Cloud-based remote maintenance method and system for IoT devices

Abstract

This invention discloses a cloud-based remote maintenance method and system for IoT devices, relating to the field of cloud computing IoT technology. The method includes: setting up a dual-track collaborative maintenance system; upon receiving a remote dimension request, reading the fault operation dataset and device connection attribute information of the IoT device to be maintained, inputting them into an adaptive decision engine for pattern decision-making, and obtaining the pattern decision result; wherein, the adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary pattern decision-making; if the pattern decision result returns empty, the adaptive decision classification network performs pattern decision-making, and remote maintenance is performed on the IoT device to be maintained according to the returned maintenance pattern. This invention solves the technical problems of existing technologies where remote maintenance modes are singular and cannot intelligently select appropriate maintenance strategies based on device status, resulting in low maintenance efficiency and insufficient reliability, thus achieving the technical effect of improving the accuracy and reliability of remote maintenance.

Description

Technical Field

[0001] This invention relates to the field of cloud computing and IoT technology, specifically to a method and system for remote maintenance of IoT devices based on cloud computing. Background Technology

[0002] In existing remote maintenance processes, devices are typically handled using fixed maintenance modes, making it impossible to differentiate maintenance based on device operating data and connectivity attributes. When device fault types are complex or operating environments vary, a single maintenance method often fails to meet timely needs, easily leading to reduced maintenance efficiency and even the risk of maintenance failure in some cases, affecting the stable operation and overall reliability of IoT devices. Summary of the Invention

[0003] This application provides a cloud-based method and system for remote maintenance of Internet of Things (IoT) devices, which addresses the technical problems of low maintenance efficiency and insufficient reliability caused by the single remote maintenance mode and the inability to intelligently select appropriate maintenance strategies based on device status in existing technologies.

[0004] In view of the above problems, this application provides a cloud computing-based method and system for remote maintenance of Internet of Things (IoT) devices.

[0005] The first aspect of this application provides a method for remote maintenance of IoT devices based on cloud computing, the method comprising:

[0006] A dual-track collaborative maintenance system is set up in the cloud decision-maker, which includes a standard maintenance mode and an enhanced maintenance mode. When the cloud decision-maker receives a remote dimension request, it reads the fault operation dataset and device connection attribute information of the IoT device to be maintained. The fault operation dataset and the device connection attribute information are input into the adaptive decision engine of the cloud decision-maker for mode decision-making, and the mode decision-making result is obtained. The mode decision-making result includes any one of the standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary mode decision-making based on multiple pre-stored security rules. If the mode decision-making result returns empty, the adaptive decision classification network performs mode decision-making based on the fault operation dataset and the device connection attribute information, and performs remote maintenance on the IoT device to be maintained according to the maintenance mode returned by the mode decision-making result.

[0007] A second aspect of this application provides a cloud-based remote maintenance system for Internet of Things (IoT) devices, the system comprising:

[0008] The system includes a maintenance system setting module for setting up a dual-track collaborative maintenance system on a cloud-based decision-maker. This system includes a standard maintenance mode and an enhanced maintenance mode. A data reading module is used to read the fault operation dataset and device connection attribute information of the IoT device to be maintained when the cloud-based decision-maker receives a remote dimension request. A mode decision module is used to input the fault operation dataset and device connection attribute information into the adaptive decision engine of the cloud-based decision-maker for mode decision-making, obtaining a mode decision result. This result includes any one of the standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary mode decision-making using multiple pre-stored security rules. If the mode decision result is empty, the adaptive decision classification network performs mode decision-making on the fault operation dataset and device connection attribute information, and remotely maintains the IoT device to be maintained according to the maintenance mode returned by the mode decision result.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application establishes a dual-track collaborative maintenance system in a cloud-based decision-maker. This system includes a standard maintenance mode and an enhanced maintenance mode. When the cloud-based decision-maker receives a remote dimension request, it reads the fault operation dataset and device connection attribute information of the IoT device to be maintained. The fault operation dataset and device connection attribute information are then input into the adaptive decision engine of the cloud-based decision-maker for mode decision-making. The mode decision result includes any one of the standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary mode decision-making based on multiple pre-stored security rules. If the mode decision result returns empty, the adaptive decision classification network performs mode decision-making based on the fault operation dataset and device connection attribute information. Remote maintenance is then performed on the IoT device to be maintained according to the maintenance mode returned by the mode decision result. This invention addresses the technical problems of low maintenance efficiency and insufficient reliability in existing remote maintenance technologies, which suffer from a single remote maintenance mode and the inability to intelligently select appropriate maintenance strategies based on equipment status. By introducing a dual-track collaborative maintenance system into the cloud decision-maker and combining it with security rule filtering and adaptive decision classification networks to achieve intelligent mode determination, the invention achieves the technical effect of automatically selecting standard maintenance, enhanced maintenance, or collaborative maintenance modes based on equipment operating data and connection attributes, thereby improving the accuracy and reliability of remote maintenance. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of a cloud-based remote maintenance method for IoT devices provided in this application embodiment;

[0013] Figure 2 This is a schematic diagram of the structure of a cloud-based IoT device remote maintenance system provided in an embodiment of this application.

[0014] Figure labeling: Maintenance system setting module 11, data reading module 12, mode decision module 13. Detailed Implementation

[0015] This application provides a cloud-based remote maintenance method and system for IoT devices. It addresses the technical problems of existing technologies, such as the single remote maintenance mode and the inability to intelligently select appropriate maintenance strategies based on device status, leading to low maintenance efficiency and insufficient reliability. By introducing a dual-track collaborative maintenance system into the cloud-based decision-maker and combining security rule filtering and adaptive decision classification networks to achieve intelligent mode determination, the system automatically selects standard maintenance, enhanced maintenance, or collaborative maintenance modes based on device operating data and connection attributes, thereby improving the accuracy and reliability of remote maintenance.

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown, this application provides a cloud-based remote maintenance method for Internet of Things (IoT) devices, the method comprising:

[0019] Step S100: Set up a dual-track collaborative maintenance system in the cloud decision-maker, the dual-track collaborative maintenance system including a standard maintenance mode and an enhanced maintenance mode.

[0020] In this embodiment, a dual-track collaborative maintenance system is first set up in the cloud-based decision-maker. This system includes a standard maintenance mode and an enhanced maintenance mode. The standard maintenance mode is used to restore the original functions and configurations of the equipment when a failure occurs. The enhanced maintenance mode is used to upgrade the equipment's functions or optimize its performance while completing fault repair.

[0021] Step S200: When the cloud decision-maker receives a remote dimension request, it reads the fault operation dataset and device connection attribute information of the IoT device to be maintained.

[0022] In this embodiment, when the cloud-based decision-maker receives a remote dimension request, it reads the fault operation dataset of the IoT device under maintenance through the data interaction channel established with the device. During this reading process, information related to the fault is extracted from the operation logs, error codes, and sensor anomaly data uploaded by the device, forming the fault operation dataset. Simultaneously, the cloud-based decision-maker reads the device connection attribute information of the IoT device under maintenance through the same interaction channel. This reading process includes obtaining the current network connection status, detecting remaining storage space, collecting battery power, and confirming whether an external power source is connected. Through these reading steps, the cloud-based decision-maker synchronously obtains the fault operation dataset and device connection attribute information when receiving the remote dimension request.

[0023] Step S300: Input the fault operation dataset and the device connection attribute information into the adaptive decision engine of the cloud decision-maker for pattern decision-making, and obtain the pattern decision result. The pattern decision result includes any one of the following modes: standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary pattern decision-making based on multiple pre-stored security rules. If the pattern decision result returns empty, the adaptive decision classification network performs pattern decision-making on the fault operation dataset and the device connection attribute information, and performs remote maintenance on the IoT device to be maintained according to the maintenance mode returned by the pattern decision result.

[0024] In this embodiment, the fault operation dataset and device connection attribute information are input into the adaptive decision engine of the cloud-based decision-maker for pattern decision-making. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary pattern decision-making on the fault operation dataset and device connection attribute information by comparing them with multiple pre-stored security rules. When the pattern decision result satisfies the security rules, it directly outputs the pattern decision result as the standard maintenance mode. When any security rule is not satisfied, the pattern decision result returns empty, and the fault operation dataset and device connection attribute information are input into the adaptive decision classification network. The adaptive decision classification network then performs pattern decision-making on the fault operation dataset and the device connection attribute information, outputting the pattern decision result. Through the aforementioned process, the obtained pattern decision result can include any one of the following modes: standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode.

[0025] Finally, the cloud-based decision-maker performs remote maintenance on the IoT devices to be maintained based on the maintenance mode corresponding to the mode decision result.

[0026] Furthermore, in the method provided in the application embodiments, the security rule filter performs preliminary pattern decisions based on multiple pre-stored security rules, and further includes:

[0027] The multiple security rules are security rules under the standard maintenance mode set based on historical prior knowledge base; the fault operation dataset and the device connection attribute information are analyzed and matched with the multiple security rules. When any security rule is satisfied, the mode decision result is returned as the standard maintenance mode; when no security rule is satisfied, the mode decision result is returned as empty, and the fault operation dataset and the device connection attribute information are transmitted to the adaptive decision classification network.

[0028] In this embodiment, multiple security rules are security rules under the standard maintenance mode based on a historical prior knowledge base. These include: when the execution function of the IoT device to be maintained belongs to the list of critical execution functions, stability must be prioritized and the device is matched to the standard maintenance mode; when the remaining storage space of the IoT device to be maintained is less than the size of the enhanced maintenance data packet in the enhanced maintenance mode, the device is determined to be unable to carry out the upgrade task and is matched to the standard maintenance mode; when the remaining power of the IoT device to be maintained is lower than a safety threshold and it is not connected to an external power source, the upgrade is determined to pose a risk of power failure and the device is matched to the standard maintenance mode.

[0029] During the analysis, the security rule filter standardizes the input fault operation dataset and device connectivity attributes. For example, it maps error codes to predefined fault types and converts storage space and battery level into percentages. These data are then compared against thresholds or conditions defined in multiple security rules. If the function of the IoT device to be maintained matches the list of critical functions, or if the remaining storage space is insufficient to support the size of the enhanced maintenance data package, or if the battery level is below a threshold and there is no external power source, then the condition is met.

[0030] When the matching result satisfies multiple security rules, the pattern decision result is directly returned as the standard maintenance mode. The cloud decision-maker then triggers the distribution of the standard maintenance data package to complete the remote maintenance of the IoT device to be maintained. If neither the fault operation dataset nor the device connection attribute information satisfies the comparison conditions of multiple security rules, the pattern decision result is returned as empty. The fault operation dataset and device connection attribute information are then transmitted to the adaptive decision classification network for pattern decision-making.

[0031] Furthermore, the method provided in the application embodiments also includes:

[0032] The multiple security rules include at least the following: the execution function of the IoT device to be maintained belongs to the list of critical execution functions; the remaining storage space of the IoT device to be maintained is less than the size of the enhanced maintenance data packet in the enhanced maintenance mode; and the remaining power of the IoT device to be maintained is lower than the safety threshold and is not connected to an external power source.

[0033] In this embodiment, the multiple security rules include at least three specific conditions. Specifically, for the execution function of the IoT device to be maintained, the execution function identifier and function category information of the IoT device to be maintained are read and matched against the function entries and category levels of the key execution function list. When the comparison result shows that the execution function of the IoT device to be maintained belongs to the key execution function list, the mode decision result returns to the standard maintenance mode.

[0034] To determine the relationship between the remaining storage space of the IoT device to be maintained and the size of the enhanced maintenance data packet in enhanced maintenance mode, the actual value of the remaining storage space of the IoT device to be maintained is obtained and compared with the size of the enhanced maintenance data packet in enhanced maintenance mode. When the comparison result shows that the remaining storage space of the IoT device to be maintained is less than the size of the enhanced maintenance data packet in enhanced maintenance mode, the mode decision result returns to standard maintenance mode.

[0035] For the IoT device under maintenance, the remaining battery power and external power supply conditions are compared with a safety threshold, and the status of whether it is connected to an external power source is also read. When the determination result is that the remaining battery power of the IoT device under maintenance is lower than the safety threshold and it is not connected to an external power source, the mode decision result returns to the standard maintenance mode.

[0036] Furthermore, in the method provided in the application embodiments, the process of making pattern decisions based on the adaptive decision classification network on the fault operation dataset and the device connection attribute information further includes:

[0037] The adaptive decision classification network is used to receive enhanced maintenance data packets issued by the adaptive decision engine based on the fault operation dataset of the IoT device to be maintained; to identify risks in the enhanced maintenance data packets and obtain a first enhanced risk probability; to identify risks in the device connection attribute information of the IoT device to be maintained and obtain a second enhanced risk probability; to calculate and determine a comprehensive enhanced risk probability using the first enhanced risk probability and the second enhanced risk probability; and when the comprehensive enhanced risk probability is greater than a first preset expected risk probability, the mode decision result is returned as the standard maintenance mode.

[0038] In this embodiment, the adaptive decision classification network receives an enhanced maintenance data packet from the adaptive decision engine based on the fault operation dataset of the IoT device to be maintained. Specifically, the adaptive decision engine takes the fault operation dataset of the IoT device to be maintained as input, completes the mapping of error codes and fault types, the location of affected components and the comparison of firmware baselines, and selects the update target and update scope according to the preset update strategy and target version list. It then assembles an enhanced maintenance data packet containing updated firmware data, incremental upgrade patches and new digital signatures, and transmits it to the adaptive decision classification network through the cloud distribution mechanism.

[0039] Next, risk identification is performed on the enhanced maintenance data package. This process begins by extracting an enhancement failure risk factor based on the historical enhancement failure probabilities of similar enhanced maintenance data packages. Then, compatibility tests are conducted between the enhanced maintenance data package and the IoT device to be maintained, checking firmware version consistency, processor architecture adaptability, and driver component matching to generate a compatibility risk factor. Subsequently, the reliability of the data source and the security of the API calls in the enhanced maintenance data package are verified, outputting a security risk factor. Finally, the enhancement failure risk factor, compatibility risk factor, and security risk factor are weighted according to preset weights to obtain the first enhancement risk probability.

[0040] Next, risk identification is performed on the device connectivity attribute information of the IoT devices to be maintained. This process first identifies the network connectivity reliability of the network environment and quantifies it as a network reliability risk factor. Then, based on the device connectivity attribute information of the IoT devices to be maintained, the number of nodes of the subordinate driving devices is extracted to determine the synchronization enhancement cost risk factor. Finally, the network reliability risk factor and the synchronization enhancement cost risk factor are weighted according to preset weights to obtain the second enhancement risk probability.

[0041] The first and second enhanced risk probabilities are then weighted according to preset weights to obtain the comprehensive enhanced risk probability. The weights of the first and second enhanced risk probabilities are preset by technical experts. After obtaining the comprehensive enhanced risk probability, it is compared with a preset first expected risk probability. When the comprehensive enhanced risk probability is greater than the first preset expected risk probability, the mode decision result returns to the standard maintenance mode.

[0042] Furthermore, the method provided in the application embodiments also includes:

[0043] When the overall enhanced risk probability is less than or equal to the first preset expected risk probability, it is determined whether the overall enhanced risk probability is greater than or equal to the second preset expected risk probability. If the overall enhanced risk probability is greater than or equal to the second preset expected risk probability, the mode decision result is returned as collaborative maintenance mode, wherein the second preset expected risk probability is less than the first preset expected risk probability; if the overall enhanced risk probability is less than the second preset expected risk probability, the mode decision result is returned as enhanced maintenance mode.

[0044] In this embodiment of the application, the comprehensive enhanced risk probability is compared with the first preset expected risk probability. When the comprehensive enhanced risk probability is less than or equal to the first preset expected risk probability, the comprehensive enhanced risk probability is compared with the preset second preset expected risk probability. If the comprehensive enhanced risk probability is greater than or equal to the second preset expected risk probability, the mode decision result is returned to the collaborative maintenance mode.

[0045] When the overall enhanced risk probability is less than the second preset expected risk probability, the mode decision result is returned to the enhanced maintenance mode. The first and second preset expected risk probabilities are both preset by technical experts, and the second preset expected risk probability is less than the first preset expected risk probability.

[0046] Furthermore, in the method provided in the application embodiments, the process of identifying risks in the enhanced maintenance data packet and obtaining a first enhanced risk probability further includes:

[0047] Identify the historical enhancement failure probability of enhancement maintenance data packets of the same category as the enhancement maintenance data packet to obtain the enhancement failure risk factor; perform compatibility testing on the firmware version, processor architecture, and driver components of the enhancement maintenance data packet and the IoT device to be maintained to obtain the compatibility risk factor; detect the security of the data source and calling interface in the enhancement maintenance data packet and output the security risk factor; perform weighted calculation according to the enhancement failure risk factor, compatibility risk factor, and security risk factor to obtain the first enhancement risk probability.

[0048] In this embodiment of the application, in order to identify the historical enhancement failure probability of enhancement maintenance data packets of the same category, the category of enhancement maintenance data packets is first used as the grouping basis, and the historical issuance and execution results are statistically analyzed to obtain the historical enhancement failure probability of the category in previous scenarios. This historical enhancement failure probability is then used as an enhancement failure risk factor.

[0049] Next, compatibility testing is performed on the enhanced maintenance data package and the IoT device to be maintained, including firmware version, processor architecture, and driver components. During this process, a compatibility checklist with three checkpoints is created to record whether firmware version consistency, processor architecture consistency, and driver component dependency satisfaction are met. A compatibility risk factor ranging from 0 to 1 is calculated by dividing the number of inconsistencies by the total number of checkpoints.

[0050] Subsequently, a compliance checklist method was applied to the data source and calling interface in the enhanced maintenance data package. On the data source side, digital signature verification and certificate trust chain verification were performed. On the calling interface side, the authentication method, permission scope and transmission encryption were verified to meet the requirements. The number of failed items was divided by the total number of checked items to obtain a security risk factor with a value range of 0 to 1.

[0051] Finally, the enhancement failure risk factor, compatibility risk factor, and security risk factor are weighted according to preset weights to obtain the first enhancement risk probability.

[0052] Furthermore, the method provided in the application embodiment, which involves risk identification of the device connection attribute information of the IoT device to be maintained and obtaining a second enhanced risk probability, further includes:

[0053] The network connection reliability of the network environment is identified based on the device connection attribute information of the IoT device to be maintained, and a network reliability risk factor is obtained; the number of nodes of the subordinate driving device of the IoT device to be maintained is obtained based on the device connection attribute information of the IoT device to be maintained, and a synchronization enhancement cost risk factor is obtained; a second enhancement risk probability is obtained by weighting the network reliability risk factor and the synchronization enhancement cost risk factor.

[0054] In this embodiment of the application, when identifying the network connection reliability of the network environment based on the device connection attribute information of the IoT device to be maintained, connection quality indicators such as packet loss rate, round-trip time (RTT), jitter, and signal strength (RSSI) in wireless scenarios are first read within a fixed time window. The threshold segmentation mapping method is used to convert each indicator into a sub-risk value in the range of 0 and 1. For example, a high packet loss rate corresponds to a high mapping value, RTT and jitter exceeding the threshold correspond to an increased mapping value, and a low RSSI correspond to an increased mapping value. Then, a simple average of these sub-risk values ​​is taken to obtain a network reliability risk factor with a value range of 0 to 1.

[0055] Next, the number of nodes of the slave driving device is obtained based on the device connection attribute information of the IoT device to be maintained. First, the number of nodes of the slave driving device is read from the device connection attribute information to confirm a unique count and remove offline or duplicate nodes. Then, the benchmark ratio method is used, with a preset benchmark size as the acceptable synchronization size, to quantify the synchronization overhead as the number of nodes / preset benchmark size, and truncates it to 1 when it exceeds 1, thus obtaining a synchronization enhancement cost risk factor with a value range of 0 to 1.

[0056] Finally, the network reliability risk factor and the synchronization enhancement cost risk factor are weighted according to the weights preset by technical experts to obtain the second enhancement risk probability.

[0057] Furthermore, the method provided in the application embodiments also includes:

[0058] When the mode decision result returns to standard maintenance mode, the hardware and software of the IoT device to be maintained are configured and restored using the standard maintenance data package issued by the adaptive decision engine. The standard maintenance data package includes a verified stable firmware image, configuration file, and execution script. When the mode decision result returns to enhanced maintenance mode, the hardware and software of the IoT device to be maintained are configured and updated using the enhanced maintenance data package issued by the adaptive decision engine. The enhanced maintenance data package includes updated firmware data, incremental upgrade patches, and new digital signatures.

[0059] In this embodiment, when the mode decision result returns to the standard maintenance mode, the adaptive decision engine performs integrity and digital signature verification on the firmware image, selects a configuration file matching the IoT device to be maintained, and generates an execution script for performing the recovery steps. The verified stable firmware image, configuration file, and execution script are combined into a standard maintenance data package. The standard maintenance data package is then sent to the IoT device to be maintained via a cloud-based distribution mechanism. The device enters a secure recovery state and sequentially executes the following steps according to the execution script: image integrity verification, writing to the system / boot partition, refilling the verified configuration file, restarting critical services, and self-testing. Through this process, a configuration restore operation is performed on the hardware and software, restoring the device to a verified stable operating state.

[0060] When the mode decision result returns to the enhanced maintenance mode, the adaptive decision engine generates updated firmware data based on the predetermined update content, combines it with the differences to form an incremental upgrade patch, and adds a new digital signature to the update content. The updated firmware data, incremental upgrade patch, and new digital signature are combined to form an enhanced maintenance data package. After the IoT device to be maintained receives the enhanced maintenance data package, it first completes the signature and integrity verification, and then applies the updated firmware data and incremental upgrade patch in a predetermined order and completes the necessary parameter migration, performing configuration update operations on the hardware and software of the IoT device to be maintained.

[0061] Furthermore, in the method provided in the application embodiments, when the mode decision result returns to the collaborative maintenance mode, it further includes:

[0062] The adaptive decision engine issues local-standard maintenance data packages and local-enhanced maintenance data packages to the IoT device to be maintained; and performs configuration restoration operations on the hardware and software of the IoT device to be maintained based on the local-standard maintenance data packages and local-enhanced maintenance data packages.

[0063] In this embodiment of the application, when the mode decision result returns to the collaborative maintenance mode, the adaptive decision engine first performs fault location based on the fault operation dataset of the IoT device to be maintained, and cross-confirms the scope of fault impact by three types of evidence: error code and module mapping, operation log and fault stack clues, and inter-component call relationship, to obtain the set of components strongly related to the fault and the set of associated components obtained by the dependency relationship, forming an initial list of component subsets that need to be processed.

[0064] After determining the scope of the fault, the adaptive decision engine evaluates the maintenance orientation of each component subset. Combining the results of the safety rule filter and the comprehensive enhancement risk probability judgment, it assigns the component subsets that need immediate stability restoration to the local standard maintenance path, and assigns the component subsets that are upgradable and whose risks are under control to the local enhanced maintenance path, thus completing the allocation decision at the component level.

[0065] Subsequently, for the subset of components included in the local standard maintenance path, the adaptive decision engine extracts the stable firmware image fragment, component-level configuration file fragment, and execution script entry corresponding to the target component. After completing integrity and signature verification, it combines them into a local-standard maintenance data package, which enables the target component to be image-written back, parameter restored, and service restored on the device according to the execution script, thereby realizing the local configuration restoration operation.

[0066] For the subset of components that are classified into the local enhancement maintenance path, the adaptive decision engine selects the corresponding update firmware data and incremental upgrade patch according to the update strategy and target version list, and generates a new digital signature for the update content. These are combined into a local-enhancement maintenance data package, which enables the component-level update writing, driver replacement and configuration migration to be completed on the device side, thereby realizing local configuration update operations.

[0067] Finally, the adaptive decision engine sends the local-standard maintenance data package and the local-enhanced maintenance data package to the IoT device to be maintained, respectively. The local standard maintenance data package is executed first to restore the stable state of the key components, and then the local enhanced maintenance data package is executed to complete the controlled upgrade. Through this process, the configuration restoration operation of the hardware and software of the IoT device to be maintained is completed.

[0068] In summary, the embodiments of this application have at least the following technical effects:

[0069] This application establishes a dual-track collaborative maintenance system in a cloud-based decision-maker. This system includes a standard maintenance mode and an enhanced maintenance mode. When the cloud-based decision-maker receives a remote dimension request, it reads the fault operation dataset and device connection attribute information of the IoT device to be maintained. The fault operation dataset and device connection attribute information are then input into the adaptive decision engine of the cloud-based decision-maker for mode decision-making. The mode decision result includes any one of the standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary mode decision-making based on multiple pre-stored security rules. If the mode decision result returns empty, the adaptive decision classification network performs mode decision-making based on the fault operation dataset and device connection attribute information. Remote maintenance is then performed on the IoT device to be maintained according to the maintenance mode returned by the mode decision result. This invention addresses the technical problems of low maintenance efficiency and insufficient reliability in existing remote maintenance technologies, which suffer from a single remote maintenance mode and the inability to intelligently select appropriate maintenance strategies based on equipment status. By introducing a dual-track collaborative maintenance system into the cloud decision-maker and combining it with security rule filtering and adaptive decision classification networks to achieve intelligent mode determination, the invention achieves the technical effect of automatically selecting standard maintenance, enhanced maintenance, or collaborative maintenance modes based on equipment operating data and connection attributes, thereby improving the accuracy and reliability of remote maintenance.

[0070] Example 2, based on the same inventive concept as the cloud-based remote maintenance method for IoT devices in the foregoing examples, such as... Figure 2 As shown, this application provides a cloud-based remote maintenance system for Internet of Things (IoT) devices. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0071] The maintenance system setting module 11 is used to set up a dual-track collaborative maintenance system in the cloud decision-maker, which includes a standard maintenance mode and an enhanced maintenance mode. The data reading module 12 is used to read the fault operation dataset and device connection attribute information of the IoT device to be maintained when the cloud decision-maker receives a remote dimension request. The mode decision module 13 is used to input the fault operation dataset and the device connection attribute information into the adaptive decision engine of the cloud decision-maker for mode decision-making, and obtain the mode decision result, which includes any one of the standard maintenance mode, enhanced maintenance mode, and collaborative maintenance mode. The adaptive decision engine includes a security rule filter and an adaptive decision classification network. The security rule filter performs preliminary mode decision-making based on multiple pre-stored security rules. If the mode decision result returns empty, the adaptive decision classification network performs mode decision-making on the fault operation dataset and the device connection attribute information, and performs remote maintenance on the IoT device to be maintained according to the maintenance mode returned by the mode decision result.

[0072] Furthermore, the system is also used to implement the following functions:

[0073] The multiple security rules are security rules under the standard maintenance mode set based on historical prior knowledge base; the fault operation dataset and the device connection attribute information are analyzed and matched with the multiple security rules. When any security rule is satisfied, the mode decision result is returned as the standard maintenance mode; when no security rule is satisfied, the mode decision result is returned as empty, and the fault operation dataset and the device connection attribute information are transmitted to the adaptive decision classification network.

[0074] Furthermore, the system is also used to implement the following functions:

[0075] The multiple security rules include at least the following: the execution function of the IoT device to be maintained belongs to the list of critical execution functions; the remaining storage space of the IoT device to be maintained is less than the size of the enhanced maintenance data packet in the enhanced maintenance mode; and the remaining power of the IoT device to be maintained is lower than the safety threshold and is not connected to an external power source.

[0076] Furthermore, the system is also used to implement the following functions:

[0077] The adaptive decision classification network is used to receive enhanced maintenance data packets issued by the adaptive decision engine based on the fault operation dataset of the IoT device to be maintained; to identify risks in the enhanced maintenance data packets and obtain a first enhanced risk probability; to identify risks in the device connection attribute information of the IoT device to be maintained and obtain a second enhanced risk probability; to calculate and determine a comprehensive enhanced risk probability using the first enhanced risk probability and the second enhanced risk probability; and when the comprehensive enhanced risk probability is greater than a first preset expected risk probability, the mode decision result is returned as the standard maintenance mode.

[0078] Furthermore, the system is also used to implement the following functions:

[0079] When the overall enhanced risk probability is less than or equal to the first preset expected risk probability, it is determined whether the overall enhanced risk probability is greater than or equal to the second preset expected risk probability. If the overall enhanced risk probability is greater than or equal to the second preset expected risk probability, the mode decision result is returned as collaborative maintenance mode, wherein the second preset expected risk probability is less than the first preset expected risk probability; if the overall enhanced risk probability is less than the second preset expected risk probability, the mode decision result is returned as enhanced maintenance mode.

[0080] Furthermore, the system is also used to implement the following functions:

[0081] Identify the historical enhancement failure probability of enhancement maintenance data packets of the same category as the enhancement maintenance data packet to obtain the enhancement failure risk factor; perform compatibility testing on the firmware version, processor architecture, and driver components of the enhancement maintenance data packet and the IoT device to be maintained to obtain the compatibility risk factor; detect the security of the data source and calling interface in the enhancement maintenance data packet and output the security risk factor; perform weighted calculation according to the enhancement failure risk factor, compatibility risk factor, and security risk factor to obtain the first enhancement risk probability.

[0082] Furthermore, the system is also used to implement the following functions:

[0083] The network connection reliability of the network environment is identified based on the device connection attribute information of the IoT device to be maintained, and a network reliability risk factor is obtained; the number of nodes of the subordinate driving device of the IoT device to be maintained is obtained based on the device connection attribute information of the IoT device to be maintained, and a synchronization enhancement cost risk factor is obtained; a second enhancement risk probability is obtained by weighting the network reliability risk factor and the synchronization enhancement cost risk factor.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] When the mode decision result returns to standard maintenance mode, the hardware and software of the IoT device to be maintained are configured and restored using the standard maintenance data package issued by the adaptive decision engine. The standard maintenance data package includes a verified stable firmware image, configuration file, and execution script. When the mode decision result returns to enhanced maintenance mode, the hardware and software of the IoT device to be maintained are configured and updated using the enhanced maintenance data package issued by the adaptive decision engine. The enhanced maintenance data package includes updated firmware data, incremental upgrade patches, and new digital signatures.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] The adaptive decision engine issues local-standard maintenance data packages and local-enhanced maintenance data packages to the IoT device to be maintained; and performs configuration restoration operations on the hardware and software of the IoT device to be maintained based on the local-standard maintenance data packages and local-enhanced maintenance data packages.

[0088] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0089] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0090] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A cloud computing-based method for remote maintenance of Internet of Things devices, characterized in that, The method comprises: The cloud decision maker sets a double-track collaborative maintenance system, which includes a standard maintenance mode and an enhanced maintenance mode; When the cloud decision maker receives a remote maintenance request, read the fault operation data set and device connection attribute information of the Internet of Things device to be maintained; Input the fault operation data set and the device connection attribute information into the adaptive decision engine of the cloud decision maker for mode decision, and obtain a mode decision result, which includes any one of the standard maintenance mode, the enhanced maintenance mode and the collaborative maintenance mode; Wherein, the adaptive decision engine includes a safety rule filter and an adaptive decision classification network, the safety rule filter makes a preliminary mode decision through a plurality of pre-stored safety rules, if the mode decision result returns empty, the adaptive decision classification network is used for mode decision based on the fault operation data set and the device connection attribute information, and the maintenance mode of the mode decision result is used for remote maintenance of the Internet of Things device to be maintained; The safety rule filter makes a preliminary mode decision through a plurality of pre-stored safety rules, which comprises: Wherein, the plurality of safety rules are safety rules in the standard maintenance mode based on historical prior knowledge base; Analyze the fault operation data set and the device connection attribute information, and match them with the plurality of safety rules, when any safety rule is met, the mode decision result returns as the standard maintenance mode; When any safety rule is not met, the mode decision result returns empty, and the fault operation data set and the device connection attribute information are transmitted to the adaptive decision classification network.

2. The method of claim 1, wherein, The plurality of safety rules at least include that the execution function of the Internet of Things device to be maintained belongs to the list of key execution functions; The remaining storage space of the Internet of Things device to be maintained is less than the enhanced maintenance data packet size in the enhanced maintenance mode; And the remaining power of the Internet of Things device to be maintained is lower than the safety threshold and is not connected to external power supply.

3. The method of claim 1, wherein, The adaptive decision classification network is used for receiving the enhanced maintenance data packet issued by the adaptive decision engine based on the fault operation data set of the Internet of Things device to be maintained; Risk identification is performed on the enhanced maintenance data packet to obtain a first enhanced risk probability; Risk identification is performed on the device connection attribute information of the Internet of Things device to be maintained to obtain a second enhanced risk probability; The first enhanced risk probability and the second enhanced risk probability are used for calculation to determine a comprehensive enhanced risk probability, when the comprehensive enhanced risk probability is greater than a first preset expected risk probability, the mode decision result returns as the standard maintenance mode. ​ 4. The method of claim 3, wherein, When the comprehensive enhanced risk probability is less than or equal to the first preset expected risk probability, it is determined whether the comprehensive enhanced risk probability is greater than or equal to a second preset expected risk probability, and if the comprehensive enhanced risk probability is greater than or equal to the second preset expected risk probability, the mode decision result is returned as a cooperative maintenance mode, wherein the second preset expected risk probability is less than the first preset expected risk probability; If the comprehensive enhanced risk probability is less than the second preset expected risk probability, the mode decision result is returned as an enhanced maintenance mode.

5. The method of claim 3, wherein, The method for performing risk identification on the enhanced maintenance data packet to obtain a first enhanced risk probability comprises: Identifying a historical enhanced failure probability of the same category enhanced maintenance data packet of the enhanced maintenance data packet to obtain an enhanced failure risk factor; Performing compatibility detection on firmware version, processor architecture and driver components of the enhanced maintenance data packet and the to-be-maintained Internet of Things device to obtain a compatibility risk factor; Detecting the security of data sources and calling interfaces in the enhanced maintenance data packet to output a security risk factor; Performing weighted calculation on the enhanced failure risk factor, the compatibility risk factor and the security risk factor to obtain the first enhanced risk probability.

6. The method of claim 3, wherein, The method for performing risk identification on device connection attribute information of the to-be-maintained Internet of Things device to obtain a second enhanced risk probability comprises: Identifying network connection reliability of a network environment according to the device connection attribute information of the to-be-maintained Internet of Things device to obtain a network reliability risk factor; According to the device connection attribute information of the to-be-maintained Internet of Things device, obtaining the number of slave driver devices of the to-be-maintained Internet of Things device to obtain a synchronization enhancement cost risk factor; Performing weighted calculation on the network reliability risk factor and the synchronization enhancement cost risk factor to obtain the second enhanced risk probability.

7. The method of claim 1, wherein, When the mode decision result is returned as a standard maintenance mode, performing configuration restoration operation on hardware and software of the to-be-maintained Internet of Things device through the standard maintenance data packet issued by the adaptive decision engine, wherein the standard maintenance data packet comprises a stable firmware image after verification, a configuration file and an execution script; When the mode decision result is returned as an enhanced maintenance mode, performing configuration update operation on hardware and software of the to-be-maintained Internet of Things device through the enhanced maintenance data packet issued by the adaptive decision engine, wherein the enhanced maintenance data packet comprises update firmware data, incremental upgrade patches and new digital signatures.

8. The method of claim 7, wherein, When the mode decision result is returned as a cooperative maintenance mode, it comprises: Issuing local-standard maintenance data packets and local-enhanced maintenance data packets of the to-be-maintained Internet of Things device through the adaptive decision engine; Performing configuration restoration operation on hardware and software of the to-be-maintained Internet of Things device according to the local-standard maintenance data packets and the local-enhanced maintenance data packets.

9. A cloud computing-based remote maintenance system for Internet of Things devices, characterized by, The system is used for executing the cloud computing-based remote maintenance method of the Internet of Things device as claimed in any one of claims 1-8, and the system comprises: The system sets a double-track collaborative maintenance system including a standard maintenance mode and an enhanced maintenance mode for the cloud decision maker; The data reading module reads the fault operation data set and the device connection attribute information of the to-be-maintained Internet of Things device when the cloud decision maker receives a remote maintenance request; The mode decision module inputs the fault operation data set and the device connection attribute information into the adaptive decision engine of the cloud decision maker for mode decision, and obtains a mode decision result, which includes any one of the standard maintenance mode, the enhanced maintenance mode, and the collaborative maintenance mode; The adaptive decision engine includes a safety rule filter and an adaptive decision classification network. The safety rule filter performs preliminary mode decision through a plurality of pre-stored safety rules. If the mode decision result is empty, the adaptive decision classification network performs mode decision on the fault operation data set and the device connection attribute information, and performs remote maintenance on the to-be-maintained Internet of Things device according to the maintenance mode of the returned mode decision result. Further, the system is also used to implement the following functions: The plurality of safety rules are safety rules in the standard maintenance mode set based on a historical prior knowledge base. The fault operation data set and the device connection attribute information are analyzed and matched with the plurality of safety rules. When any safety rule is met, the mode decision result is returned as the standard maintenance mode. When any safety rule is not met, the mode decision result is returned as empty, and the fault operation data set and the device connection attribute information are transmitted to the adaptive decision classification network.

Images