Electrical equipment real-time state live detection method and system

Abstract

The invention relates to the technical field of electrical equipment detection, and discloses an electrical equipment real-time state live detection method and system, and the system comprises a non-contact multi-source sensing array, an edge calculation unit, a high-frequency pulse excitation module, a self-adaptive installation structure, an intelligent diagnosis platform, and a self-energy-taking power supply unit. When electrical equipment state live-line detection is carried out, infrared thermodynamic characteristics, ultraviolet corona intensity and ultrasonic discharge signals are fused and collected through a non-contact multi-source sensing array, multi-parameter real-time sensing under the live-line condition is achieved, the problem of signal distortion caused by electromagnetic interference in traditional detection is solved, and the accuracy of state parameter extraction is improved; and meanwhile, the multi-dimensional feature data is subjected to standardization processing in combination with an edge calculation unit, so that the system can eliminate interference of environmental factors on detection results, the consistency of evaluation results under different working conditions is guaranteed, and state diagnosis errors are reduced.

Description

Technical Field

[0001] This invention relates to the field of electrical equipment testing technology, specifically to a method and system for real-time live-line testing of electrical equipment. Background Technology

[0002] With the rapid development of power grids and the fast pace of equipment upgrades, maintenance work is extensive. Power outages during maintenance can cause significant losses; therefore, live-line maintenance should be performed whenever possible. Live-line testing is a core technology for condition-based maintenance of power equipment. It utilizes methods such as infrared thermography, X-ray detection, and ultrasonic testing to monitor equipment in real-time while it is energized. This technology is a crucial component of the power equipment condition monitoring system, forming a technological pillar of smart grid transformation together with online monitoring. Compared to traditional accident-based and periodic maintenance, live-line testing offers advantages such as dynamically assessing equipment condition and balancing safety and economy.

[0003] Currently, since electrical equipment condition monitoring requires power outage operation, the traditional sensing system cannot efficiently collect multi-source feature data when the equipment is energized during real-time condition monitoring. When electromagnetic interference and signal attenuation occur during the detection process, the error in extracting condition parameters will be large, and the accuracy of the assessment cannot be guaranteed.

[0004] Therefore, a real-time live-line detection method and system for electrical equipment is proposed to solve the above problems. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a method and system for real-time live-line detection of electrical equipment, solving the problems mentioned in the background section.

[0006] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a method and system for real-time live-line detection of electrical equipment, the method comprising the following steps: S1. Collect multi-dimensional state characteristics of energized electrical equipment through a non-contact multi-source sensor array to generate a real-time sensing dataset; S2. Based on the edge computing unit, the real-time sensing dataset is preprocessed and features are extracted to generate a standardized state feature vector; S3. The high-frequency pulse excitation module is coupled to the surface of the electrical equipment through an adaptive mounting structure to generate an enhanced characteristic excitation signal; S4. Use the enhanced feature excitation signal to directionally excite the insulation defect area of ​​the equipment, and simultaneously collect the partial discharge response signal and generate a pulse response spectrum; S5. Input the standardized state feature vector and the impulse response spectrum into the intelligent diagnostic model for multi-source information fusion analysis to generate an equipment health status assessment index; S6. When the health status assessment index of the equipment exceeds the preset threshold, a real-time early warning command is triggered and a live-line maintenance decision plan is generated. The non-contact multi-source sensor array includes an infrared thermal imaging module, an ultraviolet corona detection module, and an ultrasonic partial discharge module. The adaptive mounting structure integrates a self-powered unit and an insulation isolation device.

[0007] Preferably, step S1 includes the following steps: S11. Collect surface temperature field distribution data of the equipment through a spatially distributed infrared sensor matrix to generate a thermodynamic feature vector; S12. Use a solar-blind ultraviolet detector to capture corona discharge spectral data and generate a corona intensity distribution map; S13. Acquire partial discharge acoustic signals based on piezoelectric ultrasonic sensor array and generate time-frequency domain discharge characteristic spectrum.

[0008] Preferably, step S2 includes the following steps: S21. Perform environmental temperature compensation correction on the thermodynamic eigenvector to generate a temperature gradient change matrix; S22. Perform spatial registration processing on the corona intensity distribution map to generate a spatiotemporal evolution model of corona intensity; S23. Perform wavelet noise reduction processing on the time-frequency domain discharge feature spectrum to extract the discharge pulse feature parameter set.

[0009] Preferably, step S3 includes the following steps: S31. The high-frequency pulse excitation module is attached to the surface of the equipment by means of a magnetically adjustable bracket, wherein the magnetically adjustable bracket includes a three-dimensional angle adjustment mechanism and a pressure feedback device. S32. Automatically adjust the tilt angle of the excitation probe based on the structural characteristics of the equipment, so that the angle between the pulse emission direction and the normal to the equipment surface is less than 5°; S33. The output impedance is adjusted to 50±2Ω through the impedance matching circuit to ensure that the signal transmission efficiency is greater than 95%.

[0010] Preferably, the operating parameters of the high-frequency pulse excitation module are: Pulse frequency range: 1MHz-10MHz adjustable; Output voltage amplitude: 0.1kV-5kV programmable; Pulse rise time: less than 10 ns; Duty cycle adjustment range: 1%-50% Preferably, step S4 includes the following steps: S41. Employs a high-speed data acquisition card to synchronously acquire pulse response signals, with a sampling rate greater than 200MSPS; S42. Locate the insulation defect using the time-domain reflection algorithm and generate a set of defect spatial coordinates; S43. Extract the discharge repetition frequency spectrum, single discharge energy distribution, phase-resolved discharge spectrum and pulse sequence correlation matrix based on pulse waveform features.

[0011] Preferably, step S5 includes the following steps: S51. Construct a deep belief network model. The input layer contains 12 feature channels, corresponding to the temperature gradient change rate, corona intensity growth rate, discharge pulse repetition rate, discharge energy entropy, pulse waveform distortion, phase distribution symmetry, and other fused feature quantities. S52. Calculate the weight coefficients of each channel using a feature importance weighting algorithm. The weight update formula is as follows: ; in Let be the weight coefficient of the i-th feature channel at iteration number t. Let be the updated weight coefficient of the i-th feature channel at iteration number t+1, and let be the learning rate. , For the loss function L with respect to Partial derivatives of the weights, For loss function, For feature channel index, This represents the number of iterations. S53, Output Device Health Status Assessment Index The formula for its calculation is: ; in This is an index for assessing the health status of equipment. For feature index, For the total number of features, For the first The fusion feature quantity of each feature, For the first The degradation factor of a feature.

[0012] Preferably, step S6 includes the following steps: S61. Establish a three-tiered early warning mechanism: the primary early warning threshold is 0.7. <0.8, the intermediate warning threshold is 0.6< <0.7, the emergency warning threshold is <0.6; S62. Generate differentiated maintenance plans based on the warning level: a primary warning triggers a 72-hour maintenance plan, a medium warning triggers a 24-hour maintenance plan, and an emergency warning triggers an immediate power outage maintenance command.

[0013] Preferably, the system includes: The non-contact multi-source sensor array uses an infrared thermal imaging module to collect infrared thermodynamic feature data of the device surface, an ultraviolet corona detection module to capture corona discharge spectrum data, and an ultrasonic partial discharge module to collect partial discharge acoustic signals to generate a real-time sensing dataset. The edge computing unit preprocesses the real-time sensing dataset through a signal preprocessing circuit and uses a feature extraction coprocessor to extract multi-dimensional features and generate a standardized state feature vector. The high-frequency pulse excitation module generates high-frequency pulse signals through a programmable pulse generator, outputs the signals to the intelligent diagnostic platform for analysis, and uses an impedance matching network for impedance matching to generate an enhanced characteristic excitation signal. The adaptive installation structure features a magnetically adjustable bracket to adjust the installation position and angle, and monitors the contact pressure through a pressure feedback device to couple the high-frequency pulse excitation module to the surface of the electrical equipment. The intelligent diagnostic platform runs a multi-source information fusion analysis algorithm, receives standardized state feature vectors and impulse response spectra, and generates equipment health status assessment indices. The self-powered unit obtains energy from the power frequency magnetic field through an induction energy harvesting coil and stores the energy using a supercapacitor energy storage device to power each unit of the system.

[0014] Preferred options include: The adaptive mounting structure includes a carbon fiber insulated connecting rod with a withstand voltage rating greater than 35kV / cm, a universal joint adjustment mechanism with an angle adjustment range of ±30°, a contact pressure sensor with an adjustable range of 0-50N, and an RF shielded housing with a shielding effectiveness greater than 80dB. The self-powered unit is implemented by inductively harvesting energy through a Rogowski coil, and outputting a power greater than 5W at a 1A power frequency current. The intelligent diagnostic platform includes an FPGA-based real-time analysis module with a processing latency of less than 10ms, a cloud-based deep training module, and a mobile terminal interaction interface.

[0015] (III) Beneficial Effects Compared with the prior art, the present invention provides a method and system for real-time live-line detection of electrical equipment, which has the following beneficial effects: 1. In this invention, when performing live-line detection of electrical equipment, infrared thermodynamic features, ultraviolet corona intensity, and ultrasonic discharge signals are collected through a non-contact multi-source sensor array. This enables real-time sensing of multiple parameters under live-line conditions, overcoming the signal distortion problem caused by electromagnetic interference in traditional detection and improving the accuracy of state parameter extraction. At the same time, the multi-dimensional feature data is standardized by combining edge computing units, enabling the system to eliminate the interference of environmental factors on the detection results, ensuring the consistency of evaluation results under different operating conditions, and reducing state diagnosis errors.

[0016] 2. In this invention, during sensor deployment, the contact angle and pressure between the excitation probe and the equipment surface are adaptively adjusted in real time through the synergistic effect of the magnetically adjustable bracket and the pressure feedback device, controlling the angle between the pulse emission direction and the equipment normal. When there is a risk of displacement in the detection position, the system automatically triggers three-dimensional attitude correction based on impedance matching feedback, enabling zero-deviation coupling of the detection module even on the surface of complex equipment structures. This solves the detection failure problem caused by traditional fixed installation methods and ensures the accuracy and reliability of data acquisition positions.

[0017] 3. In this invention, when assessing the health status of equipment, a smart diagnostic model is used to perform multi-source information fusion analysis on temperature gradient changes, corona intensity growth, and discharge pulse characteristic parameters to automatically identify weak features of insulation aging and latent partial discharge defects. At the same time, a dynamic weight optimization mechanism is used to prioritize key feature quantities, enabling the system to quantify the degree of equipment health status degradation, avoid maintenance decision deviations caused by misjudgment of a single parameter, improve the predictability and accuracy of power supply system safety management, and achieve a technological leap from passive maintenance to active protection. Attached Figure Description

[0018] Figure 1 This is a flowchart of a real-time energized detection method for electrical equipment according to the present invention; Figure 2 This is a schematic diagram of a real-time energized detection system for electrical equipment according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] For specific implementation examples, please refer to: Figures 1-2 A method and system for real-time live-line detection of electrical equipment, the method comprising the following steps: S1. Collect multi-dimensional state characteristics of energized electrical equipment through a non-contact multi-source sensor array to generate a real-time sensing dataset; S2. Based on the edge computing unit, preprocess and extract features from the real-time sensing dataset to generate a standardized state feature vector; S3. The high-frequency pulse excitation module is coupled to the surface of the electrical equipment through an adaptive mounting structure to generate an enhanced characteristic excitation signal; S4. Use the enhanced characteristic excitation signal to directionally excite the insulation defect area of ​​the equipment, and simultaneously collect the partial discharge response signal and generate the pulse response spectrum; S5. Input the standardized state feature vector and impulse response spectrum into the intelligent diagnostic model for multi-source information fusion analysis to generate the equipment health status assessment index. S6. When the equipment health status assessment index exceeds the preset threshold, a real-time early warning command is triggered and a live-line maintenance decision plan is generated. The non-contact multi-source sensor array includes an infrared thermal imaging module, an ultraviolet corona detection module, and an ultrasonic partial discharge module. The adaptive mounting structure integrates a self-powered unit and an insulation isolation device.

[0021] S1 includes the following steps: S11. Collect surface temperature field distribution data of the equipment through a spatially distributed infrared sensor matrix to generate a thermodynamic feature vector; S12. Use a solar-blind ultraviolet detector to capture corona discharge spectral data and generate a corona intensity distribution map; S13. Acquire partial discharge acoustic signals based on piezoelectric ultrasonic sensor array and generate time-frequency domain discharge characteristic spectrum.

[0022] S2 includes the following steps: S21. Perform environmental temperature compensation correction on the thermodynamic eigenvectors to generate a temperature gradient change matrix; S22. Perform spatial registration processing on the corona intensity distribution map to generate a spatiotemporal evolution model of corona intensity; S23. Perform wavelet denoising on the time-frequency domain discharge characteristic spectrum to extract the discharge pulse characteristic parameter set, including the pulse energy density. The calculation formula is: ; in For pulse energy density, For the sampling time window, Let be the time-domain discharge signal function. For time variables, It is the time differential variable.

[0023] S3 includes the following steps: S31. The high-frequency pulse excitation module is attached to the surface of the equipment by means of a magnetically adjustable bracket. The magnetically adjustable bracket includes a three-dimensional angle adjustment mechanism and a pressure feedback device. S32. Automatically adjust the tilt angle of the excitation probe based on the structural characteristics of the equipment, so that the angle between the pulse emission direction and the normal to the equipment surface is less than 5°; S33. The output impedance is adjusted to 50±2Ω through the impedance matching circuit to ensure that the signal transmission efficiency is greater than 95%.

[0024] The operating parameters of the high-frequency pulse excitation module are: Pulse frequency range: 1MHz-10MHz adjustable; Output voltage amplitude: 0.1kV-5kV programmable; Pulse rise time: less than 10 ns; Duty cycle adjustment range: 1%-50%; And it satisfies waveform fidelity constraints: ; in To allow for voltage fluctuation amplitude, To set the peak output voltage, This refers to the actual operating frequency. This is the characteristic frequency of the circuit.

[0025] S4 includes the following steps: S41. Employs a high-speed data acquisition card to synchronously acquire pulse response signals, with a sampling rate greater than 200MSPS; S42. Locate the insulation defect using the time-domain reflection algorithm and generate a set of defect spatial coordinates; S43. Based on pulse waveform features, extract the discharge repetition frequency spectrum, single discharge energy distribution, phase-resolved discharge spectrum, and pulse sequence correlation matrix. The calculation formula is as follows: ; in It is the correlation index between adjacent pulse sequences. For the first The sampled voltage values ​​of the pulse groups at the same time. pulse sequence and The arithmetic mean, This represents the number of sampling points within a single pulse cycle.

[0026] S5 includes the following steps: S51. Construct a deep belief network model. The input layer contains 12 feature channels, corresponding to the temperature gradient change rate, corona intensity growth rate, discharge pulse repetition rate, discharge energy entropy, pulse waveform distortion, phase distribution symmetry, and other fused feature quantities. S52. Calculate the weight coefficients of each channel using a feature importance weighting algorithm. The weight update formula is as follows: ; in Let be the weight coefficient of the i-th feature channel at iteration number t. Let be the updated weight coefficient of the i-th feature channel at iteration number t+1, and let be the learning rate. , For the loss function L with respect to Partial derivatives of the weights, For loss function, For feature channel index, This represents the number of iterations. S53, Output Device Health Status Assessment Index The formula for its calculation is: ; in This is an index for assessing the health status of equipment. For feature index, For the total number of features, For the first The fusion feature quantity of each feature, For the first The degradation factor of a feature.

[0027] S6 includes the following steps: S61. Establish a three-level early warning mechanism and early warning confidence factors: ; Primary warning threshold: and ; Intermediate warning threshold: and ; Emergency warning threshold: , ; in To determine the confidence level for early warning, The standard deviation of the health index H The mean of the health index H is... The attenuation coefficient is... This is an index for assessing the health status of equipment.

[0028] S62. Generate differentiated maintenance plans based on the warning level: a primary warning triggers a 72-hour maintenance plan, a medium warning triggers a 24-hour maintenance plan, and an emergency warning triggers an immediate power outage maintenance command.

[0029] The system includes: The non-contact multi-source sensor array uses an infrared thermal imaging module to collect infrared thermodynamic feature data of the device surface, an ultraviolet corona detection module to capture corona discharge spectrum data, and an ultrasonic partial discharge module to collect partial discharge acoustic signals to generate a real-time sensing dataset. The edge computing unit preprocesses the real-time sensing dataset through a signal preprocessing circuit and uses a feature extraction coprocessor to extract multi-dimensional features and generate a standardized state feature vector. The high-frequency pulse excitation module generates high-frequency pulse signals through a programmable pulse generator, outputs the signals to the intelligent diagnostic platform for analysis, and uses an impedance matching network for impedance matching to generate an enhanced characteristic excitation signal. The adaptive installation structure features a magnetically adjustable bracket to adjust the installation position and angle, and monitors the contact pressure through a pressure feedback device to couple the high-frequency pulse excitation module to the surface of the electrical equipment. The intelligent diagnostic platform runs a multi-source information fusion analysis algorithm, receives standardized state feature vectors and impulse response spectra, and generates equipment health status assessment indices. The self-powered unit obtains energy from the power frequency magnetic field through an induction energy harvesting coil and stores the energy using a supercapacitor energy storage device to power each unit of the system.

[0030] Specifically, it includes: The adaptive mounting structure includes carbon fiber insulated connecting rods with a withstand voltage rating greater than 35kV / cm, a universal joint adjustment mechanism with an angle adjustment range of ±30°, a contact pressure sensor with an adjustable range of 0-50N, and an RF shielded housing with a shielding effectiveness greater than 80dB. The self-powered unit is implemented by extracting energy through a Rogowski coil, achieving an output power greater than 5W at a 1A power frequency current, and the output power meets the following requirements: ; in For output power, For energy conversion efficiency, The power frequency magnetic field strength, The effective cross-sectional area of ​​the induction coil. The power frequency. This refers to the number of coil turns. The intelligent diagnostic platform includes an FPGA-based real-time analysis module with a processing latency of less than 10ms, a cloud-based deep training module, and a mobile terminal interaction interface.

[0031] The operation steps of this method and system for real-time live-line detection of electrical equipment are as follows: Step 1: Multi-source signal collaborative acquisition and preprocessing A spatially distributed infrared sensor matrix scans the surface temperature field distribution of charged equipment in real time, generating a dynamic thermodynamic feature vector. Simultaneously, a solar-blind ultraviolet detector captures the corona discharge spectral intensity, constructing a spatiotemporal evolution model of the corona intensity. A piezoelectric ultrasonic sensor array collects partial discharge acoustic signals, which are then processed with wavelet denoising to extract the time-frequency domain discharge feature spectrum. The three data sources are converged to an edge computing unit via a low-latency wireless transmission protocol. Environmental temperature compensation correction is applied to the thermodynamic features to eliminate environmental interference, and spatial registration processing is performed on the corona intensity distribution map to ensure spatiotemporal consistency. Finally, a standardized state feature vector is output, solving the signal distortion problem caused by electromagnetic noise in traditional detection and enabling simultaneous sensing of multiple physical quantities under charged conditions.

[0032] Step 2: Adaptive Excitation and Defect Feature Enhancement A three-dimensional angle adjustment mechanism and pressure feedback device based on a magnetically adjustable bracket are used. A high-frequency pulse excitation module, with an adjustable operating frequency of 1-10MHz and a programmable output voltage of 0.1-5kV, is adaptively attached to the equipment surface. The probe tilt angle is adjusted in real time to ensure the angle between the pulse emission direction and the equipment normal is less than 5°. An impedance matching circuit stabilizes the output impedance at 50±2Ω, ensuring a signal transmission efficiency greater than 95%. This excitation signal is used to directionally excite insulation defect areas. A 200MS / s high-speed data acquisition card synchronously captures the partial discharge response. A time-domain reflectometry algorithm is used to locate the spatial coordinates of the defect, and the discharge repetition frequency spectrum, single discharge energy distribution, and pulse sequence correlation index are extracted. Calculate the similarity of adjacent pulse waveforms to enhance the ability to capture weak features of latent defects.

[0033] Step 3: Multi-source fusion diagnosis and dynamic early warning decision-making The deep belief network model receives a 12-dimensional standardized feature vector, including the temperature gradient change rate, corona intensity growth rate, discharge pulse repetition rate, and pulse response spectrum. It dynamically optimizes the weight coefficients of each channel using a feature importance weighting algorithm, outputting a device health status assessment index H. When the H value falls below a preset threshold, a three-level early warning mechanism is triggered: combining the early warning confidence factor... Perform a graded response: H ≥ 0.7 and >0.85 triggers a 72-hour maintenance plan, 0.6≤H<0.7 and >0.75 triggers 24-hour emergency maintenance; H<0.6 ≤0.6 Immediate power outage maintenance is performed to quantify and proactively protect against equipment degradation.

[0034] Step 4: System Cooperative Control and Security Protection The self-powered unit induces power frequency magnetic field energy through a Rogowski coil, and outputs power at a current of 1A. A 5W drive carbon fiber insulated connecting rod with a withstand voltage rating greater than 35kV / cm supports the sensing-excitation module; the intelligent diagnostic platform relies on an FPGA chip to achieve real-time analysis in less than 10ms, and the cloud training module continuously updates the multi-source information fusion algorithm; the RF shielded shell has a shielding effectiveness greater than 80dB to suppress external electromagnetic interference; the universal joint adjustment mechanism has an angle range of ±30° and the contact pressure sensor has a range of 0-50N, forming dual safety monitoring to ensure zero safety accidents during live-line testing, thus forming a closed-loop technology system from data acquisition, feature enhancement, intelligent diagnosis to safety maintenance.

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0036] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time live-line detection of the status of electrical equipment, characterized in that: The method includes the following steps: S1. Collect multi-dimensional state characteristics of energized electrical equipment through a non-contact multi-source sensor array to generate a real-time sensing dataset; S2. Based on the edge computing unit, the real-time sensing dataset is preprocessed and features are extracted to generate a standardized state feature vector; S3. The high-frequency pulse excitation module is coupled to the surface of the electrical equipment through an adaptive mounting structure to generate an enhanced characteristic excitation signal; S4. Use the enhanced feature excitation signal to directionally excite the insulation defect area of ​​the equipment, and simultaneously collect the partial discharge response signal and generate a pulse response spectrum; S5. Input the standardized state feature vector and the impulse response spectrum into the intelligent diagnostic model for multi-source information fusion analysis to generate an equipment health status assessment index; S6. When the health status assessment index of the equipment exceeds the preset threshold, a real-time early warning command is triggered and a live-line maintenance decision plan is generated. The non-contact multi-source sensor array includes an infrared thermal imaging module, an ultraviolet corona detection module, and an ultrasonic partial discharge module. The adaptive mounting structure integrates a self-powered unit and an insulation isolation device.

2. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: S1 includes the following steps: S11. Collect surface temperature field distribution data of the equipment through a spatially distributed infrared sensor matrix to generate a thermodynamic feature vector; S12. Use a solar-blind ultraviolet detector to capture corona discharge spectral data and generate a corona intensity distribution map; S13. Acquire partial discharge acoustic signals based on piezoelectric ultrasonic sensor array and generate time-frequency domain discharge characteristic spectrum.

3. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: S2 includes the following steps: S21. Perform environmental temperature compensation correction on the thermodynamic eigenvector to generate a temperature gradient change matrix; S22. Perform spatial registration processing on the corona intensity distribution map to generate a spatiotemporal evolution model of corona intensity; S23. Perform wavelet noise reduction processing on the time-frequency domain discharge feature spectrum to extract the discharge pulse feature parameter set.

4. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: S3 includes the following steps: S31. The high-frequency pulse excitation module is attached to the surface of the equipment by means of a magnetically adjustable bracket, wherein the magnetically adjustable bracket includes a three-dimensional angle adjustment mechanism and a pressure feedback device. S32. Automatically adjust the tilt angle of the excitation probe based on the structural characteristics of the equipment, so that the angle between the pulse emission direction and the normal to the equipment surface is less than 5°; S33. The output impedance is adjusted to 50±2Ω through the impedance matching circuit to ensure that the signal transmission efficiency is greater than 95%.

5. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: The operating parameters of the high-frequency pulse excitation module are as follows: Pulse frequency range: 1MHz-10MHz adjustable; Output voltage amplitude: 0.1kV-5kV programmable; Pulse rise time: less than 10 ns; Duty cycle adjustment range: 1%-50%.

6. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: S4 includes the following steps: S41. Employs a high-speed data acquisition card to synchronously acquire pulse response signals, with a sampling rate greater than 200MSPS; S42. Locate the insulation defect using the time-domain reflection algorithm and generate a set of defect spatial coordinates; S43. Extract the discharge repetition frequency spectrum, single discharge energy distribution, phase-resolved discharge spectrum and pulse sequence correlation matrix based on pulse waveform features.

7. The method for real-time live-line detection of electrical equipment according to claim 1, characterized in that: S5 includes the following steps: S51. Construct a deep belief network model. The input layer contains 12 feature channels, corresponding to the temperature gradient change rate, corona intensity growth rate, discharge pulse repetition rate, discharge energy entropy, pulse waveform distortion, phase distribution symmetry, and other fused feature quantities. S52. Calculate the weight coefficients of each channel using a feature importance weighting algorithm. The weight update formula is as follows: ； in Let be the weight coefficient of the i-th feature channel at iteration number t. Let be the updated weight coefficient of the i-th feature channel at iteration number t+1, and let be the learning rate. , For the loss function L with respect to Partial derivatives of the weights, For loss function, For feature channel index, This represents the number of iterations. S53, Output Device Health Status Assessment Index The formula for its calculation is: ； in This is an index for assessing the health status of equipment. For feature index, For the total number of features, For the first The fusion feature quantity of each feature, For the first The degradation factor of a feature.

8. The method for real-time live-line detection of electrical equipment according to claim 7, characterized in that: S6 includes the following steps: S61. Establish a three-tiered early warning mechanism: the primary early warning threshold is 0.

7. <0.8, the intermediate warning threshold is 0.6< <0.7, the emergency warning threshold is <0.6; S62. Generate differentiated maintenance plans based on the warning level: a primary warning triggers a 72-hour maintenance plan, a medium warning triggers a 24-hour maintenance plan, and an emergency warning triggers an immediate power outage maintenance command.

9. A real-time live-line detection system for electrical equipment, used to implement the real-time live-line detection method for electrical equipment as described in any one of claims 1-8, characterized in that, The system includes: The non-contact multi-source sensor array uses an infrared thermal imaging module to collect infrared thermodynamic feature data of the device surface, an ultraviolet corona detection module to capture corona discharge spectrum data, and an ultrasonic partial discharge module to collect partial discharge acoustic signals to generate a real-time sensing dataset. The edge computing unit preprocesses the real-time sensing dataset through a signal preprocessing circuit and uses a feature extraction coprocessor to extract multi-dimensional features and generate a standardized state feature vector. The high-frequency pulse excitation module generates high-frequency pulse signals through a programmable pulse generator, outputs the signals to the intelligent diagnostic platform for analysis, and uses an impedance matching network for impedance matching to generate an enhanced characteristic excitation signal. The adaptive installation structure features a magnetically adjustable bracket to adjust the installation position and angle, and monitors the contact pressure through a pressure feedback device to couple the high-frequency pulse excitation module to the surface of the electrical equipment. The intelligent diagnostic platform runs a multi-source information fusion analysis algorithm, receives standardized state feature vectors and impulse response spectra, and generates equipment health status assessment indices. The self-powered unit obtains energy from the power frequency magnetic field through an induction energy harvesting coil and stores the energy using a supercapacitor energy storage device to power each unit of the system.

10. The real-time energized detection system for electrical equipment according to claim 9, characterized in that: Specifically, it includes: The adaptive mounting structure includes a carbon fiber insulated connecting rod with a withstand voltage rating greater than 35kV / cm, a universal joint adjustment mechanism with an angle adjustment range of ±30°, a contact pressure sensor with an adjustable range of 0-50N, and an RF shielded housing with a shielding effectiveness greater than 80dB. The self-powered unit is implemented by induction of energy through a Rogowski coil, and the output power is greater than 5W under a 1A power frequency current. The intelligent diagnostic platform includes an FPGA-based real-time analysis module with a processing latency of less than 10ms, a cloud-based deep training module, and a mobile terminal interaction interface.

Images