Lightning arrester parameter measurement system and method

Abstract

The invention relates to the field of lightning arrester parameter measurement, in particular to a lightning arrester parameter measurement system and method.The lightning arrester parameter measurement system comprises an induction unit, a detection unit, a data processing unit and a control and positioning unit and is used for generating and focusing laser on the surface of a lightning arrester, collecting an electrical signal and an optical signal of plasma and sending the electrical signal and the optical signal to a processor; the feature quantity is obtained through a feature extraction algorithm, and core parameters of the inversion lightning arrester are calculated through mechanical learning. Plasma is generated through laser focusing, reduction of errors caused by direct electrical connection and contact resistance with a lightning arrester is avoided, the problems of potential safety hazards and interference of traditional contact type measurement are solved, laser focus global scanning, real-time dynamic monitoring and efficient measurement are achieved through the three-dimensional electric translation platform, and the measurement precision is improved. All key areas on the surface of the lightning arrester are covered, compared with single parameter inversion in the prior art, multiple parameters and microscopic parameters are considered, and the prediction precision of key core parameters such as insulation resistance based on the random forest model is further improved.

Description

Technical Field

[0001] This invention belongs to the field of surge arrester parameter measurement, and specifically relates to a surge arrester parameter measurement system and method. Background Technology

[0002] Surge arresters are crucial protective electrical devices in power systems and are widely used. The voltage and temperature distribution characteristics of surge arresters are important parameters for their design and operation. Furthermore, as key protective equipment in power systems, the accurate measurement of core parameters such as insulation resistance and leakage current is essential for the reliability of the equipment.

[0003] For example, the surge arrester parameter measurement method and system disclosed in Chinese Invention Publication No. CN109752614A measures the small current volt-ampere characteristics of each resistor element in each surge arrester element under different surge arrester temperatures; when the surge arrester temperature is stable, it measures the temperature and current of each resistor element in each surge arrester element; it generates a surge arrester temperature distribution curve based on the temperature of each resistor element in each surge arrester element; it generates a surge arrester current distribution curve based on the current of each resistor element in each surge arrester element; it generates the voltage of each resistor element in each surge arrester element based on the current and small current volt-ampere characteristics of each resistor element; and it generates the voltage distribution non-uniformity coefficient and voltage distribution curve of each resistor element in each surge arrester element based on the voltage of each resistor element in each surge arrester element. This application improves the efficiency of surge arrester testing, reduces the workload of surge arrester testing, and improves the accuracy of parameter measurement.

[0004] However, this method relies on the direct electrical connection of the resistor element, and the contact resistance at the contact point will cause measurement errors. In addition, it may cause safety hazards in high voltage environments. The voltage distribution non-uniformity coefficient is calculated only by the correlation between current, voltage and temperature. Without integrating the microscopic characteristics of the material, it is difficult to reflect the early characteristics of insulation degradation inside the surge arrester. The measurement of temperature distribution, current distribution and voltage distribution need to be completed in stages, which cannot achieve real-time dynamic monitoring and makes it difficult to capture transient faults.

[0005] Therefore, it is crucial to design a surge arrester parameter measurement system and method that can accurately predict key core parameters such as insulation resistance based on a random forest model. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a surge arrester parameter measurement system.

[0007] To address the aforementioned technical problems, the present invention employs the following technical solution: a surge arrester parameter measurement system comprising: an induction unit for generating and focusing a laser onto the surface of the surge arrester to produce plasma; a detection unit for simultaneously acquiring the electrical and optical signals of the plasma, wherein the electrical detection subunit captures the transient current and transient voltage of the plasma channel through a current sensor and a voltage probe, and the optical detection subunit acquires the emission spectrum characteristics of the plasma using a spectrometer and a photomultiplier tube; a data processing unit for preprocessing the raw signals transmitted by the detection unit, including baseline correction, noise suppression, and normalization, and obtaining key feature quantities such as conductivity, electron density, and spectral line intensity ratio through feature extraction algorithms, and using machine learning to calculate and invert core parameters of the surge arrester such as insulation resistance and leakage current; and a control and positioning unit for regulating the laser movement of the induction unit, wavelength, pulse energy, repetition frequency, etc., and achieving precise positioning and multi-point scanning of the laser focus on the surface of the surge arrester through a three-dimensional electric translation stage, while coordinating the working sequence of the induction unit, detection unit, and data processing unit to ensure the synchronization and stability of the measurement process.

[0008] In some embodiments, the induction unit presets the single-pulse energy E, pulse width τ, and wavelength λ based on the air ionization threshold. Specifically, it includes: transmitting and focusing the laser onto the surface of the surge arrester through an optical focusing system composed of a concave mirror and a convex lens; the focused laser forms a spot with a diameter d on the surface of the surge arrester; the spot area S is obtained from the spot diameter d; and the energy density is ≥ 10 times the air ionization threshold. 11 W / cm 2 The required transmission efficiency η and the actual focused laser energy E1 are obtained until a plasma channel appears in the focusing area of ​​the surge arrester.

[0009] A further setting involves using an optical focusing system composed of a concave mirror and a convex lens to obtain the focal length f of the focusing lens, the focal length f0 of the original laser beam waist, and the laser beam waist diameter d0. Based on the theoretical formula for the laser spot area, we have:

[0010]

[0011] The actual formulas for calculating energy density exist as follows:

[0012]

[0013] Combining the above two equations, and the diffraction limit formula: Where D is the lens aperture, the theoretical single-pulse energy E, pulse width τ, and wavelength λ are solved. After being focused onto the surge arrester, and based on I ≥ air power threshold 10... 11 W / cm 2The actual transmission efficiency η of the laser is obtained by setting the laser, and the actual energy of the laser reaching the surge arrester is E1 = E·η.

[0014] In some embodiments, the detection unit captures transient signals via a current sensor and a voltage probe in the electrical detection subunit. These transient signals include a transient current I0. p and transient voltage V p The electrical detection subunit is deployed on the surge arrester using a Rogowski coil, and is controlled by the output voltage U of the Rogowski coil. I With the transient current I of the plasma channel p direct proportionality Obtain transient current I p , where k I The Rogowski coil sensitivity coefficient is determined by the number of turns, cross-sectional area, and core permeability. The rate of change of current is given by U. I The integral operation is performed to restore the transient current I. p ,Right now The electrical detection subunit also includes: a high-voltage probe deployed on the surge arrester based on the voltage divider principle, with an output voltage U V Transient voltage V at both ends of the plasma channel p The following exist: Where K V Given the probe's voltage divider ratio, the deformation has V p =K V ·U V Complete the transient voltage V p The detection unit outputs I based on the electrical detection subunit. p (t), V p (t).

[0015] In some implementations, when acquiring transient current I p and transient voltage V p Subsequently, the detection unit obtains the extension length L of the plasma channel on the surface of the surge arrester through imaging-assisted measurement in the optical detection subunit. The optical detection subunit, using the cylindrical characteristics of the plasma channel and combining it with the spot diameter d obtained by the induction unit, calculates the cross-sectional area A = πr. 2 =π(1 / 2d) 2 According to the differential form of Ohm's law:

[0016] J = σ·E2;

[0017] Where J is the current density, i.e., J = I p / A, σ is the conductivity, and E2 is the electric field strength, i.e., E2 = V p / L, thus the optical detection unit obtains the conductivity detection model, namely:

[0018]

[0019] The optical detection subunit uses the grating beam splitting principle to decompose the plasma emitted light into monochromatic light of different wavelengths, and its output spectral intensity I λ The electron density N at wavelength λ of the induction unit λ exist:

[0020]

[0021] Where G is the spectrometer gain, η is the wavelength λ output by the induction unit, and n e The electron density after decomposition;

[0022] The detection unit outputs σ(t) and n based on the optical detection subunit. e (t).

[0023] In some implementations, the data processing unit obtains features including conductivity, density, and spectral intensity features through feature extraction algorithms. A machine learning model is then used to establish a mapping relationship between these features and the core parameters of the surge arrester, achieving parameter inversion. The data processing unit combines the extracted features to construct a feature vector. A random forest regression model is used, with surge arrester calibration data of known parameters as the training set, where the input is X and the output is the insulation resistance R. i Leakage current I l Optimize model parameters by minimizing the loss function, including:

[0024]

[0025] Where M is the sample size, y m These are the actual parameter values. The values ​​are model predictions, obtained after model training regarding the core parameter insulation resistance R. i Leakage current I l Given the associated parameters θ1 and θ2, establish the parameter measurement function:

[0026]

[0027] In some embodiments, the feature extraction algorithm of the data processing unit includes: the data processing unit preprocessing the acquired I p (t), V pThe instantaneous conductivity sequence σ(t) calculated using the conductivity detection model is statistically analyzed, including: selecting the maximum value in the sequence, reflecting the strongest state of plasma conductivity, σ max =max[σ(t)]; calculates the average conductivity over the time period during which the plasma exists, reflecting the overall conductivity level. Where T is the plasma presence time, determined by the start and end times of the detected signal. Furthermore, the algorithm automatically identifies the effective signal interval and completes the integration calculation by traversing the σ(t) sequence through a sliding window. The integration process is discretized using the trapezoidal rule. N is the number of valid signal points, and Δt is the sampling interval.

[0028] In some embodiments, the control and positioning unit uses a three-dimensional electric translation stage to perform multi-point scanning of the laser focus on the surface of the surge arrester, covering key areas. This includes: establishing a three-dimensional coordinate system (x, y, z) with the center point of the surge arrester surface as the origin; identifying the edge features of the surge arrester through a visual positioning system; calculating the initial positioning deviation (Δx0, Δy0); and compensating for it through the motion of the translation stage. The control and positioning unit generates a global synchronization clock through a timing controller. The induction unit, detection unit, and data processing unit generate the clock based on the clock edge and set the working state switching sequence of the induction unit, detection unit, and data processing unit.

[0029] Another technical problem to be solved by the present invention is to provide a method for measuring the parameters of a surge arrester.

[0030] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for measuring surge arrester parameters, the method comprising the following specific steps:

[0031] S101, generates and focuses a laser onto the surface of the surge arrester to produce plasma;

[0032] S102. Collect electrical and optical signals of plasma to obtain data on transient current, transient voltage, and emission spectrum characteristics of plasma;

[0033] S103. The feature quantity of the feature extraction data is used to calculate and invert the core parameters of the surge arrester using machine learning;

[0034] S104. Regulate the laser movement of the induction unit and coordinate the working timing of the induction unit, detection unit and data processing unit.

[0035] The scope of this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this application.

[0036] Due to the application of the above technical solutions, this invention has the following advantages compared with the prior art: This invention provides a surge arrester parameter measurement system and method, which generates plasma by laser focusing, avoiding direct electrical connection with the surge arrester, reducing errors caused by contact resistance, and solving the safety hazards and interference problems of traditional contact measurement. It achieves full-domain scanning of the laser focus through a three-dimensional electric translation stage, enabling real-time dynamic monitoring and efficient measurement, covering all key areas on the surge arrester surface; it also synchronously collects transient electrical and optical signals of the plasma through a detection unit, and combines the feature extraction of conductivity and electron density. Compared with the single-parameter inversion of the prior art, it considers multiple parameters and micro-parameters, and further improves the prediction accuracy of key core parameters such as insulation resistance based on the random forest model. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] This embodiment provides a method for measuring surge arrester parameters, including the following steps:

[0040] S101, generates and focuses a laser onto the surface of the surge arrester to produce plasma;

[0041] The induction unit presets the single-pulse energy E, pulse width τ, and wavelength λ based on the air ionization threshold. Specifically, it includes transmitting and focusing the laser onto the arrester surface through an optical focusing system composed of a concave mirror and a convex lens. The focused laser forms a spot with a diameter d on the arrester surface, and the spot area S is obtained from the spot diameter d. Based on the energy density ≥ air ionization threshold 10... 11 W / cm 2 The required transmission efficiency η and the actual focused laser energy E1 are obtained until a plasma channel appears in the focusing area of ​​the surge arrester.

[0042] A further setting involves using an optical focusing system composed of a concave mirror and a convex lens to obtain the focal length f of the focusing lens, the focal length f0 of the original laser beam waist, and the laser beam waist diameter d0. Based on the theoretical formula for the laser spot area, we have:

[0043]

[0044] The actual formulas for calculating energy density exist as follows:

[0045]

[0046] Combining the above two equations, and the diffraction limit formula: Where D is the lens aperture, the theoretical single-pulse energy E, pulse width τ, and wavelength λ are solved. After being focused onto the surge arrester, and based on I ≥ air power threshold 10... 11 W / cm 2 The actual transmission efficiency η of the laser is obtained by setting the laser, and the actual energy of the laser reaching the surge arrester is E1 = E·η.

[0047] S102. Collect electrical and optical signals of plasma to obtain data on transient current, transient voltage, and emission spectrum characteristics of plasma;

[0048] The detection unit captures transient signals through the current sensor and voltage probe of the electrical detection subunit. The transient signals include transient current I. p and transient voltage V p The electrical detection subunit is deployed on the surge arrester using a Rogowski coil, and is controlled by the output voltage U of the Rogowski coil. I With the transient current I of the plasma channel p direct proportionality Obtain transient current I p , where k I The Rogowski coil sensitivity coefficient is determined by the number of turns, cross-sectional area, and core permeability. The rate of change of current is determined by U. I The integral operation is performed to restore the transient current I. p ,Right now

[0049] The electrical detection subunit also includes:

[0050] Based on the voltage divider principle, a high-voltage probe is deployed on a surge arrester, and its output voltage U V Transient voltage V at both ends of the plasma channel p The following exist: Where K V Given the probe's voltage divider ratio, the deformation has V p =K V ·U V Complete the transient voltage V p Obtain;

[0051] The detection unit outputs I based on the electrical detection subunit. p (t), V p (t).

[0052] In acquiring transient current I p and transient voltage V p Subsequently, the detection unit obtains the extension length L of the plasma channel on the surface of the surge arrester through imaging-assisted measurement in the optical detection subunit. The optical detection subunit, utilizing the cylindrical characteristics of the plasma channel and combining it with the spot diameter d obtained by the induction unit, calculates the cross-sectional area A = πr. 2 =π(1 / 2d) 2 According to the differential form of Ohm's law:

[0053] J = σ·E2;

[0054] Where J is the current density, i.e., J = I p / A, σ is the conductivity, and E2 is the electric field strength, i.e., E2 = V p / L, thus the optical detection unit obtains the conductivity detection model, namely:

[0055]

[0056] The optical detection subunit uses the grating beam splitting principle to decompose the plasma emitted light into monochromatic light of different wavelengths, and its output spectral intensity I λ The electron density N at wavelength λ of the inducing unit λ exist:

[0057]

[0058] Where G is the spectrometer gain, η is the wavelength λ output by the induction unit, and n e The electron density after decomposition;

[0059] The detection unit, based on the optical detection subunit, outputs σ(t) and n. e (t).

[0060] S103. The feature quantity of the feature extraction data is used to calculate and invert the core parameters of the surge arrester using machine learning;

[0061] The data processing unit obtains features including conductivity, density, and spectral intensity through feature extraction algorithms. It then establishes a mapping relationship between these features and the core parameters of the surge arrester through a machine learning model, thereby achieving parameter inversion.

[0062] The data processing unit combines the extracted features to construct a feature vector. A random forest regression model is used, with surge arrester calibration data of known parameters as the training set, where the input is X and the output is the insulation resistance R. i Leakage current I l Optimize model parameters by minimizing the loss function, including:

[0063]

[0064] Where M is the sample size, y m These are the actual parameter values. The values ​​are model predictions, obtained after model training regarding the core parameter insulation resistance R. i Leakage current I l Given the associated parameters θ1 and θ2, establish the parameter measurement function:

[0065]

[0066] The feature extraction algorithm of the data processing unit includes:

[0067] The data processing unit preprocesses the acquired I p (t), V p The instantaneous conductivity sequence σ(t) calculated using the conductivity detection model is statistically analyzed, including:

[0068] The maximum value in the sequence is selected, reflecting the state with the strongest plasma conductivity, σ. max =max[σ(t)];

[0069] Calculate the average conductivity over the time period of plasma presence to reflect the overall conductivity level. Where T is the plasma presence time, determined by the start and end times of the detected signal. Furthermore, the algorithm automatically identifies the effective signal interval and completes the integration calculation by traversing the σ(t) sequence through a sliding window. The integration process is discretized using the trapezoidal rule. N is the number of valid signal points, and Δt is the sampling interval.

[0070] S104. Regulate the laser movement of the induction unit and coordinate the working timing of the induction unit, detection unit and data processing unit.

[0071] The control and positioning unit uses a three-dimensional electric translation stage to perform multi-point scanning of the laser focus on the surface of the surge arrester, covering key areas, including:

[0072] A three-dimensional coordinate system (x, y, z) is established with the center point of the surge arrester surface as the origin. The edge features of the surge arrester are identified by the visual positioning system, the initial positioning deviation (Δx0, Δy0) is calculated and compensated by the motion of the translation stage. The control and positioning unit generates a global synchronization clock through the timing controller. The induction unit, detection unit, and data processing unit generate the clock based on the clock edge and set the working state switching sequence of the induction unit, detection unit, and data processing unit.

[0073] In summary, the system comprised of the solution disclosed in this embodiment has the following specific modules:

[0074] The system includes:

[0075] Induction unit; used to generate and focus laser light onto the surface of the surge arrester to produce plasma;

[0076] The detection unit is used to simultaneously acquire electrical and optical signals of the plasma. The electrical detection subunit captures the transient current and transient voltage of the plasma channel through a current sensor and a voltage probe, while the optical detection subunit obtains the emission spectrum characteristics of the plasma with the help of a spectrometer and a photomultiplier tube.

[0077] The data processing unit is used to process the raw signal transmitted by the preprocessing detection unit, including baseline correction, noise suppression and normalization, and to obtain key feature quantities such as conductivity, electron density and spectral line intensity ratio through feature extraction algorithms, and to calculate and invert core parameters such as insulation resistance and leakage current of the surge arrester using machine learning.

[0078] The control and positioning unit is used to regulate the laser movement of the induction unit and to achieve precise positioning and multi-point scanning of the laser focus on the surface of the surge arrester through a three-dimensional electric translation stage. At the same time, it coordinates the working sequence of the induction unit, detection unit and data processing unit to ensure the synchronization and stability of the measurement process.

[0079] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0080] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0082] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0084] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0085] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A surge arrester parameter measurement system, characterized in that, The system includes: Induction unit; used to generate and focus laser light onto the surface of the surge arrester to produce plasma; The detection unit is used to simultaneously acquire electrical and optical signals of the plasma. The electrical detection subunit captures the transient current and transient voltage of the plasma channel, while the optical detection subunit acquires the emission spectrum characteristics of the plasma. The data processing unit is used to process the raw signal transmitted by the preprocessing detection unit and obtain feature quantities through feature extraction algorithms, so as to calculate and invert the core parameters of the surge arrester through machine learning. The control and positioning unit is used to regulate the laser movement of the induction unit and coordinate the working timing of the induction unit, detection unit, and data processing unit.

2. The surge arrester parameter measurement system according to claim 1, characterized in that, The induction unit presets the single-pulse energy E, pulse width τ, and wavelength λ according to the air ionization threshold. Specifically, it includes: transmitting and focusing the laser onto the surface of the surge arrester through an optical focusing system composed of a concave mirror and a convex lens; the focused laser forms a spot with a diameter d on the surface of the surge arrester; the spot area S is obtained from the spot diameter d; and the energy density is ≥ 10 times the air ionization threshold. 11 W / cm 2 The required transmission efficiency η and the actual focused laser energy E1 are obtained until a plasma channel appears in the focusing area of ​​the surge arrester.

3. The surge arrester parameter measurement system according to claim 1, characterized in that, The detection unit captures transient signals through the current sensor and voltage probe of the electrical detection subunit. The transient signals include transient current I. p and transient voltage V p The electrical detection subunit is deployed on the surge arrester using a Rogowski coil, and is controlled by the output voltage U of the Rogowski coil. I With the transient current I of the plasma channel p direct proportionality Obtain transient current I p , where k I The Rogowski coil sensitivity coefficient is determined by the number of turns, cross-sectional area, and core permeability. The rate of change of current is determined by U. I The integral operation is performed to restore the transient current I. p ,Right now The electrical detection subunit also includes: a high-voltage probe deployed on the surge arrester based on the voltage divider principle, with an output voltage U V Transient voltage V at both ends of the plasma channel p The following exist: Where K V Given the probe's voltage divider ratio, the deformation has V p =K V ·U V Complete the transient voltage V p The detection unit outputs I based on the electrical detection subunit. p (t), V p (t).

4. The surge arrester parameter measurement system according to claim 3, characterized in that, In acquiring transient current I p and transient voltage V p Subsequently, the detection unit obtains the extension length L of the plasma channel on the surface of the surge arrester through imaging-assisted measurement in the optical detection subunit. The optical detection subunit, using the cylindrical characteristics of the plasma channel and combining it with the spot diameter d obtained by the induction unit, calculates the cross-sectional area A = πr. 2 =π(1 / 2d) 2 According to the differential form of Ohm's law: J = σ·E²; where J is the current density, i.e., J = I p / A, σ is the conductivity, and E2 is the electric field strength, i.e., E2 = V p / L, thus the optical detection unit obtains the conductivity detection model, namely: The optical detection subunit uses the grating beam splitting principle to decompose the plasma emitted light into monochromatic light of different wavelengths, and its output spectral intensity I λ The electron density N at wavelength λ of the induction unit λ exist: Where G is the spectrometer gain, η is the wavelength λ output by the induction unit, and n e The density of electrons after decomposition; the detection unit outputs σ(t) and n based on the optical detection subunit. e (t).

5. The surge arrester parameter measurement system according to claim 1, characterized in that, The data processing unit obtains features including conductivity, density, and spectral intensity through a feature extraction algorithm. It then establishes a mapping relationship between these features and the core parameters of the surge arrester using a machine learning model, achieving parameter inversion. The data processing unit combines the extracted features to construct a feature vector. A random forest regression model is used, with surge arrester calibration data of known parameters as the training set, where the input is X and the output is the insulation resistance R. i Leakage current I l Optimize model parameters by minimizing the loss function, including: Where M is the sample size, y m These are the actual parameter values. The values ​​are model predictions, obtained after model training regarding the core parameter insulation resistance R. i Leakage current I l Given the associated parameters θ1 and θ2, establish the parameter measurement function:

6. The surge arrester parameter measurement system according to claim 5, characterized in that, The feature extraction algorithm of the data processing unit includes: the I obtained by the data processing unit through preprocessing. p (t), V p The instantaneous conductivity sequence σ(t) calculated using the conductivity detection model is statistically analyzed, including selecting the maximum value in the sequence, which reflects the strongest state of plasma conductivity, i.e., σ(t). max =max[σ(t)]; calculates the average value of the plasma over the time period in which it exists, reflecting the overall conductivity level, i.e. Where T is the plasma existence time, which is determined by the start and end times of the detection signal.

7. The surge arrester parameter measurement system according to claim 1, characterized in that, The control and positioning unit uses a three-dimensional electric translation stage to perform multi-point scanning of the laser focus on the surface of the surge arrester, covering key areas. This includes: establishing a three-dimensional coordinate system (x, y, z) with the center point of the surge arrester surface as the origin; identifying the edge features of the surge arrester through a visual positioning system; calculating the initial positioning deviation (Δx0, Δy0); and compensating for it through the motion of the translation stage. The control and positioning unit generates a global synchronization clock through a timing controller. The induction unit, detection unit, and data processing unit generate the clock based on the clock edge and set the working state switching sequence of the induction unit, detection unit, and data processing unit.

8. A method for measuring surge arrester parameters, applied to the surge arrester parameter measurement system according to any one of claims 1-7, characterized in that, The method includes the following specific steps: S101, generates and focuses a laser onto the surface of the surge arrester to produce plasma; S102. Collect electrical and optical signals of plasma to obtain data on transient current, transient voltage, and emission spectrum characteristics of plasma; S103. The feature quantity of the feature extraction data is used to calculate and invert the core parameters of the surge arrester using machine learning; S104. Regulate the laser movement of the induction unit and coordinate the working timing of the induction unit, detection unit and data processing unit.

Images