Batch equipment upgrading scheduling system and method based on strategy management

Abstract

The invention discloses a batch equipment upgrading scheduling system and method based on strategy management, and relates to the technical field of industrial Internet of Things. The coverage range of equipment to be upgraded is divided into a plurality of sub-regions, historical data of gray-scale and non-gray-scale upgrading equipment are collected respectively to construct an analysis set, and the equipment is accurately analyzed to predict the upgrading trend; the scientificity of batch equipment upgrading scheduling is improved; a machine learning algorithm is utilized to simulate ideal state upgrading data, two types of analysis sets are combined to perform disassembly calculation on historical data, an upgrading data fluctuation value is obtained, a fluctuation deviation curve is constructed, a time sequence prediction model and a neural network model are combined to derive standard upgrading data, and upgrading prediction accuracy and strategy adaptability are improved; a three-level cache architecture is constructed based on a cloud platform and edge computing, standard upgrade data is analyzed, the total resource amount is determined, a cache pool is initialized, efficient transmission and safe storage of the upgrade data are achieved, and the resource utilization efficiency and the system operation stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial Internet of Things, and in particular to a batch equipment upgrade scheduling system and method based on policy management. Background Art

[0002] Driven by digital transformation, the number of devices in the Internet of Things, Industrial Internet, and intelligent systems is rapidly increasing. Mass device upgrades are crucial for maintaining efficient and stable system operation and driving technological innovation. Mass device upgrades enable timely deployment of new features to meet increasingly complex business needs; rapid security patch updates effectively defend against new cyber threats, ensuring data security and system reliability; and unified optimization of device performance to improve overall resource utilization efficiency. Furthermore, the large-scale operation of mass upgrades significantly reduces operation and maintenance costs and mitigates the risk of manual intervention, providing solid support for the intelligent development of smart cities, intelligent manufacturing, and other fields, helping businesses and society accelerate their transition to a high-quality digital and intelligent development stage.

[0003] Traditional batch device upgrade methods lack systematic policy management and are unable to meet the needs of complex scenarios. During the upgrade process, they often face difficulties in ensuring device compatibility and lack of scientific resource scheduling planning, which can easily lead to serious consequences such as device failures, interruptions to business continuity, and even loss of critical data. Therefore, a policy-based batch device upgrade scheduling method is urgently needed. Summary of the Invention

[0004] The purpose of the present invention is to provide a batch equipment upgrade scheduling system and method based on policy management to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a batch device upgrade scheduling method based on policy management, the batch device upgrade scheduling method comprising the following steps: Step S1: Divide the device range covered by the devices to be upgraded into multiple sub-areas, obtain the upgrade data of the devices with historical grayscale upgrades in each sub-area to construct a device upgrade compatibility analysis set, and obtain the historical upgrade data of the devices without grayscale upgrades in each sub-area to construct a device upgrade analysis set; Step S1-1: Divide the area by the number of devices to be upgraded, and evenly distribute the total amount of devices to be upgraded to each sub-area; Step S1-2: Build a data collection platform and obtain historical upgrade data for each grayscale upgrade device from the sub-region through an API interface. The grayscale upgrade device represents a portion of the devices that have been upgraded first, selected from all the devices to be upgraded. The historical upgrade data represents the upgrade time, failure rate, resource usage, and corresponding upgrade timestamp data generated by the grayscale upgrade device during the initial upgrade process. Step S1-3: construct a device upgrade compatibility analysis set based on the acquired grayscale upgrade device historical upgrade data as a value and the grayscale upgrade device unique identifier as an index; Step S1-4: construct a device upgrade analysis set based on the acquired historical upgrade data of non-grayscale upgrade devices as a value and the unique identifier of the non-grayscale upgrade device as an index. The non-grayscale upgrade device is represented as a device that has not been screened for prior upgrade and directly participates in subsequent full upgrades or historical regular upgrades.

[0006] When evenly dividing the area by the number of devices to be upgraded, clustering algorithms such as the DBSCAN algorithm, based on the geographic information system (GIS) and network topology, are used to analyze the spatial and network proximity of the devices. While ensuring a balanced number of devices in each sub-area, devices with adjacent physical locations and close network connections are prioritized for placement in the same sub-area. This avoids data transmission delays and inefficient resource scheduling caused by dispersed device distribution, thereby improving the overall efficiency of subsequent data collection and upgrade scheduling. By evenly dividing the sub-areas of equipment to be upgraded by quantity and collecting historical data of grayscale and non-grayscale upgraded equipment to build analysis sets, we can achieve classified integration and orderly management of equipment upgrade data, providing a comprehensive and structured data foundation for subsequent accurate analysis of equipment upgrade compatibility, prediction of upgrade data fluctuation trends, and optimization of upgrade resource scheduling. This effectively avoids analysis bias caused by data mixing and improves the scientificity and accuracy of batch equipment upgrade scheduling.

[0007] Step S2: Analyze the device upgrade compatibility analysis set to obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device in the sub-region; analyze the device upgrade analysis set to obtain the standard upgrade data fluctuation value of the device to be upgraded in the sub-region; Step S2-1: Use a machine learning algorithm to simulate and obtain upgrade time data, failure rate data, and resource usage data of the device to be upgraded under ideal conditions; Step S2-2: traverse and read the upgrade data in the device upgrade compatibility analysis set, use the upgrade time data, failure rate data, and resource usage data under ideal conditions, subtract the upgrade time data, failure rate data, and resource usage data read through traversal, and obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device through normalization processing combined with weighted fusion calculation; The data difference and normalization calculations use the following formula: ; Where, Expressed as the normalized fluctuation value of the i-th data; It represents the i-th upgrade data under ideal conditions, and the upgrade data includes upgrade time data, failure rate data, and resource usage data; Represents the i-th item of historical data of the grayscale upgrade device; The weighted fusion calculation uses the following formula: ; Where W gary It represents the grayscale upgrade data fluctuation value of the grayscale upgrade device during the upgrade; w i Expressed as the weight coefficient of the i-th upgrade data, satisfying ; n represents the dimension of the upgraded data, n=3; Step S2-2-1: Decompose the upgrade data in the device upgrade compatibility analysis set to obtain upgrade time consumption data, failure rate data, and resource usage data corresponding to the grayscale upgrade device upgrade timestamp; Step S2-2-2: After subtracting the upgrade time data, failure rate data, and resource usage data, perform absolute value processing on the difference, and normalize the data after absolute value processing; Step S2-2-3: After normalizing all data, the grayscale upgrade data fluctuation value of the grayscale upgrade device is obtained through weighted fusion calculation; Step S2-3: Synchronously use the upgrade data fluctuation value calculation method of step S2-2 to process the upgrade data in the device upgrade analysis set to obtain the grayscale upgrade data fluctuation value of the non-grayscale upgrade device.

[0008] By simulating ideal upgrade data through machine learning algorithms and based on the device upgrade compatibility analysis set and the device upgrade analysis set, the historical data of grayscale and non-grayscale upgrade devices are disassembled, difference calculated, normalized and weighted fused to accurately obtain the upgrade data fluctuation values of the two types of devices, providing a quantitative basis for subsequent fluctuation deviation analysis and prediction of upgrade trends. It can effectively capture the difference between actual performance and ideal status during the device upgrade process, thereby laying the foundation for formulating batch device upgrade strategies that are more in line with actual needs.

[0009] Step S3: Perform a fluctuation deviation analysis using the grayscale upgrade data fluctuation value and the standard upgrade data fluctuation value to obtain a fluctuation deviation curve for the sub-region. This fluctuation deviation curve analysis is performed by analyzing the upgrade data of each grayscale upgrade of the sub-region in history. Combined with the real-time upgrade data of the devices undergoing grayscale upgrades in the sub-region, the standard upgrade data for the devices to be upgraded in the sub-region is obtained. Step S3-1, controlling the grayscale fluctuation deviation curve and the non-grayscale fluctuation deviation curve to perform time alignment through the timestamp data; Step S3-2: Subtract the standard upgrade data fluctuation value from the grayscale upgrade data fluctuation value corresponding to the timestamp data to obtain the fluctuation deviation value between the single upgrade of the grayscale upgrade device in the sub-region and the single upgrade of the non-grayscale upgrade device; Step S3-3: Perform time sequence analysis based on the timestamp data to obtain the fluctuation difference between the grayscale upgrade devices and the non-grayscale upgrade devices in the sub-region during the upgrade; Step S3-4: construct a fluctuation deviation curve after arranging in time sequence, wherein the horizontal axis of the fluctuation deviation curve is the timestamp data corresponding to the single upgrade of the sub-region grayscale upgrade device and the upgrade of the non-grayscale upgrade device, and the vertical axis of the fluctuation deviation curve is the fluctuation difference value corresponding to the upgrade; Step S3-5: Input the fluctuation gap value and timestamp data of the fluctuation deviation curve into the time series prediction model for training, and obtain the upgrade fluctuation trend of the devices to be upgraded and the grayscale upgrade devices in the sub-region, which is recorded as the upgrade fluctuation prediction trend; Step S3-6: Obtain the grayscale upgrade data fluctuation value of the grayscale upgrade devices in the sub-area, and obtain the upgrade data fluctuation value of the non-grayscale upgrade devices in the sub-area based on the upgrade fluctuation prediction trend; Step S3-7: Using the fluctuation gap value and timestamp data of the fluctuation deviation curve as input data, and combining the historical upgrade time data, failure rate data, and resource usage data in the device upgrade analysis set as target variables, the neural network model is trained to obtain a mapping relationship between the upgrade data fluctuation value of the device to be upgraded and the upgrade data; Step S3-8: Using the upgrade data fluctuation value obtained in step S3-6, the upgrade data of the sub-region is obtained through a mapping relationship and recorded as standard upgrade data.

[0010] By aligning the fluctuation values of grayscale and non-grayscale upgrade devices in time series, calculating the difference and arranging them in time series, constructing a fluctuation deviation curve and combining the time series prediction model with the neural network model, we can achieve upgrade fluctuation trend prediction and mapping conversion between fluctuation values and upgrade data, accurately derive standard upgrade data for devices to be upgraded, provide a scientific basis for subsequent resource scheduling and cache management, and effectively improve the prediction accuracy and strategy adaptability of batch device upgrades.

[0011] Step S4: Create buffer zones for each sub-area based on the cloud platform's managed cache, schedule resources in the cache zone using standard upgrade data, deploy lightweight cache services on edge nodes using distributed cache technology and the edge computing platform, and build and manage data cache pools using edge computing protocols combined with standard upgrade data. Step S4-1: Through the global resource view of the cloud platform hosted cache, a three-level cache architecture is constructed according to the network topology of the sub-region: the first level is the cloud center-level distributed cache cluster, which is specifically a high-performance distributed cache system deployed in the cloud data center, responsible for storing global device upgrade metadata, load balancing policy configuration between sub-regions, and providing data preheating and disaster recovery support for edge nodes; the second level is the sub-region edge node cache layer, which is specifically a lightweight cache service deployed on the regional edge computing node close to the terminal device, responsible for storing the upgrade data packets accessed in this region and the device upgrade priority queue for real-time calculation; the third level is the device-side local cache, which is specifically a micro-memory database embedded in the terminal of the device to be upgraded, responsible for temporarily storing the upgrade data fragments related to this device, the intermediate state information during the upgrade process, and the digital signature authentication information generated based on the device's unique identifier; Step S4-2: Analyze the standard upgrade data of the device to be upgraded obtained through the output of step S3, and extract resource occupancy data as basic resource occupancy data of the device to be upgraded; Step S4-3: derive the total amount of resources required for device upgrade in the sub-region by detecting the product of the number of devices to be upgraded in the current sub-region and the basic resource occupancy data; The total amount of sub-region resources is calculated using the following formula: ; Where R total It is represented by the total amount of resources required to upgrade the devices to be upgraded in the sub-area; N is represented by the total number of devices to be upgraded in the sub-area; r d Expressed as the basic resource usage of device d; c d It is represented as the concurrent upgrade coefficient of device d, usually 1; Step S4-4: constructing a data cache pool based on the total amount of sub-region resources obtained in step S4-3, and initializing the total capacity of the data cache pool; Step S4-5: Fill the data cache pool of each sub-area with data through the cloud center-level distributed cache cluster. The edge node cache layer pushes the cached upgrade data packet to the terminal local cache of the corresponding device to be upgraded according to the upgrade priority queue of the device to be upgraded. After verifying the data integrity through the digital signature authentication information in the local cache, the upgrade operation is performed. The intermediate status information during the upgrade process is stored in the local cache in real time. After the device upgrade is completed, the result is fed back to the edge node and cloud center, triggering the release of cache resources and status update.

[0012] Based on the cloud platform and edge computing, a three-level cache architecture and data cache pool management system are constructed. By analyzing the standard upgrade data, the total amount of resources is determined and the cache pool is initialized, realizing layered collaborative caching of the cloud center, edge nodes and device ends; the cloud center-level cache coordinates the global metadata and load balancing strategy, the edge node caches the storage area data packets and schedules the priority, and the local cache on the device end ensures data security and upgrade continuity. The three work together to realize the efficient transmission, precise scheduling and secure storage of upgrade data, effectively improving the resource utilization efficiency and system stability during the batch equipment upgrade process.

[0013] Furthermore, a batch device upgrade scheduling system based on policy management includes a region division module, a fluctuation analysis module, a cache construction module and an upgrade execution module; The area division module is used to divide the coverage area of the device to be upgraded into sub-areas and collect and construct an upgrade data set; the fluctuation analysis module is used to analyze the upgrade data fluctuation and predict the standard upgrade data of the device to be upgraded; the curve modeling unit is used to construct a fluctuation deviation curve based on the fluctuation value and train the prediction model to obtain the standard upgrade data; the cache construction module is used to build and manage the cache architecture based on the standard upgrade data; The output end of the region division module is electrically connected to the input end of the fluctuation analysis module; the output end of the fluctuation analysis module is electrically connected to the input end of the cache construction module; the output end of the cache construction module is electrically connected to the input end of the upgrade execution module; The area division module includes a device partitioning unit and a data acquisition unit; the device partitioning unit is used to evenly divide the devices to be upgraded into each sub-area according to the number of devices; the data acquisition unit is used to obtain the historical upgrade data of the grayscale and non-grayscale upgrade devices in the sub-area and construct a corresponding analysis set; The fluctuation analysis module includes a data calculation unit and a curve modeling unit; the data calculation unit is used to calculate the data fluctuation value of the grayscale and non-grayscale upgrade devices; the curve modeling unit is used to construct a fluctuation deviation curve based on the fluctuation value and train the prediction model to obtain standard upgrade data; The cache construction module includes an architecture deployment unit and a resource allocation unit; the architecture deployment unit is used to build a three-level cache architecture at the cloud center level, edge node, and device end; the resource allocation unit is used to schedule cache resources according to standard upgrade data and build a management cache pool; The upgrade execution module includes a data transmission unit and an upgrade feedback unit; the data transmission unit is used to push the upgrade data packet to the local cache of the device and verify the data integrity; the upgrade feedback unit is used to feedback the result after the device completes the upgrade and trigger the release of cache resources and status update; Compared with the prior art, the present invention has the following beneficial effects: 1. This invention divides the coverage area of devices to be upgraded into multiple sub-areas. Using a clustering algorithm based on geographic information systems and network topology, this method balances the number of devices while ensuring proximity. It also collects historical data for both grayscale and non-grayscale upgrade devices to construct an analysis set. This enables the categorized integration of device upgrade data, providing a structured data foundation for subsequent, precise analysis of device compatibility, prediction of upgrade trends, and optimized resource scheduling, effectively improving the scientific nature of batch device upgrade scheduling.

[0014] 2. This invention utilizes machine learning algorithms to simulate ideal upgrade data. Combining a device upgrade compatibility analysis set with a device upgrade analysis set, this method deconstructs, calculates, and integrates historical data from both grayscale and non-grayscale upgrade devices to accurately capture upgrade data fluctuations. Furthermore, by constructing a fluctuation deviation curve and combining a time series prediction model with a neural network model, this method derives standard upgrade data, providing a scientific basis for resource scheduling and cache management, significantly improving upgrade prediction accuracy and policy adaptability.

[0015] 3. This invention builds a three-level cache architecture based on a cloud platform and edge computing. This architecture analyzes standard upgrade data to determine total resources and initializes the cache pool, enabling hierarchical collaborative cache management among the cloud center, edge nodes, and device terminals. The cloud center coordinates global strategies, edge nodes schedule regional resources, and the device terminal ensures data security. Together, these three enable efficient transmission and storage of upgrade data, effectively improving resource utilization and system stability during batch device upgrades. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a batch device upgrade scheduling method based on policy management according to the present invention; Figure 2 This is a structural diagram of a batch equipment upgrade scheduling system based on policy management of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] Example 1: Figure 1 As shown, the present invention provides a technical solution, a batch device upgrade scheduling method based on policy management, the batch device upgrade scheduling method comprises the following steps: Step S1: Divide the device range covered by the devices to be upgraded into multiple sub-areas, obtain the upgrade data of the devices with historical grayscale upgrades in each sub-area to construct a device upgrade compatibility analysis set, and obtain the historical upgrade data of the devices without grayscale upgrades in each sub-area to construct a device upgrade analysis set; Step S1-1: Divide the area by the number of devices to be upgraded, and evenly distribute the total amount of devices to be upgraded to each sub-area; Step S1-2: Build a data collection platform and obtain historical upgrade data for each grayscale upgrade device from the sub-region through an API interface. The grayscale upgrade device represents a portion of the devices that have been upgraded first, selected from all the devices to be upgraded. The historical upgrade data represents the upgrade time, failure rate, resource usage, and corresponding upgrade timestamp data generated by the grayscale upgrade device during the initial upgrade process. Step S1-3: construct a device upgrade compatibility analysis set based on the acquired grayscale upgrade device historical upgrade data as a value and the grayscale upgrade device unique identifier as an index; Step S1-4: construct a device upgrade analysis set based on the acquired historical upgrade data of non-grayscale upgrade devices as a value and the unique identifier of the non-grayscale upgrade device as an index. The non-grayscale upgrade device is represented as a device that has not been screened for prior upgrade and directly participates in subsequent full upgrades or historical regular upgrades.

[0019] In specific implementation, using production equipment upgrades in smart factories as an example, the DBSCAN algorithm was used to cluster over a thousand devices to be upgraded based on their physical location and network connectivity, ensuring that network latency for devices within the same sub-region did not exceed 5ms. The data collection platform established a persistent connection with each sub-region's API via the MQTT protocol, continuously acquiring historical upgrade data for both grayscale and non-grayscale devices every minute. Blockchain technology was also used to ensure that the data could not be tampered with.

[0020] Step S2: Analyze the device upgrade compatibility analysis set to obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device in the sub-region; analyze the device upgrade analysis set to obtain the standard upgrade data fluctuation value of the device to be upgraded in the sub-region; Step S2-1: Use a machine learning algorithm to simulate and obtain upgrade time data, failure rate data, and resource usage data of the device to be upgraded under ideal conditions; Step S2-2: traverse and read the upgrade data in the device upgrade compatibility analysis set, use the upgrade time data, failure rate data, and resource usage data under ideal conditions, subtract the upgrade time data, failure rate data, and resource usage data read through traversal, and obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device through normalization processing combined with weighted fusion calculation; Step S2-2-1: Decompose the upgrade data in the device upgrade compatibility analysis set to obtain upgrade time consumption data, failure rate data, and resource usage data corresponding to the grayscale upgrade device upgrade timestamp; Step S2-2-2: After subtracting the upgrade time data, failure rate data, and resource usage data, perform absolute value processing on the difference, and normalize the data after absolute value processing; Step S2-2-3: After normalizing all data, the grayscale upgrade data fluctuation value of the grayscale upgrade device is obtained through weighted fusion calculation; Step S2-3: Synchronously use the upgrade data fluctuation value calculation method of step S2-2 to process the upgrade data in the device upgrade analysis set to obtain the grayscale upgrade data fluctuation value of the non-grayscale upgrade device.

[0021] In specific implementation, using the distributed meter upgrade in a smart grid as an example, the Python pandas library was used to resample timestamp data, unifying upgrade data of varying frequencies into 15-minute time series data. To construct the fluctuation deviation curve, the Prophet time series forecasting model was used, with the seasonality parameter set to weekly. The upgrade fluctuation trend over the next 72 hours was predicted. When training the neural network model, 5,000 historical data points from the device upgrade analysis set were used as the target variable, and the model parameters were optimized using the mean squared error loss function.

[0022] Step S3: Perform a fluctuation deviation analysis using the grayscale upgrade data fluctuation value and the standard upgrade data fluctuation value to obtain a fluctuation deviation curve for the sub-region. This fluctuation deviation curve analysis is performed by analyzing the upgrade data of each grayscale upgrade of the sub-region in history. Combined with the real-time upgrade data of the devices undergoing grayscale upgrades in the sub-region, the standard upgrade data for the devices to be upgraded in the sub-region is obtained. Step S3-1, controlling the grayscale fluctuation deviation curve and the non-grayscale fluctuation deviation curve to perform time alignment through the timestamp data; Step S3-2: Subtract the standard upgrade data fluctuation value from the grayscale upgrade data fluctuation value corresponding to the timestamp data to obtain the fluctuation deviation value between the single upgrade of the grayscale upgrade device in the sub-region and the single upgrade of the non-grayscale upgrade device; Step S3-3: Perform time sequence analysis based on the timestamp data to obtain the fluctuation difference between the grayscale upgrade devices and the non-grayscale upgrade devices in the sub-region during the upgrade; Step S3-4: construct a fluctuation deviation curve after arranging in time sequence, wherein the horizontal axis of the fluctuation deviation curve is the timestamp data corresponding to the single upgrade of the sub-region grayscale upgrade device and the upgrade of the non-grayscale upgrade device, and the vertical axis of the fluctuation deviation curve is the fluctuation difference value corresponding to the upgrade; Step S3-5: Input the fluctuation gap value and timestamp data of the fluctuation deviation curve into the time series prediction model for training, and obtain the upgrade fluctuation trend of the devices to be upgraded and the grayscale upgrade devices in the sub-region, which is recorded as the upgrade fluctuation prediction trend; Step S3-6: Obtain the grayscale upgrade data fluctuation value of the grayscale upgrade devices in the sub-area, and obtain the upgrade data fluctuation value of the non-grayscale upgrade devices in the sub-area based on the upgrade fluctuation prediction trend; Step S3-7: Using the fluctuation gap value and timestamp data of the fluctuation deviation curve as input data, and combining the historical upgrade time data, failure rate data, and resource usage data in the device upgrade analysis set as target variables, the neural network model is trained to obtain a mapping relationship between the upgrade data fluctuation value of the device to be upgraded and the upgrade data; Step S3-8: Using the upgrade data fluctuation value obtained in step S3-6, the upgrade data of the sub-region is obtained through a mapping relationship and recorded as standard upgrade data.

[0023] In practice, the cloud-center cache is deployed using a Redis cluster, employing a master-slave replication architecture to ensure high availability. Edge node caches utilize the lightweight Memcached service, and device-side local caches integrate with a SQLite database. The Prometheus monitoring system collects real-time metrics such as cache hit rate and response time. A data pre-warming mechanism is automatically triggered when the edge node cache hit rate is low. When pushing upgrade data packages, a dynamic priority algorithm based on device importance and remaining battery power is used to ensure that critical devices are upgraded first.

[0024] Step S4: Create buffer zones for each sub-area based on the cloud platform's managed cache, schedule resources in the cache zone using standard upgrade data, deploy lightweight cache services on edge nodes using distributed cache technology and the edge computing platform, and build and manage data cache pools using edge computing protocols combined with standard upgrade data. Step S4-1: Through the global resource view of the cloud platform hosted cache, a three-level cache architecture is constructed according to the network topology of the sub-region: the first level is the cloud center-level distributed cache cluster, which is specifically a high-performance distributed cache system deployed in the cloud data center, responsible for storing global device upgrade metadata, load balancing policy configuration between sub-regions, and providing data preheating and disaster recovery support for edge nodes; the second level is the sub-region edge node cache layer, which is specifically a lightweight cache service deployed on the regional edge computing node close to the terminal device, responsible for storing the upgrade data packets accessed in this region and the device upgrade priority queue for real-time calculation; the third level is the device-side local cache, which is specifically a micro-memory database embedded in the terminal of the device to be upgraded, responsible for temporarily storing the upgrade data fragments related to this device, the intermediate state information during the upgrade process, and the digital signature authentication information generated based on the device's unique identifier; Step S4-2: Analyze the standard upgrade data of the device to be upgraded obtained through the output of step S3, and extract resource occupancy data as basic resource occupancy data of the device to be upgraded; Step S4-3: derive the total amount of resources required for device upgrade in the sub-region by detecting the product of the number of devices to be upgraded in the current sub-region and the basic resource occupancy data; Step S4-4: constructing a data cache pool based on the total amount of sub-region resources obtained in step S4-3, and initializing the total capacity of the data cache pool; Step S4-5: Fill the data cache pool of each sub-area with data through the cloud center-level distributed cache cluster. The edge node cache layer pushes the cached upgrade data packet to the terminal local cache of the corresponding device to be upgraded according to the upgrade priority queue of the device to be upgraded. After verifying the data integrity through the digital signature authentication information in the local cache, the upgrade operation is performed. The intermediate status information during the upgrade process is stored in the local cache in real time. After the device upgrade is completed, the result is fed back to the edge node and cloud center, triggering the release of cache resources and status update.

[0025] In specific implementation, the global configuration strategy is stored through the cloud center-level cache, the edge node cache layer manages the local data queue, and the terminal cache records data fragments and signatures; based on the standard upgrade data, resource parameters are extracted, the product of the number of devices and resource occupancy is calculated to determine the total amount, and the cache pool capacity is initialized; the cloud center pushes data packets to the edge nodes, and the edge nodes distribute them to the terminals according to priority. After verifying the signature, the terminal temporarily stores the data and executes the upgrade, feedbacks the status in real time and dynamically releases cache resources. After the upgrade is completed, the global cache status is synchronously updated.

[0026] Example 2, as Figure 2As shown, the present invention provides a batch device upgrade scheduling system based on policy management, which includes a region division module, a fluctuation analysis module, a cache construction module and an upgrade execution module; The area division module is used to divide the coverage area of the device to be upgraded into sub-areas and collect and construct an upgrade data set; the fluctuation analysis module is used to analyze the upgrade data fluctuation and predict the standard upgrade data of the device to be upgraded; the curve modeling unit is used to construct a fluctuation deviation curve based on the fluctuation value and train the prediction model to obtain the standard upgrade data; the cache construction module is used to build and manage the cache architecture based on the standard upgrade data; The output end of the region division module is electrically connected to the input end of the fluctuation analysis module; the output end of the fluctuation analysis module is electrically connected to the input end of the cache construction module; the output end of the cache construction module is electrically connected to the input end of the upgrade execution module; The area division module includes a device partitioning unit and a data acquisition unit; the device partitioning unit is used to evenly divide the devices to be upgraded into each sub-area according to the number of devices; the data acquisition unit is used to obtain the historical upgrade data of the grayscale and non-grayscale upgrade devices in the sub-area and construct a corresponding analysis set; The fluctuation analysis module includes a data calculation unit and a curve modeling unit; the data calculation unit is used to calculate the data fluctuation value of the grayscale and non-grayscale upgrade devices; the curve modeling unit is used to construct a fluctuation deviation curve based on the fluctuation value and train the prediction model to obtain standard upgrade data; The cache construction module includes an architecture deployment unit and a resource allocation unit; the architecture deployment unit is used to build a three-level cache architecture at the cloud center level, edge node, and device end; the resource allocation unit is used to schedule cache resources according to standard upgrade data and build a management cache pool; The upgrade execution module includes a data transmission unit and an upgrade feedback unit; the data transmission unit is used to push the upgrade data packet to the local cache of the device and verify the data integrity; the upgrade feedback unit is used to feedback the results after the device completes the upgrade and trigger the cache resource release and status update.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A batch device upgrade scheduling method based on policy management, characterized by: The batch equipment upgrade scheduling method comprises the following steps: Step S1: Divide the device range covered by the devices to be upgraded into multiple sub-areas, obtain the upgrade data of the devices with historical grayscale upgrades in each sub-area to construct a device upgrade compatibility analysis set, and obtain the historical upgrade data of the devices without grayscale upgrades in each sub-area to construct a device upgrade analysis set; Step S2: Analyze the device upgrade compatibility analysis set to obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device in the sub-region; analyze the device upgrade analysis set to obtain the standard upgrade data fluctuation value of the device to be upgraded in the sub-region; Step S3: Perform a fluctuation deviation analysis using the grayscale upgrade data fluctuation value and the standard upgrade data fluctuation value to obtain a fluctuation deviation curve for the sub-region. This fluctuation deviation curve analysis is performed by analyzing the upgrade data of each grayscale upgrade of the sub-region in history. Combined with the real-time upgrade data of the devices undergoing grayscale upgrades in the sub-region, the standard upgrade data for the devices to be upgraded in the sub-region is obtained. Step S4: Create buffer zones for each sub-area based on the cloud platform hosted cache, schedule the resources in the cache zone through standard upgrade data, and deploy lightweight cache services on edge nodes through distributed cache technology and edge computing platform. Build and manage the data cache pool using edge computing protocols combined with standard upgrade data.

2. The method for batch device upgrade scheduling based on policy management according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Divide the area by the number of devices to be upgraded, and evenly distribute the total amount of devices to be upgraded to each sub-area; Step S1-2: Build a data collection platform and obtain historical upgrade data for each grayscale upgrade device from the sub-region through an API interface. The grayscale upgrade device represents a portion of the devices that have been upgraded first, selected from all the devices to be upgraded. The historical upgrade data represents the upgrade time, failure rate, resource usage, and corresponding upgrade timestamp data generated by the grayscale upgrade device during the initial upgrade process. Step S1-3: construct a device upgrade compatibility analysis set based on the acquired grayscale upgrade device historical upgrade data as a value and the grayscale upgrade device unique identifier as an index; Step S1-4: construct a device upgrade analysis set based on the acquired historical upgrade data of non-grayscale upgrade devices as a value and the unique identifier of the non-grayscale upgrade device as an index. The non-grayscale upgrade device is represented as a device that has not been screened for prior upgrade and directly participates in subsequent full upgrades or historical regular upgrades.

3. The method for batch device upgrade scheduling based on policy management according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: Use a machine learning algorithm to simulate and obtain upgrade time data, failure rate data, and resource usage data of the device to be upgraded under ideal conditions; Step S2-2: traverse and read the upgrade data in the device upgrade compatibility analysis set, use the upgrade time data, failure rate data, and resource usage data under ideal conditions, subtract the upgrade time data, failure rate data, and resource usage data read through traversal, and obtain the grayscale upgrade data fluctuation value of the grayscale upgrade device through normalization processing combined with weighted fusion calculation; Step S2-2-1: Disassemble the upgrade data in the device upgrade compatibility analysis set to obtain upgrade time consumption data, failure rate data, and resource usage data corresponding to the grayscale upgrade device upgrade timestamp; Step S2-2-2: After subtracting the upgrade time data, failure rate data, and resource usage data, perform absolute value processing on the difference, and normalize the data after absolute value processing; Step S2-2-3: After normalizing all data, the grayscale upgrade data fluctuation value of the grayscale upgrade device is obtained through weighted fusion calculation; Step S2-3: Synchronously use the upgrade data fluctuation value calculation method of step S2-2 to process the upgrade data in the device upgrade analysis set to obtain the grayscale upgrade data fluctuation value of the non-grayscale upgrade device.

4. The method for batch device upgrade scheduling based on policy management according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1, controlling the grayscale fluctuation deviation curve and the non-grayscale fluctuation deviation curve to perform time alignment through the timestamp data; Step S3-2: Subtract the standard upgrade data fluctuation value from the grayscale upgrade data fluctuation value corresponding to the timestamp data to obtain the fluctuation deviation value between the single upgrade of the grayscale upgrade device in the sub-region and the single upgrade of the non-grayscale upgrade device; Step S3-3: Perform time sequence arrangement based on the timestamp data analysis to obtain the fluctuation difference value between the grayscale upgraded devices and the non-grayscale upgraded devices in the sub-region during the upgrade.

5. The method for batch device upgrade scheduling based on policy management according to claim 4, characterized in that: In step S3, it also includes: Step S3-4: construct a fluctuation deviation curve after arranging in time sequence, wherein the horizontal axis of the fluctuation deviation curve is the timestamp data corresponding to the single upgrade of the sub-region grayscale upgrade device and the upgrade of the non-grayscale upgrade device, and the vertical axis of the fluctuation deviation curve is the fluctuation difference value corresponding to the upgrade; Step S3-5: Input the fluctuation gap value and timestamp data of the fluctuation deviation curve into the time series prediction model for training, and obtain the upgrade fluctuation trend of the devices to be upgraded and the grayscale upgrade devices in the sub-region, which is recorded as the upgrade fluctuation prediction trend; Step S3-6: Obtain the grayscale upgrade data fluctuation value of the grayscale upgrade devices in the sub-area, and obtain the upgrade data fluctuation value of the non-grayscale upgrade devices in the sub-area based on the upgrade fluctuation prediction trend; Step S3-7: Using the fluctuation gap value and timestamp data of the fluctuation deviation curve as input data, and combining the historical upgrade time data, failure rate data, and resource usage data in the device upgrade analysis set as target variables, the neural network model is trained to obtain a mapping relationship between the upgrade data fluctuation value of the device to be upgraded and the upgrade data; Step S3-8: Using the upgrade data fluctuation value obtained in step S3-6, the upgrade data of the sub-region is obtained through a mapping relationship and recorded as standard upgrade data.

6. The method for batch device upgrade scheduling based on policy management according to claim 5, characterized in that: The specific steps of step S4 are as follows: Step S4-1: Through the global resource view of the cloud platform hosted cache, a three-level cache architecture is constructed according to the network topology of the sub-region: the first level is the cloud center-level distributed cache cluster, which is specifically a high-performance distributed cache system deployed in the cloud data center, responsible for storing global device upgrade metadata, load balancing policy configuration between sub-regions, and providing data preheating and disaster recovery support for edge nodes; the second level is the sub-region edge node cache layer, which is specifically a lightweight cache service deployed on the regional edge computing node close to the terminal device, responsible for storing the upgrade data packets accessed in this region and the device upgrade priority queue for real-time calculation; the third level is the device-side local cache, which is specifically a micro-memory database embedded in the terminal of the device to be upgraded, responsible for temporarily storing the upgrade data fragments related to this device, the intermediate state information during the upgrade process, and the digital signature authentication information generated based on the device's unique identifier; Step S4-2: Analyze the standard upgrade data of the device to be upgraded obtained through the output of step S3, and extract resource occupancy data as basic resource occupancy data of the device to be upgraded.

7. The method for batch device upgrade scheduling based on policy management according to claim 6, characterized in that: In step S4, it also includes: Step S4-3: derive the total amount of resources required for device upgrade in the sub-region by detecting the product of the number of devices to be upgraded in the current sub-region and the basic resource occupancy data; Step S4-4: constructing a data cache pool based on the total amount of sub-region resources obtained in step S4-3, and initializing the total capacity of the data cache pool; Step S4-5: Fill the data cache pool of each sub-area with data through the cloud center-level distributed cache cluster. The edge node cache layer pushes the cached upgrade data packet to the terminal local cache of the corresponding device to be upgraded according to the upgrade priority queue of the device to be upgraded. After verifying the data integrity through the digital signature authentication information in the local cache, the upgrade operation is performed. The intermediate status information during the upgrade process is stored in the local cache in real time. After the device upgrade is completed, the result is fed back to the edge node and cloud center, triggering the release of cache resources and status update.

8. A policy-based batch equipment upgrade scheduling system, applied to a policy-based batch equipment upgrade scheduling method according to any one of claims 1 to 7, characterized in that: The batch equipment upgrade scheduling system includes a region division module, a fluctuation analysis module, a cache construction module and an upgrade execution module; The area division module is used to divide the coverage area of the device to be upgraded into sub-areas and collect and construct an upgrade data set; the fluctuation analysis module is used to analyze the upgrade data fluctuation and predict the standard upgrade data of the device to be upgraded; the curve modeling unit is used to construct a fluctuation deviation curve based on the fluctuation value and train the prediction model to obtain the standard upgrade data; the cache construction module is used to build and manage the cache architecture based on the standard upgrade data; The output end of the region division module is electrically connected to the input end of the fluctuation analysis module; the output end of the fluctuation analysis module is electrically connected to the input end of the cache construction module; the output end of the cache construction module is electrically connected to the input end of the upgrade execution module.

9. The policy-based batch equipment upgrade scheduling system according to claim 8, characterized in that: The area division module includes a device partitioning unit and a data acquisition unit; the device partitioning unit is used to evenly divide the devices to be upgraded into each sub-area according to the number of devices; the data acquisition unit is used to obtain the historical upgrade data of the grayscale and non-grayscale upgrade devices in the sub-area and construct a corresponding analysis set; The fluctuation analysis module includes a data calculation unit and a curve modeling unit; the data calculation unit is used to calculate the data fluctuation values of grayscale and non-grayscale upgrade devices; the curve modeling unit is used to construct a fluctuation deviation curve through the fluctuation values and train a prediction model to obtain standard upgrade data.

10. The policy management-based batch equipment upgrade scheduling system according to claim 8, characterized in that: The cache construction module includes an architecture deployment unit and a resource allocation unit; the architecture deployment unit is used to build a three-level cache architecture at the cloud center level, edge node, and device end; the resource allocation unit is used to schedule cache resources according to standard upgrade data and build a management cache pool; The upgrade execution module includes a data transmission unit and an upgrade feedback unit; the data transmission unit is used to push the upgrade data packet to the local cache of the device and verify the data integrity; The upgrade feedback unit is used to feed back the result after the device completes the upgrade and trigger the release of cache resources and status update.

Images