AI technology-based ultrahigh pressure equipment health state dynamic evaluation system and method

Abstract

The invention provides an ultrahigh-voltage equipment health state dynamic evaluation system and method based on an AI technology, and relates to the technical field of intelligent operation and maintenance of power equipment. Comprising a sensing layer, an edge calculation layer, an evaluation layer, a block chain evidence storage layer and an application layer. The method comprises the following steps: collecting operation environment and state information of ultrahigh-voltage equipment, and performing real-time preprocessing and preliminary analysis; the method specifically combines time sequence modeling and a multi-dimensional feature recognition algorithm, constructs a hybrid evaluation model, comprehensively analyzes collected real-time information and corresponding historical data, recognizes a potential equipment fault trend, quantifies an equipment health index, and generates an operation risk evaluation report. An alliance chain structure is adopted, and a multi-party trust mechanism is established among a power grid, an equipment manufacturer and a third-party mechanism; the risk assessment report is subjected to uplink storage after being encrypted; through a preset logic rule, alarm is automatically triggered, a work order is generated, operation and maintenance resources are scheduled, and a complete closed-loop processing mechanism is formed.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for power equipment, and in particular to a dynamic health status assessment system and method for ultra-high voltage equipment based on AI technology. Background Technology

[0002] With the continuous expansion of power systems and the increasing demands for power supply reliability, ultra-high voltage (UHV) transmission equipment, as a crucial hub in power grid operation, plays a vital role in ensuring the safe and stable operation of the power grid by effectively monitoring its operational status and enhancing the ability to identify potential faults. However, current equipment operation and maintenance methods still rely primarily on periodic inspections and experience-based judgment. This model suffers from shortcomings such as untimely response, limited risk identification capabilities, and high operation and maintenance costs, making it difficult to meet the needs of modern power grids for real-time perception, accurate diagnosis, and predictive maintenance of equipment status.

[0003] In recent years, although some operation and maintenance units have gradually applied online monitoring and condition assessment systems, most existing technologies are still at the stage of basic data collection and static analysis. They are insufficient in revealing the dynamic evolution of equipment conditions under complex operating conditions, making it difficult to form in-depth data fusion and intelligent assessment capabilities. Given the increasingly diverse types of power system equipment and the increasingly complex operating environment, assessment methods relying solely on single-parameter monitoring or traditional algorithms are insufficient to identify potential risks in a timely manner, posing certain risks.

[0004] To address the technical bottlenecks in the current operation of ultra-high voltage (UHV) power equipment, such as untimely health assessments, insufficient fault prediction capabilities, poor data reliability, and over-reliance on manual judgment, this invention proposes an artificial intelligence-based dynamic health status assessment method for UHV equipment. This invention utilizes deep learning, data mining, and edge computing technologies to integrate multi-dimensional and multi-temporal scale input information, achieving efficient processing and accurate modeling of multi-source monitoring data. It comprehensively reflects the dynamic changes in equipment health status, improves the accuracy and reliability of assessment results, and provides a more scientific decision-making basis for power equipment operation and maintenance, possessing significant application value and promotional significance.

[0005] Ultra-high voltage power equipment is characterized by high unit value, high operational risk, and long maintenance cycles. Its operational stability has a decisive impact on the safety and reliability of the entire power system. Currently, traditional power operation and maintenance models are unable to grasp the true operating status of equipment in a timely manner. It is necessary to introduce new technologies with continuous sensing and dynamic assessment capabilities to achieve accurate identification and predictive management of equipment health status. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a dynamic health status assessment system and method for ultra-high voltage equipment based on AI technology.

[0007] On the one hand, the present invention provides a dynamic health status assessment system for ultra-high voltage equipment based on AI technology, including a perception layer, an edge computing layer, an assessment layer, a blockchain storage layer, and an application layer;

[0008] The sensing layer is used to collect information on the operating environment and status of ultra-high voltage equipment. By deploying current transformers, voltage transformers, online oil chromatography monitoring devices, infrared thermal imagers, acceleration sensors, and temperature and humidity sensors, it can achieve real-time acquisition of multi-dimensional information.

[0009] The edge computing layer performs real-time preprocessing and preliminary analysis of the multi-dimensional information collected by the perception layer; specifically, it preprocesses the multi-dimensional information through noise reduction filtering and feature extraction; and then performs anomaly detection through edge computing.

[0010] The evaluation layer is used to dynamically model and quantify the health of equipment. Specifically, it combines time series modeling and multi-dimensional feature recognition algorithms to construct a hybrid evaluation model, which comprehensively analyzes the collected real-time information and corresponding historical data to identify potential equipment failure trends, quantify equipment health indicators, and generate an operational risk assessment report.

[0011] The blockchain evidence storage layer adopts a consortium blockchain structure to establish a multi-party trust mechanism among the power grid, equipment manufacturers, and third-party institutions; the risk assessment report is encrypted and then stored on the blockchain.

[0012] The application layer uses preset logical rules to automatically trigger alarms, generate work orders, and schedule operation and maintenance resources, forming a complete closed-loop processing mechanism.

[0013] The perception layer, edge computing layer, evaluation layer, blockchain evidence storage layer, and application layer operate collaboratively through an industrial IoT protocol.

[0014] On the other hand, the present invention provides a method for dynamic assessment of the health status of ultra-high voltage equipment based on AI technology, which is implemented through the aforementioned dynamic assessment system for the health status of ultra-high voltage equipment based on AI technology, and specifically includes the following steps:

[0015] Step 1: Construct a multi-source heterogeneous sensor network;

[0016] Specifically, current transformers, voltage transformers, online oil chromatography monitoring devices, infrared thermal imagers, acceleration sensors, and temperature and humidity sensors are deployed on different types of ultra-high voltage equipment, and the sensors are networked together to form a multi-source heterogeneous sensor network.

[0017] Step 2: Edge node preprocessing;

[0018] Data cleaning is performed on the multi-source heterogeneous data collected by the multi-source heterogeneous sensor network to remove sensor outliers and noise, and then feature extraction is performed.

[0019] Step 3: Evaluation using a hybrid evaluation model;

[0020] The hybrid assessment model combines time series analysis and nonlinear characteristics to output a comprehensive health index, predict future trends, and generate an operational risk assessment report.

[0021] Specifically, the model is based on time-series feature extraction to identify potential degradation trajectories during equipment operation; while the nonlinear model captures complex failure modes under the interaction of multiple factors.

[0022] The comprehensive health index is as follows:

[0023] ;

[0024] in, The key state variables are the normalized values ​​of current, voltage offset, oil chromatography key gas concentration, partial discharge amplitude vibration acceleration, temperature and humidity, insulation temperature rise, and infrared hotspot temperature difference ΔT. For weights, The degradation trend score is extracted by the time-series feature extraction model. Anomaly risk factors identified for nonlinear models It is a state mapping function;

[0025] Step 4: Evaluate the results and store them on the blockchain;

[0026] After the hybrid assessment model outputs the operational risk assessment report, the risk assessment report is stored in an immutable, anti-counterfeiting, and traceable manner through hash encryption and distributed ledger mechanisms. At the same time, a trust mechanism for data sharing is established among multiple parties through on-chain evidence storage, enabling operation and maintenance units, regulatory agencies, and equipment manufacturers to make collaborative decisions based on unified and trustworthy data.

[0027] Step 5: The smart contract triggers an operation and maintenance response;

[0028] Based on the equipment risk assessment report and blockchain evidence, by setting a warning threshold, when the health index exceeds the safe range of the threshold, the smart contract will automatically trigger an alarm and generate an operation and maintenance work order, and dispatch relevant resources to respond.

[0029] Thirdly, this application proposes a computer-readable storage medium storing executable instructions that, when executed, cause a processor to perform the dynamic health status assessment system for ultra-high voltage equipment.

[0030] Fourthly, this application proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the dynamic health status assessment system for ultra-high voltage equipment.

[0031] The beneficial effects of adopting the above technical solution are as follows:

[0032] This invention provides a dynamic health status assessment system and method for ultra-high voltage (UHV) equipment based on AI technology. The system employs a multi-layered architecture design of "perception—processing—assessment—evidence storage—application," forming a complete closed-loop operation mechanism. Each layer has a clear division of labor while maintaining close integration, thereby achieving systematic operation of data collection, processing, analysis, evidence storage, and application. This system architecture effectively meets the needs of full lifecycle health management for UHV equipment, balancing real-time performance, reliability, and traceability, and effectively addresses the challenges of large data volumes, complex statuses, and high safety requirements in equipment operation and maintenance. Attached Figure Description

[0033] Figure 1 This invention provides a structural block diagram of an AI-based dynamic health status assessment system for ultra-high voltage equipment. Detailed Implementation

[0034] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0035] Example 1:

[0036] On the one hand, this invention provides a dynamic health status assessment system for ultra-high voltage equipment based on AI technology, such as... Figure 1 As shown, it includes a perception layer, an edge computing layer, an evaluation layer, a blockchain evidence storage layer, and an application layer. It realizes multi-dimensional real-time acquisition, hierarchical processing, and intelligent analysis of the operating status of ultra-high voltage equipment, which can significantly enhance the timeliness and accuracy of fault identification while improving data quality and processing efficiency. At the same time, it uses blockchain technology to ensure the security and trustworthiness of evaluation results, realizes standardized management of the entire equipment life cycle, effectively shortens the time cycle from fault discovery to handling, promotes the transformation of operation and maintenance management from passive response to proactive defense, and comprehensively improves the safety, reliability, and intelligent operation and maintenance level of ultra-high voltage equipment.

[0037] The sensing layer, as the lowest-level functional module of the system, is used to collect information on the operating environment and status of ultra-high voltage equipment. By deploying current transformers, voltage transformers, online oil chromatography monitoring devices, infrared thermal imagers, accelerometers, and temperature and humidity sensors, it achieves real-time acquisition of multi-dimensional information on electrical, thermal, mechanical, and environmental aspects. The sensing layer possesses all-weather, continuous data acquisition capabilities, enabling it to promptly capture key parameters and environmental changes during equipment operation and transform them into raw data suitable for processing by higher layers. Its core function is to ensure the comprehensiveness and continuity of data sources, providing reliable data support for subsequent analysis and modeling.

[0038] The edge computing layer performs real-time preprocessing and preliminary analysis of the multi-dimensional information collected by the perception layer. Specifically, it preprocesses the multi-dimensional information through noise reduction filtering and feature extraction. Then, it performs anomaly detection through edge computing, effectively eliminating redundant and interfering information and improving data quality and transmission efficiency. Edge computing has the capability of task decentralization, moving some computational tasks to edge nodes for execution, achieving rapid local response. This reduces the computational and transmission pressure on the central system, shortens data processing latency, and enhances local intelligent analysis capabilities, thereby improving the overall system's operational efficiency and stability.

[0039] The evaluation layer is positioned as the core analysis module of the system, used for dynamic modeling and health quantification of equipment status. Specifically, it combines time series modeling and multi-dimensional feature recognition algorithms to construct a hybrid evaluation model. This model comprehensively analyzes the collected real-time information and corresponding historical data to identify potential equipment failure trends, quantify equipment health indicators, and generate operational risk assessment reports. The evaluation layer enables the transformation from data to information and then to knowledge, providing maintenance personnel with reliable decision-making support and ensuring that potential equipment problems can be detected early and intervened in a timely manner.

[0040] The blockchain evidence storage layer ensures the security, reliability, and traceability of data and assessment results. This layer employs a consortium blockchain structure, establishing a multi-party trust mechanism among the power grid, equipment manufacturers, and third-party institutions. Risk assessment reports are encrypted and stored on the blockchain, achieving tamper-proof, traceable, and shareable functionality. Its main functions are to enhance the transparency of the data flow process, ensure the integrity of assessment results, and provide reliable evidence for subsequent audits, liability determination, and compliance checks, thereby strengthening the credibility and authority of the system.

[0041] The application layer transforms the analysis and results from the previous layers into executable operation and maintenance procedures. This layer, through preset logical rules, automatically triggers alarms, generates work orders, and schedules operation and maintenance resources, forming a complete closed-loop processing mechanism. Its core function is to quickly respond to equipment failures, achieving fully automated management from problem discovery to resolution. Simultaneously, the application layer also supports the optimization of the operation and maintenance management system, improving the efficiency of personnel scheduling, spare parts management, and maintenance decision-making, enabling a shift from a passive response to a proactive defense operation and maintenance model, and comprehensively enhancing the system's intelligence level and operation and maintenance collaboration.

[0042] On the other hand, the present invention provides a method for dynamic assessment of the health status of ultra-high voltage equipment based on AI technology, which is implemented through the aforementioned dynamic assessment system for the health status of ultra-high voltage equipment based on AI technology, and specifically includes the following steps:

[0043] Step 1: Construct a multi-source heterogeneous sensor network;

[0044] At the power system equipment level, constructing a multi-source heterogeneous sensor network is key to achieving intelligent sensing. Specifically, this involves deploying current transformers, voltage transformers, online oil chromatography monitoring devices, infrared thermal imagers, accelerometers, and temperature and humidity sensors on different types of ultra-high voltage equipment. Electrical sensors monitor current, voltage, and partial discharge signals in real time, thermal sensors capture temperature distribution and hotspot changes, and mechanical sensors monitor vibration and noise. This multi-source heterogeneous sensing network enables continuous, real-time data acquisition, avoiding the information gaps associated with single-mode sensing. The networked deployment of sensors forms a multi-source heterogeneous sensing network; these sensors can collaborate and complement each other, enhancing the completeness and robustness of monitoring. This comprehensive, multi-dimensional sensing provides a comprehensive and reliable data source for subsequent edge computing and fusion evaluation, laying the foundation for the health management of power equipment throughout its entire lifecycle.

[0045] Step 2: Edge node preprocessing;

[0046] In multi-source heterogeneous data collected by multi-source heterogeneous sensor networks, direct transmission to the central server would lead to a surge in storage and computing pressure. Edge node preprocessing allows for initial data processing on-site, improving the system's real-time performance and flexibility. Data cleaning of the multi-source heterogeneous data collected by the multi-source heterogeneous sensor network removes sensor outliers and noise, ensuring data accuracy. Feature extraction then compresses high-dimensional complex signals into representative parameters, improving model training efficiency. Data normalization unifies the representation of different modalities and data formats, enhancing consistency in cross-modal analysis. Through this two-layer processing system of "rapid on-site response – in-depth central analysis," the system can maintain immediate response to emergencies while ensuring in-depth analysis of complex fault mechanisms, providing dual protection for intelligent operation and maintenance.

[0047] Step 3: Evaluation using a hybrid evaluation model;

[0048] The hybrid assessment model is a core component in transforming multi-source heterogeneous data into health status information. This model combines time-series analysis and nonlinear characteristics to output a comprehensive health index and predict future trends. Specifically, a time-series-based model identifies potential degradation trajectories during equipment operation, while a nonlinear model captures complex failure modes under multi-factor interactions. The fusion model, through the complementarity of different algorithms, improves the accuracy and sensitivity of early fault detection, avoiding missed detections and misjudgments due to the limitations of a single model. Simultaneously, the assessment results can be converted into trend curves, helping maintenance personnel intuitively grasp the equipment's health evolution path. This process not only strengthens preventative maintenance capabilities but also realizes a shift from "post-event repair" to "pre-event prevention," improving the initiative and economy of power equipment operation and maintenance.

[0049] In this embodiment of the invention, the Health Index (HI) of ultra-high voltage power equipment is defined as a comprehensive index composed of the health status of key state parameters of ultra-high voltage equipment, the time-series degradation trend score, and nonlinear anomaly risk factors. By standardizing multi-source heterogeneous data and determining the weights of each index based on entropy weighting, AHP, or AI-based feature importance, the following health index formula is constructed:

[0050] ;

[0051] in, Normalized values ​​for key state variables (current, voltage offset, oil chromatography key gas concentration (H2C2H2, etc.), partial discharge amplitude vibration acceleration, temperature and humidity, insulation temperature rise, infrared hotspot temperature difference ΔT), The weights of each parameter, The degradation trend score extracted from the time series model. Anomaly risk factors identified for nonlinear models This is a state mapping function. This health index can reflect the health status and degradation degree of ultra-high voltage equipment in real time, providing a quantitative basis for subsequent risk assessment, alarm and operation and maintenance scheduling.

[0052] At the model implementation level, this invention employs existing publicly available temporal feature extraction models (such as those based on time series decomposition, temporal convolution, or recurrent neural networks) to identify potential degradation trajectories during device operation. Simultaneously, it introduces common nonlinear modeling frameworks (such as random forests, gradient boosting trees, or deep nonlinear representation models) to capture complex fault modes under multi-factor interactions. If some functions cannot fully rely on existing models, relevant algorithm modules are supplemented through secondary development to meet the identification needs of specific scenarios.

[0053] Step 4: Evaluate the results and store them on the blockchain;

[0054] After the hybrid assessment model generates an operational risk assessment report, ensuring the reliability of the results and the security of their sharing becomes a critical issue. Blockchain technology provides an effective solution. Through hash encryption and distributed ledger mechanisms, the risk assessment report is stored in an immutable, counterfeit-proof, and traceable manner. Simultaneously, on-chain notarization establishes a trust mechanism for data sharing among multiple parties, enabling maintenance units, regulatory agencies, and equipment manufacturers to make collaborative decisions based on unified and reliable data. This not only improves data flow efficiency but also reduces potential disputes and compliance risks. In practical applications, the introduction of blockchain strengthens the transparency and credibility of the power equipment operation and maintenance system, providing a trusted foundation for the subsequent automatic execution of smart contracts.

[0055] Step 5: The smart contract triggers an operation and maintenance response;

[0056] Based on equipment risk assessment reports and blockchain-based evidence storage, the introduction of smart contracts enables automated closed-loop management from monitoring to operation and maintenance. By setting early warning thresholds, when equipment status indicators exceed the safe range of the threshold, the smart contract automatically triggers an alarm and generates an operation and maintenance work order, scheduling relevant resources to respond. This mechanism avoids the delays and subjectivity of manual decision-making, achieving efficient linkage between "monitoring—assessment—decision-making—execution." Simultaneously, smart contracts are programmable, allowing for flexible customization based on different equipment types, operation and maintenance strategies, and security levels. Their automated execution capabilities improve operation and maintenance efficiency, enhance the system's rapid recovery capability in the face of sudden failures, and promote the development of power equipment operation and maintenance towards intelligence and adaptability.

[0057] Example 2:

[0058] This embodiment is applicable to health status assessment and predictive maintenance of critical equipment such as ultra-high voltage substations, transmission lines, and switching stations. The method includes the following steps: By deploying current transformers, voltage transformers, fiber optic temperature sensors, vibration sensors, online gas chromatography monitoring devices, and infrared imaging equipment on transformers, circuit breakers, surge arresters, and other equipment, multi-dimensional acquisition of electrical, thermal, mechanical, and environmental parameters is achieved, forming a multi-modal dataset of equipment operating characteristics. Subsequently, the edge computing gateway performs noise reduction, cleaning, feature extraction, and modal fusion on the raw data, improving data quality and transmission efficiency through wavelet transform and principal component analysis (PCA) algorithms. The system utilizes an AI fusion model combining LSTM and CNN to learn the health evolution patterns of the equipment, outputting a comprehensive health index (HI) and predicting potential fault trends. The assessment results are hash-encrypted and written to the consortium blockchain ledger, ensuring data tamper-proofing and trusted sharing through a multi-node consensus mechanism. When the HI value falls below a set threshold, the smart contract automatically generates an alarm work order and schedules maintenance resources, achieving closed-loop management of "monitoring—early warning—response." This method significantly improves the accuracy and response speed of status identification for ultra-high voltage equipment, enabling a shift from "post-event maintenance" to "pre-event early warning".

[0059] The system described in this embodiment includes a perception layer, an edge computing layer, an evaluation layer, a blockchain evidence storage layer, and an application layer. These layers collaborate via industrial IoT protocols such as MQTT or OPC UA. The perception layer deploys various types of sensors on equipment with voltage levels of 220kV and above to achieve comprehensive monitoring of current, voltage, temperature, vibration, and oil / gas composition, supporting both wired and wireless hybrid communication. The edge computing layer, based on NVIDIA Jetson or Huawei Atlas 200 modules, performs local data processing and model inference, and features emergency caching and analysis capabilities in case of communication interruptions. The evaluation layer runs a deep learning model integrating XGBoost and LSTM, outputting equipment health indices and risk level reports. The blockchain evidence storage layer adopts a Hyperledger Fabric consortium blockchain architecture, enabling encrypted on-chain storage and traceable management of evaluation reports and operation logs among power grid companies, operation and maintenance centers, and third-party institutions. The application layer constructs a visualized operation and maintenance platform, displaying real-time health status, trend curves, and fault heatmaps, and supports integration with SCADA and EMS systems.

[0060] This embodiment further optimizes model performance and system synergy. At the algorithm level, it integrates the Transformer structure and Graph Neural Network (GNN) to achieve inter-device correlation feature learning and anomaly propagation path identification. A transfer learning strategy is employed to pre-train historical data, improving prediction accuracy in small-sample scenarios. At the diagnostic level, Bayesian confidence intervals and probabilistic inference are introduced to quantify the uncertainty of the health index of ultra-high voltage equipment, enhancing the robustness of risk assessment. Regarding operation and maintenance feedback, the system automatically records handling results and back-optimizes model parameters, forming an adaptive cycle of "assessment—execution—learning—re-optimization." The system supports private cloud and hybrid cloud architectures, ensuring reliability and flexibility in different operation and maintenance environments. This preferred solution further enhances the system's intelligence, adaptability, and security, making it suitable for comprehensive health management and remote intelligent monitoring of multi-regional and multi-type ultra-high voltage equipment.

[0061] Example 3:

[0062] This embodiment proposes a computer-readable storage medium that stores executable instructions. When these instructions are executed, if they are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0063] The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the dynamic health status assessment method for ultra-high voltage equipment according to various embodiments of this application.

[0064] The aforementioned storage media include: flash memory, hard disk, multimedia card, card-type memory (e.g., SD (Secure Digital Memory Card) or DX (Memory Data Register, MDR) memory, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, disk, optical disk, server, APP (Application) application store, and other media capable of storing program verification codes, on which computer programs are stored. When the computer programs are executed by the processor, they can implement the various steps of the above-mentioned dynamic health status assessment method for ultra-high voltage equipment.

[0065] Example 4:

[0066] This embodiment proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the dynamic health status assessment method for ultra-high voltage equipment.

[0067] Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a computer program product.

[0068] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0069] The scope of protection of this application is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of the methods disclosed herein and their equivalents, then the intent of this disclosure also includes such modifications and variations.

Claims

1. An AI technology-based ultra-high voltage equipment health state dynamic evaluation system, characterized in that, The application comprises a perception layer, an edge computing layer, an evaluation layer, a blockchain storage layer and an application layer. The perception layer is used for collecting information of the running environment and state of the ultra-high voltage equipment, and realizes real-time acquisition of multi-dimensional information by deploying current transformers, voltage transformers, online oil chromatographic monitoring devices, infrared thermal imagers, acceleration sensors and temperature and humidity sensors. The edge computing layer performs real-time preprocessing and preliminary analysis on the multi-dimensional information collected by the perception layer. The evaluation layer is used for dynamic modeling and health quantification of the equipment state. The blockchain storage layer adopts a consortium chain structure to establish a multi-party trust mechanism among the power grid, equipment manufacturers and third-party institutions. The risk assessment report is stored on the chain after encryption. The application layer realizes automatic alarm triggering, work order generation and dispatching of operation and maintenance resources through preset logical rules, forming a complete closed-loop processing mechanism. The perception layer, edge computing layer, evaluation layer, blockchain storage layer and application layer realize collaborative operation through industrial Internet of Things protocols.

2. A method for dynamically evaluating the health state of an ultra-high voltage equipment based on AI technology, implemented by the system for dynamically evaluating the health state of an ultra-high voltage equipment based on AI technology of claim 1, characterized in that, The method comprises the following steps: Step 1: Construct a multi-source heterogeneous sensor network. Step 2: Edge node preprocessing; the multi-source heterogeneous data collected by the multi-source heterogeneous sensor network is cleaned to eliminate sensor outliers and noise, and then feature extraction is performed. Step 3: Hybrid evaluation model evaluation. The hybrid evaluation model combines time series analysis and nonlinear features to output a comprehensive health index and predict future trends, generating a running risk assessment report. Step 4: Store the evaluation results on the chain. Step 5: Trigger the operation and maintenance response through the smart contract.

3. The AI technology-based dynamic health state evaluation method for ultra-high voltage equipment according to claim 2, characterized in that, In step 1, current transformers, voltage transformers, online oil chromatographic monitoring devices, infrared thermal imagers, acceleration sensors and temperature and humidity sensors are arranged on different types of ultra-high voltage equipment.

4. The AI technology-based dynamic health state evaluation method of an ultra-high voltage equipment according to claim 2, characterized in that, In step 3, the time series feature extraction model is used to identify potential degradation trajectories during equipment operation, and the nonlinear model captures complex failure modes under multi-factor interaction. In step 4, after the hybrid evaluation model outputs the running risk assessment report, the report is stored in a tamper-proof, counterfeit-proof and traceable manner through hash encryption and distributed ledger mechanism. ； wherein, is a key state variable, i.e. a normalized value of current, voltage offset, oil chromatographic key gas concentration, partial discharge amplitude vibration acceleration, temperature and humidity, insulation temperature rise, infrared hotspot temperature difference ΔT, is a weight, is a degradation trend score extracted by a time series feature extraction model, is an abnormal risk factor identified by a nonlinear model, is a state mapping function.

5. The AI technology-based dynamic health state evaluation method of an ultra-high voltage equipment according to claim 2, characterized in that, In step 5, based on the device risk assessment report and blockchain storage, a warning threshold is set, and when the health index exceeds the safety range of the threshold, the smart contract automatically triggers an alarm and generates an operation and maintenance work order to dispatch relevant resources for response.

6. The AI technology-based dynamic health state evaluation method of an ultra-high voltage equipment according to claim 2, characterized in that, ​ 7. The ultra-high pressure equipment energy efficiency intelligent control system based on big data analysis according to claim 6, implemented based on a computer readable storage medium, characterized in that, The system stores executable instructions that, when executed, cause the processor to perform the AI-based dynamic health status assessment method for ultra-high voltage equipment. 8.The ultra-high voltage equipment energy efficiency intelligent control system based on big data analysis of claim 6, realized by a computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the AI-based dynamic health status assessment method for ultra-high voltage equipment.

Images