Intelligent management method and system for equipment operation and maintenance

Abstract

The invention discloses an intelligent management method and system for equipment operation and maintenance, relates to the technical field of data management, and solves the problems that equipment data are difficult to analyze from multiple angles, an equipment relation graph and potential fault data relevance analysis are lacked, and the relevance between fault equipment and other equipment cannot be clearly known, so that the equipment operation and maintenance efficiency is improved. And the influence range of the fault is difficult to accurately evaluate. A multi-dimensional sensor, an operation log collector and an edge computing node are deployed on an edge layer, so that equipment state data, equipment operation video stream data and operation log data can be primarily processed and analyzed locally. A multi-dimensional data analysis model for risk data and key equipment node data is established in a cloud layer, and equipment data can be analyzed from multiple angles. The influence range of the corresponding potential fault is analyzed through the equipment relation graph, and the incidence relation between the fault equipment and other equipment can be clearly known.

Description

Technical Field

[0001] The present invention belongs to the field of data management, and specifically relates to an intelligent management method and system for equipment operation and maintenance. Background Art

[0002] Traditional equipment operation and maintenance relies heavily on manual inspections, requiring personnel to regularly visit equipment for inspections. For example, in large factories, where equipment is numerous and widely distributed, inspectors spend a significant amount of time traversing the factory floor, checking each device's operating status. For example, in a factory with hundreds of devices, a comprehensive inspection can take days. This is not only inefficient, but can also lead to poor inspection quality due to factors like fatigue, resulting in missed or false detections. When equipment malfunctions occur, traditional operations often rely on on-site personnel to detect and report them. If a malfunction occurs during off-hours or in a remote area, it may not be addressed promptly.

[0003] The rise of the Internet of Things (IoT) provides a foundation for intelligent equipment operation and maintenance. Therefore, an intelligent equipment operation and maintenance management solution is needed. By installing various sensors on equipment, real-time equipment operating data can be collected. Using big data technology, this massive amount of equipment operation and maintenance data can be stored, managed, and analyzed.

[0004] Existing intelligent equipment operation and maintenance management solutions struggle to analyze device data from multiple perspectives. They rely solely on single factors or simple rules to identify faults, which can easily lead to misjudgments or omissions. The lack of a device relationship map and analysis of the correlations between potential fault data prevents a clear understanding of the relationships between the faulty device and other devices, making it difficult to accurately assess the scope of the fault's impact. For example, in a complex production system, a failure in a key device could impact the normal operation of other connected devices. However, the lack of a device relationship map makes it impossible to quickly determine the scope of affected devices. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent management method and system for equipment operation and maintenance, which is used to solve the technical problems that it is difficult to analyze equipment data from multiple angles, which is prone to misjudgment or omission, and there is a lack of equipment relationship maps and analysis of the correlation of potential fault data, which makes it impossible to clearly understand the correlation between the faulty equipment and other equipment, and it is difficult to accurately assess the impact scope of the fault.

[0006] To solve the above problems, the first aspect of the present invention provides an intelligent management method and system for equipment operation and maintenance, including: Deploy multi-dimensional sensors, operation log collectors, and edge computing nodes at the edge layer. The edge computing nodes manage the device links within the signal connection range, and set up a time series database on the edge computing nodes to store device status data, device operation video stream data, and operation log data. The edge computing nodes establish a device relationship map based on the device nodes of the device links they manage, and screen key device nodes. A risk data detection model is established at the edge layer to analyze device data in the time series database. The risk data and key device node data are uploaded to the cloud layer. The cloud layer extracts the hash value of the operation log of the key device node and stores it on the blockchain. Establish a multi-dimensional data analysis model for risk data and key equipment node data at the cloud layer to analyze equipment data, identify potential fault data of equipment nodes, and generate common maintenance work orders; The impact range of potential faults is analyzed through the device relationship map. At the same time, when potential fault data is detected in multiple device nodes in the same device link, the correlation of the potential fault data is analyzed. When the fault impact range is greater than the threshold or there is correlation between the potential fault data, an alarm is triggered and an expedited maintenance work order is generated.

[0007] Optionally, in an example of the above aspect, establishing a device relationship graph based on the device nodes of the device links managed by the edge computing node and screening key device nodes includes the following steps: Add device nodes to the device relationship graph in the device link and add device attribute data, as well as device load and alarm records; Establishing edges between device nodes that have connection or data transmission relationships; Calculate the degree centrality and betweenness centrality of each device node in the device relationship graph, and take the weighted average of the degree centrality and betweenness centrality to obtain the importance coefficient of the device node; Among them, degree centrality Cd=d(v) / (N−1), N−1 is the number of directly connected edges of node v, and N is the total number of nodes in the graph; betweenness centrality ,σst is the total number of all paths from node s to t, σst(v) is the number of paths from node s to t that pass through node v; In the same device link, the importance coefficients of the device nodes are sorted from high to low, and the top 10% of the device nodes are selected as key device nodes.

[0008] Optionally, in an example of the above aspect, establishing a risk data detection model at the edge layer and analyzing device data in a time series database includes the following steps: Establish a risk data detection model at the edge layer, including three-level risk detection and analysis; A three-level risk detection and analysis detection strategy is set up, including: the three levels detect the device's operation log data, device status data, and device operation video stream data in turn. If risk data is detected at any level, the data of the device in the same period will be judged as risk data, and the detection is completed. Otherwise, the risk detection and analysis of the next level will be continued.

[0009] Optionally, in an example of the above aspect, the first-level risk detection includes: The edge computing node obtains the standard operating procedure data from the device attribute data of the management device node, extracts the operation keywords and timestamps of the operation log data in the time series database, arranges the operation keywords according to the operation time sequence, generates the actual operation process, compares the actual operation process with the standard operation process data, and detects the actual operation process in the standard operation process data; The second level of risk detection includes: using a machine learning model pre-trained with historical fault data to detect abnormalities in equipment status data and determine risk data; The third level of risk detection includes: using the FasterR-CNN recognition model pre-trained with historical operating image data from equipment failures, capturing images of the equipment operation video stream data every 50 frames, and detecting whether there are fault points in the captured image data; and determining risk data based on whether there are fault points.

[0010] Optionally, in an example of the above aspect, the cloud layer extracts the hash value of the operation log of the key device node and stores it in the blockchain, including the following steps: Extract the operation logs of key equipment nodes and merge the operation logs whose operation time interval is less than the threshold; Add a hash value tag = Hash (operation content + timestamp + device ID) to the operation log and store it in the blockchain.

[0011] Optionally, in an example of the above aspect, a multi-dimensional data analysis model for risk data and key device node data is established at the cloud layer to analyze the device data and identify potential fault data of the device nodes, including the following steps: Targeting risk data and key equipment node data, a multi-dimensional data analysis model for equipment nodes is established, including: a pre-trained Faster R-CNN recognition model replicated by the trained edge layer, an operation log data risk factor check model, and a multi-dimensional analysis algorithm for equipment status data; Based on the detection results of the model and algorithm, a weighted average is performed. If the weighted average result is greater than the threshold, it is determined that there is a potential fault in the device node. Otherwise, it is determined that there is no potential fault.

[0012] Optionally, in an example of the above aspect, analyzing the impact range of the corresponding potential fault through a device relationship map includes the following steps: Extract the data of device node failures with a time interval less than a preset threshold from the historical fault data of the same device link, count the number of failures Sn of the corresponding device node, and the number of failures Sc of the device node with a time interval less than the preset threshold with the corresponding device node, and calculate the number of failed nodes in the corresponding failure times, and calculate the mean value pc of the number of failed nodes in the corresponding failure times. Obtain the importance coefficient of the corresponding device node in the device relationship graph and analyze the impact range coefficient of the corresponding potential fault: Among them, Pb is the impact range coefficient of the corresponding potential fault, p0 is the total number of devices in the same device link, Ft is the importance coefficient of the corresponding device node in the device relationship graph, and Fu is the analysis result of the multi-dimensional data analysis model of the device node.

[0013] Optionally, in an example of the above aspect, when potential fault data is detected in multiple device nodes in the same device link, analyzing the correlation of the potential fault data includes the following steps: By extracting historical fault data of the same device link, the data whose device node failure time interval is less than the preset threshold is found, and the data whose failure time interval is less than the preset threshold is marked with correlation; The LSTM time series model is trained with labeled data, and the LSTM time series model is used to detect whether there is a correlation between the potential fault data of device nodes.

[0014] According to another aspect of the present disclosure, there is provided an intelligent equipment operation and maintenance management system, characterized in that the system adopts the above-mentioned intelligent equipment operation and maintenance management method to realize intelligent equipment operation and maintenance management. Compared with the prior art, the present invention has the following beneficial effects: The present invention establishes a multi-dimensional data analysis model for risk data and key equipment node data at the cloud layer, which can analyze equipment data from multiple angles. By comprehensively analyzing these data, the potential fault data of the equipment node can be identified more accurately, avoiding misjudgments or omissions that may be caused by single-factor analysis. By analyzing the impact range of the corresponding potential fault through the equipment relationship map, the relationship between the faulty equipment and other equipment can be clearly understood. Accurately assessing the impact range of the fault helps operation and maintenance personnel to reasonably arrange maintenance resources and time. For faults with a smaller impact range, maintenance can be arranged during non-production peak periods; for faults with a larger impact range, immediate measures need to be taken to deal with them to reduce the impact on production.

[0015] The present invention deploys multi-dimensional sensors, operation log collectors, and edge computing nodes at the edge layer, enabling preliminary local processing and analysis of device status data, device operation video stream data, and operation log data. The edge computing nodes perform preliminary screening and analysis of the data, uploading only risk data and key device node data to the cloud layer. This reduces the amount of data uploaded to the cloud, reduces the data processing pressure and network bandwidth usage on the cloud, and improves the operating efficiency of the entire system. Establishing a device relationship map based on the device nodes of the managed device links can clearly display the associations between devices, facilitating subsequent analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 Schematic diagram of the process of the present invention; Figure 2 The figure is a flow chart of the method for identifying potential fault data of a device node according to the present invention. DETAILED DESCRIPTION

[0018] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] See also Figure 1-Figure 2 The first embodiment of the present invention provides an intelligent management method and system for equipment operation and maintenance, including: Deploy multi-dimensional sensors, operation log collectors, and edge computing nodes at the edge layer. Edge computing nodes manage device links within the signal connection range and set up an InfluxDB time series database within the edge computing nodes to store device status data, device operation video stream data, and operation log data. Establish a device relationship map based on the device nodes in the device links managed by the edge computing nodes to screen key device nodes. A risk data detection model is established at the edge layer to analyze device data in the time series database. The risk data and key device node data are uploaded to the cloud layer. The cloud layer extracts the hash value of the operation log of the key device node and stores it on the blockchain. Establish a multi-dimensional data analysis model for risk data and key equipment node data at the cloud layer to analyze equipment data, identify potential fault data of equipment nodes, and generate common maintenance work orders; The impact range of potential faults is analyzed through the device relationship map. At the same time, when potential fault data is detected in multiple device nodes in the same device link, the correlation of the potential fault data is analyzed. When the fault impact range is greater than the threshold or there is correlation between the potential fault data, an alarm is triggered and an expedited maintenance work order is generated.

[0020] Specifically, in this embodiment, multi-dimensional sensors, operation log collectors and edge computing nodes are deployed at the edge layer, so that device status data, device operation video stream data and operation log data can be preliminarily processed and analyzed locally.

[0021] The edge computing nodes perform preliminary screening and analysis of the data, and only upload risk data and key equipment node data to the cloud layer, reducing the amount of data uploaded to the cloud, reducing the data processing pressure and network bandwidth usage on the cloud, and improving the operating efficiency of the entire system.

[0022] By creating a device relationship map based on the device nodes in the managed device links, you can clearly display the relationships between devices. This helps analyze the overall system architecture and the dependencies between devices, facilitating more accurate impact analysis later. By screening out key device nodes, you can focus on analyzing the operating status of key devices, rationally allocate O&M resources, and improve O&M efficiency.

[0023] Establishing a risk data detection model at the edge layer enables timely detection of potential risks during equipment operation. Because detection is performed locally, it's unaffected by factors like network latency, enabling faster risk identification and actionable measures. InfluxDB, a time series database specifically designed for storing and processing time series data, is deployed at edge computing nodes to store device data. This database offers efficient data writing and querying. This allows for convenient querying and analysis of historical device data, providing strong support for fault diagnosis, performance evaluation, and predictive maintenance.

[0024] The cloud layer extracts hash values ​​from the operation logs of key equipment nodes and stores them on the blockchain, which is decentralized, immutable, and traceable. Storing the hash values ​​of operation logs on the blockchain ensures their integrity and authenticity, preventing data tampering. For example, in the event of a device failure or safety incident, the authenticity of the operation logs can be verified using the hash values ​​on the blockchain, providing a reliable basis for accident investigation and accountability determination.

[0025] The cloud layer receives risk data and key equipment node data uploaded by the edge layer, enabling more comprehensive and in-depth analysis and processing. Comprehensive analysis of data uploaded by multiple edge computing nodes can identify potential equipment operation issues, providing a more scientific basis for enterprise equipment operation and maintenance decisions.

[0026] A multi-dimensional data analysis model for risk data and key equipment node data is established at the cloud layer, enabling analysis of equipment data from multiple perspectives. This comprehensive analysis of this data allows for more accurate identification of potential fault data for equipment nodes, avoiding misjudgments or omissions that can result from single-factor analysis. The system automatically generates standard maintenance work orders based on identified potential fault data, reducing manual intervention and improving work efficiency. Operations and maintenance personnel can schedule maintenance work promptly based on these work orders, preventing further escalation of faults and reducing equipment downtime and repair costs.

[0027] By analyzing the impact scope of potential faults using the device relationship map, we can clearly understand the relationships between the faulty device and other devices. Accurately assessing the impact scope of a fault helps maintenance personnel rationally allocate maintenance resources and time. For faults with a smaller impact scope, maintenance can be scheduled during off-peak production periods. However, for faults with a larger impact scope, immediate action is required to minimize the impact on production.

[0028] When potential fault data is detected for multiple device nodes within a single device link, analyzing the correlation of this data can reveal underlying connections between the faults. For example, multiple device failures may be caused by the same cause, such as a power issue, network failure, or external environmental factors. Correlation analysis helps identify potential complex failure patterns in advance. When potential fault data from multiple devices is correlated, it may indicate an impending, more serious system failure. By triggering an alarm, operations and maintenance personnel can take preventive and maintenance measures to prevent the failure from escalating.

[0029] Set a threshold for the scope of a fault's impact. When the scope exceeds the threshold or when there's correlation between potential fault data, the system automatically triggers an alert and generates an expedited maintenance work order. This mechanism ensures that operations personnel promptly address and address major faults, minimizing the impact on production.

[0030] Generating expedited maintenance work orders allows operations personnel to prioritize urgent troubleshooting tasks, improving the timeliness and effectiveness of troubleshooting. Compared to standard maintenance work orders, expedited maintenance work orders have a higher priority, ensuring that critical equipment failures are resolved promptly and maintaining normal production operations.

[0031] Multi-dimensional data analysis and correlation analysis at the cloud layer provide rich data support for operation and maintenance decision-making. Based on the analysis results, operation and maintenance personnel can formulate more scientific and reasonable operation and maintenance strategies. By accurately identifying potential failures and promptly addressing them, preventive maintenance is achieved, reducing the probability of sudden equipment failures, equipment maintenance costs, and production losses.

[0032] In this embodiment, the threshold corresponding to the impact range of the potential fault is set in the following manner: Statistics are collected to determine the historical detection data of simultaneous failures of key equipment nodes when the time interval between failures of multiple equipment nodes is less than a preset threshold. The impact range coefficient of the corresponding potential failures is calculated, and the minimum value is taken as the threshold.

[0033] In one embodiment of the present invention, the edge computing node establishes a device relationship map based on the device nodes of the managed device links, and selects key device nodes, including the following steps: Add device nodes to the device relationship graph in the device link and add device attribute data, as well as device load and alarm records; Device attribute data includes: device ID, type, model, IP address, geographic location, system, and standard operating procedures; Establishing edges between device nodes that have connection or data transmission relationships; Calculate the degree centrality and betweenness centrality of each device node in the device relationship graph, and take the weighted average of the degree centrality and betweenness centrality to obtain the importance coefficient of the device node; Among them, degree centrality Cd=d(v) / (N−1), N−1 is the number of directly connected edges of node v, and N is the total number of nodes in the graph; betweenness centrality ,σst is the total number of all paths from node s to t, σst(v) is the number of paths from node s to t that pass through node v; In the same device link, the importance coefficients of the device nodes are sorted from high to low, and the top 10% of the device nodes are selected as key device nodes.

[0034] In this embodiment, the connection relationship between device nodes includes: Physical connections, such as switch-server connections in a network topology.

[0035] Logical dependencies, such as microservice A depends on database B.

[0036] Data transmission relationships include: Data stream transmission, such as sensor → gateway → cloud platform.

[0037] The number of direct neighbors of a device node is calculated through degree centrality to identify the most connected devices.

[0038] Betweenness centrality measures the frequency of a node acting as a "bridge," and is applicable to critical path nodes in logical dependencies. Degree centrality and betweenness centrality are weighted averages, with weights of 0.4 and 0.6, respectively.

[0039] In one embodiment of the present invention, a risk data detection model is established at the edge layer to analyze device data in a time series database, including the following steps: Establish a risk data detection model at the edge layer, including three-level risk detection and analysis; A three-level risk detection and analysis detection strategy is set up, including: the three levels detect the device's operation log data, device status data, and device operation video stream data in turn. If risk data is detected at any level, the data of the device in the same period will be judged as risk data, and the detection is completed. Otherwise, the risk detection and analysis of the next level will be continued.

[0040] In one embodiment of the present invention, the first level risk detection includes: The edge computing node obtains the standard operating procedure data from the device attribute data of the management device node, extracts the operation keywords and timestamps of the operation log data in the time series database, arranges the operation keywords according to the operation time sequence, generates the actual operation process, compares the actual operation process with the standard operation process data, and detects the actual operation process in the standard operation process data; The second level of risk detection includes: using a machine learning model pre-trained with historical fault data to detect abnormalities in equipment status data and determine risk data; The third level of risk detection includes: using the FasterR-CNN recognition model pre-trained with historical operating image data from equipment failures, capturing images of the equipment operation video stream data every 50 frames, and detecting whether there are fault points in the captured image data; and determining risk data based on whether there are fault points.

[0041] In one embodiment of the present invention, the cloud layer extracts the hash value of the operation log of the key device node and stores it in the blockchain, including the following steps: Extract the operation logs of key equipment nodes and merge the operation logs whose operation time interval is less than the threshold; Add a hash value tag = Hash (operation content + timestamp + device ID) to the operation log and store it in the blockchain; Hash uses the SHA-256 algorithm.

[0042] In one embodiment of the present invention, a multi-dimensional data analysis model for risk data and key device node data is established at the cloud layer to analyze the device data and identify potential fault data of the device nodes, including the following steps: Targeting risk data and key equipment node data, a multi-dimensional data analysis model for equipment nodes is established, including: a pre-trained Faster R-CNN recognition model replicated by the trained edge layer, an operation log data risk factor check model, and a multi-dimensional analysis algorithm for equipment status data; Based on the detection results of the model and algorithm, a weighted average is performed. If the weighted average result is greater than the threshold, it is determined that there is a potential fault in the device node. Otherwise, it is determined that there is no potential fault.

[0043] In one embodiment of the present invention, the pre-trained Faster R-CNN recognition model, the operation log data risk factor check model, and the device status data multi-dimensional analysis algorithm replicated by the verification training edge layer are constructed in the following manner: The pre-trained Faster R-CNN recognition model copied from the edge layer is validated and trained using historical operating image data of the corresponding device node with annotated abnormal areas. If the detected abnormal area in the video stream exists for longer than a preset threshold, the risk factor of the device's operating video stream is set to 1; otherwise, it is set to 0. The operation log data risk factor checking model includes: defining risky operation steps for device nodes, summarizing continuous operation steps based on the operation log data and corresponding time series features within a preset time period through a large language model, performing cosine similarity tests on the summarized continuous operation steps and the defined risky operation steps, and taking the maximum cosine similarity as the operation risk factor; The multidimensional analysis algorithm for equipment status data is: Among them, St is the equipment status risk coefficient, Ei is the information entropy of the i-th type of equipment status data detection sensor, M0 is the mean of the historical detection data of the i-th type of equipment status data detection sensor when the equipment is operating normally, and Mi is the detection value of the i-th type of equipment status data detection sensor.

[0044] In this embodiment, the pre-trained Faster R-CNN recognition model, the operation log data risk coefficient inspection model, and the device status data multidimensional analysis algorithm copied from the edge layer are weighted averaged according to the detection results of the model and algorithm, including: the weight of the device operation video stream risk coefficient is set to 0.3, the operation risk coefficient and the device status risk coefficient are normalized respectively, and the weight of the normalized operation risk coefficient is set to 0.3, and the weight of the device status risk coefficient is set to 0.4; when the device is operating normally, the mean and the minimum value of the historical detection data are used as the threshold. If the weighted average is greater than the threshold, it is judged that the device node has a potential fault. Otherwise, it is judged that there is no potential fault.

[0045] In one embodiment of the present invention, analyzing the impact range of a potential fault using a device relationship graph includes the following steps: Extract the data of device node failures with a time interval less than a preset threshold from the historical fault data of the same device link, count the number of failures Sn of the corresponding device node, and the number of failures Sc of the device node with a time interval less than the preset threshold with the corresponding device node, and calculate the number of failed nodes in the corresponding failure times, and calculate the mean value pc of the number of failed nodes in the corresponding failure times. Obtain the importance coefficient of the corresponding device node in the device relationship graph and analyze the impact range coefficient of the corresponding potential fault: Among them, Pb is the impact range coefficient of the corresponding potential fault, p0 is the total number of devices in the same device link, Ft is the importance coefficient of the corresponding device node in the device relationship graph, and Fu is the analysis result of the multi-dimensional data analysis model of the device node.

[0046] In this embodiment, the threshold of the impact range coefficient corresponding to the potential fault is set in the following manner: Statistics are collected to determine the historical detection data of simultaneous failures of key equipment nodes when the time interval between failures of multiple equipment nodes is less than a preset threshold. The impact range coefficient of the corresponding potential failures is calculated, and the minimum value is taken as the threshold.

[0047] In one embodiment of the present invention, when potential fault data is detected in multiple device nodes in the same device link, analyzing the correlation of the potential fault data includes the following steps: By extracting historical fault data of the same device link, the data whose device node failure time interval is less than the preset threshold is found, and the data whose failure time interval is less than the preset threshold is marked with correlation; The LSTM time series model is trained with labeled data, and the LSTM time series model is used to detect whether there is a correlation between the potential fault data of device nodes.

[0048] In an optional embodiment, an intelligent equipment operation and maintenance management system is provided, characterized in that the system adopts the above-mentioned intelligent equipment operation and maintenance management method to implement intelligent equipment operation and maintenance management, including: Edge layer: Deploy multi-dimensional sensors, operation log collectors, and edge computing nodes. Edge computing nodes manage device links within the signal connection range, and set up a time series database on the edge computing nodes to store device status data, device operation video stream data, and operation log data. Establish a device relationship map based on the device nodes in the device links managed by the edge computing nodes to screen key device nodes. Establish a risk data detection model, analyze device data in the time series database, and upload risk data and key device node data to the cloud layer; Cloud layer: Extract the hash value of the operation log of key equipment nodes and store it in the blockchain; establish a multi-dimensional data analysis model for risk data and key equipment node data, analyze the equipment data, identify potential fault data of equipment nodes, and generate common maintenance work orders; The impact range of potential faults is analyzed through the device relationship map. At the same time, when potential fault data is detected in multiple device nodes in the same device link, the correlation of the potential fault data is analyzed. When the fault impact range is greater than the threshold or there is correlation between the potential fault data, an alarm is triggered and an expedited maintenance work order is generated.

[0049] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for intelligent management of equipment operation and maintenance, characterized in that: include: Deploy multi-dimensional sensors, operation log collectors, and edge computing nodes at the edge layer. The edge computing nodes manage the device links within the signal connection range, and set up a time series database on the edge computing nodes to store device status data, device operation video stream data, and operation log data. The edge computing nodes establish a device relationship map based on the device nodes of the device links they manage, and screen key device nodes. A risk data detection model is established at the edge layer to analyze device data in the time series database. The risk data and key device node data are uploaded to the cloud layer. The cloud layer extracts the hash value of the operation log of the key device node and stores it on the blockchain. Establish a multi-dimensional data analysis model for risk data and key equipment node data at the cloud layer to analyze equipment data, identify potential fault data of equipment nodes, and generate common maintenance work orders; The impact range of potential faults is analyzed through the device relationship map. At the same time, when potential fault data is detected in multiple device nodes in the same device link, the correlation of the potential fault data is analyzed. When the fault impact range is greater than the threshold or there is correlation between the potential fault data, an alarm is triggered and an expedited maintenance work order is generated.

2. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: Establish a device relationship graph based on the device nodes of the managed device links for edge computing nodes and select key device nodes, including the following steps: Add device nodes to the device relationship graph in the device link and add device attribute data, as well as device load and alarm records; Establishing edges between device nodes that have connection or data transmission relationships; Calculate the degree centrality and betweenness centrality of each device node in the device relationship graph, and take the weighted average of the degree centrality and betweenness centrality to obtain the importance coefficient of the device node; Among them, degree centrality Cd=d(v) / (N−1), N−1 is the number of directly connected edges of node v, and N is the total number of nodes in the graph; betweenness centrality ,σst is the total number of all paths from node s to t, σst(v) is the number of paths from node s to t that pass through node v; In the same device link, the importance coefficients of the device nodes are sorted from high to low, and the top 10% of the device nodes are selected as key device nodes.

3. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: Establishing a risk data detection model at the edge layer and analyzing device data in the time series database involves the following steps: Establish a risk data detection model at the edge layer, including three-level risk detection and analysis; A three-level risk detection and analysis detection strategy is set up, including: the three levels detect the device's operation log data, device status data, and device operation video stream data in turn. If risk data is detected at any level, the data of the device in the same period will be judged as risk data, and the detection is completed. Otherwise, the risk detection and analysis of the next level will be continued.

4. The intelligent management method for equipment operation and maintenance according to claim 3, characterized in that: The first level of risk detection includes: The edge computing node obtains the standard operating procedure data from the device attribute data of the management device node, extracts the operation keywords and timestamps of the operation log data in the time series database, arranges the operation keywords according to the operation time sequence, generates the actual operation process, compares the actual operation process with the standard operation process data, and detects the actual operation process in the standard operation process data; The second level of risk detection includes: using a machine learning model pre-trained with historical fault data to detect abnormalities in equipment status data and determine risk data; The third level of risk detection includes: using the Faster R-CNN recognition model pre-trained with historical operating image data from equipment failures, capturing images of the equipment operation video stream data every 50 frames, and detecting whether there are fault points in the captured image data; and determining risk data based on the presence of fault points.

5. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: The cloud layer extracts the hash value of the operation log of the key device node and stores it in the blockchain, including the following steps: Extract the operation logs of key equipment nodes and merge the operation logs whose operation time interval is less than the threshold; Add a hash value tag = Hash (operation content + timestamp + device ID) to the operation log and store it in the blockchain.

6. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: A multi-dimensional data analysis model for risk data and key equipment node data is established at the cloud layer to analyze the equipment data and identify potential fault data of the equipment nodes. The following steps are included: Targeting risk data and key equipment node data, a multi-dimensional data analysis model for equipment nodes is established, including: a pre-trained Faster R-CNN recognition model replicated by the trained edge layer, an operation log data risk factor check model, and a multi-dimensional analysis algorithm for equipment status data; Based on the detection results of the model and algorithm, a weighted average is performed. If the weighted average result is greater than the threshold, it is determined that there is a potential fault in the device node. Otherwise, it is determined that there is no potential fault.

7. The intelligent management method for equipment operation and maintenance according to claim 6, characterized in that: The pre-trained Faster R-CNN recognition model, the risk factor check model for operation log data, and the multi-dimensional analysis algorithm for device status data replicated in the edge layer of the verification training are constructed in the following way: The pre-trained Faster R-CNN recognition model copied from the edge layer is validated and trained using historical operating image data of the corresponding device node with annotated abnormal areas. If the detected abnormal area in the video stream exists for longer than a preset threshold, the risk factor of the device's operating video stream is set to 1; otherwise, it is set to 0. The operation log data risk factor checking model includes: defining risky operation steps for device nodes, summarizing continuous operation steps based on the operation log data and corresponding time series features within a preset time period through a large language model, performing cosine similarity tests on the summarized continuous operation steps and the defined risky operation steps, and taking the maximum cosine similarity as the operation risk factor; The multidimensional analysis algorithm for equipment status data is: Among them, St is the equipment status risk coefficient, Ei is the information entropy of the i-th type of equipment status data detection sensor, M0 is the mean of the historical detection data of the i-th type of equipment status data detection sensor when the equipment is operating normally, and Mi is the detection value of the i-th type of equipment status data detection sensor.

8. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: Analyzing the impact of potential faults using the device relationship diagram includes the following steps: Extract the data of device node failures with a time interval less than a preset threshold from the historical fault data of the same device link, count the number of failures Sn of the corresponding device node, and the number of failures Sc of the device node with a time interval less than the preset threshold with the corresponding device node, and calculate the number of failed nodes in the corresponding failure times, and calculate the mean value pc of the number of failed nodes in the corresponding failure times. Obtain the importance coefficient of the corresponding device node in the device relationship graph and analyze the impact range coefficient of the corresponding potential fault: Among them, Pb is the impact range coefficient of the corresponding potential fault, p0 is the total number of devices in the same device link, Ft is the importance coefficient of the corresponding device node in the device relationship graph, and Fu is the analysis result of the multi-dimensional data analysis model of the device node.

9. The intelligent management method for equipment operation and maintenance according to claim 1, characterized in that: When potential fault data is detected on multiple device nodes in the same device link, the correlation between the potential fault data is analyzed, including the following steps: By extracting historical fault data of the same device link, the data whose device node failure time interval is less than the preset threshold is found, and the data whose failure time interval is less than the preset threshold is marked with correlation; The LSTM time series model is trained with labeled data, and the LSTM time series model is used to detect whether there is a correlation between the potential fault data of device nodes.

10. An intelligent management system for equipment operation and maintenance, characterized in that: The system implements intelligent management of equipment operation and maintenance by using an intelligent management method for equipment operation and maintenance according to any one of claims 1 to 9, including: Edge layer: Deploy multi-dimensional sensors, operation log collectors, and edge computing nodes. Edge computing nodes manage device links within the signal connection range, and set up a time series database on the edge computing nodes to store device status data, device operation video stream data, and operation log data. Establish a device relationship map based on the device nodes in the device links managed by the edge computing nodes to screen key device nodes. Establish a risk data detection model, analyze device data in the time series database, and upload risk data and key device node data to the cloud layer; Cloud layer: Extract the hash value of the operation log of key equipment nodes and store it in the blockchain; establish a multi-dimensional data analysis model for risk data and key equipment node data, analyze the equipment data, identify potential fault data of equipment nodes, and generate common maintenance work orders; The impact range of potential faults is analyzed through the device relationship map. At the same time, when potential fault data is detected in multiple device nodes in the same device link, the correlation of the potential fault data is analyzed. When the fault impact range is greater than the threshold or there is correlation between the potential fault data, an alarm is triggered and an expedited maintenance work order is generated.

Images