Remote operation and maintenance method, system and equipment for wide-area substation automation equipment

Abstract

The invention relates to the technical field of power system automation, discloses a remote operation and maintenance method, system and equipment for wide-area substation automation equipment, and aims to solve the problems of data islanding, inaccurate diagnosis, strategy stiffness, low safety cooperation efficiency and the like in remote operation and maintenance of the wide-area substation automation equipment. According to the method and the system, a comprehensive platform is constructed, and multi-source heterogeneous data acquisition, intelligent fusion analysis, fault diagnosis and prediction, adaptive strategy generation, security remote control and block chain evidence storage and collaborative operation and maintenance are integrated. The method can solve the problem of data islands, improve the diagnosis and prediction precision and timeliness, optimize the operation and maintenance strategy, prolong the service life of equipment, and enhance the remote operation and maintenance safety and cooperation efficiency.

Description

Technical Field

[0001] This invention belongs to the field of power system automation, and specifically relates to a remote operation and maintenance method, system and equipment for wide-area substation automation equipment. Background Technology

[0002] Substations are the core hubs in power transmission and distribution networks. The various automated equipment deployed inside them undertakes key functions such as monitoring, control, protection, and communication. With the continuous growth in the number of substations and their increasingly wide distribution, efficient and reliable operation and maintenance management of these widely distributed automated equipment has become an increasingly arduous challenge.

[0003] Traditional substation equipment operation and maintenance primarily relies on manual inspections and on-site operations. Its inherent limitations include low response efficiency, high labor costs, and difficulty in achieving real-time, continuous monitoring of widely distributed equipment. Although some substations have introduced rudimentary remote monitoring systems capable of transmitting some equipment operating data, these systems typically focus on passive data collection and status display rather than in-depth intelligent analysis, proactive fault warnings, and remote diagnosis and intervention, resulting in limited improvements in operation and maintenance efficiency.

[0004] Existing technologies still have many shortcomings in the remote operation and maintenance of automation equipment in wide-area substations. For example, current systems suffer from poor flexibility and scalability in information exchange between the master station and substations. Most remote monitoring systems lack unified compatibility with communication protocols for different manufacturers and models of equipment, resulting in severe data silos and hindering the formation of a comprehensive equipment health view and cross-regional collaborative management. For intermittent or hidden faults, existing systems often struggle to provide real-time and accurate early warnings and location, requiring significant manual intervention in fault diagnosis and handling, leading to long response times and greatly impacting power supply reliability. Furthermore, due to the wide distribution and geographical dispersion of equipment, remote operation and maintenance strategies struggle to effectively coordinate with on-site operation and maintenance resources, resulting in low maintenance efficiency and difficulty in achieving preventative maintenance and fault trend prediction based on big data analysis, further increasing the risk of unplanned equipment downtime. In the context of increasingly severe cybersecurity challenges, existing remote operation and maintenance solutions also have potential vulnerabilities in data transmission encryption, access control, and authentication, failing to meet high-level security requirements. The aforementioned problems severely restrict the intelligent and refined development of wide-area substation operation and maintenance, and there is an urgent need for a more advanced, efficient, and safe remote operation and maintenance method, system, and equipment for wide-area substation automation equipment. Summary of the Invention

[0005] To address the aforementioned technical problems, this application is proposed. Embodiments of this application provide a remote operation and maintenance method, system, and equipment for wide-area substation automation equipment. This involves constructing a comprehensive platform integrating multi-source heterogeneous data acquisition, intelligent fusion analysis, fault diagnosis and prediction, adaptive strategy generation, and secure remote control. The platform utilizes advanced data processing technology to uniformly aggregate and deeply fuse real-time operating data, alarm logs, configuration information, and environmental parameters from different types and manufacturers of automation equipment within the substation. Then, a multimodal deep learning model is used to intelligently analyze the fused data, enabling early warning, precise location, and health status assessment of potential equipment faults. Based on the intelligent analysis results, combined with a pre-set operation and maintenance strategy library and reinforcement learning mechanism, adaptive generation or optimization of maintenance, repair, parameter adjustment, and firmware upgrade schemes are achieved. Simultaneously, blockchain technology and multiple encryption mechanisms are introduced to ensure the security, integrity, and traceability of remote operation and maintenance command transmission and data interaction, and to provide multi-party collaborative capabilities, thereby comprehensively improving the operation and maintenance efficiency, reliability, and security of wide-area substation automation equipment.

[0006] According to one aspect of this application, a remote operation and maintenance method for wide-area substation automation equipment is provided, comprising: By deploying equipment data acquisition units in multiple wide-area substations, real-time data collection is achieved from all automation equipment within the substations, including operational status data, alarm event data, configuration data, and environmental monitoring data. The collected heterogeneous data undergoes preliminary format conversion and protocol adaptation to achieve unified data access. The collected operational status data includes, but is not limited to, real-time analog and digital data such as voltage, current, frequency, power, temperature, vibration, and relay operation counts. The collected alarm event data includes, but is not limited to, relay protection operation reports, abnormal events in monitoring and control devices, communication link interruption alarms, and equipment over-temperature alarms. The collected configuration data includes the automation equipment's model, serial number, firmware version, operating parameter settings, and network address. The collected environmental monitoring data includes the substation's internal ambient temperature, humidity, and dust concentration. The equipment data acquisition units, for example, employ embedded industrial-grade processors and are equipped with multiple communication interfaces, such as Ethernet, RS485, and RS232 interfaces, supporting various mainstream communication protocols in substation automation, including IEC 61850, Modbus, and DNP3 protocols.

[0007] The data preprocessing module performs deep cleaning, standardization, feature extraction, and multimodal fusion on heterogeneous data from different wide-area substations and various types of automation equipment. Deep cleaning includes outlier detection, missing value imputation, and data denoising, for example, using statistical methods or machine learning-based anomaly detection algorithms to identify and correct errors or inconsistencies in the data. Standardization maps data with different dimensions and numerical ranges to a unified scale, such as using min-max normalization or Z-score normalization. Feature extraction targets time-series data, extracting time-domain features such as mean, variance, kurtosis, and skewness, and frequency-domain features such as power spectral density, harmonic components, and wavelet transform coefficients. Multimodal fusion aligns and integrates real-time operating data, alarm event data, configuration data, and environmental monitoring data in both time and space dimensions to construct a unified equipment health status representation vector or multidimensional tensor, for example, using attention mechanisms or convolutional neural networks for deep fusion of multimodal features.

[0008] The intelligent diagnosis and prediction module, based on the fused equipment health status representation, utilizes deep learning models to perform fault diagnosis, health assessment, and lifespan prediction for automated equipment. Fault diagnosis, for example, employs Long Short-Term Memory networks, gated recurrent unit networks, or Transformer networks to learn temporal features to identify and classify potential fault modes; or it uses graph neural networks combined with substation equipment topology to analyze fault propagation paths and locate root causes. Health assessment, for example, uses support vector machines or Gaussian mixture models to map the current equipment state to a health index or level, assessing the equipment's operational health status in real time. Lifespan prediction, for example, uses methods based on degradation models or deep neural networks to predict the remaining service life of the equipment. The intelligent diagnosis and prediction module provides detailed outputs of the identified fault type, fault level, fault location, and remaining lifespan, along with confidence level indicators.

[0009] The adaptive operation and maintenance strategy generation module intelligently generates or optimizes operation and maintenance strategies for specific automated equipment based on the output of the intelligent diagnosis and prediction module, combined with a pre-set operation and maintenance strategy library and a reinforcement learning mechanism. The pre-set operation and maintenance strategy library includes standardized maintenance procedures, maintenance plans, parameter adjustment suggestions, and firmware upgrade guidelines for different equipment types and fault modes. The reinforcement learning mechanism aims to maximize equipment operational reliability and minimize operation and maintenance costs. Based on real-time diagnostic results and historical operation and maintenance effects, it dynamically adjusts and optimizes the strategy generation rules, for example, by learning the optimal sequence of operation and maintenance actions in a simulated environment using deep Q-networks or actor-critic algorithms. The operation and maintenance strategies include, but are not limited to, detailed maintenance plans, preventative maintenance tasks, parameter tuning instructions, firmware version update suggestions, and fault repair steps.

[0010] The secure remote control and execution module performs security verification and permission approval on the generated operation and maintenance policies. Upon approval, control commands or configuration update packages are sent to the target wide-area substation automation equipment via an encrypted communication channel. The security verification, for example, uses formal verification methods or expert systems to check the legality, compatibility, and potential risks of the commands. The permission approval adopts a multi-level approval process, such as tiered authorization by operation and maintenance personnel or administrators with different roles. The encrypted communication channel uses end-to-end encryption technology, such as based on transport layer security protocols or virtual private network protocols, to ensure the confidentiality and integrity of commands during transmission. At the device end, a security gateway decrypts the commands, verifies digital signatures, and authenticates the identity to ensure the credibility of the command source. After successful verification, the device executor executes the commands.

[0011] The blockchain-based evidence storage and traceability module stores all important operation and maintenance records, equipment status changes, diagnostic results, and policy execution logs in an immutable manner using distributed ledger technology. This module may employ a consortium blockchain architecture, with substations, operation and maintenance units, and equipment suppliers as participating nodes. Each operation and maintenance operation or data change generates a transaction, attaches a timestamp and digital signature, and is stored on the blockchain. The module supports rapid retrieval and full-chain traceability of any historical operation and maintenance event, ensuring the authenticity, transparency, and non-repudiation of operation and maintenance data. The module can also utilize smart contracts to automate certain condition-triggered permission management, data sharing, or event notification logic.

[0012] The human-computer interaction and collaborative operation and maintenance platform provides operation and maintenance personnel with a unified graphical user interface and a multi-party collaborative working environment. This platform displays the real-time operating status, health assessment results, fault alarm information, and recommended operation and maintenance strategies for all automated equipment in the wide-area substation. The platform offers functions such as equipment ledger management, historical data query, report generation, work order management, and an operation and maintenance knowledge base. The multi-party collaborative working environment integrates tools such as online chat, video conferencing, and document sharing, facilitating remote collaboration, technical exchange, and experience sharing among operation and maintenance personnel from different specialties and departments, thereby improving the efficiency of handling complex faults.

[0013] According to another aspect of this application, a remote operation and maintenance system for wide-area substation automation equipment is provided, comprising: The equipment data acquisition unit is deployed across multiple wide-area substations to collect real-time operational status data, alarm event data, configuration data, and environmental monitoring data from all automation equipment within the substations. It performs preliminary format conversion and protocol adaptation on the collected heterogeneous data to achieve unified data access. The operational status data includes, but is not limited to, real-time analog and digital data such as voltage, current, frequency, power, temperature, vibration, and relay operation counts. The alarm event data includes, but is not limited to, relay protection operation reports, abnormal events in monitoring and control devices, communication link interruption alarms, and equipment over-temperature alarms. The configuration data includes the model, serial number, firmware version, operating parameter settings, and network address of the automation equipment. The environmental monitoring data includes the substation's internal ambient temperature, humidity, and dust concentration. The equipment data acquisition unit, for example, uses an embedded industrial-grade processor and is equipped with multiple communication interfaces, such as Ethernet, RS485, and RS232 interfaces, supporting various mainstream communication protocols in substation automation, including IEC 6180, Modbus, and DNP3 protocols.

[0014] The data preprocessing module performs deep cleaning, standardization, feature extraction, and multimodal fusion on heterogeneous data from different wide-area substations and various types of automation equipment. Deep cleaning includes outlier detection, missing value imputation, and data denoising, for example, using statistical methods or machine learning-based anomaly detection algorithms to identify and correct errors or inconsistencies in the data. Standardization maps data with different dimensions and numerical ranges to a unified scale, such as using min-max normalization or Z-score normalization. Feature extraction targets time-series data, extracting time-domain features such as mean, variance, kurtosis, and skewness, and frequency-domain features such as power spectral density, harmonic components, and wavelet transform coefficients. Multimodal fusion aligns and integrates real-time operating data, alarm event data, configuration data, and environmental monitoring data in both time and space dimensions to construct a unified equipment health status representation vector or multidimensional tensor, for example, using attention mechanisms or convolutional neural networks for deep fusion of multimodal features.

[0015] The intelligent diagnosis and prediction module is used to perform fault diagnosis, health assessment, and lifespan prediction of automated equipment based on the fused equipment health status representation and employing deep learning models. Fault diagnosis, for example, uses Long Short-Term Memory networks, gated recurrent unit networks, or Transformer networks to learn temporal features to identify and classify potential fault modes; or it uses graph neural networks combined with substation equipment topology to analyze fault propagation paths and locate root causes. Health assessment, for example, uses Support Vector Machines or Gaussian Mixture Models to map the current state of the equipment to a health index or level, assessing the equipment's operational health status in real time. Lifespan prediction, for example, uses methods based on degradation models or deep neural networks to predict the remaining service life of the equipment. The intelligent diagnosis and prediction module outputs the identified fault type, fault level, fault location, and remaining lifespan, and provides a confidence index.

[0016] The adaptive operation and maintenance strategy generation module is used to intelligently generate or optimize operation and maintenance strategies for specific automated equipment based on the output results of the intelligent diagnosis and prediction module, combined with a preset operation and maintenance strategy library and a reinforcement learning mechanism. The preset operation and maintenance strategy library contains standardized maintenance procedures, maintenance plans, parameter adjustment suggestions, and firmware upgrade guidelines for different equipment types and fault modes. The reinforcement learning mechanism uses maximizing equipment operational reliability and minimizing operation and maintenance costs as its objective function. Based on real-time diagnostic results and historical operation and maintenance effects, it dynamically adjusts and optimizes the strategy generation rules, for example, by learning the optimal sequence of operation and maintenance actions in a simulated environment through deep Q-networks or actor-critic algorithms. The operation and maintenance strategies include, but are not limited to, detailed maintenance plans, preventative maintenance tasks, parameter tuning instructions, firmware version update suggestions, and fault repair steps.

[0017] The secure remote control and execution module performs security verification and permission approval on the generated operation and maintenance policies. Upon approval, it issues control commands or configuration update packages to the target wide-area substation automation equipment via an encrypted communication channel. The security verification, for example, uses formal verification methods or expert systems to check the legality, compatibility, and potential risks of the commands. The permission approval employs a multi-level approval process, such as tiered authorization by operation and maintenance personnel or administrators with different roles. The encrypted communication channel uses end-to-end encryption technology, such as based on transport layer security protocols or virtual private network protocols, to ensure the confidentiality and integrity of commands during transmission. The secure remote control and execution module integrates a security gateway for decrypting commands, verifying digital signatures, and authenticating identities at the device end, ensuring the credibility of the command source. After successful verification, the device executor executes the commands.

[0018] The blockchain-based evidence storage and traceability module is used to immutably store all important operation and maintenance records, equipment status changes, diagnostic results, and policy execution logs using distributed ledger technology. This module may employ a consortium blockchain architecture, with substations, operation and maintenance units, and equipment suppliers as participating nodes. The module treats each operation and data change as a transaction, attaching a timestamp and digital signature before storing it on the blockchain. The module supports rapid retrieval and full-chain traceability of any historical operation and maintenance event, ensuring the authenticity, transparency, and non-repudiation of operation and maintenance data. The module can also utilize smart contracts to automate certain condition-triggered permission management, data sharing, or event notification logic.

[0019] The human-machine interaction and collaborative operation and maintenance platform provides operation and maintenance personnel with a unified graphical user interface and a multi-party collaborative working environment. This platform displays the real-time operating status, health assessment results, fault alarm information, and recommended operation and maintenance strategies for all automated equipment in the wide-area substation. The platform provides functions such as equipment ledger management, historical data query, report generation, work order management, and an operation and maintenance knowledge base. The multi-party collaborative working environment integrates tools such as online chat, video conferencing, and document sharing, facilitating remote collaboration, technical exchange, and experience sharing among operation and maintenance personnel from different specialties and departments, thereby improving the efficiency of handling complex faults.

[0020] A remote operation and maintenance device for wide-area substation automation equipment includes a storage device for data storage and a central processing unit for data processing.

[0021] Compared with the prior art, the advantages and positive effects of this application are as follows: This application completely solves the data silo problem caused by inconsistent data formats and communication protocols among different manufacturers and models of automated equipment by introducing a unified acquisition and deep fusion mechanism for multi-source heterogeneous data. Through the construction of a standardized data access layer and multimodal fusion algorithm, seamless integration and deep correlation analysis of operational status data, alarm event data, configuration data, and environmental monitoring data are achieved, providing a comprehensive and high-quality data foundation for subsequent intelligent decision-making.

[0022] This application achieves significant improvements in accuracy and timeliness in fault diagnosis and prediction. Unlike existing technologies that primarily rely on preset thresholds or simple rule matching, this application utilizes advanced deep learning models such as Long Short-Term Memory (LSTM) networks and graph neural networks. These models enable deep learning of the fused temporal, topological, and multimodal features, effectively capturing the early characteristics and evolutionary patterns of complex and hidden faults. This allows for proactive early warning and precise location of potential faults, shifting from passive repair to proactive preventative maintenance, significantly reducing unplanned downtime and operational costs.

[0023] The adaptive operation and maintenance strategy generation mechanism provided in this application overcomes the shortcomings of existing rigid and inflexible operation and maintenance strategies. By combining intelligent diagnostic and prediction results, a preset operation and maintenance strategy library, and reinforcement learning algorithms, this application can dynamically generate or optimize the most suitable and efficient maintenance, repair, parameter adjustment, or firmware upgrade scheme based on the actual health status of the equipment, environmental changes, and historical operation and maintenance results. This adaptive strategy ensures optimal allocation of operation and maintenance resources, effectively extends equipment lifespan, and improves power supply reliability.

[0024] This application offers significant advantages in terms of security and collaborative efficiency for remote operation and maintenance. By introducing end-to-end encrypted communication, multi-level permission approval, digital signature verification, and blockchain-based evidence storage and traceability mechanisms, this application ensures the confidentiality, integrity, authenticity, and traceability of remote control commands, effectively preventing risks of misoperation and malicious attacks. Simultaneously, the human-machine interaction and collaborative operation and maintenance platform integrates rich collaborative tools and information display functions, significantly improving the efficiency of remote collaboration among multi-disciplinary and multi-level operation and maintenance personnel, shortening response and handling times for complex faults, and comprehensively enhancing the remote operation and maintenance level of wide-area substation automation equipment. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall technical solution architecture of the remote operation and maintenance method for wide-area substation automation equipment proposed in this invention; Figure 2 This is a schematic diagram of the core principle framework for generating intelligent diagnosis and adaptive operation and maintenance strategies in this invention. Detailed Implementation

[0026] This invention proposes a remote operation and maintenance method for automation equipment in wide-area substations. This method integrates multi-source heterogeneous data and utilizes advanced artificial intelligence technology to construct a deep perception, intelligent diagnosis, and adaptive operation and maintenance decision-making system for equipment operating status. The aim is to improve operation and maintenance efficiency, reduce operation and maintenance costs, and enhance the reliability and security of power grid operation. The method uses intelligent acquisition terminals deployed at the wide-area substation site to acquire real-time operating data, environmental data, and event logs from the automation equipment. This data is transmitted to a remote operation and maintenance center via a secure and reliable communication network, where it undergoes in-depth data processing, feature extraction, and intelligent analysis. The core component involves using machine learning models to diagnose and predict equipment status in real time, and combining historical operating records and expert knowledge to dynamically generate optimal operation and maintenance strategies. Finally, the operation and maintenance strategies are issued to the field equipment for execution in the form of remote commands, forming a closed-loop intelligent operation and maintenance process.

[0027] The remote operation and maintenance method for wide-area substation automation equipment proposed in this invention has the following overall technical architecture: Figure 1 As shown, the core principle framework for intelligent diagnosis and adaptive operation and maintenance strategy generation is as follows: Figure 2 As shown. This method includes the following main steps: First, real-time monitoring of the operation data of the substation automation equipment is conducted. This step aims to comprehensively and accurately capture the operating status information of various automation equipment within the substation. Specifically, various types of sensors and intelligent electronic devices are deployed on-site, such as voltage sensors, current sensors, temperature sensors, humidity sensors, circuit breaker status indicators, switchgear position sensors, and various relay protection devices and measurement and control devices. These devices can continuously collect real-time operating parameters, including but not limited to three-phase voltage, three-phase current, active power, reactive power, frequency, power factor, internal and external ambient temperature, equipment humidity, gas pressure, insulation status parameters, vibration amplitude, noise intensity, circuit breaker opening and closing status, switch position, and valve opening degree. Simultaneously, various alarm information, event records, operation logs, diagnostic logs, and communication link status information generated by the equipment are also collected.

[0028] The data acquisition process features high sampling rate and high precision. For example, for electrical parameters, the sampling frequency can reach tens of points or even higher per cycle to ensure the capture of transient changes. The analog signals output by the sensors are converted into digital signals by a high-resolution analog-to-digital converter and undergo preliminary data preprocessing, such as noise removal, data smoothing, and unit conversion. This raw data is then locally aggregated and initially integrated on-site through intelligent acquisition terminals. The intelligent acquisition terminal is typically an industrial computer or dedicated data acquisition unit with edge computing capabilities. Its built-in software modules are responsible for reading data from various device interfaces according to a preset acquisition strategy, either periodically or triggered by events. Configuration parameters for the acquisition strategy include the data acquisition cycle, data type filtering rules, and abnormal data reporting thresholds. These parameters can be dynamically adjusted based on the device type and importance. Data verification mechanisms, such as cyclic redundancy check codes or parity bits, are used to ensure data integrity. When data loss or anomalies are detected, the intelligent acquisition terminal initiates a data retransmission mechanism or marks the data quality, and notifies maintenance personnel through a preset alarm mechanism. Furthermore, to ensure data time consistency, the intelligent data acquisition terminal synchronizes with the standard time of the National Time Service Center, and marks all collected data with high-precision timestamps, typically at the millisecond level, for subsequent accurate time-series analysis. Data transmission employs secure industrial protocols, such as the revised IEC 61850 protocol, to ensure reliable data transmission within the local network.

[0029] Subsequently, secure transmission of remote operation and maintenance data is carried out. This step aims to transmit the massive amounts of operational data collected from the field at a wide-area substation to a remote operation and maintenance center in a secure, efficient, and reliable manner. Since substations are typically geographically dispersed, and data transmission involves core power grid information, data security and transmission stability are of paramount importance.

[0030] Data transmission utilizes multiple encryption and authentication mechanisms to ensure data security. First, at the intelligent acquisition terminal or local aggregation gateway in the wide-area substation, all data to be transmitted is encrypted. Commonly used encryption algorithms include 256-bit encryption of the Advanced Encryption Standard (AES) or the Chinese national cryptographic algorithm series. A strict certificate system is employed for encryption key management to ensure the uniqueness and validity of certificates. The data transmission channel typically uses Virtual Private Network (VPN) technology or a dedicated network link based on Multiprotocol Label Switching (MPLS). VPN technology establishes an encrypted tunnel over a public network, encapsulating data transmission between the substation and the operation and maintenance center within a secure tunnel, effectively preventing data from being eavesdropped on or tampered with during transmission. MPLS, through predefined paths and label switching, provides a high-quality, low-latency dedicated transmission channel, suitable for critical data with high real-time requirements. At the packet level, Transport Layer Security (TLS) or Secure Sockets Layer (SSL) protocols are used for end-to-end encryption to further enhance transmission security.

[0031] Data transmission reliability is achieved through various mechanisms. For example, data fragmentation and retransmission mechanisms are employed; when data packet loss or corruption is detected, retransmission is automatically triggered to ensure complete data delivery. Real-time monitoring of the communication link is a crucial means of ensuring reliability. This involves monitoring network latency, packet loss rate, and bandwidth utilization through heartbeat mechanisms and link quality detection. When link quality degrades or is interrupted, the system can automatically switch to a backup link, such as using satellite communication or 4G / 5G mobile networks as supplementary transmission channels, achieving link redundancy. Data compression technologies, such as Gzip or Lempel-Ziv-Welch compression algorithms, compress data before transmission, reducing the amount of data transmitted, thereby reducing bandwidth consumption and accelerating transmission speed, which is particularly important in bandwidth-constrained wide-area network environments. Standardized data transmission interfaces, such as using RESTful APIs based on the TCP / IP protocol suite or Message Queuing Telemetry Transport Protocol (MQTT), ensure interoperability between different types of intelligent acquisition terminals and the data receiving module of the operation and maintenance center. After receiving data at the operation and maintenance center, the data receiving module performs integrity verification and decryption to ensure the originality and confidentiality of the data. In addition, all data transmission activities are recorded in detail in the security audit log so that any abnormal behavior can be traced and analyzed.

[0032] Building upon this foundation, further data management and preprocessing are conducted at the remote operation and maintenance center. This step aims to efficiently store, standardize, clean, and transform the massive amounts of heterogeneous data transmitted from wide-area substations, providing a high-quality, structured data foundation for subsequent intelligent diagnostics and strategy generation.

[0033] The data management module is first responsible for classifying and storing the received raw data. Raw data typically includes real-time telemetry data, remote signaling data, event alarm data, operation log data, and diagnostic messages. This data is stored in a distributed file system such as HDFS or a database specifically optimized for time-series data such as InfluxDB or OpenTSDB to handle its high write speed and large-scale storage requirements. Simultaneously, data metadata, including data source identifiers, acquisition time, data type, data unit, and data checksum, is stored independently in a relational database such as PostgreSQL for easy data querying and indexing. The data lake architecture is used to integrate data with different structures and formats, providing a unified storage and access interface.

[0034] The data preprocessing module performs a series of processes on the stored data to improve its quality. Specifically, this is broken down into the following sub-steps: I. Data Cleaning. This sub-step is used to identify and handle errors, missing values, outliers, and duplicates in the data. Missing value handling methods include mean imputation, median imputation, mode imputation, forward imputation, or backward imputation, or predictive imputation based on machine learning models such as the K-nearest neighbors algorithm. Outlier detection uses statistical methods such as the three Sigma criterion and box plot analysis, or machine learning methods such as isolated forests and local anomaly algorithms for identification, and corrects or removes data according to preset rules. Duplicate data is deleted using unique identifiers.

[0035] II. Data Standardization and Normalization. Data collected by different sensors and devices may have different dimensions and numerical ranges. To eliminate the influence of dimensions on model training, the data needs to be standardized or normalized. Standardization typically uses the Z-score standardization method to transform the data into a distribution with a mean of zero and a variance of one. Normalization uses Min-Max scaling to linearly transform the data to a fixed range, such as [0,1] or [-1,1]. The choice of method depends on the specific requirements of the subsequent machine learning model.

[0036] III. Feature Engineering. This is a crucial step in data preprocessing, aiming to extract or construct features from raw data that are more meaningful for intelligent diagnosis and strategy generation. For time-series data, statistical features such as mean, variance, standard deviation, maximum, minimum, skewness, and kurtosis can be extracted; frequency domain features such as spectral energy and dominant frequency components obtained through Fourier transform (FFT); and time domain features such as zero-crossing rate, waveform kurtosis, and impulse factor can also be generated. Difference features, ratio features, and rolling window statistical features such as rolling average and rolling standard deviation can also be generated to capture data trends and periodicity. For non-time-series data, such as equipment model and installation location, discrete feature encoding such as one-hot encoding and tag encoding can be performed. All these features are organized into a unified feature vector and stored in a feature library for subsequent machine learning models to access. The data flow is clear: from the raw data storage area to the preprocessing module, and then to the feature library. All operation logs and data transformation rules during preprocessing are recorded to ensure the traceability and transparency of the data processing process.

[0037] Furthermore, automated intelligent diagnostics of the equipment are performed. This step is a core component of this remote maintenance methodology, aiming to identify anomalies, potential faults, or signs of performance degradation in equipment operation and predict its future health trends through in-depth analysis of preprocessed and feature-engineered data.

[0038] The intelligent diagnostic unit comprises a series of sophisticated machine learning models and an expert knowledge base. Its specific process is broken down as follows: I. Anomaly Detection and Fault Early Warning. This sub-step utilizes supervised and unsupervised learning models to identify abnormal patterns in equipment operation data. Supervised learning models, such as Support Vector Machines, Random Forests, and Gradient Boosting Trees, build classification models by learning the feature differences between historical fault samples and normal operation samples. When new equipment data is input, the model outputs a fault probability or classification result. Unsupervised learning models, such as K-means clustering, Gaussian mixture models, and autoencoders, learn the inherent structure of normal operation data and identify equipment data deviating from this structure as anomalies, even without explicit fault labels. Setting the anomaly threshold is crucial, as it determines the sensitivity and false alarm rate of the early warning. The threshold parameter is dynamically adjusted based on historical operation data, expert experience, and the severity of the fault. For example, when the partial discharge signal strength of the equipment exceeds a preset threshold or the insulation resistance shows a continuous downward trend, the system will immediately trigger an early warning and indicate the possible fault type, such as insulation aging.

[0039] II. Fault Location and Root Cause Analysis. Once an anomaly is detected and an alert is issued, the goal of this sub-step is to determine the specific location of the fault and its root cause. This typically involves more complex diagnostic models, such as Bayesian networks, causal graph models, or deep learning-based fault diagnosis networks (e.g., convolutional neural networks for pattern recognition of vibration signals, and long short-term memory networks for trend analysis of time-series electrical data). These models can comprehensively analyze data from multiple sensors, event logs, and device topology information to pinpoint the fault source through inference chains or pattern matching. For example, if a temperature sensor indicates localized overheating while a current sensor detects abnormal current fluctuations in a certain phase, the system may infer that winding insulation breakdown or poor contact is the root cause of the fault. The accuracy of root cause analysis depends on the quality and diversity of the model training data, as well as the incorporation of expert knowledge.

[0040] III. Remaining Life Prediction and Degradation Assessment. This sub-step uses advanced prediction models to assess the current health status of equipment and predict its remaining life under current operating conditions. Commonly used models include recurrent neural networks (RNNs), long short-term memory networks (LSTMs), gated recurrent units (GRUs), and autoregressive integral moving average (ARIMA) models, among other time-series prediction models. These models establish a mapping relationship between equipment health indicators and time by learning degradation patterns from historical equipment operating data, such as vibration signal spectrum changes caused by bearing wear and increased dielectric losses due to insulation aging. Health indicators are typically dimensionless values ​​that aggregate multiple key parameters, directly reflecting the degree of equipment degradation. The output of the prediction model is the expected failure time or the time point at which the equipment reaches the failure threshold, along with the prediction interval and confidence level. For example, by monitoring the concentration of characteristic gases in dissolved gas analysis data of transformer oil, the existence of latent oil faults in the transformer can be predicted, and their development trend can be assessed. A simplified remaining life prediction model can be expressed as: in, Indicates the remaining service life of the equipment. Representing the One health characteristic indicator, These are the weight coefficients for the corresponding features. The number of health characteristics is used. This model combines multiple health characteristics into a single prediction of remaining life expectancy using a nonlinear exponential function.

[0041] All diagnostic results, including anomaly type, faulty component, root cause analysis, and remaining life prediction, are stored in a structured data format and updated in real time to the equipment health record. The output of the diagnostic unit serves as an important input for generating the next adaptive operation and maintenance strategy. Uncertainties or low-confidence results encountered during the diagnostic process will be reviewed and intervened by human experts through a dedicated interactive interface to achieve human-machine collaborative diagnosis.

[0042] Subsequently, an adaptive operation and maintenance strategy is generated. This step, based on the results of intelligent diagnosis and combined with multiple factors, intelligently formulates and recommends the optimal operation and maintenance action plan, realizing the transformation from reactive maintenance to predictive and preventive maintenance.

[0043] After receiving the equipment health status, fault warning information, root cause analysis results, and remaining life prediction values ​​output by the intelligent diagnostic unit, the adaptive operation and maintenance strategy generation unit will comprehensively consider the following key factors: I. Equipment Importance and Operating Environment. The position of equipment within the power grid, such as main transformers and ordinary distribution equipment, necessitates different maintenance priorities. Specific characteristics of the operating environment, such as high altitude, high humidity, and highly corrosive areas, also influence strategy selection.

[0044] II. Operation and Maintenance Resources and Cost-Effectiveness. This includes available spare parts inventory, maintenance personnel skills and scheduling, historical maintenance costs, and economic losses caused by power outages. The system will perform a cost-benefit analysis to evaluate the economics of different operation and maintenance solutions.

[0045] III. Safety Production Requirements and Procedures. All generated strategies must comply with the safety production standards, technical procedures, and relevant laws and regulations of the power industry.

[0046] IV. Historical Operation and Maintenance Experience and Expert Knowledge. The tacit knowledge of domain experts and historical operation and maintenance cases are digitized and formalized into rules to build an expert rule base and case base, providing guidance for strategy generation.

[0047] Based on the above comprehensive considerations, the adaptive operation and maintenance strategy generation unit employs a multi-objective optimization algorithm and a rule-based reasoning engine to generate strategies. Multi-objective optimization algorithms, such as the Non-Dominated Sorting Genetic Algorithm (NSGA-II), can find an optimal balance between maximizing equipment availability, minimizing operation and maintenance costs, and minimizing risk. The rule-based reasoning engine maps diagnostic results to specific operation and maintenance actions based on a pre-defined operation and maintenance knowledge graph and decision tree. For example, if the diagnostic results indicate a slight jamming in a circuit breaker mechanism and the remaining life prediction shows a high risk of failure within the next three months, and the substation is currently in a low-load period with similar spare parts available, the system may recommend immediately arranging preventative maintenance, providing detailed maintenance steps, required tools, and estimated man-hours. If the diagnostic results indicate that excessively high ambient temperatures are causing equipment overload risk, the system may recommend adjusting the load distribution strategy or activating auxiliary cooling equipment.

[0048] A core aspect of operation and maintenance strategy generation is the dynamic adjustment of charging voltage. This is particularly important in scenarios involving energy storage devices or automated equipment requiring regulation of their charging and discharging behavior. Based on the equipment's health status, such as battery health, internal impedance, and temperature, as well as the real-time demands of the power grid and the operation and maintenance strategy, the system can generate precise charging voltage regulation commands. For example, when early signs of degradation are detected in the energy storage battery pack, or when the battery temperature is too high, the system will intelligently reduce the charging voltage to slow down the charging rate, thereby extending battery life and preventing thermal runaway. Conversely, when the power grid requires emergency support or when the equipment's health status allows, the system can appropriately increase the charging voltage to accelerate the charging process.

[0049] The strategy generation process produces multi-dimensional outputs, including recommended operational actions, execution priorities, estimated completion times, required resource lists, risk assessment reports, and performance indicator predictions. All strategies undergo internal simulation evaluation before official deployment to verify their feasibility, effectiveness, and security. A simplified strategy optimization objective function can be expressed as: in, The objective function value representing the optimal strategy The cost is due to equipment failure. It is the probability of equipment failure, while This represents the cost of implementing the operation and maintenance strategy. The objective function aims to minimize the total expected cost to balance failure risk and maintenance investment.

[0050] All generated operation and maintenance strategies will be archived and correlated with actual execution results to form a feedback loop for continuous optimization of the strategy generation model and rule base.

[0051] Finally, remote operation and maintenance strategy execution and feedback are performed. This step involves securely and accurately distributing the adaptive operation and maintenance strategy generated in the previous steps to the automation equipment in the wide-area substation for execution, and monitoring the execution results in real time, thus forming an operation and maintenance closed loop.

[0052] The remote execution unit is responsible for translating policy instructions into control commands or configuration parameters that the device can recognize. For example, for devices that require adjustments to operating parameters (such as adjusting protection settings or modifying controller parameters), the system generates control messages conforming to the IEC 61850 MMS protocol or the DNP3 protocol. For devices requiring software upgrades or firmware updates, the system packages the corresponding upgrade files and distributes them through a secure transmission channel. All remote operation commands must undergo a strict authentication and authorization process before being distributed, ensuring that only approved users or the system can execute critical operations. Command data is digitally signed and encrypted to prevent tampering or forgery during transmission.

[0053] After the command is issued, the system monitors the response of the field equipment in real time. This includes receiving execution confirmation information returned by the equipment, new operating status data, operation logs, and any abnormal alarms. For example, when a remote closing command is issued, the system continuously monitors the circuit breaker's closing status feedback signal and confirms successful closing within a preset time. If the operation fails or times out, the system will immediately trigger an abnormal alarm, record the reason for the failure, and notify maintenance personnel to intervene manually or initiate a rollback mechanism. All data during the execution process, including command content, issuance time, execution results, and equipment responses, will be recorded in detail in the maintenance operation log database for auditing and tracking.

[0054] The feedback mechanism is crucial in this step. It compares the actual execution results with the expected strategy effects, forming a learning loop. The executed equipment operation data is re-entered into the data management and preprocessing module, and then sent to the intelligent diagnostic unit for further analysis to evaluate the effectiveness of the maintenance strategy. For example, if a predictive maintenance strategy successfully avoids an equipment failure, or remote parameter optimization significantly improves equipment operating efficiency, this positive feedback will enhance the confidence of the strategy generation model. Conversely, if a strategy fails to achieve the expected results or causes new problems, the system will analyze the reasons and adjust the strategy generation logic or optimize model parameters. This process supports online learning and continuous improvement of the model, ensuring the adaptability and optimality of the maintenance strategy. For example, the model parameters can be fine-tuned using a gradient descent optimization algorithm. in, These are the updated model parameters. These are the parameters of the old model. It's the learning rate. It is a loss function Regarding parameters The gradient, and This represents feedback data after the operation and maintenance are performed.

[0055] Through this closed-loop management of remote execution and feedback, this method enables continuous optimization of operation and maintenance strategies, thereby continuously improving the intelligence level, efficiency, and reliability of operation and maintenance of automation equipment in wide-area substations.

[0056] This invention also proposes a remote operation and maintenance system for wide-area substation automation equipment. This system serves as the physical carrier and software platform for implementing the aforementioned remote operation and maintenance methods. The overall system architecture consists of three main layers: a field data acquisition layer, a data transmission network layer, and a remote operation and maintenance center layer.

[0057] The field data acquisition layer is deployed within the wide-area substation and mainly consists of various automated equipment, sensors, and intelligent acquisition terminals. Automated equipment includes relay protection devices, measurement and control devices, intelligent switchgear, transformers, and circuit breakers. Sensors are responsible for real-time monitoring of physical parameters such as voltage, current, temperature, vibration, humidity, and oil chromatography. The intelligent acquisition terminal is the core component of this layer, integrating a data acquisition module, an edge computing module, and a local communication module. The data acquisition module connects to the field automated equipment and sensors via industrial communication protocols (such as IEC 61850, Modbus, and DNP3) for data polling, subscription, and event-triggered acquisition. The edge computing module has data preprocessing capabilities, enabling local data cleaning, preliminary analysis, and feature extraction, reducing the data processing burden on the remote operation and maintenance center. The local communication module is responsible for transmitting the processed data to the aggregation gateway via the local network. Furthermore, this layer also includes a field aggregation gateway, responsible for integrating and encrypting data from multiple intelligent acquisition terminals and connecting them to the remote operation and maintenance center via a secure communication channel.

[0058] The data transmission network layer, built over a wide geographical area, serves as a bridge connecting the field data acquisition layer and the remote operation and maintenance center layer. This layer primarily consists of fiber optic communication networks, wireless communication networks (such as 5G and LTE), satellite communication links, and VPN tunnels. Its core function is to provide a high-bandwidth, low-latency, highly reliable, and highly secure data transmission channel. The transmission channel features a redundant design, automatically switching to a backup channel when the primary channel fails, ensuring continuous data transmission. Data encryption devices and firewalls function at this layer, ensuring the confidentiality and integrity of data during transmission and preventing unauthorized access and data leakage.

[0059] The remote operation and maintenance center layer is the core processing and decision-making hub of the entire system. This layer consists of a series of functional modules and high-performance computing resources.

[0060] I. Data Access and Storage Module: This module is responsible for receiving all data from the wide-area substation, performing data verification, decryption, and timestamp synchronization, and storing the raw data and preprocessed data in a distributed database or time-series database. This module supports concurrent writing and efficient querying of massive amounts of data.

[0061] II. Data Management and Preprocessing Module: This module cleans, deduplicates, standardizes, and normalizes the stored data. It also includes a feature engineering submodule, capable of extracting statistical features, frequency domain features, time domain features, and high-level abstract features from the raw data to build a feature library for model training.

[0062] III. Intelligent Diagnostic Module: This module integrates various machine learning and deep learning models, such as convolutional neural networks, recurrent neural networks, support vector machines, and decision trees, to achieve equipment fault detection, anomaly identification, root cause analysis, and remaining life prediction. This module has a model inference engine that can process input data and output diagnostic results in real time.

[0063] IV. Adaptive Operation and Maintenance Strategy Generation Module: Based on intelligent diagnostic results, equipment health records, operating environment parameters, operation and maintenance costs, safety procedures, and expert knowledge base, the module dynamically generates optimal operation and maintenance strategies and control instructions using multi-objective optimization algorithms, rule reasoning engines, and reinforcement learning models.

[0064] V. Remote Control and Execution Module: This module is responsible for converting generated operation and maintenance policies into standardized control commands and distributing them to the target substation equipment for execution via a secure communication channel. It also includes a command execution monitoring submodule, which provides real-time feedback on command execution status and results.

[0065] VI. Human-Computer Interaction Module: Provides operation and maintenance personnel with an intuitive and user-friendly interface, including data visualization, equipment health dashboard, alarm management, historical event query, report generation, and remote operation command approval and issuance functions.

[0066] VII. Security Management Module: Responsible for user authentication, access control, operation auditing, intrusion detection, and security incident response, ensuring the security of the entire system.

[0067] 8. Model Training and Update Module: Provides offline and online model training functions, and continuously optimizes the performance of intelligent diagnosis and strategy generation models by utilizing new runtime data and operation and maintenance feedback.

[0068] Through close collaboration among its modules, the system forms an end-to-end intelligent operation and maintenance closed loop, comprehensively improving the intelligence, automation, and efficiency of operation and maintenance of automated equipment in wide-area substations.

[0069] A remote operation and maintenance device for wide-area substation automation equipment includes a storage device for data storage and a central processing unit for data processing.

Claims

1. A method of remote operation and maintenance of a wide-area substation automation device, characterized by, The application relates to a real-time operation and maintenance method and system for wide-area substation automation equipment. The method comprises the following steps: Real-time acquisition of operation state data, alarm event data, configuration data and environment monitoring data of all automation equipment in a substation through a data acquisition unit deployed in the substation, to realize comprehensive coverage of multi-source heterogeneous data; Preliminary format conversion and protocol adaptation of the collected heterogeneous data, and secure transmission to a remote operation and maintenance center through multiple encryption and authentication mechanisms; Deep cleaning, standardization processing, feature extraction and multi-modal fusion of the heterogeneous data in the remote operation and maintenance center, to construct a unified equipment health state representation vector or multi-dimensional tensor; Fault diagnosis, health assessment and life prediction of the automation equipment based on the equipment health state representation, and output of detailed results including fault type, fault level, fault location and remaining life; Adaptive generation or optimization of operation and maintenance strategies for specific automation equipment based on the fault diagnosis, health assessment and life prediction results, in combination with a preset operation and maintenance strategy library and a reinforcement learning mechanism; 2. The wide-area substation automation device remote operation method according to claim 1, characterized in that, Security check and permission approval of the generated operation and maintenance strategies, and issuance of control instructions or configuration update packages to target wide-area substation automation equipment through an encrypted communication channel after the approval, to realize remote precise control of the automation equipment. The real-time acquisition of operation state data, alarm event data, configuration data and environment monitoring data of all automation equipment in a substation comprises the following steps: Real-time acquisition of voltage, current, frequency, power, temperature, vibration or relay action frequency real-time analog and digital data of the automation equipment as the operation state data; Real-time acquisition of relay protection action reports, measurement and control device abnormal events, communication link interruption alarms or equipment over-temperature alarms of the automation equipment as the alarm event data; Real-time acquisition of model number, serial number, firmware version, running parameter set value or network address of the automation equipment as the configuration data; 3. The wide-area substation automation device remote operation method according to claim 2, characterized in that, Real-time acquisition of internal environment temperature, humidity or dust concentration of the substation as the environment monitoring data. The preliminary format conversion and protocol adaptation of the collected heterogeneous data, and secure transmission to a remote operation and maintenance center through multiple encryption and authentication mechanisms, comprise the following steps: Protocol adaptation of the collected heterogeneous data according to IEC 61850, Modbus or DNP3 protocols, and conversion into a unified format; Encryption processing of the unified format data according to AES or SM algorithm; 4. The wide-area substation automation device remote operation method according to claim 3, characterized in that, Data transmission through virtual private network technology or a private network link based on multi-protocol label switching, and end-to-end encryption through TLS or SSL protocol. The deep cleaning, standardization processing, feature extraction and multi-modal fusion of the heterogeneous data comprise the following steps: Deep cleaning of the heterogeneous data, including identification and correction of abnormal values by statistical methods or abnormal detection algorithms based on machine learning models, or missing value processing by mean filling, median filling or prediction based on machine learning models; The heterogeneous data is standardized, including converting the data into a distribution with a mean of zero and a variance of one or mapping to a fixed range by using a Z-score standardization method or a Min-Max scaling method; Feature extraction is performed on the heterogeneous data, including extracting time domain features or frequency domain features for time series data; Multi-modal fusion is performed on the operation state data, alarm event data, configuration data and environmental monitoring data, including deep fusion of multi-modal features by using an attention mechanism or a convolutional neural network.

5. The wide-area substation automation device remote operation method according to claim 4, characterized in that, Based on the device health state representation, deep learning models are used for fault diagnosis, health assessment and life prediction of the automated device, including: Long short-term memory networks, gated recurrent unit networks or Transformer networks are used to learn time series features to identify potential fault patterns and perform fault classification; Graph neural networks are used in combination with substation device topology to analyze fault propagation paths and locate root causes; Support vector machines or Gaussian mixture models are used to map the current state of the device to a health index or grade to assess the operational health of the device in real time; Degradation models or deep neural networks are used to predict the remaining useful life of the device.

6. The wide-area substation automation device remote operation method according to claim 5, characterized in that, According to the results of fault diagnosis, health assessment and life prediction, combined with a pre-set operation and maintenance strategy library and a reinforcement learning mechanism, operation and maintenance strategies for specific automated devices are generated or optimized, including: The pre-set operation and maintenance strategy library includes standardized repair procedures, maintenance plans, parameter adjustment suggestions or firmware upgrade guidelines for different device types or different fault modes; The reinforcement learning mechanism takes maximizing device operation reliability or minimizing operation and maintenance cost as the objective function, and dynamically adjusts and optimizes the strategy generation rules through deep Q networks or actor-critic algorithms; The operation and maintenance strategy includes detailed repair plans, preventive maintenance tasks, parameter tuning instructions, firmware version update suggestions, fault repair steps, or further includes dynamic adjustment of charging voltage instructions.

7. The wide-area substation automation device remote operation method according to claim 6, characterized in that, The generated operation and maintenance strategy is subjected to security verification and permission approval, and after approval, control instructions or configuration update packages are issued to the target wide-area substation automation device through an encrypted communication channel, including: The legality, compatibility or potential risks of the control instructions or configuration update packages are checked by formal verification methods or expert systems for security verification; Multi-level approval processes are used for hierarchical authorization by operation and maintenance personnel or administrators of different roles; The encrypted communication channel uses the Transport Layer Security protocol or the Virtual Private Network protocol to ensure the confidentiality, integrity and authenticity of the instructions during transmission; At the target wide-area substation automation device end, the security gateway is used for decryption, digital signature verification and identity authentication of the instructions, and after verification, the device executor executes the instructions.

8. The wide-area substation automation device remote operation method according to claim 7, characterized in that, Through the blockchain storage and traceability module, important data such as operation and maintenance operation records, device state changes, diagnosis results and strategy execution logs are stored and traced in the whole chain by distributed ledger technology, including: Adopt a consortium chain architecture, in which the wide-area substation, operation and maintenance unit or equipment supplier serve as different participating nodes; Each operation or data change is treated as a transaction, with a timestamp and digital signature, and is stored on the chain; Support for quick retrieval and authenticity verification of any historical operation event, and automatic execution of permission management, data sharing or event notification logic based on specific conditions using smart contracts.

9. A system for a remote operation and maintenance method of the wide-area substation automation apparatus according to any one of claims 1 to 8, characterized by, Comprise: A device data acquisition unit for real-time acquisition of the operating state data, alarm event data, configuration data and environmental monitoring data of all automation devices in the substation by deploying in multiple wide-area substations to achieve comprehensive coverage of multi-source heterogeneous data; A data transmission and adaptation module for preliminary format conversion and protocol adaptation of the collected heterogeneous data, and secure transmission to the remote operation and maintenance center through multiple encryption and authentication mechanisms; A data management and preprocessing module for deep cleaning, standardization processing, feature extraction and multi-modal fusion of the heterogeneous data at the remote operation and maintenance center to construct a unified device health state representation vector or multi-dimensional tensor; An intelligent diagnosis and prediction module for fault diagnosis, health assessment and life prediction of automation devices based on the device health state representation using deep learning models, and output of detailed results including fault type, fault level, fault location and remaining life; An adaptive operation and maintenance strategy generation module for adaptive generation or optimization of operation and maintenance strategies for specific automation devices based on the fault diagnosis, health assessment and life prediction results, combined with a pre-set operation and maintenance strategy library and reinforcement learning mechanism; A secure remote control and execution module for security verification and permission approval of the generated operation and maintenance strategies, and after approval, issuing control instructions or configuration update packages to the target wide-area substation automation devices through an encrypted communication channel to achieve remote precise control of the automation devices; A blockchain storage and traceability module for storing and tracing the operation records, device state changes, diagnosis results and strategy execution logs important data in a non-tamperable manner through distributed ledger technology.

10. A device for operating a remote operation and maintenance system of a wide-area substation automation device according to claim 9, characterized in that Comprise a storage for data storage and a central processing unit for data processing.

Images