability:

[0129] For scenarios where the number of equipment and tasks is inconsistent, the present invention adapts to actual application requirements through dynamic constraint adjustment and scheduling optimization algorithms, avoiding the efficiency reduction caused by quantity mismatch in traditional systems.

[0130] Intelligent scheduling strategy:

[0131] The genetic algorithm combines multi-dimensional factors such as equipment health status and task urgency, and improves the accuracy of the scheduling scheme and resource utilization rate by dynamically adjusting and optimizing the fitness function.

[0132] Real-time feedback and adjustment:

[0133] The present invention has the ability of real-time response, and can quickly regenerate a scheduling scheme according to emergencies (such as new tasks, equipment failures) to ensure task continuity.

[0134] From the perspective of the optimization logic, urgent tasks are preferentially assigned to low-load healthy equipment to increase the F value and promote the exploration of optimal solutions. Assume the task urgency is [8, 6, 4], the equipment load rate is [0.2, 0.6, 0.8], and the health degree is [0.9, 0.7, 0.6]. Reasonably assign task 1 to equipment 1, task 2 to equipment 2, and task 3 to equipment 3. Compared with randomly assigning task 1 to equipment 3, etc., the matching of urgent tasks and low-load equipment is better, the average urgency decreases, the load balance increases, the health risk is controlled, the F value is high, guiding the algorithm to evolve towards optimization, driving the precise adaptation of tasks and equipment, and the overall optimization of the system. Embodiment 3

[0135] Disclosed is an intelligent scheduling system based on equipment usage frequency for implementing the intelligent scheduling method based on equipment usage frequency described in Embodiment 1 or Embodiment 2, including:

[0136] A data acquisition module for the basic information of tasks and relevant information of all equipment to be used;

[0137] The main responsibility of the data acquisition module is to collect the current state of the equipment and task requirement data in real time. The data obtained by this module from equipment sensors and management platforms includes:

[0138] Device usage frequency data: Record the number of times each device is used within a specific time window to generate usage frequency records.

[0139] Device health status data: Record the operating conditions of the device, including health indicators such as temperature, vibration, and energy consumption.

[0140] Task request data: Capture the basic information of the task, including parameters such as the start time, preset priority, and remaining time of the task.

[0141] Logical relationship: The data acquisition module is connected to the device sensors through a multi-sensor data interface to obtain the status parameters of the device. All data is first preliminarily screened and preprocessed by the edge computing node and then transmitted to the cloud for use by subsequent modules.

[0142] The device usage frequency analysis module is used to calculate the usage frequency of the device through historical data analysis, identify high-frequency devices with a usage frequency higher than the set threshold and low-frequency devices with a usage frequency lower than the set threshold, and provide data support for scheduling decisions;

[0143] This module analyzes the usage frequency of the device within a specific time period to identify high-frequency and low-frequency used devices, thus providing a reference for scheduling optimization.

[0144] The device usage frequency is used to measure the workload and usage trend of the device. Frequency analysis is performed based on the historical usage data of the device to obtain the frequency index Fd of the device. The specific analysis process is as follows:

[0145] Historical data extraction: Extract the usage records of the device in the past period from the database.

[0146] Usage frequency calculation: According to the usage records of the device, calculate the usage frequency , the formula is:

[0147] ;

[0148] T represents the time window, represents the number of times the device is used within the time window T, represents the usage frequency of the device.

[0149] Among them, T is the set time window period (such as one week, one month, etc.). The usage trend of the device is obtained through periodic frequency calculation, thus providing a reference for scheduling optimization.

[0150] Usage frequency update: The system will regularly update the usage frequency information of each device to ensure the timeliness of the data.

[0151] The analysis of device usage frequency enables the system to identify devices with high loads, so as to preferentially use low-frequency devices in scheduling, avoiding resource waste and overload.

[0152] A task urgency evaluation module, which is used to evaluate the task urgency in real time according to the remaining time and priority of the task, ensuring that high-priority tasks are processed first;

[0153] The task urgency evaluation module mainly calculates the urgency of the task in real time based on the remaining time and priority of the task. The urgency calculation process is as follows:

[0154] Obtaining task priority: Read the priority information of the task, which is used to measure the importance of the task.

[0155] Calculating the remaining time: Calculate the remaining time from the start time and the preset completion time of the task .

[0156] Urgency calculation formula:

[0157] ;

[0158] where represents the urgency value of the task. Tasks with higher urgency will be assigned higher priorities to ensure that urgent tasks can be processed in a timely manner.

[0159] After the urgency calculation is completed, the tasks are sorted according to the urgency, and the system preferentially assigns high-urgency tasks to ensure the rationality of task scheduling.

[0160] An intelligent scheduling optimization module, which is used to combine device usage frequency and task urgency, and use intelligent optimization algorithms to achieve dynamic allocation of devices to obtain the optimal scheduling scheme;

[0161] The intelligent scheduling optimization module is the core of the system. By combining device usage frequency, device health status and task urgency, it uses reinforcement learning algorithms or genetic algorithms to dynamically optimize device allocation.

[0162] In each iteration, the genetic algorithm selects the optimal individuals based on the fitness value and performs crossover and mutation operations until the fitness converges to the best scheduling scheme.

[0163] The specific steps for scheduling optimization using the genetic algorithm are as follows:

[0164] Initializing the scheduling model: Define the initial population, and each individual is an allocation combination of devices and tasks.

[0165] Fitness function setting: The fitness function is used to evaluate the quality of the scheduling combination. To improve the scheduling efficiency, the present invention uses a genetic algorithm to optimize the scheduling strategy. This algorithm is based on parameters such as device usage frequency, task urgency, and device health status, and generates an optimal device allocation plan through operations such as selection, crossover, and mutation in the genetic algorithm. The fitness function for scheduling optimization is designed as follows:

[0166] ;

[0167] The fitness function aims to ensure that high-urgency tasks are assigned to low-load devices, and the health status of the devices meets the task execution requirements.

[0168] Selection, crossover, and mutation: The scheduling combination is optimized through the selection, crossover, and mutation operations of the genetic algorithm to generate a new scheduling plan. The specific implementation process is as follows:

[0169] Selection: Select the scheduling plan according to the fitness value to ensure that high-quality plans are preferentially retained.

[0170] Crossover: Perform a crossover operation on the selected plans to generate new plans, making the combination of devices and tasks more diverse.

[0171] Mutation: Randomly mutate some combinations of devices and tasks to enhance the system's adaptability to sudden tasks.

[0172] Scheduling result output: After multiple iterations, when the fitness value reaches the preset convergence condition, the optimal scheduling plan is output. This plan can maximize the device utilization efficiency and meet the real-time requirements of tasks at the same time.

[0173] Scheduling adjustment: During the actual execution process, if there are sudden tasks or device failures, the system can automatically start an emergency adjustment process to regenerate the scheduling plan to ensure task continuity.

[0174] A scheduling execution module is used to transmit the optimal scheduling plan to the device control system to ensure that the device can execute tasks according to the optimal scheduling plan.

[0175] The task of the scheduling execution module is to apply the optimized scheduling plan to the actual device operation. The specific process is as follows:

[0176] Scheduling plan transmission: Transmit the optimization result to the device control system to ensure that the device executes tasks according to the system instructions;

[0177] Task tracking and feedback: Real-time track the device execution situation to ensure the continuity and correctness of the scheduling process;

[0178] Fault feedback and scheduling correction: During the execution of the device, if a device fault or a change in task requirements occurs, the system automatically feeds back to the intelligent scheduling module to re-optimize the scheduling to ensure the smooth completion of the scheduling task.

[0179] The design of the scheduling execution module realizes the rapid application and real-time adjustment of the scheduling plan, ensuring the stable operation of the entire scheduling system.

[0180] Explanation of action relationships and process logic:

[0181] 1. Data collection and preprocessing: The data collection module first collects data and transmits it to the analysis module after preprocessing. This stage provides reliable real-time data support for device status evaluation.

[0182] 2. Analysis of device usage frequency and task urgency: The system first calculates the device usage frequency based on historical data and then evaluates the urgency of the task. These data are the basis for intelligent scheduling optimization.

[0183] 3. Execution of intelligent scheduling optimization: By comprehensively considering the device usage frequency and task urgency through the genetic algorithm, the dynamic allocation of devices is realized. The intelligent scheduling optimization module plays a core control role in the system.

[0184] 4. Execution and monitoring feedback of the scheduling plan: The scheduling execution module transmits the optimized plan to the device control system, and at the same time monitors the device status in real time to ensure the smooth implementation of the scheduling result and provide feedback data for the next round of scheduling.

[0185] This intelligent scheduling system realizes the reasonable allocation of device resources and the priority control of tasks through frequency analysis and urgency evaluation, and dynamically adapts to environmental changes. Through the genetic algorithm, the system can automatically generate the optimal scheduling plan in a complex environment to ensure the efficient operation of the device. Embodiment 4

[0186] An electronic device is disclosed, which is characterized by including at least one processor and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute an intelligent scheduling method based on device usage frequency as described in Embodiment 1 or 2. Embodiment 5

[0187] A storage medium is disclosed, storing a computer program, and when the computer program is executed by a processor, it implements an intelligent scheduling method based on device usage frequency as described in Embodiment 1 or 2.

[0188] The specific embodiments described above further elaborate on the objective, technical solution and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent scheduling method based on device usage frequency, characterized in that, Including: Collecting the basic information of the acquisition tasks and the relevant information of all devices to be used; Establishing a scheduling model for multiple tasks of the same type and all devices to be used corresponding to the tasks of this type respectively, and optimizing the scheduling model to obtain an optimal scheduling plan; The specific process is as follows: A1. Set the number of iterations of the genetic algorithm and the number of individuals to be solved in each generation of the population. Among them, one individual to be solved is a chromosome, and one chromosome includes a combination of different devices and different tasks of corresponding types. Among them, one gene position of the chromosome is a combination of one device and one task, and the devices of different gene positions are different, and the tasks of different gene positions are different; A2. Establish a scheduling model, and the scheduling model includes an objective function and total constraint conditions. Among them, the objective function is: ； Among them, F is the objective function, is the number of tasks to be scheduled, , , are the weight factors of task urgency, equipment health status, and equipment usage frequency respectively, , , are the urgency of the task after normalization, the health status of the equipment , and the usage frequency of the equipment respectively; A3. Encode the combination of the devices and tasks to obtain the encoded value of the individual to be solved, and randomly generate an initial population formed by several individuals to be solved based on the total constraint conditions. Let this initial population be the parental population, and let the individuals to be solved in the parental population be parental individuals. The encoded value of each parental individual includes different devices and different tasks of the same type; A4. Perform scheduling simulation according to the encoded values of the parental individuals in the parental population to obtain a simulated scheduling result; A5. Calculate the fitness function according to the simulated scheduling result, and calculate the fitness value of each parental individual in the parental population according to the simulated scheduling result and sort them; A6. Save the first H parental individuals with the largest fitness values in the parental population. Select parental individuals by roulette wheel selection from all parental individuals except the first H parental individuals with the largest fitness values to perform crossover and mutation operations to obtain offspring individuals, and then calculate the fitness values of the offspring individuals after crossover and mutation and sort them. Reinsert the offspring individuals into the parental population according to the fitness values, select a set number of individuals to be solved to form a new parental population, and then return to A4; A7. Repeat A4 - A6 until the number of iterations is reached or the objective function value converges. The parental individual with the largest fitness value in the finally obtained parental population is the optimal scheduling plan; Perform scheduling allocation of devices and tasks according to the optimal scheduling plan.

2. The intelligent scheduling method based on device usage frequency according to claim 1, wherein The basic information of the tasks includes the type of the task, the start time of the task, the remaining time, and the preset priority.

3. The intelligent scheduling method based on equipment usage frequency according to claim 1, characterized in that: The relevant information of the devices to be used includes the health status of the devices, the task types corresponding to the devices, and the usage frequency of the devices.

4. An intelligent scheduling method based on device usage frequency according to claim 1, characterized in that The method for optimizing the scheduling model is to use a reinforcement learning algorithm to optimize the scheduling model. Among them, in the reinforcement learning algorithm, the state is the scheduling plan, the action is the corresponding switching of the device and the task, and the reward is the reward value considering the health status, usage frequency, and preset priority.

5. The intelligent scheduling method based on equipment usage frequency according to claim 1, characterized in that: The total constraint conditions include general constraints, the first constraint when the number of devices is greater than or equal to the number of tasks, and the second constraint when the number of devices is less than the number of tasks. Among them, The general constraint is: Each task must be assigned to a device, and the execution time of the tasks assigned to the device shall not exceed the remaining execution time of the device; The specific first constraint is: Each task must be assigned a device, and ; Among them, represents the status value of whether the judgment device is assigned, represents the device is assigned to the task , represents the device is not assigned, is the number of devices, represents all devices corresponding to the tasks of this type; The specific second constraint is: Sort the task urgencies from high to low, and obtain the top m tasks as the tasks for the genetic algorithm.

6. The intelligent scheduling method based on equipment usage frequency according to claim 1, characterized in that: During the process of device scheduling, when there are sudden tasks or device failures, the genetic algorithm is used to dynamically adjust the scheduling plan again, and a new scheduling plan that meets the set conditions is selected for device scheduling. The set conditions are: ； in, It represents the optimal scheduling solution after re-adjustment through genetic algorithm after an emergency task or equipment failure occurs. It represents the optimal scheduling plan before an emergency task or equipment failure occurs.

7. An intelligent scheduling system based on device usage frequency, for implementing an intelligent scheduling method based on device usage frequency according to any one of claims 1-6, characterized in that, Including: A data acquisition module, which is used for the basic information of tasks and the relevant information of all devices to be used; A device usage frequency analysis module, which is used to calculate the usage frequency of devices through historical data analysis, identify high-frequency devices with a usage frequency higher than the set threshold and low-frequency devices with a usage frequency lower than the set threshold, and provide data support for scheduling decisions; A task urgency evaluation module, which is used to evaluate the task urgency in real time according to the remaining time and priority of the task to ensure that high-priority tasks are processed first; An intelligent scheduling optimization module, which is used to combine the device usage frequency and task urgency, and use intelligent optimization algorithms to achieve dynamic allocation of devices to obtain the optimal scheduling plan; A scheduling execution module, which is used to transmit the optimal scheduling plan to the device control system to ensure that the device can execute tasks according to the optimal scheduling plan.

8. An electronic device, characterized in that: Including at least one processor and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute an intelligent scheduling method based on device usage frequency according to any one of claims 1 to 6.

9. A storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements an intelligent scheduling method based on device usage frequency according to any one of claims 1 to 6.

Images