Terminal equipment intelligent fault diagnosis method based on electric power internet-of-things edge calculation

Abstract

The invention relates to a terminal equipment intelligent fault diagnosis method based on electric power internet of things edge computing. The method comprises the following steps: constructing a cloud edge interconnection architecture; in a data acquisition stage, real-time denoising and standardization processing are carried out by utilizing an edge computing layer, and data availability and access speed are guaranteed by adopting an edge caching technology; data from different sensors are integrated by using a data fusion technology, the accuracy of fault detection and diagnosis is improved, multi-source data association analysis is applied, and multi-angle early warning is performed for abnormal behaviors; event mode identification and processing are realized on an edge calculation layer by using CEP, and potential fault signals are quickly responded; all transmission data of the communication layer uses an end-to-end encryption protocol to ensure data security, and edge device intrusion detection and firewall protection are implemented to prevent hostile attacks. According to the invention, the accuracy and efficiency of power equipment fault diagnosis are improved, and the safety and operation stability of the system are remarkably enhanced.

Description

Technical Field

[0001] This invention relates to the field of terminal fault diagnosis, and in particular to an intelligent fault diagnosis method for terminal devices based on power IoT edge computing. Background Technology

[0002] In modern power systems, with the development of Internet of Things (IoT) technology, the Power IoT has become a new trend. It integrates sensor networks, communication technologies, edge computing, cloud computing, and big data analytics to achieve comprehensive, real-time monitoring and intelligent management of the operating status of power equipment. Traditional power equipment operation and maintenance mainly relies on periodic inspections and manual monitoring, which is not only costly but also unable to predict and handle equipment failures in a timely manner.

[0003] By introducing edge computing, the Internet of Things (IoT) for power systems can preprocess and analyze data at edge nodes close to the data source. This computing architecture reduces bandwidth requirements and latency issues for data transmission to the cloud, improving response efficiency. However, edge devices typically have much smaller computing and storage capabilities than cloud servers. This limits the amount and complexity of data they can handle, especially when facing large volumes of high-frequency data streams, potentially requiring more complex system optimizations. Furthermore, as the number of edge nodes increases, managing and maintaining these devices becomes more complex. Ensuring the stability and consistency of each node remains an ongoing challenge, especially under different device and environmental conditions. Summary of the Invention

[0004] To address the aforementioned issues, the present invention aims to provide an intelligent fault diagnosis method for terminal devices based on power IoT edge computing. This method not only improves the accuracy and efficiency of power equipment fault diagnosis but also significantly enhances the system's security and operational stability, enabling the entire system to perform exceptionally well in handling complex power equipment management problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for intelligent fault diagnosis of terminal devices based on power IoT edge computing includes the following steps:

[0007] Construct a cloud-edge interconnect architecture, including the terminal device layer, edge computing layer, communication layer, and cloud layer;

[0008] During the data acquisition phase, edge computing layer is used for real-time noise reduction and standardization, and edge caching technology is adopted to ensure data availability and access speed.

[0009] Data fusion technology is used to integrate data from different sensors to improve the accuracy of fault detection and diagnosis. Multi-source data correlation analysis is applied to provide multi-angle early warning for abnormal behavior.

[0010] By leveraging CEP to identify and process event patterns at the edge computing layer, we can quickly respond to potential fault signals.

[0011] All data transmitted in the communication layer uses an end-to-end encryption protocol to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks.

[0012] Furthermore, the edge device layer includes multi-functional sensors installed on each power device to achieve multi-modal data collection; the edge computing layer deploys edge gateways to support real-time analysis and has good scalability, supporting the deployment of edge applications based on virtualization or containerization; the communication layer adopts the MQTT communication protocol to ensure low latency and high reliability of data transmission; the cloud layer provides big data storage and high-performance computing resources for model training, data analysis, and result visualization, and supports dynamic adjustment of edge device configuration and management through cloud-edge collaboration.

[0013] Furthermore, edge caching technology is employed to ensure data availability and access speed, specifically as follows: Fast-access flash memory is configured on the edge computing layer for temporary data storage, reducing data retrieval latency; RAM is used as a temporary buffer to store recently used data blocks, improving data processing response speed; the edge computing layer includes several edge nodes, and a distributed caching system is used to share the cache across multiple edge nodes, improving the overall data access speed and reliability of the system; a consistent hashing algorithm is used to evenly distribute data across nodes, preventing cache hotspots; and a data prefetching mechanism is implemented: based on historical data access patterns, future potential data demands are predicted and pre-loaded into the cache, improving the system's pre-response speed.

[0014] Furthermore, data fusion technology is used to integrate data from different sensors to improve the accuracy of fault detection and diagnosis. Multi-source data correlation analysis is applied to provide multi-angle early warnings for abnormal behavior, as detailed below:

[0015] Ensure time synchronization of data from different sensors by using a timestamp correction mechanism and calibrating the measurement units and accuracy of each sensor;

[0016] Fusion is achieved at the signal level by processing real-time signals through weighted averaging to reduce noise and redundancy and improve the representativeness of single-variable signals;

[0017] The Apriori algorithm is used to mine association rules and identify potential relationships between data from different sensors.

[0018] The DBSCAN anomaly detection algorithm is used to mark anomalous data points and analyze anomaly patterns in the sensor cluster.

[0019] Scenario models are built based on Bayesian networks, and the logical relationships of multi-sensor data are integrated to predict and warn of potential faults.

[0020] Furthermore, a scenario model is constructed based on Bayesian networks, integrating the logical relationships of multi-sensor data to predict and warn of potential faults, as detailed below:

[0021] Identify the types of faults that need to be identified and the associated sensor data; set risk levels and corresponding warning thresholds for different faults based on historical data and expert knowledge;

[0022] Select key sensor parameters related to the fault as random variables and potential fault nodes; determine whether to use discrete, continuous, or hybrid Bayesian networks based on the data type.

[0023] Initially define nodes and their causal relationships using expert knowledge; for example, a rise in temperature may cause equipment to overheat.

[0024] The structure learning algorithm is used to automatically learn the relationships between nodes from historical data and perform cross-validation to ensure that the network structure has good generalization ability.

[0025] Network parameters are learned using the training dataset via maximum likelihood estimation, and for mixed data types, the parameters are optimized using the expectation-maximization algorithm.

[0026] Gibbs sampling is used to infer the probability of a fault occurring in a Bayesian network. Dynamic Bayesian networks are used to process time series data and capture the trend of sensor data changing over time.

[0027] Based on the current sensor input, the conditional probability distribution of each potential fault node is calculated, compared with the preset risk threshold, and high-risk fault nodes are automatically identified and marked.

[0028] Furthermore, CEP is used to identify and process event patterns at the edge computing layer, enabling rapid response to potential fault signals;

[0029] Collect and synchronize data streams from sensors to provide a unified input for CEP analysis; utilize the message queuing system Apache Kafka as the data stream transmission pipeline to ensure efficient data transmission and processing;

[0030] Define key event patterns that represent different faults, and create and orchestrate complex event patterns using a graphical IDE;

[0031] Set time windows for different types of event patterns and combine them with event sequence analysis to identify early signs of failure through order event sequences;

[0032] The CEP engine processes incoming data streams in real time, identifies defined event patterns, and implements state management to track events and state transitions and identify abnormal changes.

[0033] Once a potential failure mode is identified, the CEP system triggers alerts and automated actions via API or messaging system to execute emergency response measures directly at the edge node, reducing latency.

[0034] Furthermore, all data transmitted at the communication layer uses an end-to-end encryption protocol to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks, as detailed below:

[0035] Establish a secure channel between the device and the gateway using the TLS protocol, and protect the confidentiality of data through symmetric encryption; use HMAC for data integrity verification to prevent data from being tampered with during transmission; deploy Snort to monitor transmission traffic and detect suspicious behavior in real time according to a predefined set of rules; configure a firewall on each edge device to control allowed and denied network connections, and protect the device according to inbound and outbound rules.

[0036] The present invention has the following beneficial effects:

[0037] 1. This invention integrates data from different sensors to improve the ability to identify fault modes. This enables the system to obtain more comprehensive fault status information by comprehensively analyzing multiple data sources, thereby improving the accuracy of fault detection and diagnosis. Furthermore, by analyzing the correlation and patterns of different data sources, it can identify the characteristics of complex faults earlier, provide multi-faceted early warning information, and provide decision support for fault elimination.

[0038] 2. The present invention uses CEP technology at the edge computing layer to quickly identify and process event patterns, ensuring that fault signals are accurately captured and responded to on the first attempt, thereby improving the system's ability to respond to sudden events.

[0039] 3. This invention not only improves the accuracy and efficiency of power equipment fault diagnosis, but also significantly enhances the system's safety and operational stability, enabling the entire system to perform excellently in dealing with complex power equipment management problems. Attached Figure Description

[0040] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0041] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0042] refer to Figure 1In this embodiment, a method for intelligent fault diagnosis of terminal devices based on power IoT edge computing is provided, characterized by the following steps:

[0043] Construct a cloud-edge interconnect architecture, including the terminal device layer, edge computing layer, communication layer, and cloud layer;

[0044] During the data acquisition phase, edge computing layer is used for real-time noise reduction and standardization, and edge caching technology is adopted to ensure data availability and access speed.

[0045] Data fusion technology is used to integrate data from different sensors to improve the accuracy of fault detection and diagnosis. Multi-source data correlation analysis is applied to provide multi-angle early warning for abnormal behavior.

[0046] By leveraging CEP to identify and process event patterns at the edge computing layer, we can quickly respond to potential fault signals.

[0047] All data transmitted in the communication layer uses an end-to-end encryption protocol to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks.

[0048] In this embodiment, the end device layer includes multi-functional sensors (temperature, vibration, current, voltage, etc.) installed on each power device to achieve multi-modal data collection; the edge computing layer deploys edge gateways to support real-time analysis and has good scalability, supporting the deployment of edge applications based on virtualization or containerization; the communication layer adopts the MQTT communication protocol to ensure low latency and high reliability of data transmission; the cloud layer provides big data storage and high-performance computing resources for model training, data analysis, and result visualization, and supports dynamic adjustment of edge device configuration and management through cloud-edge collaboration.

[0049] In this embodiment, edge caching technology is employed to ensure data availability and access speed, specifically as follows: Fast-access flash memory (such as SSD) is configured on the edge computing layer for temporary data storage, reducing data retrieval latency; RAM is used as a temporary buffer to store recently used data blocks, improving data processing response speed; the edge computing layer includes several edge nodes, and a distributed caching system (such as Redis or Memcached) is used to share the cache across multiple edge nodes, improving the overall data access speed and reliability of the system; a consistent hashing algorithm is used to evenly distribute data across nodes, preventing cache hotspots; and a data prefetching mechanism is implemented: based on historical data access patterns, future potential data demands are predicted and pre-loaded into the cache, improving the system's pre-response speed.

[0050] In this embodiment, data fusion technology is used to integrate data from different sensors to improve the accuracy of fault detection and diagnosis. Multi-source data correlation analysis is applied to provide multi-angle early warning for abnormal behavior, as detailed below:

[0051] Ensure time synchronization of data from different sensors by using a timestamp correction mechanism and calibrating the measurement units and accuracy of each sensor;

[0052] Fusion is achieved at the signal level by processing real-time signals through weighted averaging to reduce noise and redundancy and improve the representativeness of single-variable signals;

[0053] The Apriori algorithm is used to mine association rules and identify potential relationships between data from different sensors.

[0054] The DBSCAN anomaly detection algorithm is used to mark anomalous data points and analyze anomaly patterns in the sensor cluster.

[0055] Scenario models are built based on Bayesian networks, and the logical relationships of multi-sensor data are integrated to predict and warn of potential faults.

[0056] In this embodiment, a scenario model is constructed based on a Bayesian network. By integrating the logical relationships of multi-sensor data, possible faults are predicted and warned, as follows:

[0057] Determine the types of faults to be identified (such as overheating, mechanical failure, etc.) and the related sensor data (such as temperature, vibration, current, etc.); set the risk levels and corresponding warning thresholds for different faults based on historical data and expert knowledge;

[0058] Select key sensor parameters related to the fault as random variables and potential fault nodes; determine whether to use discrete, continuous, or hybrid Bayesian networks based on the data type.

[0059] Initially define nodes and their causal relationships using expert knowledge; for example, a rise in temperature may cause equipment to overheat.

[0060] The structure learning algorithm is used to automatically learn the relationships between nodes from historical data and perform cross-validation to ensure that the network structure has good generalization ability.

[0061] Network parameters are learned using the training dataset via maximum likelihood estimation, and for mixed data types, the parameters are optimized using the expectation-maximization algorithm.

[0062] Gibbs sampling is used to infer the probability of a fault occurring in a Bayesian network. Dynamic Bayesian networks are used to process time series data and capture the trend of sensor data changing over time.

[0063] Based on the current sensor input, the conditional probability distribution of each potential fault node is calculated, compared with the preset risk threshold, and high-risk fault nodes are automatically identified and marked.

[0064] In this embodiment, CEP is used to identify and process event patterns at the edge computing layer, enabling rapid response to potential fault signals;

[0065] Collect and synchronize data streams from sensors to provide a unified input for CEP analysis; utilize the message queuing system Apache Kafka as the data stream transmission pipeline to ensure efficient data transmission and processing;

[0066] Define key event patterns that represent different faults, and create and orchestrate complex event patterns using a graphical IDE;

[0067] Set time windows for different types of event patterns and combine them with event sequence analysis to identify early signs of failure through order event sequences;

[0068] The CEP engine processes incoming data streams in real time, identifies defined event patterns, and implements state management to track events and state transitions and identify abnormal changes.

[0069] Once a potential failure mode is identified, the CEP system triggers alerts and automated actions via API or messaging systems (such as MQTT) to directly implement emergency response measures (such as load shedding and shutdown) at the edge nodes, reducing latency.

[0070] In this implementation, all data transmitted at the communication layer uses an end-to-end encryption protocol to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks. Specifically, the following measures are taken: a secure channel is established between the device and the gateway using the TLS protocol, and data confidentiality is protected through symmetric encryption; HMAC is used for data integrity verification to prevent data from being tampered with during transmission; Snort is deployed to monitor transmission traffic and detect suspicious behavior in real time according to a predefined set of rules; and a firewall is configured on each edge device to control allowed and denied network connections, protecting the device according to inbound and outbound rules.

[0071] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0072] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0073] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0074] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0075] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing, characterized in that: The following steps are involved: Build a cloud-edge interconnection architecture, including the terminal device layer, edge computing layer, communication layer, and cloud layer; During the data collection phase, the edge computing layer is used for real-time denoising and standardization, and edge caching technology is used to ensure data availability and access speed; Use data fusion technology to integrate data from different sensors to improve the accuracy of fault detection and diagnosis, and apply multi-source data correlation analysis to provide multi-angle warnings for abnormal behaviors; Use CEP to identify and process event patterns at the edge computing layer and quickly respond to potential fault signals; All transmitted data in the communication layer uses end-to-end encryption protocols to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks.

2. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 1 is characterized in that: The end device layer includes multifunctional sensors installed on each power equipment to realize multimodal data collection; the edge computing layer deploys edge gateways to support real-time analysis, and has good scalability and supports edge application deployment based on virtualization or containerization; the communication layer adopts the communication protocol MQTT to ensure low latency and high reliability of data transmission; the cloud layer provides big data storage and high-performance computing resources for model training, data analysis and result visualization, and supports dynamic adjustment of edge device configuration and management through cloud-edge collaboration.

3. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 1 is characterized in that: The edge caching technology is used to ensure data availability and access speed, as follows: fast-access flash memory is configured on the edge computing layer for temporary data storage to reduce data acquisition delays; RAM is used as a temporary buffer to store recently used data blocks to improve the response speed of data processing; the edge computing layer includes several edge nodes, and a distributed cache system is used to share cache on multiple edge nodes to improve the overall data access speed and reliability of the system; and a consistent hashing algorithm is used to evenly distribute data to each node to prevent cache hot spots; a data pre-fetching mechanism is implemented: based on historical data access patterns, possible future data needs are predicted, and the data is loaded into the cache in advance to improve the system's pre-response speed.

4. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 1 is characterized in that: The data fusion technology is used to integrate data from different sensors to improve the accuracy of fault detection and diagnosis, and multi-source data association analysis is applied to provide multi-angle warnings for abnormal behaviors, as follows: Ensure time synchronization of different sensor data, use timestamp correction mechanisms, and calibrate the measurement units and accuracy of each sensor; Fusion is achieved at the signal level, where real-time signals are processed by weighted averaging to reduce noise and redundancy and improve the representativeness of single variable signals; Use the Apriori algorithm to mine association rules and identify potential relationships between different sensor data; Use the anomaly detection algorithm DBSCAN to mark abnormal data points and analyze abnormal patterns in the sensor cluster; A scenario model is constructed based on the Bayesian network, and the logical relationship of multi-sensor data is integrated to predict and warn possible failures.

5. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 4 is characterized in that: The scenario model is constructed based on the Bayesian network, and the logical relationship of multi-sensor data is integrated to predict and warn possible faults, as follows: Determine the fault types that need to be identified and the sensor data associated with them; set the risk levels of different faults and corresponding warning thresholds based on historical data and expert knowledge; Select important sensor parameters related to faults as random variables and potential fault nodes; use discrete, continuous or mixed Bayesian networks according to the data type; Initially define nodes and their causal relationships through expert knowledge; for example, a temperature rise may cause a device to overheat; Use structural learning algorithms to automatically learn the relationship between nodes from historical data and perform cross-validation to ensure that the network structure has good generalization ability. Use the training data set to learn network parameters through maximum likelihood estimation, and use the expectation maximization algorithm to optimize parameters for mixed data types; Gibbs sampling is used to reason about the Bayesian network to infer the possibility of fault occurrence, and a dynamic Bayesian network is used to process time series data to capture the trend of sensor data changing over time. Based on the current sensor input, the conditional probability distribution of each potential fault node is calculated, compared with the preset risk threshold, and high-risk fault nodes are automatically identified and marked.

6. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 1 is characterized in that: The use of CEP to identify and process event patterns at the edge computing layer can quickly respond to potential fault signals; Collect and synchronize data streams from sensors to provide unified input for CEP analysis; use the message queue system Apache Kafka as a transmission pipeline for data streams to ensure efficient transmission and processing of data; Define key event patterns representing different faults, and create and orchestrate complex event patterns using a graphical IDE; Set time windows for different types of event patterns and combine them with event sequence analysis to identify the precursors of failures through order event sequences; Incoming data streams are processed in real time in the CEP engine to identify defined event patterns; and Implement state management, track events and state transitions, and identify abnormal changes; Once a potential failure mode is identified, the CEP system triggers alerts and automated actions through APIs or messaging systems to execute emergency response measures directly at the edge node, reducing latency.

7. The terminal equipment intelligent fault diagnosis method based on power Internet of Things edge computing according to claim 1 is characterized in that: All data transmitted in the communication layer uses end-to-end encryption protocol to ensure data security, and implements edge device intrusion detection and firewall protection to prevent malicious attacks, as follows: Use the TLS protocol to establish a secure channel between the device and the gateway, and protect the confidentiality of the data through symmetric encryption; use HMAC to verify data integrity to prevent data tampering during transmission; deploy Snort to monitor transmission traffic and detect suspicious behavior in real time based on predefined rule sets; configure a firewall on each edge device to control allowed and denied network connections, and protect the device based on inbound and outbound rules.

Images