Remote operation and maintenance monitoring system for intelligent power electrical cabinet
Through the remote operation and maintenance monitoring system combined with GAT, DRL and blockchain technology, real-time monitoring and fault prediction of power and electrical cabinets are realized, solving the problems of inefficiency and insufficient data management of traditional operation and maintenance methods, and providing an efficient and secure intelligent operation and maintenance solution.
Patent Information
- Application Number
- CN202510588128.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-08
- Publication Date
- 2025-08-12
AI Technical Summary
The traditional power electrical cabinet operation and maintenance methods are inefficient, slow fault response, high cost and insufficient data management. The existing degree of intelligence is limited and there is a lack of comprehensive solutions, making it difficult to meet the efficient and intelligent operation and maintenance needs of modern power systems.
The remote operation and maintenance monitoring system is adopted, combined with graph attention network (GAT), deep reinforcement learning (DRL), blockchain technology and distributed storage, to realize intelligent management of real-time data acquisition, status detection, fault prediction and operation and maintenance planning, ensure data security through blockchain, and efficient management of distributed storage, and use intelligent algorithms to optimize operation and maintenance strategies.
Real-time monitoring of power and electrical cabinets, accurate fault prediction and dynamic operation and maintenance planning adjustments have been achieved, which improves operation and maintenance efficiency and system reliability, reduces operation and maintenance costs, and provides safe and efficient intelligent operation and maintenance solutions.
Smart Images
Figure CN120474180A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power system operation and maintenance, and in particular to a remote operation and maintenance monitoring system for an intelligent power electrical cabinet. Background Art
[0002] Electrical cabinets are essential, critical equipment in power systems, widely used in substations, distribution rooms, industrial production lines, and other fields. Their operational status is directly related to the safety and reliability of the power system. However, traditional O&M methods, which rely primarily on manual inspections and scheduled maintenance, suffer from inefficiency, slow fault response, high costs, and insufficient data management. With the growth of electricity demand and the increasing complexity of power systems, traditional methods are no longer able to meet the requirements of modern power systems for efficient and intelligent O&M.
[0003] In recent years, the rapid development of technologies such as the Internet of Things, artificial intelligence, and big data has provided new solutions for the intelligent operation and maintenance of power distribution cabinets. Through real-time data collection, intelligent analysis, and prediction, intelligent operation and maintenance systems enable 24 / 7 monitoring, fault prediction, dynamic adjustment of operation and maintenance plans, and data-driven decision-making, significantly improving operation and maintenance efficiency and reliability. However, existing technologies still suffer from insufficient data security, limited intelligence, and a lack of comprehensive solutions, making them unable to fully meet the actual needs of power systems. Summary of the Invention
[0004] The present invention provides a remote operation and maintenance monitoring system for an intelligent power electrical cabinet, which is used to solve the problems existing in the prior art such as insufficient data security, limited intelligence and lack of comprehensive solutions.
[0005] In order to achieve the above-mentioned objectives, an embodiment of the present invention provides a remote operation and maintenance monitoring system for an intelligent power electrical cabinet. The remote operation and maintenance monitoring system includes: a remote operation and maintenance server, which is used to generate and arrange an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet; a data acquisition module, which is used to collect real-time operation data of each unit in the electrical cabinet; a terminal computing device, which is used to perform status detection and fault prediction on each unit in the electrical cabinet based on the collected real-time operation data and using a preset intelligent monitoring model, and adjust the preset operation and maintenance monitoring plan according to the results and execute the operation and maintenance monitoring plan.
[0006] Optionally, the remote operation and maintenance monitoring system also includes a communication module for data interaction between the data acquisition module and the terminal computing device, wherein blockchain technology is used to construct a secure channel between the data acquisition module and the terminal computing device, encrypt the transmitted data, and verify the integrity and legitimacy of the transmitted data through a consensus mechanism.
[0007] Optionally, the terminal computing device includes a storage module for storing operating data within a preset time period and corresponding status detection data and fault prediction data. The storage module is configured to distribute the operating data through distributed storage nodes, divide the operating data into multiple small data blocks, store them on different nodes, and use smart contract technology to manage access rights to the operating data.
[0008] Optionally, the remote operation and maintenance server is also used to train an intelligent monitoring model based on historical operation data, and send the trained intelligent monitoring model to the terminal computing device. When training the intelligent monitoring model, a topological graph structure reflecting the connection relationship between each unit in the electrical cabinet is constructed based on the units in the electrical cabinet and the connection relationship between them, and the historical operation data of each unit in the electrical cabinet is converted into corresponding node features, wherein the intelligent monitoring model is constructed based on the GAT algorithm.
[0009] Optionally, the status detection and fault prediction of each unit in the electrical cabinet based on the collected real-time operation data and using a preset intelligent monitoring model includes: inputting the collected real-time operation data into a trained intelligent monitoring model; the intelligent monitoring model converts the real-time operation data of each unit in the electrical cabinet into corresponding node features; and using the GAT algorithm to judge the status of each unit in the electrical cabinet and perform fault prediction based on the node features and the topological graph structure.
[0010] Optionally, the use of the GAT algorithm to judge the status of each unit in the electrical cabinet and perform fault prediction based on node features and topological graph structure includes: using the GAT algorithm to combine the node features into a feature vector; based on the feature vector, using a preset classifier, classifying the status of each unit in the electrical cabinet to obtain the status of each unit in the electrical cabinet; based on the feature vector, using a preset regression model, predicting the probability of failure of each unit in the electrical cabinet to obtain a corresponding fault probability value; if the fault probability value is greater than a preset threshold, the unit corresponding to the fault probability value has a fault risk; and determining the faulty unit based on the topological graph structure.
[0011] Optionally, the generating and arranging of an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet includes: generating an operation and maintenance monitoring plan using a trained decision generation model, wherein the decision generation model is constructed and trained based on a deep reinforcement learning algorithm, and the deep reinforcement learning algorithm is configured to: use the operating status of the electrical cabinet as the environment, the operation and maintenance operation as the intelligent agent action, and reduce the number of failures and reduce the operation and maintenance cost as the reward function, and generate an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet.
[0012] Optionally, the remote operation and maintenance monitoring system further includes: a data processing module, which is used to obtain the operating data of the electrical cabinet through the data acquisition module, clean the obtained operating data, remove abnormal values and noise data, and perform normalization processing.
[0013] Optionally, the terminal computing device adjusts the preset operation and maintenance monitoring plan including: establishing a data model based on the historical operation data of the electrical cabinet; simulating the impact of different adjustment schemes on the overall operation of the electrical cabinet through the data model based on the results of status detection and fault prediction; determining the optimal adjustment scheme based on the determined impact, and adjusting the preset operation and maintenance monitoring plan according to the determined adjustment scheme.
[0014] Optionally, during training, the intelligent monitoring model uses the Adam optimizer to adjust the parameters of the intelligent monitoring model, and optimizes the intelligent monitoring model through a preset weighted loss function, wherein the weighted loss function is configured as a loss function obtained by fusing the cross entropy loss function and the mean square error loss function and assigning different weights.
[0015] The remote operation and maintenance monitoring system for smart power electrical cabinets provided by the present invention realizes real-time monitoring of the operating status of smart power electrical cabinets, accurate fault prediction, and dynamic adjustment of operation and maintenance plans through intelligent monitoring models and decision-making generation models, significantly improving operation and maintenance efficiency and system reliability; at the same time, it uses blockchain technology to ensure the security of data transmission and storage, combines distributed storage and smart contracts to achieve efficient data management and access control, reduces operation and maintenance costs, and reduces manual intervention, providing a comprehensive, safe and efficient solution for the intelligent operation and maintenance of power systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions of the present invention or the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0017] Figure 1 This is a flow chart of a remote operation and maintenance monitoring system provided by an embodiment of the present invention;
[0018] Figure 2 This is a schematic diagram of the distributed storage structure of the blockchain provided by an embodiment of the present invention;
[0019] Figure 3 This is a flowchart of the operation of the intelligent monitoring model provided by an embodiment of the present invention;
[0020] Figure 4 Schematic diagram of the GAT algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0021] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.
[0022] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.
[0023] Power electrical cabinets are core equipment in power systems and are widely used in scenarios such as substations, distribution rooms, industrial production lines, and data centers. Their primary function is to centrally control, protect, and distribute power to power equipment, ensuring the stable operation of the power system. However, traditional electrical cabinet operation and maintenance methods rely primarily on manual inspections and regular maintenance, which can lead to low efficiency, delayed fault response, high operation and maintenance costs, and insufficient data management. As power systems expand in size and complexity, traditional operation and maintenance methods can no longer meet the demands of modern power systems for efficient and intelligent operation and maintenance. Therefore, it is particularly important to develop a safe, efficient, and intelligent remote operation and maintenance monitoring system.
[0024] To address this issue, the present invention provides a remote operation and maintenance monitoring system for intelligent power electrical cabinets. This system combines advanced technologies such as graph attention networks (GAT), deep reinforcement learning (DRL), blockchain technology, and distributed storage to achieve real-time monitoring of the cabinet's operating status, accurate fault prediction, and intelligent management of operation and maintenance plans. By introducing blockchain technology to ensure the security of data transmission and storage, utilizing distributed storage and smart contracts to achieve efficient data management and access control, and combining intelligent algorithms to optimize operation and maintenance strategies, the present invention provides a comprehensive, secure, and efficient solution for the intelligent operation and maintenance of power electrical cabinets, promoting the transformation of power systems towards digitalization and intelligence.
[0025] The following combination Figures 1-4 The present invention will be described in detail.
[0026] like Figure 1As shown, an embodiment of the present invention provides a remote operation and maintenance monitoring system for an intelligent power electrical cabinet, comprising: a remote operation and maintenance server, for generating and arranging an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet; a data acquisition module, for collecting real-time operation data of each unit in the electrical cabinet; a terminal computing device, for performing status detection and fault prediction on each unit in the electrical cabinet based on the collected real-time operation data and using a preset intelligent monitoring model, and adjusting the preset operation and maintenance monitoring plan according to the results and executing the operation and maintenance monitoring plan.
[0027] The remote operation and maintenance server can generate and arrange operation and maintenance monitoring plans based on the historical operation data of each unit inside the electrical cabinet, realize remote monitoring of the electrical cabinet and automatic generation of operation and maintenance plans. At the same time, the data acquisition module can collect the operation data of each unit in the electrical cabinet in real time. These operation data can be calculated by the terminal computing device to obtain the results of status detection and fault prediction of the electrical cabinet, and the terminal computing device can adjust the existing operation and maintenance monitoring plan according to the results and execute the adjusted operation and maintenance monitoring plan, realize real-time regulation of the operation of the electrical cabinet, and increase the efficiency of the operation and maintenance of the electrical cabinet.
[0028] like Figure 2 As shown, the remote operation and maintenance monitoring system preferably also includes a communication module for data exchange between the data acquisition module and the terminal computing device. Blockchain technology is used to establish a secure channel, encrypt the transmitted data, and verify the integrity and legitimacy of the data through a consensus mechanism. The terminal computing device includes a storage module for storing operating data within a preset time period, as well as corresponding status detection data and fault prediction data. The storage module is configured to distribute the data through distributed storage nodes, dividing the data into multiple small data blocks and storing them on different nodes. Smart contract technology is also used to manage data access rights.
[0029] In a preferred embodiment of the present invention, blockchain is a distributed ledger technology that uses encryption algorithms and consensus mechanisms to ensure data security, integrity, and immutability. In the intelligent power cabinet remote operation and maintenance monitoring system of the present invention, blockchain technology is primarily used to address security issues in data transmission and storage, ensuring data integrity and credibility. In this invention, blockchain technology is first used to encrypt collected data, using an asymmetric encryption algorithm to generate a unique encryption key for each data block. The encrypted data is then packaged according to specific rules to form data blocks containing information such as the data content, timestamp, and the hash value of the previous data block. These packaged data blocks are then transmitted to each node. During data transmission, the encrypted channel of the blockchain ensures that the data cannot be stolen or tampered with. After receiving the data block, each node initiates a consensus mechanism to verify the data block's integrity, legitimacy, and consistency with data on other nodes. Only data blocks that pass verification by a majority of nodes are considered authentic. Verified data blocks are added to a storage module and stored on each node. Each node maintains a complete or partial copy of the data, achieving distributed storage. This storage method not only improves data security but also enhances data availability, ensuring data is not lost even if some nodes fail. When operations personnel or related systems need to access data, permission verification is first performed through a smart contract. Only authorized users can read, query, and perform other operations on the blockchain within their permissions. Data modifications, such as updating operations and maintenance records, also require strict permission verification and consensus confirmation to ensure data security and consistency.
[0030] For example, a system consists of a data acquisition module (node A), a terminal computing device (node B), and multiple blockchain network verification nodes (nodes C, D, and E, etc.). Node A collects real-time operating data from various units within an electrical cabinet, such as current of 5A, voltage of 220V, and temperature of 30°C, and packages this data into a data block. Node A generates a random symmetric encryption key K and encrypts the data block with it, producing a ciphertext α. Node A then obtains node B's public key and encrypts the symmetric key K with it, producing the encrypted key K'. Node A then transmits the ciphertext α and the encrypted key K' over the network to node B. While transmitting the data to node B, node A also broadcasts the transaction, which contains the encrypted data, sender identity, and timestamp, to the other verification nodes in the blockchain network (nodes C, D, and E, etc.). Upon receiving the data, node B decrypts the encrypted key K' with its own private key, obtaining the symmetric key K. It then uses K to decrypt the ciphertext α, recovering the original data.
[0031] Preferably, the remote operation and maintenance server is also used to train an intelligent monitoring model based on historical operation data, and send the trained intelligent monitoring model to the terminal computing device. When training the intelligent monitoring model, a topological graph structure reflecting the connection relationship between each unit in the electrical cabinet is constructed based on the units in the electrical cabinet and the connection relationship between them, and the historical operation data of each unit in the electrical cabinet is converted into corresponding node features, wherein the intelligent monitoring model is constructed based on the GAT algorithm.
[0032] In a preferred embodiment of the present invention, the Graph Attention Network (GAT) is an advanced algorithm based on a graph neural network, which is specifically used to process graph structured data. By introducing an attention mechanism, it can dynamically calculate the relationship weights between nodes in the graph, thereby capturing the complex dependencies between nodes. The GAT algorithm performs well in tasks such as node classification, graph classification, and link prediction, and is particularly suitable for processing data with complex topological structures. In the present invention, the GAT algorithm is mainly used to construct an intelligent monitoring model to achieve status detection and fault prediction of each unit in the electrical cabinet. When training the intelligent monitoring model, first, based on the connection relationship between the units in the electrical cabinet, a graph reflecting the topological structure between devices is constructed. The nodes in the graph represent each unit in the electrical cabinet, and the edges represent the connection relationship between the units. The historical operating data of each unit in the electrical cabinet is then converted into node features, which are used as input to the GAT algorithm to train the intelligent monitoring model.
[0033] like Figure 3 As shown, Figure 3 The process of detecting the status and predicting faults of each unit in the electrical cabinet by the intelligent monitoring model is demonstrated. Preferably, based on the collected real-time operation data and using the preset intelligent monitoring model, detecting the status and predicting faults of each unit in the electrical cabinet includes: inputting the collected real-time operation data into the trained intelligent monitoring model; the intelligent monitoring model converts the real-time operation data of each unit in the electrical cabinet into corresponding node features; the GAT algorithm judges the status of each unit in the electrical cabinet and predicts faults based on the node features and the topological graph structure. Using the GAT algorithm to judge the status of each unit in the electrical cabinet and predict faults based on the node features and the topological graph structure includes: using the GAT algorithm to combine the node features into a feature vector; based on the feature vector, using a preset classifier to classify the status of each unit in the electrical cabinet to obtain the status of each unit in the electrical cabinet; based on the feature vector, using a preset regression model to predict the probability of failure of each unit in the electrical cabinet to obtain a corresponding fault probability value. If the fault probability value is greater than a preset threshold, the unit corresponding to the fault probability value has a fault risk; and determining the faulty unit based on the topological graph structure.
[0034] In a preferred embodiment of the present invention, when monitoring an electrical cabinet, the real-time operating data of each unit in the electrical cabinet is first converted into node features as input to the GAT algorithm. The GAT algorithm then calculates the relationship weights between nodes through an attention mechanism, captures the mutual influence between devices, and aggregates the information of neighboring nodes based on the attention weights to update the node features. Finally, the GAT algorithm determines the operating status of each unit in the electrical cabinet based on the updated node features and predicts potential faults. The dynamic weight calculation and topology structure perception capabilities adopted by the present invention significantly improve the intelligence level and prediction accuracy of the system, and provide strong technical support for the safe operation and efficient operation and maintenance of power electrical cabinets.
[0035] Taking a circuit breaker as an example, we first use the GAT algorithm to obtain node feature representations. The node feature vectors are then input into a Softmax classifier, which categorizes the electrical cabinet unit status into three categories: normal, abnormal warning, and fault. For example, if the output probability vector for a circuit breaker is [0.1, 0.7, 0.2], it indicates an abnormal warning state. For fault prediction, a three-layer multilayer perceptron, for example, is used to input the node feature vectors and output the probability of a fault. For example, a circuit breaker with a predicted failure probability of 0.85 indicates a high risk of failure.
[0036] like Figure 4 As shown, for example, there are nodes i and j in the electrical cabinet. The steps of the GAT algorithm to monitor the electrical cabinet are as follows: First, calculate the attention coefficient e between node i and node j ij :
[0037] e ij =a(Wh i ,Wh j ) (1)
[0038] Where W is the weight matrix, a is the shared attention mechanism function, and h i and h j Represent the real-time running data of node i and node j respectively. Subsequently, the softmax function is used to normalize the attention coefficient:
[0039]
[0040] Among them, α ij represents the normalized attention weight, i.e., the contribution of node j to node i, and N(i) represents the set of neighboring nodes of node i. Then, according to the attention weight, the information of neighboring nodes is aggregated and the node features are updated. The updated feature h of node i is i ' is expressed as:
[0041]
[0042] Here, σ represents a nonlinear activation function. Finally, based on the updated node features, the operating status of each unit in the electrical cabinet is determined and potential faults are predicted.
[0043] Preferably, an operation and maintenance monitoring plan is generated based on the historical operation data of each unit in the electrical cabinet, including: using a trained decision generation model to generate an operation and maintenance monitoring plan, the decision generation model is constructed and trained based on a deep reinforcement learning algorithm, and the deep reinforcement learning algorithm is configured to: use the operating status of the electrical cabinet as the environment, the operation and maintenance operation as the intelligent agent action, and reduce the number of failures and reduce the operation and maintenance cost as the reward function, and generate an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet.
[0044] In a preferred embodiment of the present invention, deep reinforcement learning (DRL) is a combination of reinforcement learning (RL) and deep learning (DL), which aims to learn the optimal strategy through the interaction between the agent and the environment. In the present invention, the deep reinforcement learning algorithm is used to generate and optimize the operation and maintenance monitoring plan. In the present invention, the deep reinforcement learning algorithm models the operating state of the electrical cabinet as a reinforcement learning environment, trains the agent to learn the optimal operation and maintenance strategy, and generates an operation and maintenance monitoring plan based on the trained agent. Deep reinforcement learning realizes the intelligent optimization of operation and maintenance strategies through environment modeling, agent training, operation and maintenance plan generation and dynamic adjustment, providing strong technical support for the intelligent operation and maintenance of power electrical cabinets.
[0045] Preferably, the remote operation and maintenance monitoring system further includes: a data processing module for cleaning the acquired operating data of the electrical cabinet after obtaining the operating data through the data acquisition module, removing abnormal values and noise data, and performing normalization processing.
[0046] In a preferred embodiment of the present invention, after the acquisition module collects data, the data processing module performs cleaning and normalization to remove outliers and noise data, ensuring the accuracy and reliability of the data, and then uses the processed data as input to the intelligent monitoring model. Among them, the normalization process can use min-max normalization:
[0047]
[0048] Among them, x is the original data, X is the data set, and x' is the normalized data.
[0049] Preferably, the data acquisition module includes a plurality of sensors for collecting operating parameters of the electrical cabinet and a camera for acquiring images of equipment inside the electrical cabinet.
[0050] In a preferred embodiment of the present invention, data acquisition is a fundamental component of the remote operation and maintenance monitoring system for intelligent power cabinets. The operating status of the cabinet requires comprehensive monitoring using a variety of sensors and cameras to obtain comprehensive operational data and equipment status information. Sensors and cameras can provide a variety of data types (such as electrical parameters, environmental data, and image data), providing rich information for subsequent analysis and decision-making. Image data, combined with sensor data, enables more accurate diagnosis of equipment faults. The data collected by sensors and cameras serves as input to intelligent monitoring models for status detection and fault prediction.
[0051] Preferably, the communication module is configured to use 5G communication technology to send the data collected by the data acquisition module to the terminal computing device.
[0052] In a preferred embodiment of the present invention, efficient and reliable communication between the data acquisition module and the terminal computing device in a remote operation and maintenance monitoring system is key to achieving real-time monitoring and rapid response. This invention leverages the high bandwidth characteristics of 5G communication technology to achieve rapid transmission of large-scale data. Furthermore, 5G's low latency ensures real-time data transmission. This system meets the transmission requirements for electrical cabinet operating data and image data, is suitable for large-scale power systems, and can rapidly transmit fault warning information, ensuring timely processing.
[0053] Preferably, the terminal computing device adjusts the preset operation and maintenance monitoring plan including: establishing a data model based on the historical operation data of the electrical cabinet; simulating the impact of different adjustment schemes on the overall operation of the electrical cabinet through the data model based on the results of status detection and fault prediction; determining the optimal adjustment scheme based on the determined impact, and adjusting the preset operation and maintenance monitoring plan according to the determined adjustment scheme.
[0054] In a preferred embodiment of the present invention, a terminal computing device builds a data model based on historical data, combined with the operational characteristics of the electrical cabinet and relevant domain knowledge. For example, a model based on time series analysis is constructed to describe the temporal trends of the electrical cabinet's operating parameters. Once the data model is established, it is used to evaluate the impact of different O&M monitoring plan adjustments on the overall operation of the electrical cabinet. For each possible adjustment plan, such as preemptively maintaining a unit, increasing the monitoring frequency of a unit, or replacing a certain component, the relevant parameters are input into the data model. Based on its internal algorithms and learned knowledge, the data model simulates the changes in the electrical cabinet's operating status after implementing these adjustment plans. For example, by simulating the preemptive replacement of a key unit's components, it is possible to predict whether the electrical cabinet's failure rate will decrease over a period of time, whether its operational stability will improve, and the extent of its impact on the overall power supply. In this way, various adjustment plans can be quantitatively evaluated, comparing their performance on various metrics (such as reliability, cost-effectiveness, and maintenance costs), and ultimately determining the optimal adjustment plan.
[0055] For example, if status monitoring indicates that the operating parameters of a unit in an electrical cabinet are within the normal range but approaching critical values, while a fault warning has not yet been triggered, a potential risk exists. The terminal computing device will fine-tune the pre-set O&M monitoring plan, for example, increasing the monitoring frequency of that unit from hourly to half-hourly to more closely monitor its operating status. It will also move up the next preventive maintenance schedule for that unit and schedule a comprehensive inspection and commissioning by professional O&M personnel to ensure continued stable operation. If fault prediction results indicate a high probability of failure for that unit in the near future, the terminal computing device will quickly make significant adjustments to the O&M monitoring plan. For example, it will immediately suspend operations related to that unit to prevent the failure from escalating and causing further losses. The O&M team will then be notified to prioritize addressing the potential fault and dispatch appropriate maintenance personnel and equipment to the site. Simultaneously, the O&M monitoring schedule for other units will be replanned to ensure that while the faulty unit is addressed, normal operation of other units is not affected.
[0056] Preferably, when training the intelligent monitoring model, the Adam optimizer is used to adjust the parameters of the intelligent monitoring model, and the intelligent monitoring model is optimized through a preset weighted loss function, wherein the weighted loss function is configured as a loss function obtained by fusing the cross entropy loss function and the mean square error loss function and assigning different weights.
[0057] In a preferred embodiment of the present invention, training the intelligent monitoring model is a core component of the remote operation and maintenance monitoring system. By designing the optimizer and loss function, the model's predictive accuracy and generalization capabilities can be improved. This invention utilizes the Adam optimizer, which combines the advantages of momentum and adaptive learning rates, making it suitable for large-scale data and complex models, consistent with power systems. Furthermore, this invention employs a fusion loss function approach to optimize the intelligent monitoring model.
[0058] Because the intelligent monitoring model in the present invention performs state detection and fault prediction on the electrical cabinet, which belong to classification tasks and regression tasks respectively, and the cross entropy loss function is used for classification tasks, and the output is a discrete category label or probability distribution, and the mean square error loss function is used for regression tasks, and the output is a continuous value. The cross entropy loss function and the mean square error loss function have different task types and output forms, so the two loss functions cannot be used directly at the same time. The present invention combines these two loss functions through multi-task learning to obtain a fusion loss function, which is expressed as:
[0059] Total Loss=αCross-Entropy Loss+βMSE Loss (5)
[0060] Among them, α and β are the weight coefficients of the cross-entropy loss function and the mean square error loss function, MSELoss. By fusing the loss function, the present invention can more comprehensively optimize the intelligent monitoring model and improve the accuracy of the intelligent monitoring model output.
[0061] In summary, the present invention provides a remote operation and maintenance monitoring system for smart power electrical cabinets, which realizes real-time monitoring of the operating status of smart power electrical cabinets, accurate fault prediction, and intelligent management of operation and maintenance plans by integrating advanced technologies such as 5G communication, blockchain, distributed storage, graph attention network (GAT) and deep reinforcement learning (DRL); uses blockchain technology to ensure the security of data transmission and storage, combines distributed storage and smart contracts to achieve efficient management and access control of data, and optimizes operation and maintenance strategies through intelligent algorithms, significantly improving operation and maintenance efficiency, system reliability and security, and providing a comprehensive and efficient solution for the intelligent operation and maintenance of smart power electrical cabinets.
[0062] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0063] Additionally, the terms "system" and "network" are often used interchangeably. The term "and / or" is simply used to describe a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates an "or" relationship between the related objects.
[0064] It should be understood that in the embodiments of the present invention, "B corresponding to A" means that B is associated with A and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information.
[0065] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0066] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0067] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.
[0068] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of the embodiments of the present invention.
[0069] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.
[0070] Through the description of the above embodiments, it will be clear to those skilled in the art that the present invention can be implemented using hardware, firmware, or a combination thereof. When implemented using software, the above functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media includes any medium that facilitates the transmission of computer programs from one location to another. Storage media can be any available medium that can be accessed by a computer. By way of example and not limitation, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer. In addition, any connection can be appropriately considered a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, wherein disks usually reproduce data magnetically, while discs use lasers to reproduce data optically. The above combinations should also be included in the scope of protection of computer-readable media.
[0071] In short, the above is only a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A remote operation and maintenance monitoring system for intelligent power electrical cabinets, characterized in that: The remote operation and maintenance monitoring system includes: Remote operation and maintenance server, used to generate and arrange operation and maintenance monitoring plans based on the historical operation data of each unit in the electrical cabinet; Data acquisition module, used to collect real-time operating data of each unit in the electrical cabinet; The terminal computing device is used to perform status detection and fault prediction on each unit in the electrical cabinet based on the collected real-time operation data and using the preset intelligent monitoring model, and adjust and execute the preset operation and maintenance monitoring plan according to the results.
2. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The remote operation and maintenance monitoring system also includes a communication module for data exchange between the data acquisition module and the terminal computing device. Among them, blockchain technology is used to build a secure channel between the data acquisition module and the terminal computing device, encrypt the transmitted data, and verify the integrity and legitimacy of the transmitted data through a consensus mechanism.
3. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The terminal computing device includes a storage module for storing operation data within a preset time period and corresponding status detection data and fault prediction data. The storage module is configured to perform distributed storage of the operation data through distributed storage nodes, divide the operation data into multiple small data blocks, store them on different nodes, and use smart contract technology to manage the access rights to the operation data.
4. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The remote operation and maintenance server is also used to train the intelligent monitoring model based on historical operation data and send the trained intelligent monitoring model to the terminal computing device. When training the intelligent monitoring model, a topological structure reflecting the connection relationship between the units in the electrical cabinet is constructed according to the connection relationship between the units in the electrical cabinet. Convert the historical operation data of each unit in the electrical cabinet into corresponding node features, Among them, the intelligent monitoring model is built based on the GAT algorithm.
5. The remote operation and maintenance monitoring system according to claim 4, characterized in that: Based on the collected real-time operating data and using the preset intelligent monitoring model, the status of each unit in the electrical cabinet is detected and fault prediction is carried out, including: Input the collected real-time operation data into the trained intelligent monitoring model; The intelligent monitoring model converts the real-time operating data of each unit in the electrical cabinet into corresponding node features; The GAT algorithm is used to determine the status of each unit in the electrical cabinet and perform fault prediction based on node characteristics and topological graph structure.
6. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The method of using the GAT algorithm to judge the status of each unit in the electrical cabinet and perform fault prediction based on node characteristics and topological graph structure includes: Using the GAT algorithm, the node features are combined into a feature vector; According to the feature vector, the status of each unit in the electrical cabinet is classified by a preset classifier to obtain the status of each unit in the electrical cabinet; Based on the characteristic vector, a preset regression model is used to predict the probability of failure of each unit in the electrical cabinet to obtain a corresponding failure probability value. If the failure probability value is greater than a preset threshold, the unit corresponding to the failure probability value has a failure risk; According to the topology structure, a faulty unit is determined.
7. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The generation and arrangement of an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet includes: Use the trained decision generation model to generate operation and maintenance monitoring plans, The decision generation model is built and trained based on a deep reinforcement learning algorithm, which is configured to: The operating status of the electrical cabinet is used as the environment, the operation and maintenance operation is used as the agent action, and the reward function is to reduce the number of failures and reduce the operation and maintenance costs. Generate an operation and maintenance monitoring plan based on the historical operation data of each unit in the electrical cabinet.
8. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The remote operation and maintenance monitoring system also includes: The data processing module is used to obtain the operating data of the electrical cabinet through the data acquisition module, clean the obtained operating data, remove abnormal values and noise data, and perform normalization processing.
9. The remote operation and maintenance monitoring system according to claim 1, characterized in that: The terminal computing device adjusts the preset operation and maintenance monitoring plan including: Establish a data model based on the historical operating data of the electrical cabinet; Based on the results of status detection and fault prediction, the data model is used to simulate the impact of different adjustment schemes on the overall operation of the electrical cabinet; Based on the determined impact, the optimal adjustment plan is determined, and the preset operation and maintenance monitoring plan is adjusted according to the determined adjustment plan.
10. The remote operation and maintenance monitoring system according to claim 1, characterized in that: During training, the intelligent monitoring model uses the Adam optimizer to adjust the parameters of the intelligent monitoring model and optimizes the intelligent monitoring model through a preset weighted loss function. Among them, the weighted loss function is configured as a loss function obtained by fusing the cross entropy loss function and the mean square error loss function and assigning different weights.