Intelligent safety protection electric power operation and maintenance system

Abstract

The invention discloses an intelligent safety protection electric power operation and maintenance system, and relates to the technical field of electric power operation and maintenance, and the system comprises a microscopic self-repairing module which is integrated in equipment and is located between a multi-mode sensor layer and an insulating medium of electrical equipment. And the micro self-repairing module comprises a nano robot group and a trigger type self-healing microcapsule, and realizes closed-loop self-healing of micro damage, immediate repairing and re-detection. By designing the microscopic self-repairing module, real-time scanning, positioning activation and in-situ polymerization self-healing functions are achieved, the technical problems that traditional micro-damage cannot be repaired online, manual repairing is high in risk and equipment is prone to aging and failure are solved, the service life of the equipment is prolonged, shutdown is reduced, the maintenance cost is reduced, and the safety of maintenance personnel is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power operation and maintenance, and specifically to an intelligent safety protection power operation and maintenance system. Background Art

[0002] Traditional power operation and maintenance rely on manual regular inspections and centralized monitoring. The repair speed of manual inspections is slow and the repair risk coefficient is high, making it difficult to meet the growing requirements for equipment reliability and safety protection, resulting in low repair efficiency and poor safety during the power operation and maintenance protection process.

[0003] Patent CN113724470B discloses a multifunctional power operation and maintenance inspection safety protective clothing. The above patent realizes reminding maintenance personnel to keep a safe distance from live equipment and ensures the personal safety of operators.

[0004] The above patent uses a voltage sensor arranged on the protective clothing to detect the voltage in the surrounding environment in real time, and connects the voltage sensor to an alarm device. When the voltage sensor monitors that the voltage value in the surrounding environment exceeds the preset voltage value, the alarm device gives an early warning to remind maintenance personnel to keep a safe distance from live equipment. However, relying on manual inspections and repairs during the inspection and repair process has problems of low efficiency and low safety.

[0005] Therefore, this application proposes an intelligent safety protection power operation and maintenance system that can achieve self-healing repair. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent safety protection power operation and maintenance system to solve the technical problems of low repair efficiency and poor safety proposed in the above background art.

[0007] To achieve the above purpose, the present invention provides the following technical solution: An intelligent safety protection power operation and maintenance system includes a microscopic self-repair module. The microscopic self-repair module is integrated inside the equipment, located between the multi-modal sensor layer and the insulating medium of the electrical equipment. The microscopic self-repair module includes:

[0008] A group of nano-robots that are suspended in the insulating medium and self-organize into formations through a multi-modal driving unit of magnetic coupling, acoustic propulsion, and optical induction;

[0009] Triggered self-healing microcapsules filled with a two-component self-healing resin and a catalyst;

[0010] The group of nano-robots uses integrated electrical detection sensors and optical scattering sensors to detect microscopic cracks and discharge channels inside the electrical equipment in real time. After detecting micro-damage, the corresponding self-healing microcapsules are ruptured at a fixed point, so that the resin and the catalyst are mixed and polymerized on-site, realizing a closed-loop self-healing of "micro-damage - instant repair - re-detection".

[0011] Preferably, the multi-modal sensor layer is deployed on the surface and internal structure of the electrical equipment body, including:

[0012] Current transformers and voltage sensors for collecting the power flow state under real-time operating conditions;

[0013] Optical fiber temperature sensors and infrared thermal imaging arrays for realizing the reconstruction of temperature distribution maps;

[0014] A discharge warning unit jointly composed of partial discharge sensors and ultrasonic sensors for identifying precursors of electrical breakdown;

[0015] Gas sensor arrays for monitoring changes in gas concentrations in transformer oil and realizing the traceability of fault gases.

[0016] Preferably, the operation and maintenance system includes an edge computing gateway installed inside the electrical equipment terminal box or switch cabinet, integrating an embedded AI processing unit, an edge cache module, and a local encryption module, for performing fusion analysis on the data collected by the multi-modal sensor layer, and deploying a federated learning engine to realize device health state modeling and dynamic fault risk scoring under privacy protection.

[0017] Preferably, the operation and maintenance system includes a blockchain security module built on a lightweight distributed ledger architecture, adopting a PBFT consensus mechanism, and built-in smart contracts for automatically recording the abnormal detection time, self-healing behavior logs, and execution paths of operation and maintenance instructions, ensuring the traceability, credibility, and immutability of the system.

[0018] Preferably, the operation and maintenance system includes a cloud digital twin platform that fuses the three-dimensional geometric structure, electromagnetic-thermal coupling state, and sensor data of the electrical equipment through digital modeling tools to construct a virtual twin. The cloud digital twin platform has the capabilities of model iteration, self-learning synchronization, and state visualization, and is linked with the historical database to realize trend prediction and calculation of the remaining structural life.

[0019] Preferably, the operation and maintenance system includes an augmented reality operation and maintenance terminal worn on the head of on-site operation and maintenance personnel, which synchronously displays the hidden danger information of key parts inside the equipment, the distribution of risk hot zones, and the visualized operation guidance path based on the digital twin model and real-time sensor data, and supports gesture interaction, voice control, and automatic calibration functions.

[0020] Preferably, the operation and maintenance system further includes an adaptive maintenance scheduler that, based on the joint inference results of the edge computing gateway and the cloud platform, uses a reinforcement learning optimization algorithm to generate inspection plans, dispatch paths, and human resource allocation strategies in real time, comprehensively considering the self-healing state, risk level, personnel accessibility, and task timeliness.

[0021] Preferably, the nanorobot swarm is suspended in insulating oil or composite insulating resin. Each nanorobot has a length of no more than 500 nanometers and is coated with a multimodal driving layer, including a superparamagnetic response coating for magnetic navigation, a pressure material layer for acoustic wave propulsion, and a photosensitive particle array for directional light manipulation. At the same time, an electric field gradient recognition chip and a fluorescence imaging unit are integrated;

[0022] Trigger-type self-healing microcapsules evenly distributed in the insulating medium. Each microcapsule is composed of multiple polymer shell layers. The outer shell has an electric field-responsive microporous channel, which only ruptures when a nanorobot approaches and provides mechanical disturbance or ultrasonic energy of a specific frequency; Inside the microcapsule, two groups of non-reactive resins and catalyst monomers, namely A and B, are encapsulated. After rupture, they are instantaneously mixed to achieve in-situ polymerization self-healing at the damage position;

[0023] Nanorobots scan the internal microscopic damage area of the insulator in real time. After detecting microcracks or discharge paths, they converge into a repair formation, activate the nearest microcapsule at a fixed point, and complete crack sealing and dielectric constant recovery;

[0024] The results of the repair behavior are verified twice through a machine vision and electric field reconstruction mechanism. If the repair fails to be recognized, a secondary repair will be automatically initiated.

[0025] Preferably, the nanorobots adjust the formation spacing through a tunable magnetic field gradient, with a response frequency range of 0.1 Hz to 100 Hz, for migration control in different medium viscosity environments; The outer shell material of the microcapsule adopts a three-layer structure. The outermost layer is a hydrophobic polymer to improve suspension stability, the middle layer is an electric field-responsive conductive shell, and the inner layer is a controllable rupture polyurethane chain polymer.

[0026] Preferably, the electrical equipment includes but is not limited to power transformers, high-voltage switchgears, cable joints, GIS equipment, and reactor systems. The microscopic autonomous repair module can be embedded in the insulating oil, solid insulation structure, or gel layer of the electrical equipment as needed; The scale of the nanorobot swarm supports a dynamic loading and unloading mechanism. At the initial operation, a basic repair swarm is injected by the edge control system, and then the density of nanorobots is automatically scheduled according to the damage distribution to achieve resource conservation and enhanced centralized repair capabilities.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. By designing a microscopic autonomous repair module, the present invention realizes functions of real-time scanning, positioning activation, and in-situ polymerization self-healing, solves the technical problems of traditional microdamage that cannot be repaired online, high risk of manual repair, and easy aging and failure of equipment, extends the equipment life, reduces downtime, lowers maintenance costs, and improves the safety of maintenance personnel;

[0029] 2. The present invention realizes the function of capturing abnormal events of electrical equipment by designing a multi-modal sensor layer, solves the problems of high missed detection rate and early warning lag of a single sensor, identifies potential fault hazards in advance, and improves the reliability of operation and maintenance.

[0030] 3. The present invention realizes the functions of dynamic health modeling and real-time risk scoring under privacy protection by designing an edge federated learning engine, solves the problems of data privacy, centralized model delay, and poor adaptation to environmental differences, and improves the level of local rapid decision-making, user data protection, and continuous model optimization.

[0031] 4. The present invention realizes the functions of optimal inspection and resource allocation by designing blockchain, digital twin, and AI scheduling methods, solves the problems of easy tampering of operation and maintenance records, low guidance efficiency, and uneven scheduling, and improves operation efficiency, safety, and collaboration. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0033] Figure 2 It is a schematic diagram of the operation and maintenance process of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0035] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", "front end", "backend", "both ends", "one end", "the other end", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0036] In the description of the present invention, it should be noted that, unless otherwise clearly specified and defined, terms such as "installation", "equipped with", "connection", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] Please refer to Figure 1 and Figure 2 For an embodiment provided by the present invention: An intelligent safety protection power operation and maintenance system, including a microscopic self-repair module, the microscopic self-repair module is integrated inside the device, located between the multi-modal sensor layer and the insulating medium of the electrical equipment, and the microscopic self-repair module includes:

[0038] A group of nano-robots that are suspended in the insulating medium and self-organize into formations through a multi-modal driving unit of magnetic coupling, acoustic propulsion, and optical induction;

[0039] Trigger-type self-healing microcapsules, encapsulated with a two-component self-healing resin and a catalyst;

[0040] The group of nano-robots, through the integrated electrical detection sensors and optical scattering sensors, can detect microscopic cracks and discharge channels inside the electrical equipment in real time. After detecting micro-damage, the corresponding self-healing microcapsules are ruptured at a fixed point, so that the resin and the catalyst are mixed and polymerized on-site, realizing the closed-loop self-healing of "micro-damage - instant repair - re-detection";

[0041] Furthermore, the size of the nano-robots: about 300 nm in length and about 50 nm in diameter, with a superparamagnetic response coating, a piezoelectric PZT layer, and a photosensitive silicon-based particle array coated on the surface. Each robot integrates thermoelectric and piezoelectric vibration units, which are sufficient to maintain 1000 times of driving and communication. The nano-robots and the trigger-type self-healing microcapsules are mixed into the insulating oil at a volume ratio of 0.1%, and ultrasonically oscillated and premixed for 5 minutes to ensure uniform suspension;

[0042] The group of nano-robots cruise in a grid pattern at a speed of 1 mm / s under the drive of a magnetic field gradient and acoustic waves in the insulating oil, synchronously trigger the electrical and optical sensors to collect signals, and are analyzed in real time by the edge computing network associated federated learning engine to identify the crack signal characteristics or partial discharge channels; if the sudden change of the electric field strength in a certain area > 0.5 kV / mm or the sudden change of the scattering signal amplitude > 15%, it is determined as "micro-damage" and the XYZ coordinates are recorded;

[0043] The edge computing gateway issues positioning instructions through optical communication. The instructions include the target coordinates and the formation radius. The nano-robots respond to the magnetic coupling field and acoustic pulses and self-organize a "repair formation" within ≤50 ms to surround the microcracks. The nano-robot closest to the crack in the repair formation uses an acoustic excitation unit to apply local stress to the outer shell of the targeted microcapsule, triggering the opening of the outer micropores. The two-component epoxy-based prepolymer self-healing resin and the catalyst are instantaneously mixed at the crack. At an ambient temperature of 25°C, the in-situ polymerization reaction is completed within 60 s, forming a polymer repair bridge and restoring the local dielectric constant to over 95% of its original value.

[0044] The fluorescence imaging unit carried by the nano-robots scans the repair area to confirm that the polymer is completely filled. The electrical sensor measures the local electric field distribution again. If the deviation is <5%, it is considered a successful repair; otherwise, the above repair process is automatically repeated or reported to the edge computing gateway.

[0045] Please refer to Figure 1 and Figure 2 For an embodiment provided by the present invention: an intelligent safety protection power operation and maintenance polyamine system, the multimodal sensor layer is deployed on the surface and internal structure of the electrical equipment body and includes:

[0046] Current transformers and voltage sensors for collecting the power flow state under real-time working conditions;

[0047] Optical fiber temperature sensors and infrared thermal imaging arrays for realizing the reconstruction of the temperature distribution map;

[0048] A discharge warning unit composed of a combination of partial discharge sensors and ultrasonic sensors for identifying the precursors of electrical breakdown;

[0049] A gas sensor array for monitoring the change in the gas concentration in transformer oil and realizing the traceability of fault gases;

[0050] Furthermore, a current transformer with a ratio of 1000:5A is selected and installed on the primary side busbar. Ratio error calibration is performed through a standard injected current (100A), with an error ≤ 0.2%. A high-precision capacitive voltage divider is used and installed on the side of the equipment cabinet. The output voltage ratio to the primary voltage is 1:1000. After calibration by a DC voltage source (5kV), the error ≤ 0.1%. Fiber optic temperature sensor: 10 fiber Bragg gratings are arranged along the winding oil duct at an interval of 200mm, with a central wavelength of 1550nm, a temperature sensitivity of 10pm / ℃, a calibrated temperature range of –40℃ to +150℃. Infrared array: 640×480 pixels, a field of view (FOV) of 60°×45°, a frame rate of 30fps, adjustable emissivity, calibrated using a blackbody calibration plate (50℃, 80℃), with a temperature measurement accuracy of ±1℃. Partial discharge sensor: operating frequency band 300MHz to 1.5GHz, sensitivity –70dBm, installed on the top of the oil tank. Ultrasonic sensor: bandwidth 20kHz to 100kHz, sensitivity –40dB, attached to the side wall of the oil tank. Calibration: By artificially exciting a partial discharge source (100pC pulse), adjusting the gain and threshold to ensure a detection rate ≥ 95%. Gas types: H2, CO, C2H2, C2H4. Sensor types: electrochemistry (H2, CO) and semiconductor MOS type (C2H2, C2H4). Detection range 0–2000ppm, resolution 1ppm. Calibration: Zero and range calibration are performed using a standard gas mixture (500ppm H2, 200ppm CO, 50ppm C2H2, 50ppm C2H4);

[0051] Current transformer / voltage sensor: 10kHz sampling rate, 50Hz ± 5Hz band-pass filtering. Fiber optic temperature sensor: 1Hz wavelength scanning, temperature demodulation. Infrared array: 30fps thermal image acquisition, non-uniformity correction (NUC) and radiation correction. PD-UHF: 200MS / s high-speed sampling, digital band-pass filtering (300MHz–1.5GHz). Ultrasonic: 200kS / s sampling, 20kHz–100kHz filtering. Gas: Sampling once per minute, temperature compensation algorithm. Current transformer / voltage sensor: Removing power frequency harmonics, calculating instantaneous active / reactive power and current unbalance. Temperature: Fiber optic temperature sensor and infrared data fusion, using cubic spline interpolation to reconstruct the temperature distribution map of the equipment surface and windings. Partial discharge sensor: UHF pulse counting and ultrasonic energy envelope analysis, extracting partial discharge pulse characteristics (amplitude, width, repetition rate) after denoising. Gas: Preliminary fault type discrimination based on the IEC 60599 fault gas ratio method (Duval triangle diagram);

[0052] All sensors achieve clock synchronization within ±1μs through the IEEE 1588 PTP protocol. For current imbalance, a warning is issued when it is >5%. For temperature rise hot spots, a warning is issued when the local temperature exceeds the baseline value by 10°C. For partial discharge sensor events, a warning is issued when it is >200 pulses / s or the single-pulse amplitude is >–50 dBm. For gas concentration, H2>300 ppm, CO>150 ppm, C2H2>50 ppm. When any two indicators exceed the threshold simultaneously, or a single indicator continuously exceeds the threshold for >5 min, the edge computing gateway generates an "early fault warning"; abnormal events are pushed to the cloud and local HMI through the MQTT encrypted channel, triggering the AR terminal to highlight the risk area, and at the same time, the preprocessed raw data is stored in the edge cache for subsequent blockchain module packaging and accounting.

[0053] Please refer to Figure 1 and Figure 2 An embodiment provided by the present invention: An intelligent safety protection power operation and maintenance system, the operation and maintenance system includes an edge computing gateway, which is installed inside the electrical equipment terminal box or switch cabinet, and integrates an embedded AI processing unit, an edge cache module, and a local encryption module, and is used for fusion analysis of the data collected by the multimodal sensor layer, and deploys a federated learning engine to realize device health state modeling and dynamic fault risk scoring under privacy protection;

[0054] The operation and maintenance system includes a blockchain security module, which is built on a lightweight distributed ledger architecture, adopts a PBFT consensus mechanism, and has an intelligent contract built-in, and is used for automatically recording the abnormal detection time, self-healing behavior logs, and the execution path of operation and maintenance instructions, ensuring the traceability, trustworthiness, and immutability of the system;

[0055] The operation and maintenance system includes a cloud digital twin platform, which fuses the three-dimensional geometric structure, electromagnetic-thermal coupling state, and sensor data of electrical equipment through a digital modeling tool to construct a virtual twin body. The cloud digital twin platform has the capabilities of model iteration, self-learning synchronization, and state visualization, and is linked with the historical database to realize trend prediction and calculation of the remaining life of the structure;

[0056] The operation and maintenance system includes an augmented reality operation and maintenance terminal, which is worn on the head of on-site operation and maintenance personnel, and synchronously displays the hidden danger information of key parts inside the device, the distribution of risk hot areas, and the visual operation guidance path based on the digital twin model and real-time sensor data, and supports gesture interaction, voice control, and automatic calibration functions;

[0057] Furthermore, the edge computing gateway extracts key features: Electrical features: Instantaneous active / reactive power, harmonic content, unbalance degree; Thermal features: Winding hot spot temperature gradient, oil temperature distribution entropy value; Discharge features: PD pulse count, energy envelope, ultrasonic energy integral; Gas features: Duval triangle diagram ratio, concentration change rate; Local model: Time series prediction based on LSTM + classification model of LightGBM, which is trained once an hour on local data regularly. Gradient encryption: Using a homomorphic encryption library, only the encrypted gradient is uploaded to protect the privacy of the original data. Participate in the whole network's federated learning: The gateway uploads the encrypted gradient to the cloud federated aggregation server, receives the global model update, and updates the local model parameters. Risk scoring: Use the updated model to infer the latest features, output a risk score of 0–100, and classify according to the thresholds <30 safe / 30–60 warning / > 60 dangerous; When the risk score > 60 or the key features exceed the limit: Trigger a local alarm: Drive an audible and visual alarm through GPIO; Push an "abnormal detection event" request to the blockchain module; Push high-priority messages to the cloud platform and AR terminal through MQTT;

[0058] The cloud digital twin platform imports the device geometric model from the CAD / PLM system, performs electromagnetic-thermal coupling simulation using ANSYS or COMSOL to generate the baseline physical field distribution, matches the simulation results with the sensor calibration points, and completes the construction of the initial virtual twin; Receives the raw and preprocessed data reported by the edge computing gateway through MQTT / TCP, pushes the data to the time series database and spatio-temporal database using Kafka, and maps the data to the 3D model through WebGL / WebSocket to dynamically update the thermal field, partial discharge hot spots, and self-healing positions; Runs the model correction process every night: Compare the differences between the simulation field and the actual field, and adjust the material thermal conductivity and conductivity parameters; Use the Bayesian optimization algorithm to update the simulation model to improve the prediction accuracy of the next cycle; Historical database linkage: Extract the key indicator sequences of the past year, and call the machine learning service to estimate the remaining life; Based on the ARIMA+LSTM hybrid model, predict the trends of temperature, PD frequency, and risk score within the next 7 days. When the predicted risk score exceeds the threshold, automatically generate a "preventive maintenance work order" and push it to the adaptive maintenance scheduler;

[0059] The operation and maintenance personnel wear the AR headset HoloLens2, start the SLAM module to scan the on-site environment, generate a dense point cloud, match the point cloud with the external geometric model of the digital twin in the cloud using the ICP algorithm, complete the coordinate system registration with an error <5mm, receive the risk heat map, PD hot spots, self-healing coordinates, etc. pushed by the cloud through WebSocket, overlay a semi-transparent heat map, flashing marks and arrow indicators on the actual device scene, display "highest risk area", "micro self-healing position", etc., gesture interaction: wave to call out the "risk list", pinch to zoom the model, click in the air to select the fault point, voice control: support commands such as "next step", "mark completed", "query history", operation process: gradually guide according to the predefined process, including "confirm device power-off → attach detection probe → observe data → trigger nano repair → verify result"; the operation and maintenance personnel mark "good repair effect" or "need to repair again" in the AR interface, and the data is immediately synchronized to the edge computing gateway and the cloud platform, and the system automatically updates the work order status according to the feedback and records it in the blockchain log to ensure the whole process is traceable.

[0060] Please refer to Figure 1 and Figure 2 For an embodiment provided by the present invention: an intelligent safety protection power operation and maintenance system, the operation and maintenance system further includes an adaptive maintenance scheduler, which based on the joint inference results of the edge computing gateway and the cloud platform, uses a reinforcement learning optimization algorithm to generate a patrol plan, a dispatching path and a human resource allocation strategy in real time, comprehensively considering the self-healing state, risk level, personnel accessibility and task timeliness;

[0061] The electrical equipment includes but is not limited to power transformers, high-voltage switchgears, cable joints, GIS equipment and reactor systems. The micro-autonomous repair module can be embedded in the insulating oil, solid insulation structure or gel layer of the electrical equipment as needed; the scale of the nano-robot group supports a dynamic loading and unloading mechanism. At the initial operation, the edge control system injects a basic repair group, and then automatically adjusts the nano-robot density according to the damage distribution to achieve resource conservation and enhanced centralized repair capabilities;

[0062] Furthermore, the joint data of the edge computing gateway and the cloud platform includes: the real-time risk score output by the edge computing gateway for each device; the log of the micro-autonomous repair module, including the number of damages to be repaired, the repair success rate, the current nano-robot density; the location, skill level and current task status of the operation and maintenance personnel; the latest processing time for each risk point; the data is summarized in a structured JSON array form to the memory database of the adaptive maintenance scheduler;

[0063] Construct a state vector for each device i

[0064]

[0065] where r iis the risk score, d i is the number of damages to be repaired, is the estimated formation time from person j to device i, a j is the personnel availability (0 / 1), τ i is the remaining time limit of the task;

[0066] Using the deep deterministic policy gradient algorithm, which supports continuous action spaces, to simultaneously determine personnel allocation, nanorobot density, and scheduling time; for each device i, the action a i =(p ij , ρ i , t i ), p ij ∈{0, 1}: whether to assign person j; ρ i ∈[ρmin, ρmax]: the density of injected nanorobots; t i : the start time of the task;

[0067] Reward function:

[0068] Αβγδ are weight coefficients. The first term penalizes high-risk delays, the second term penalizes long-distance scheduling, the third term penalizes overuse of nanorobots, and the last term rewards on-time completion; every hour or when a new warning is triggered, the adaptive maintenance scheduler inputs all current s i into the policy network and outputs the optimal a i combination to form the current inspection and self-healing plan;

[0069] The adaptive maintenance scheduler calls the Dijkstra algorithm to calculate the shortest path for personnel walking or vehicle movement based on the internal network topology map of the substation, generates route instructions, the operation and maintenance personnel receive the task list and path navigation through the AR terminal, and the edge computing gateway receives the nanorobot density and injection time, and controls the micro-injection pump and solenoid valve to inject or recover nanorobots into the device insulation medium at the specified time;

[0070] When the system starts or a new device is put into operation, the edge network controls the micro-main device to inject a basic nanorobot population into the insulating oil or solid insulating gel layer. When the adaptive maintenance scheduler instructs ρ i＞ ρ base , the injection pump increases the nanorobot density to ρ i before the start of the task. The injection position is guided to the preset micro-injection port through the AR terminal. After the repair is completed, the edge computing gateway activates the acoustic drive and magnetic field reverse gradient to guide the nanorobots to converge towards the recovery port, and the recovery pump sucks and recovers them to the local nanorobot storage unit for the next scheduling use. The edge computing gateway real-time detects the rainfall in the storage unit, and when the total amount is lower than the threshold, it triggers a "nanorobot replenishment" reminder or automatically issues a replenishment work order to the central warehouse;

[0071] The operation and maintenance personnel mark "task completed" on the AR terminal and upload on-site photos and repair logs. The edge computing gateway records the actual consumption of nanorobots, the repair success rate, and the time, and writes them into the blockchain through a smart contract. The adaptive maintenance scheduler regularly uses the task execution records and the actual system performance as experience for follow-up visits.

[0072] Please refer to Figure 1 and Figure 2 For an embodiment provided by the present invention: an intelligent security protection power operation and maintenance system, in which the nanorobot swarm floats in insulating oil or composite insulating resin. Each nanorobot has a length of no more than 500 nanometers and is coated with a multi-modal driving layer, including a superparamagnetic response coating for magnetic navigation, a pressure material layer for acoustic wave propulsion, and a photosensitive particle array for directional light manipulation. At the same time, an electric field gradient identification chip and a fluorescence imaging unit are integrated;

[0073] Trigger-type self-healing microcapsules evenly distributed in the insulating medium. Each microcapsule is composed of a multi-layer polymer shell layer. The outer shell has an electric field-responsive microporous channel and only ruptures when a nanorobot approaches and provides mechanical disturbance or ultrasonic energy of a specific frequency; Inside the microcapsule, two groups of non-reactive resins and catalyst monomers, A and B, are encapsulated. After rupture, they are instantaneously mixed to achieve in-situ polymerization self-healing at the damage position;

[0074] The nanorobots continuously scan the microscopic damage area inside the insulator. After detecting microcracks or discharge paths, they converge into a repair formation, activate the nearest microcapsule at a fixed point, and complete crack sealing and dielectric constant recovery;

[0075] The results of the repair behavior are verified twice through a machine vision and electric field reconstruction mechanism. If a repair failure is identified, a secondary repair will be automatically initiated;

[0076] The nanorobots adjust the formation spacing through a tunable magnetic field gradient, and the response frequency range is 0.1Hz - 100Hz, which is used to adapt to migration control in different medium viscosity environments; The microcapsule outer shell material adopts a three-layer structure. The outermost layer is a hydrophobic polymer to improve suspension stability. The middle layer is an electric field-responsive conductive shell, and the inner layer is a controllable rupture polyurethane ester chain polymer;

[0077] Furthermore, by applying a tunable magnetic field gradient and acoustic waves, the nanorobots perform grid cruising, synchronously excite a 650nm light beam for directional light manipulation. The electric field gradient identification chip continuously monitors local electric field mutations, and the fluorescence imaging unit captures the fluorescence marker distribution to generate a real-time three-dimensional position map; The edge computing gateway sends a "formation coordinates + spacing 500nm" instruction to the nanorobots in the target area through optical communication. The nanorobots respond to the instruction and self-organize into a "repair formation" to surround the microcrack or discharge channel within ≤50ms;

[0078] The nano-robots at the formation center activate the micro-acoustic excitation unit to apply local stress to the outer layer of the nearest microcapsule. The conductive shell of the middle layer responds to the electric field and forms a channel under the local field strength. The micropores on the outer shell open, and the shell ruptures under mechanical disturbance, releasing the internal components A and B. Components A and B are instantaneously mixed at the crack position. At an ambient temperature of 25°C, the polymer cures within 60 s to form a dense repair bridge, restoring the dielectric constant to more than 95% of its original value. The nano-robots carry a fluorescence imaging unit to scan the repair area and detect the filling integrity. The electric field gradient recognition chip re-measures the local field strength. If the deviation is <5%, the repair is determined to be successful. If the verification fails, the edge computing gateway automatically issues a "re-repair" instruction until it is successful or exceeds the limit 3 times, and records the results on the blockchain.

[0079] Working principle: The system realizes the synchronous, real-time acquisition and fusion preprocessing of power flow, temperature field, discharge precursors, and fault gases through a multi-modal sensor layer deployed on the surface and inside of the electrical equipment body, providing a high-precision, multi-dimensional perception basis for subsequent analysis.

[0080] The edge computing gateway deploys embedded AI and federated learning engines locally to extract features and perform dynamic risk scoring on multi-source sensor data. The cloud digital twin platform combines the historical database and the electromagnetic-thermal coupling simulation model to perform iterative modeling of the equipment state and predict the remaining life. The adaptive maintenance scheduler uses reinforcement learning algorithms to jointly reason about elements such as risk assessment, self-healing status, and personnel accessibility to generate an optimal inspection and repair plan.

[0081] After detecting microscopic damage, the microscopic autonomous repair module completes a nano-level closed-loop self-healing process of "scanning, formation, fixed-point activation, in-situ polymerization, and secondary verification". The blockchain security module records the abnormal time, self-healing logs, and operation and maintenance instructions in an immutable manner throughout the process. The augmented reality operation and maintenance terminal combines with the digital twin model to provide visual guidance for on-site personnel. The system continuously optimizes the sensing, repair, and scheduling strategies based on the feedback.

[0082] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. An intelligent security protection power operation and maintenance system, including a microscopic self-repair module, characterized in that: The microscopic autonomous repair module is integrated inside the device and is located between the multimodal sensor layer and the insulating medium of the electrical device. The microscopic autonomous repair module includes: A group of nanorobots, suspended in an insulating medium and self-organized into formations through magnetic coupling, acoustic propulsion, and optically induced multimodal drive units; Trigger-type self-healing microcapsules contain two-component self-healing resin and catalyst; The nanorobot swarm uses integrated electrical detection sensors and optical scattering sensors to detect microscopic cracks and discharge channels inside electrical equipment in real time. After detecting micro-damage, it ruptures the corresponding self-healing microcapsules at a fixed point, allowing the resin and catalyst to mix and polymerize on-site, achieving a closed-loop self-healing of "micro-damage-instant repair-re-detection".

2. An intelligent security protection power operation and maintenance system according to claim 1, characterized in that: The multimodal sensor layer is deployed on the surface and internal structure of the electrical equipment body, including: Current transformers and voltage sensors are used to collect power flow status under real-time operating conditions; Fiber optic temperature sensor and infrared thermal imaging array are used to reconstruct temperature distribution maps; A discharge warning unit composed of a partial discharge sensor and an ultrasonic sensor is used to identify precursors to electrical breakdown; The gas sensor array is used to monitor changes in gas concentration in transformer oil and trace the source of faulty gas.

3. An intelligent safety protection power operation and maintenance system according to claim 1, characterized in that: The operation and maintenance system includes an edge computing gateway, which is installed in the terminal box or switch cabinet of the electrical equipment, and integrates an embedded AI processing unit, an edge cache module and a local encryption module, and is used to perform fusion analysis on the data collected by the multimodal sensor layer, and deploy a federated learning engine to achieve equipment health status modeling and dynamic fault risk scoring under privacy protection.

4. An intelligent security protection power operation and maintenance system according to claim 1, characterized in that: The operation and maintenance system includes a blockchain security module, which is built on a lightweight distributed ledger architecture, adopts a PBFT consensus mechanism, and has a built-in smart contract for automatically recording anomaly detection time, self-healing behavior logs, and the execution path of operation and maintenance instructions to ensure that the system is traceable, trustworthy, and cannot be tampered with.

5. An intelligent safety protection power operation and maintenance system according to claim 1, characterized in that: The operation and maintenance system includes a cloud-based digital twin platform, which uses digital modeling tools to integrate the three-dimensional geometric structure, electromagnetic thermal coupling status and sensor data of electrical equipment to construct a virtual twin. The cloud-based digital twin platform has the capabilities of model iteration, self-learning synchronization and status visualization, and is linked with the historical database to realize trend prediction and calculation of the remaining life of the structure.

6. An intelligent safety protection power operation and maintenance system according to claim 1, characterized in that: The operation and maintenance system includes an augmented reality operation and maintenance terminal, which is worn on the head of on-site operation and maintenance personnel. Based on the digital twin model and real-time sensor data, it synchronously displays hidden danger information of key parts inside the equipment, risk hot zone distribution and visual operation guidance paths, and supports gesture interaction, voice control and automatic calibration functions.

7. An intelligent security protection power operation and maintenance system according to claim 1, characterized in that: The operation and maintenance system also includes an adaptive maintenance scheduler, which uses a reinforcement learning optimization algorithm to generate inspection plans, work dispatch paths and human resource allocation strategies in real time based on the joint reasoning results of the edge computing gateway and the cloud platform, and comprehensively considers self-healing status, risk level, personnel accessibility and task timeliness.

8. An intelligent and secure power operation and maintenance system according to claim 1, characterized in that: The swarm of nanorobots is suspended in insulating oil or composite insulating resin. Each nanorobot has a length of no more than 500 nanometers and is coated with a multimodal driving layer, including a superparamagnetic response coating for magnetic navigation, a pressure material layer for acoustic wave propulsion, and a photosensitive particle array for directional light manipulation. At the same time, an electric field gradient recognition chip and a fluorescence imaging unit are integrated; Trigger-type self-healing microcapsules evenly distributed in the insulating medium. Each microcapsule is composed of multiple polymer shell layers. The outer shell has an electric field-responsive microporous channel and only ruptures when a nanorobot approaches and provides mechanical perturbation or ultrasonic energy at a specific frequency; Inside the microcapsule, two groups of non-reactive resins and catalyst monomers, A and B, are encapsulated. After rupture, they are instantaneously mixed to achieve in-situ polymerization self-healing at the damage location; The nanorobots continuously scan the microscopic damage area inside the insulator. After detecting microcracks or discharge paths, they converge into a repair formation, activate the nearest microcapsule at a fixed point, and complete crack sealing and dielectric constant restoration; The results of the repair actions are verified a second time through a machine vision and electric field reconstruction mechanism. If a repair failure is identified, a second repair will be automatically initiated.

9. An intelligent safety protection power operation and maintenance system according to claim 8, characterized in that: The nanorobots adjust the formation spacing through a tunable magnetic field gradient, with a response frequency range of 0.1 Hz to 100 Hz, for migration control in different medium viscosity environments; The microcapsule outer shell material has a three-layer structure. The outermost layer is a hydrophobic polymer to improve suspension stability, the middle layer is an electric field-responsive conductive shell, and the inner layer is a controllable rupture polyurethane ester chain polymer.

10. An intelligent safety protection power operation and maintenance system according to claim 1, characterized in that: The electrical equipment includes but is not limited to power transformers, high-voltage switchgear, cable joints, GIS equipment, and reactor systems. The microscopic autonomous repair module can be embedded in the insulating oil, solid insulation structure, or gel layer of the electrical equipment as needed; The scale of the nanorobot swarm supports a dynamic loading and unloading mechanism. At the initial operation, a basic repair swarm is injected by the edge control system, and then the density of the nanorobots is automatically scheduled according to the damage distribution to achieve resource conservation and enhanced centralized repair capabilities.

Images