A remote maintenance system based on an internet of things cloud platform device

Abstract

The application provides a kind of based on internet of things cloud platform equipment remote maintenance system, comprising: data acquisition module, for based on the multi-source heterogeneous data of sensor acquisition equipment;Cloud platform processing module, for based on the data twin of equipment is constructed with multi-source heterogeneous data, and based on the real-time running data of equipment to the data twin is updated with same frequency;Fault analysis module, for comparing and analyzing the real-time state data of data twin after same frequency update with the historical fault vector in fault vector database, and determining fault prediction result and maintenance strategy based on comparison analysis result;Maintenance management module, for based on fault prediction result and maintenance strategy to equipment is remotely maintained and managed. Precise fault prediction and remote maintenance can be realized, equipment operation and maintenance efficiency is significantly improved, downtime is reduced, and maintenance resource allocation is optimized, which guarantees the effect and reliability of remote maintenance of equipment.

Description

Technical Field

[0001] This invention relates to the field of equipment condition monitoring and management technology, and in particular to a remote equipment maintenance system based on an Internet of Things (IoT) cloud platform. Background Technology

[0002] Currently, with the rapid development of IoT technology, various devices such as industrial equipment and smart home devices are increasingly connected to the network. These devices may experience malfunctions or require maintenance operations such as parameter adjustments during operation.

[0003] Traditional equipment maintenance mainly relies on on-site maintenance personnel for regular maintenance or post-failure repairs. Regular maintenance is costly and may lead to over-maintenance, while post-failure repairs result in long downtimes and significant production losses due to the suddenness of failures, consuming a large amount of manpower and resources. Moreover, existing technologies are mostly limited to local data collection and simple threshold alarms, lacking high-precision virtual models for in-depth analysis, and the prediction and maintenance actions are disconnected, failing to form an automated closed loop, thus greatly reducing the effectiveness of equipment operation and maintenance.

[0004] Therefore, in order to overcome the above-mentioned defects, the present invention provides a remote maintenance system for devices based on an Internet of Things cloud platform. Summary of the Invention

[0005] This invention provides a remote maintenance system for equipment based on an Internet of Things (IoT) cloud platform. It integrates sensor data to build a virtual model that is synchronously updated with the physical equipment in real time. Combined with a historical fault feature database for intelligent analysis, it can achieve accurate fault prediction and remote maintenance, significantly improve equipment operation and maintenance efficiency, reduce downtime, optimize maintenance resource allocation, and ensure the effectiveness and reliability of remote equipment maintenance.

[0006] This invention provides a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, comprising:

[0007] The data acquisition module is used to acquire multi-source heterogeneous data from sensor-based devices;

[0008] The cloud platform processing module is used to build a data twin of the device based on multi-source heterogeneous data, and to update the data twin in sync with the device’s real-time operating data.

[0009] The fault analysis module is used to compare and analyze the real-time status data of the data twin updated at the same frequency with the historical fault vectors in the fault vector database, and determine the fault prediction results and maintenance strategies based on the comparison and analysis results.

[0010] The maintenance management module is used to remotely maintain and manage equipment based on fault prediction results and maintenance strategies.

[0011] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a data acquisition module, comprising:

[0012] The sensor configuration unit is used to acquire the data types collected by the device and configure the corresponding sensors according to the data types.

[0013] The adaptation unit is used to adapt the device and the sensor according to the communication protocol based on the configuration result, and to perform corresponding business monitoring on each device based on the adaptation result to obtain multi-source heterogeneous data corresponding to the device.

[0014] The data processing unit is used to unify the format of the multi-source heterogeneous operating data of the equipment, obtain standard data collected by each sensor, and process the obtained standard data.

[0015] Preferably, an edge data processing unit in a remote maintenance system for devices based on an IoT cloud platform includes:

[0016] The data area is a molecular unit used to extract the data source of each standard data and to classify the obtained standard data based on the data source;

[0017] The data processing subunit is used for:

[0018] Based on the category distinction results, the business attributes of the device under the current category are determined according to the data source, and the data benchmark indicators of the corresponding category are determined based on the business attributes;

[0019] Data cleaning is performed on standard data under the corresponding categories based on data benchmark indicators, and category labels are generated for each category based on the business attributes of each category after data cleaning is completed.

[0020] The standard data of each category after data cleaning is labeled and distinguished based on category labels.

[0021] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a data acquisition module, comprising:

[0022] Data sorting unit, used for:

[0023] Each sensor is assigned a communication interface and communication route based on a preset gateway, and a local area network is built between the sensors based on the assignment results.

[0024] Based on management requirements, the communication priorities of each sensor are sorted in the local area network, and the multi-source heterogeneous data uploaded by different sensors at the same time are sorted into transmission queues based on the sorting results.

[0025] Transmission unit, used for:

[0026] The data traversal is performed on the transmission queue sorting results at each time step, and the communication resources required for uplink communication of the preset gateway are determined based on the data traversal results.

[0027] The system dynamically schedules the communication resources required at different times, and uploads the heterogeneous source data at each time to the cloud platform based on the sorting results of the transmission queue according to the dynamic scheduling results.

[0028] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a cloud platform processing module comprising:

[0029] The data retrieval unit is used to retrieve multi-source heterogeneous data from the equipment and split the multi-source heterogeneous data into equipment structure data and equipment operation data;

[0030] Data twin building blocks, used for:

[0031] The equipment structure data is analyzed to determine the relative positional relationships of different points on the equipment, and a basic framework model of the equipment is constructed based on the relative positional relationships.

[0032] Based on the physical mechanism attributes, the parameters related to physical rules in the equipment structure data are analyzed, and the basic framework model is structurally modified based on the analysis results;

[0033] Simultaneously, the timestamps of the device operation data are extracted, and the device operation data is aligned based on the timestamps;

[0034] The actual operating status of the device at the corresponding time is determined based on the timestamp, and the alignment result is associated with the actual operating status;

[0035] The associated device operation data is analyzed using a deep learning network to extract the action features and the execution amount of each action feature during the device operation process. Based on the action features and the execution amount, the operating behavior pattern of the device in different operating stages is determined.

[0036] Based on operational behavior patterns, construct data-driven models for devices at different operational stages;

[0037] The system acquires the fault monitoring dimensions of the equipment and the monitoring rules corresponding to each fault monitoring dimension based on the management terminal, and constructs the rule model corresponding to the equipment based on the fault monitoring dimensions and monitoring rules.

[0038] The basic framework model, data-driven model, and rule model after structural modification are linked and fused together, and a data twin of the device is obtained based on the linkage and fusion results.

[0039] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a cloud platform processing module comprising:

[0040] Deployment management unit, used for:

[0041] The obtained data twin is used to determine the resources required for operation based on the basic operational requirements of the data twin;

[0042] The necessary resources for operation are scheduled and configured in the cloud platform to build a virtual operating environment for the data twin, and the resulting data twin is deployed on the cloud platform based on the construction results.

[0043] Update snap-in, used for:

[0044] Based on the deployment results, a real-time data flow pipeline is built between the data twin and the device, and the real-time operation data of the device is transmitted to the data twin based on the real-time data flow pipeline;

[0045] The real-time status of the data twin is updated based on the transmission results, and the status synchronization between the data twin and the device is completed based on the update results.

[0046] Preferably, a fault analysis module in a remote maintenance system for devices based on an IoT cloud platform includes:

[0047] The condition monitoring unit is used for:

[0048] Access the script file of the data twin based on the synchronous update result, and extract the mapping parameters of the data twin to different project indicators of the device from the script file of the data twin;

[0049] The mapping parameters of different project indicators are summarized to obtain the real-time status data of the data twin;

[0050] Fault diagnosis unit, used for:

[0051] The real-time status data of the data twin is vectorized to obtain the real-time status vector. At the same time, the fault vector database is accessed to retrieve the historical fault vectors of the equipment.

[0052] The real-time status vector is matched with the historical fault vector based on similarity, and the target historical fault vector is determined based on the similarity matching results.

[0053] The similarity between the real-time state vector and the target historical fault vector is divided into steps based on a preset similarity step threshold. Based on the step division results, the faults corresponding to the target historical fault vectors at the corresponding step thresholds are determined as potential faults or existing faults.

[0054] The maintenance strategy determination unit is used to obtain fault prediction results based on the judgment results, and to determine the maintenance strategy for equipment faults based on the fault prediction results.

[0055] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a maintenance strategy determination unit, comprising:

[0056] The result acquisition subunit is used to acquire the obtained fault prediction results and extract the fault emergency maintenance strategy corresponding to the target historical fault vector based on the fault prediction results.

[0057] Strategy-determining subunits are used for:

[0058] When an existing fault exists, the fault emergency maintenance strategy corresponding to the target historical fault vector is determined as the first maintenance strategy for the equipment fault.

[0059] Meanwhile, when a potential fault exists, the nodes where the real-time status vector differs from the target historical fault vector are identified, and the associated strategy nodes of the nodes where the differences exist in the fault emergency maintenance strategy corresponding to the target historical fault vector are identified.

[0060] Based on the difference nodes, the associated strategy nodes in the fault emergency maintenance strategy are adjusted accordingly, and a second maintenance strategy for equipment faults is obtained based on the difference adjustment results.

[0061] Preferably, a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform includes a maintenance management module comprising:

[0062] Instruction generation unit, used for:

[0063] Based on the fault prediction results, the emergency response equipment to be activated is determined. At the same time, based on the maintenance strategy, the emergency response control items and control parameters of the emergency response equipment to be activated are determined, and the instruction elements are retrieved from the instruction database based on the emergency response control items.

[0064] The execution parameters in the instruction element are adaptively adjusted based on the control parameters, and the control instructions are obtained based on the adaptive adjustment results.

[0065] The remote maintenance management unit is used for:

[0066] The system starts and controls the emergency response equipment to be activated based on control commands, and performs remote maintenance on the equipment based on the startup and control results.

[0067] Meanwhile, remote maintenance is monitored in real time, and if the emergency response equipment fails to resolve the equipment failure, a maintenance notification is sent to the maintenance personnel's smart terminal, and maintenance feedback is received from the maintenance personnel's smart terminal in real time.

[0068] Preferably, a remote maintenance management unit for devices based on an Internet of Things (IoT) cloud platform includes:

[0069] The data acquisition subunit is used to acquire the entire remote maintenance process data of equipment failure after receiving maintenance feedback from the smart terminal of the maintenance personnel, and to generate an index identifier for the entire remote maintenance process data of each equipment failure based on the maintenance time.

[0070] The recording sub-unit is used to associate the index identifier with the remote maintenance process data for each equipment failure and record the association results in the preset maintenance report.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] By integrating sensor data to build a virtual model that is synchronized with the physical equipment in real time, and combining it with a historical fault feature database for intelligent analysis, accurate fault prediction and remote maintenance can be achieved. This significantly improves equipment operation and maintenance efficiency, reduces downtime, optimizes maintenance resource allocation, and ensures the effectiveness and reliability of remote equipment maintenance.

[0073] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0074] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0075] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0076] Figure 1 This is a structural diagram of a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, as described in an embodiment of the present invention.

[0077] Figure 2 This is a structural diagram of a data acquisition module in a remote maintenance system for IoT cloud platform devices according to an embodiment of the present invention;

[0078] Figure 3 This is a structural diagram of a fault analysis module in a remote maintenance system for IoT cloud platform devices according to an embodiment of the present invention. Detailed Implementation

[0079] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0080] Example 1:

[0081] This embodiment provides a remote device maintenance system based on an Internet of Things (IoT) cloud platform, such as... Figure 1 As shown, it includes:

[0082] The data acquisition module is used to acquire multi-source heterogeneous data from sensor-based devices;

[0083] The cloud platform processing module is used to build a data twin of the device based on multi-source heterogeneous data, and to update the data twin in sync with the device’s real-time operating data.

[0084] The fault analysis module is used to compare and analyze the real-time status data of the data twin updated at the same frequency with the historical fault vectors in the fault vector database, and determine the fault prediction results and maintenance strategies based on the comparison and analysis results.

[0085] The maintenance management module is used to remotely maintain and manage equipment based on fault prediction results and maintenance strategies.

[0086] In this embodiment, multi-source heterogeneous data refers to the structural data of the device and the different types of operational data corresponding to the device, such as temperature data, energy consumption data and power data during device operation.

[0087] In this embodiment, the data twin is a dynamic virtual mapping of the actual device.

[0088] In this embodiment, synchronous updates refer to the real-time synchronization of the virtual model with the operating status of the actual device.

[0089] In this embodiment, the fault vector database is a dedicated library for storing historical fault feature data, used to store historical fault vectors, that is, all fault information that has occurred in the equipment in the past.

[0090] The beneficial effects of the above technical solution are: by integrating sensor data to build a virtual model that is synchronously updated with the physical equipment in real time, and combining it with a historical fault feature database for intelligent analysis, accurate fault prediction and remote maintenance can be achieved, significantly improving equipment operation and maintenance efficiency, reducing downtime, optimizing maintenance resource allocation, and ensuring the effectiveness and reliability of remote equipment maintenance.

[0091] Example 2:

[0092] Based on Example 1, this example provides a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, such as... Figure 2 As shown, the data acquisition module includes:

[0093] The sensor configuration unit is used to acquire the data types collected by the device and configure the corresponding sensors according to the data types.

[0094] The adaptation unit is used to adapt the device and the sensor according to the communication protocol based on the configuration result, and to perform corresponding business monitoring on each device based on the adaptation result to obtain multi-source heterogeneous data corresponding to the device.

[0095] The data processing unit is used to unify the format of the multi-source heterogeneous operating data of the equipment, obtain standard data collected by each sensor, and process the obtained standard data.

[0096] In this embodiment, adapting the device and sensor according to the communication protocol refers to configuring the communication link and communication route between the device and the sensor according to the known communication protocol.

[0097] In this embodiment, standard data refers to standardized data that has undergone format unification and processing.

[0098] The beneficial effects of the above technical solution are: by automatically identifying collection needs and flexibly configuring sensors, it can be compatible with multiple communication protocols to achieve rapid device access and data aggregation, and perform standardized cleaning and processing on raw data, thereby significantly improving the efficiency and accuracy of data collection and providing a high-quality, standardized data foundation.

[0099] Example 3:

[0100] Based on Embodiment 2, this embodiment provides a remote maintenance system for IoT cloud platform devices, including an edge data processing unit:

[0101] The data area is a molecular unit used to extract the data source of each standard data and to classify the obtained standard data based on the data source;

[0102] The data processing subunit is used for:

[0103] Based on the category distinction results, the business attributes of the device under the current category are determined according to the data source, and the data benchmark indicators of the corresponding category are determined based on the business attributes;

[0104] Data cleaning is performed on standard data under the corresponding categories based on data benchmark indicators, and category labels are generated for each category based on the business attributes of each category after data cleaning is completed.

[0105] The standard data of each category after data cleaning is labeled and distinguished based on category labels.

[0106] In this embodiment, the data source refers to the specific sensor corresponding to different standard data.

[0107] In this embodiment, business attributes refer to the business types corresponding to different categories of standard data, as well as the corresponding data structure characteristics and value range limitations.

[0108] In this embodiment, the data benchmark index refers to the data quality and standardization standards formulated based on different business attributes.

[0109] In this embodiment, the category label is a marker symbol used to identify the business category to which the data belongs after cleaning.

[0110] The beneficial effects of the above technical solution are: by automatically identifying and classifying the sources of standardized data, and by setting benchmark indicators based on the business characteristics of various types of data for intelligent cleaning and labeling, data quality and consistency can be effectively improved, manual intervention can be reduced, and a clear, reliable and highly structured data foundation can be provided for subsequent analysis and application.

[0111] Example 4:

[0112] Based on Example 1, this example provides a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, including a data acquisition module comprising:

[0113] Data sorting unit, used for:

[0114] Each sensor is assigned a communication interface and communication route based on a preset gateway, and a local area network is built between the sensors based on the assignment results.

[0115] Based on management requirements, the communication priorities of each sensor are sorted in the local area network, and the multi-source heterogeneous data uploaded by different sensors at the same time are sorted into transmission queues based on the sorting results.

[0116] Transmission unit, used for:

[0117] The data traversal is performed on the transmission queue sorting results at each time step, and the communication resources required for uplink communication of the preset gateway are determined based on the data traversal results.

[0118] The system dynamically schedules the communication resources required at different times, and uploads the heterogeneous source data at each time to the cloud platform based on the sorting results of the transmission queue according to the dynamic scheduling results.

[0119] In this embodiment, the preset gateway refers to a pre-configured network gateway device responsible for connecting the sensor network and the upper-layer cloud platform.

[0120] In this embodiment, the management requirements are known in advance and are used to define the communication priorities of each sensor.

[0121] In this embodiment, transmission queue sorting refers to the sequential arrangement of data packets to be uploaded according to priority rules.

[0122] In this embodiment, the preset gateway uplink communication refers to the process of transmitting data to the cloud platform through the preset gateway.

[0123] In this embodiment, communication resources refer to bandwidth and other resources required for data transmission.

[0124] The beneficial effects of the above technical solution are: by intelligently allocating network paths and building an internal communication network, it is possible to automatically sort data transmission priorities according to management needs and dynamically schedule uplink bandwidth resources, thereby significantly improving the efficiency and timeliness of multi-source data upload, ensuring the real-time transmission of critical data, and optimizing network resource utilization.

[0125] Example 5:

[0126] Based on Example 1, this example provides a remote maintenance system for devices based on an IoT cloud platform, including a cloud platform processing module:

[0127] The data retrieval unit is used to retrieve multi-source heterogeneous data from the equipment and break down the multi-source heterogeneous data into equipment structure data and equipment operation data;

[0128] Data twin building blocks, used for:

[0129] The equipment structure data is analyzed to determine the relative positional relationships of different points on the equipment, and a basic framework model of the equipment is constructed based on the relative positional relationships.

[0130] Based on the physical mechanism attributes, the parameters related to physical rules in the equipment structure data are analyzed, and the basic framework model is structurally modified based on the analysis results;

[0131] Simultaneously, the timestamps of the device operation data are extracted, and the device operation data is aligned based on the timestamps;

[0132] The actual operating status of the device at the corresponding time is determined based on the timestamp, and the alignment result is associated with the actual operating status;

[0133] The associated device operation data is analyzed using a deep learning network to extract the action features and the execution amount of each action feature during the device operation process. Based on the action features and the execution amount, the operating behavior pattern of the device in different operating stages is determined.

[0134] Based on operational behavior patterns, construct data-driven models for devices at different operational stages;

[0135] The system acquires the fault monitoring dimensions of the equipment and the corresponding monitoring rules for each fault monitoring dimension based on the management terminal, and constructs the corresponding rule model for the equipment based on the fault monitoring dimensions and monitoring rules.

[0136] The basic framework model, data-driven model, and rule model after structural modification are linked and fused together, and a data twin of the device is obtained based on the linkage and fusion results.

[0137] In this embodiment, the basic framework model refers to the initial three-dimensional digital framework composed of the physical structure of the equipment and the positional relationships between its components.

[0138] In this embodiment, physical mechanism attributes are the basis or standards that need to be followed to characterize the actual physical connection relationship between different device components.

[0139] In this embodiment, the deep learning network is pre-trained.

[0140] In this embodiment, action characteristics refer to the specific steps or operations performed by the device during operation.

[0141] In this embodiment, the amount of action execution refers to the specific degree of execution corresponding to each action feature of the device, such as the rotation speed per second.

[0142] In this embodiment, the operation phase includes the equipment startup phase, the normal operation phase, and the shutdown phase.

[0143] In this embodiment, the operating behavior mode refers to the specific operating conditions or operating schemes of the device at different operating stages.

[0144] In this embodiment, the data-driven model refers to a model composed of features and behavioral patterns extracted from historical operational data through deep learning.

[0145] In this embodiment, the rule model refers to a logical judgment model constructed based on equipment management experience and fault diagnosis rules.

[0146] The beneficial effects of the above technical solution are as follows: by deeply integrating the static structural information and dynamic operation data of the equipment, and combining physical mechanisms and data-driven analysis, a high-precision, dynamically updatable virtual model can be automatically constructed. This model can not only accurately reflect the real-time status and behavior patterns of the equipment, but also perform multi-dimensional fault monitoring based on preset rules. This enables the accurate and effective construction of a data twin of the equipment, thereby achieving a comprehensive understanding, accurate prediction, and intelligent decision support for the health status of the equipment, significantly improving the level of intelligence in equipment management and operational efficiency.

[0147] Example 6:

[0148] Based on Example 1, this example provides a remote maintenance system for devices based on an IoT cloud platform, including a cloud platform processing module:

[0149] Deployment management unit, used for:

[0150] The obtained data twin is used to determine the resources required for operation based on the basic operational requirements of the data twin;

[0151] The necessary resources for operation are scheduled and configured in the cloud platform to build a virtual operating environment for the data twin, and the resulting data twin is deployed on the cloud platform based on the construction results.

[0152] Update snap-in, used for:

[0153] Based on the deployment results, a real-time data flow pipeline is built between the data twin and the device, and the real-time operation data of the device is transmitted to the data twin based on the real-time data flow pipeline;

[0154] The real-time status of the data twin is updated based on the transmission results, and the status synchronization between the data twin and the device is completed based on the update results.

[0155] In this embodiment, the basic operating requirements refer to the standards or configurations that the data twin needs to achieve during runtime, such as the running speed of the algorithm.

[0156] In this embodiment, the real-time data stream pipeline refers to a dedicated communication channel used for continuous, low-latency transmission of real-time operating data from the device to the data twin.

[0157] The beneficial effects of the above technical solution are: through automated deployment and dynamic resource allocation, a stable and efficient operating environment can be quickly created for the virtual model, and a real-time data channel can be established to ensure continuous synchronization between physical devices and virtual models, thereby significantly improving the model deployment efficiency and resource utilization, ensuring that the virtual model always truly reflects the latest status of the equipment, and providing timely and reliable data support for remote monitoring and decision-making.

[0158] Example 7:

[0159] Based on Example 1, this example provides a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, such as... Figure 3 As shown, the fault analysis module includes:

[0160] The condition monitoring unit is used for:

[0161] Access the script file of the data twin based on the synchronous update result, and extract the mapping parameters of the data twin to different project indicators of the device from the script file of the data twin;

[0162] The mapping parameters of different project indicators are summarized to obtain the real-time status data of the data twin;

[0163] Fault diagnosis unit, used for:

[0164] The real-time status data of the data twin is vectorized to obtain the real-time status vector. At the same time, the fault vector database is accessed to retrieve the historical fault vectors of the equipment.

[0165] The real-time status vector is matched with the historical fault vector based on similarity, and the target historical fault vector is determined based on the similarity matching results.

[0166] The similarity between the real-time state vector and the target historical fault vector is divided into steps based on a preset similarity step threshold. Based on the step division results, the faults corresponding to the target historical fault vectors at the corresponding step thresholds are determined as potential faults or existing faults.

[0167] The maintenance strategy determination unit is used to obtain fault prediction results based on the judgment results, and to determine the maintenance strategy for equipment faults based on the fault prediction results.

[0168] In this embodiment, the script file refers to the program file corresponding to the data twin in the computer.

[0169] In this embodiment, the mapping parameter refers to the limit value of the data twin for different indicators of the device.

[0170] In this embodiment, the real-time state vector refers to the digital feature representation composed of the current operating indicators of the device.

[0171] In this embodiment, the historical fault vector refers to the feature data extracted from past fault cases that is used to characterize a specific fault mode.

[0172] In this embodiment, the target historical fault vector refers to the historical fault vector that meets the similarity requirement with the real-time state vector after the real-time state vector and the historical fault vector are matched for similarity.

[0173] In this embodiment, the preset similarity ladder threshold is a pre-set standard used to divide similarity into ladders, thereby determining the similarity level.

[0174] In this embodiment, determining the fault corresponding to the target historical fault vector of the corresponding step threshold as a potential fault or an existing fault based on the step division result refers to determining the similarity level corresponding to different threshold intervals according to the preset similarity step threshold. For example, when the similarity is at the first step threshold, that is, the real-time state vector is completely consistent with the historical fault vector, it is determined that a fault has occurred (i.e., an existing fault). When the similarity is at other step thresholds, that is, the real-time state vector is not completely consistent with the historical fault vector, it is determined that a potential fault exists.

[0175] In this embodiment, determining the maintenance strategy for equipment failure based on the failure prediction result means retrieving the maintenance strategy or plan corresponding to the historical failure as the specific plan for maintaining the current equipment failure.

[0176] The beneficial effects of the above technical solution are: by automatically extracting key state parameters of the virtual model and intelligently comparing and classifying them with the historical fault feature library, it is possible to achieve early and accurate prediction and diagnosis of equipment faults and automatically generate corresponding maintenance strategies, thereby significantly improving the automation level of equipment condition monitoring, the accuracy of fault identification and the efficiency of maintenance response.

[0177] Example 8:

[0178] Based on Embodiment 7, this embodiment provides a remote maintenance system for devices based on an Internet of Things cloud platform, including a maintenance strategy determination unit, comprising:

[0179] The result acquisition subunit is used to acquire the obtained fault prediction results and extract the fault emergency maintenance strategy corresponding to the target historical fault vector based on the fault prediction results.

[0180] Strategy-determining subunits are used for:

[0181] When an existing fault exists, the fault emergency maintenance strategy corresponding to the target historical fault vector is determined as the first maintenance strategy for the equipment fault.

[0182] Meanwhile, when a potential fault exists, the nodes where the real-time status vector differs from the target historical fault vector are identified, and the associated strategy nodes of the nodes where the differences exist in the fault emergency maintenance strategy corresponding to the target historical fault vector are identified.

[0183] Based on the difference nodes, the associated strategy nodes in the fault emergency maintenance strategy are adjusted accordingly, and a second maintenance strategy for equipment faults is obtained based on the difference adjustment results.

[0184] In this embodiment, the fault emergency maintenance strategy refers to the standard maintenance operation plan that has been formulated in advance for historical faults.

[0185] In this embodiment, the difference node refers to a specific parameter or indicator that shows a difference between the current device status and historical fault characteristics.

[0186] In this embodiment, the associated strategy node refers to the specific parameters or indicators associated with the difference node in the fault emergency maintenance strategy corresponding to the target historical fault vector.

[0187] The beneficial effects of the above technical solution are: it can automatically call historical emergency plans based on the fault prediction conclusions, and dynamically generate or precisely adjust maintenance strategies according to the different characteristics of existing and potential faults, significantly improving the intelligence and refinement of maintenance response, ensuring the timeliness and applicability of strategies, effectively preventing fault escalation and optimizing maintenance resource investment.

[0188] Example 9:

[0189] Based on Example 1, this example provides a remote maintenance system for devices based on an IoT cloud platform, including a maintenance management module:

[0190] Instruction generation unit, used for:

[0191] Based on the fault prediction results, the emergency response equipment to be activated is determined. At the same time, based on the maintenance strategy, the emergency response control items and control parameters of the emergency response equipment to be activated are determined, and the instruction elements are retrieved from the instruction database based on the emergency response control items.

[0192] The execution parameters in the instruction element are adaptively adjusted based on the control parameters, and the control instructions are obtained based on the adaptive adjustment results.

[0193] The remote maintenance management unit is used for:

[0194] The system starts and controls the emergency response equipment to be activated based on control commands, and performs remote maintenance on the equipment based on the startup and control results.

[0195] Meanwhile, remote maintenance is monitored in real time, and if the emergency response equipment fails to resolve the equipment failure, a maintenance notification is sent to the maintenance personnel's smart terminal, and maintenance feedback is received from the maintenance personnel's smart terminal in real time.

[0196] In this embodiment, the emergency response control item refers to the specific business that controls the emergency response equipment to be activated, such as increasing the rotation speed or cooling down.

[0197] In this embodiment, the instruction element refers to a pre-set, callable basic instruction template.

[0198] In this embodiment, the execution parameter refers to the specific execution parameter defined by the instruction element, such as the adjustment amount of the rotation speed.

[0199] In this embodiment, the control parameter refers to the specific adjustment value or parameter determined based on the actual fault situation.

[0200] The beneficial effects of the above technical solution are: it can automatically generate and issue precise control commands based on diagnostic conclusions, realize intelligent start-up and remote control of emergency equipment, and conduct real-time tracking and intelligent early warning during maintenance. When automatic failure is handled, it can promptly notify manual intervention, thereby greatly improving the automation level, execution efficiency and reliability of remote maintenance, and effectively ensuring the stable operation of equipment.

[0201] Example 10:

[0202] Based on Example 9, this example provides a remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, including a remote maintenance management unit comprising:

[0203] The data acquisition subunit is used to acquire the entire remote maintenance process data of equipment failure after receiving maintenance feedback from the smart terminal of the maintenance personnel, and to generate an index identifier for the entire remote maintenance process data of each equipment failure based on the maintenance time.

[0204] The recording sub-unit is used to associate the index identifier with the remote maintenance process data for each equipment failure and record the association results in the preset maintenance report.

[0205] In this embodiment, remote maintenance full-process data refers to all data parameters when performing operation and maintenance management on equipment faults.

[0206] In this embodiment, the index identifier refers to a keyword generated based on the maintenance time, used to uniquely identify and quickly retrieve a maintenance record.

[0207] In this embodiment, remote maintenance full-process data refers to the complete process data from fault occurrence, diagnosis, strategy formulation to remote operation and feedback.

[0208] In this embodiment, the preset maintenance report refers to a standardized report template with pre-designed format and content.

[0209] The beneficial effects of the above technical solution are: by automatically collecting and indexing the information of each remote maintenance process, a complete and traceable maintenance archive can be established, significantly improving the organization efficiency and retrieval speed of operation and maintenance data, providing reliable data support for subsequent fault analysis, strategy optimization and performance evaluation, while reducing manual recording errors and improving the level of intelligent knowledge management.

[0210] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A remote maintenance system for devices based on an Internet of Things (IoT) cloud platform, characterized in that, include: The data acquisition module is used to acquire multi-source heterogeneous data from sensor-based devices; The cloud platform processing module is used to build a data twin of the device based on multi-source heterogeneous data, and to update the data twin in sync with the device’s real-time operating data. The fault analysis module is used to compare and analyze the real-time status data of the data twin updated at the same frequency with the historical fault vectors in the fault vector database, and determine the fault prediction results and maintenance strategies based on the comparison and analysis results. The maintenance management module is used for remote maintenance management of equipment based on fault prediction results and maintenance strategies; The cloud platform processing module includes: The data retrieval unit is used to retrieve multi-source heterogeneous data from the equipment and split the multi-source heterogeneous data into equipment structure data and equipment operation data; Data twin building blocks, used for: The equipment structure data is analyzed to determine the relative positional relationships of different points on the equipment, and a basic framework model of the equipment is constructed based on the relative positional relationships. Based on the physical mechanism attributes, the parameters related to physical rules in the equipment structure data are analyzed, and the basic framework model is structurally modified based on the analysis results; Simultaneously, the timestamps of the device operation data are extracted, and the device operation data is aligned based on the timestamps; The actual operating status of the device at the corresponding time is determined based on the timestamp, and the alignment result is associated with the actual operating status; The associated device operation data is analyzed using a deep learning network to extract the action features and the execution amount of each action feature during the device operation process. Based on the action features and the execution amount, the operating behavior pattern of the device in different operating stages is determined. Based on operational behavior patterns, construct data-driven models for devices at different operational stages; The system acquires the fault monitoring dimensions of the equipment and the monitoring rules corresponding to each fault monitoring dimension based on the management terminal, and constructs the rule model corresponding to the equipment based on the fault monitoring dimensions and monitoring rules. The basic framework model, data-driven model, and rule model after structural modification are linked and fused together, and a data twin of the device is obtained based on the linkage and fusion results.

2. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 1, characterized in that, The data acquisition module includes: The sensor configuration unit is used to acquire the data types collected by the device and configure the corresponding sensors according to the data types. The adaptation unit is used to adapt the device and the sensor according to the communication protocol based on the configuration result, and to perform corresponding business monitoring on each device based on the adaptation result to obtain multi-source heterogeneous data corresponding to the device. The data processing unit is used to unify the format of the multi-source heterogeneous operating data of the equipment, obtain standard data collected by each sensor, and process the obtained standard data.

3. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 2, characterized in that, The edge data processing unit includes: The data area is a molecular unit used to extract the data source of each standard data and to classify the obtained standard data based on the data source; The data processing subunit is used for: Based on the category distinction results, the business attributes of the device under the current category are determined according to the data source, and the data benchmark indicators of the corresponding category are determined based on the business attributes; Data cleaning is performed on standard data under the corresponding categories based on data benchmark indicators, and category labels are generated for each category based on the business attributes of each category after data cleaning is completed. The standard data of each category after data cleaning is labeled and distinguished based on category labels.

4. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 1, characterized in that, The data acquisition module includes: Data sorting unit, used for: Each sensor is assigned a communication interface and communication route based on a preset gateway, and a local area network is built between the sensors based on the assignment results. Based on management requirements, the communication priorities of each sensor are sorted in the local area network, and the multi-source heterogeneous data uploaded by different sensors at the same time are sorted into transmission queues based on the sorting results. Transmission unit, used for: The data traversal is performed on the transmission queue sorting results at each time step, and the communication resources required for uplink communication of the preset gateway are determined based on the data traversal results. The system dynamically schedules the communication resources required at different times, and uploads the heterogeneous source data at each time to the cloud platform based on the sorting results of the transmission queue according to the dynamic scheduling results.

5. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 1, characterized in that, The cloud platform processing module includes: Deployment management unit, used for: The obtained data twin is used to determine the resources required for operation based on the basic operational requirements of the data twin; The necessary resources for operation are scheduled and configured in the cloud platform to build a virtual operating environment for the data twin, and the resulting data twin is deployed on the cloud platform based on the construction results. Update snap-in, used for: Based on the deployment results, a real-time data flow pipeline is built between the data twin and the device, and the real-time operation data of the device is transmitted to the data twin based on the real-time data flow pipeline; The real-time status of the data twin is updated based on the transmission results, and the status synchronization between the data twin and the device is completed based on the update results.

6. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 1, characterized in that, The fault analysis module includes: The condition monitoring unit is used for: Access the script file of the data twin based on the synchronous update result, and extract the mapping parameters of the data twin to different project indicators of the device from the script file of the data twin; The mapping parameters of different project indicators are summarized to obtain the real-time status data of the data twin; The fault diagnosis unit is used for: The real-time status data of the data twin is vectorized to obtain the real-time status vector. At the same time, the fault vector database is accessed to retrieve the historical fault vectors of the equipment. The real-time status vector is matched with the historical fault vector based on similarity, and the target historical fault vector is determined based on the similarity matching results. The similarity between the real-time state vector and the target historical fault vector is divided into steps based on a preset similarity step threshold. Based on the step division results, the faults corresponding to the target historical fault vectors at the corresponding step thresholds are determined as potential faults or existing faults. The maintenance strategy determination unit is used to obtain fault prediction results based on the judgment results, and to determine the maintenance strategy for equipment faults based on the fault prediction results.

7. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 6, characterized in that, The maintenance strategy determination unit includes: The result acquisition subunit is used to acquire the obtained fault prediction results and extract the fault emergency maintenance strategy corresponding to the target historical fault vector based on the fault prediction results. Strategy-determining subunits are used for: When an existing fault exists, the fault emergency maintenance strategy corresponding to the target historical fault vector is determined as the first maintenance strategy for the equipment fault. Meanwhile, when a potential fault exists, the nodes where the real-time status vector differs from the target historical fault vector are identified, and the associated strategy nodes of the nodes where the differences exist in the fault emergency maintenance strategy corresponding to the target historical fault vector are identified. Based on the difference nodes, the associated strategy nodes in the fault emergency maintenance strategy are adjusted accordingly, and a second maintenance strategy for equipment faults is obtained based on the difference adjustment results.

8. The remote maintenance system for devices based on an Internet of Things cloud platform according to claim 1, characterized in that, The maintenance and management module includes: Instruction generation unit, used for: Based on the fault prediction results, the emergency response equipment to be activated is determined. At the same time, based on the maintenance strategy, the emergency response control items and control parameters of the emergency response equipment to be activated are determined, and the instruction elements are retrieved from the instruction database based on the emergency response control items. The execution parameters in the instruction element are adaptively adjusted based on the control parameters, and the control instructions are obtained based on the adaptive adjustment results. The remote maintenance management unit is used for: The system starts and controls the emergency response equipment to be activated based on control commands, and performs remote maintenance on the equipment based on the startup and control results. Meanwhile, remote maintenance is monitored in real time, and if the emergency response equipment fails to resolve the equipment failure, a maintenance notification is sent to the maintenance personnel's smart terminal, and maintenance feedback is received from the maintenance personnel's smart terminal in real time.

9. A remote maintenance system for devices based on an Internet of Things cloud platform according to claim 8, characterized in that, The remote maintenance management unit includes: The data acquisition subunit is used to acquire the entire remote maintenance process data of equipment failure after receiving maintenance feedback from the smart terminal of the maintenance personnel, and to generate an index identifier for the entire remote maintenance process data of each equipment failure based on the maintenance time. The recording sub-unit is used to associate the index identifier with the remote maintenance process data for each equipment failure and record the association results in the preset maintenance report.

Images