Device and method for testing damage of lightning arrester under action of strong electromagnetic pulse

Abstract

The application discloses a lightning arrester damage test analysis device and method under strong electromagnetic pulse action, and relates to the technical field of electromagnetic compatibility test of power equipment, and comprises the following: a strong electromagnetic pulse injection module, which is used for applying a strong electromagnetic pulse to the lightning arrester; a human damage construction module, which is used for constructing a controllable damage on the surface of the lightning arrester; a multi-modal sensing detection module, which is used for synchronously collecting multi-dimensional physical quantity data related to the damage of the lightning arrester under the strong electromagnetic pulse action; and a data processing and analysis module, which is used for coupling processing the multi-dimensional physical quantity data and analyzing and evaluating the damage degree of the lightning arrester. The application effectively solves the problems of single sensing detection dimension, great randomness of damage scene and one-sided damage evaluation in the prior art, can comprehensively represent the damage mechanism of the lightning arrester under the strong electromagnetic pulse action, and accurately evaluates the damage degree of the lightning arrester.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic compatibility testing technology for power equipment, and in particular to a device and method for testing and analyzing damage to surge arresters under strong electromagnetic pulses. Background Technology

[0002] Surge arresters are core protective devices in power distribution networks, limiting lightning overvoltages and switching overvoltages and protecting the insulation safety of lines and equipment. Their performance and reliability are directly related to the stable operation of the power distribution network. High-altitude nuclear electromagnetic pulse (HEMP), as a strong transient electromagnetic environment, applies high-steep pulse voltages and high-frequency energy injections to surge arresters, causing problems such as deterioration of nonlinear volt-ampere characteristics, increased residual voltage, increased leakage current, localized thermal damage, and even permanent failure, seriously threatening the safety of the power distribution network.

[0003] Existing test methods for surge arrester damage characteristics under HEMP have the following technical shortcomings: 1. Limited Sensing Dimensions: Strong electromagnetic pulses (SEMs) are rapid pulse waves on the order of 10ns. When they act on surge arresters, the energy is concentrated instantaneously and the propagation path is complex. It is difficult to determine whether the discharge path is the internal zinc oxide diaphragm or the surface skirt using a single electrical quantity. Therefore, single-dimensional detection is prone to errors in damage location and severity assessment. Existing tests mostly only collect electrical parameters such as voltage and current, ignoring key physical quantities such as acoustic signals, equipment temperature rise, and surface contamination during the discharge process. This makes it difficult to comprehensively characterize the damage mechanism of surge arresters and achieve complementary verification of multi-dimensional damage information. Specifically, the acoustic signals generated by the discharge are directly related to the discharge intensity and dielectric breakdown type. The acoustic characteristics of internal diaphragm breakdown and surface skirt flashover are significantly different, which can intuitively reflect the location of the discharge. Equipment temperature rise can characterize the energy dissipation path. Internal discharge will cause a sharp increase in local temperature in the diaphragm area, while surface discharge will cause heat to diffuse faster, resulting in a completely different temperature rise distribution. Surface contamination is an important inducing factor for surface discharge. Its accumulation directly affects the flashover probability and changes the morphology of the surface discharge path.

[0004] Existing detection methods that rely solely on electrical parameters lack the corroborating evidence of the aforementioned key physical quantities, making it impossible to construct a complete damage evolution chain. This not only makes it difficult to accurately distinguish between internal and external discharge faults under strong electromagnetic pulses, but also misses early-stage damage risks such as contamination accumulation and localized temperature rise. Consequently, the assessment of the surge arrester's health status is one-sided, failing to provide comprehensive and reliable data support for fault prediction and lifespan assessment, thus affecting the safety and stability of power system operation.

[0005] 2. Poor controllability of damage scenarios: Under natural conditions, the damage to surge arresters is random, and the distribution of discharge channels is uncertain, making it difficult to effectively collect positioning data such as ultrasonic data. The repeatability of the test is low, and it is impossible to accurately analyze the characteristics of specific damage locations. 3. Lack of data coupling analysis: Existing technologies have not established a coupled analysis model for electrical parameters, acoustic parameters, and pollution parameters. Damage assessment based on only a single parameter is one-sided and cannot accurately reflect the true damage status of the surge arrester. 4. The correlation between pollution degree and damage is not close: Although the existing standards include methods for measuring the pollution degree of insulators, they are not combined with the HEMP damage effect system, and it is impossible to quantify the influence of pollution degree on the damage evolution of surge arresters.

[0006] Therefore, there is an urgent need to build an experimental platform that integrates multimodal sensing, controllable artificial damage scenarios, and multi-parameter coupled analysis to solve the problems of poor test repeatability, incomplete damage characterization, and low evaluation accuracy in existing technologies, and to provide scientific support for the reliability design and protection of surge arresters in strong electromagnetic environments. Summary of the Invention

[0007] The purpose of this invention is to overcome the technical defects of existing test methods for surge arrester damage characteristics under strong electromagnetic pulse (ESP) conditions, such as single sensing detection dimension and poor controllability of damage scenarios, resulting in low test evaluation accuracy. This invention provides a device and method for testing and analyzing surge arrester damage under ESP conditions. By integrating modules for ESP injection, artificial damage construction, multimodal sensing detection, synchronous data acquisition, and data processing and analysis, it achieves synchronous acquisition and coupled analysis of multimodal data on electrical, acoustic, temperature, and pollution levels of the surge arrester under ESP conditions, enabling accurate and quantitative assessment of the degree of surge arrester damage under ESP environments.

[0008] The objective of this invention is achieved through the following technical solution: A surge arrester damage test and analysis device under strong electromagnetic pulse includes: A high electromagnetic pulse injection module is used to apply a high electromagnetic pulse to the surge arrester; Artificial damage construction module, used to construct controllable damage on the surface of surge arrester; The multimodal sensing and detection module is used to synchronously acquire multidimensional physical quantity data related to damage to the surge arrester under the action of strong electromagnetic pulse; The data processing and analysis module is used to couple and process multidimensional physical quantity data to analyze and evaluate the damage level of surge arresters.

[0009] Preferably, the artificial damage construction module includes a contamination layer application unit and a discharge channel optimization unit. The contamination layer application unit is used to apply contamination to the surface of the surge arrester, and the discharge channel optimization unit is used to set a discharge channel in the contamination layer or to arrange conductive wires in the surge arrester.

[0010] Preferably, the multimodal sensing and detection module includes an electrical parameter sensing unit, an acoustic parameter sensing unit, a temperature parameter sensing unit, and a pollution level detection unit. The electrical parameter sensing units are located at both ends of the surge arrester and in the grounding circuit, and are used to collect the voltage and current of the surge arrester. The acoustic parameter sensing units are evenly distributed around the surge arrester and are used to collect acoustic signals. The temperature parameter sensing units are located on the surface of the surge arrester and are used to collect temperature rise change data.

[0011] Preferably, the data processing and analysis module includes a data preprocessing unit, a parameter extraction unit, a database construction unit, and a damage assessment unit. The data preprocessing unit is used to preprocess the multidimensional physical quantity data collected by the multimodal sensing and detection module. The parameter extraction unit is used to extract feature parameters from the preprocessed multidimensional physical quantity data. The database construction unit is used to construct a coupled model through the feature parameters. The damage assessment unit is used to assess the damage degree of the surge arrester through the coupled model.

[0012] Preferably, the output terminal of the strong electromagnetic pulse injection module is connected to the standard electrical connection port on the high-voltage side of the surge arrester. The strong electromagnetic pulse injection module applies a strong electromagnetic pulse to the surge arrester by directly injecting a fast-leading pulse into the standard electrical connection port on the high-voltage side of the surge arrester, and the low-voltage end of the surge arrester is grounded.

[0013] A method for analyzing surge arrester damage under strong electromagnetic pulse, comprising the following steps: Step 1: Build a surge arrester damage test and analysis device under strong electromagnetic pulse action. Connect the strong electromagnetic pulse injection module, surge arrester, multi-mode sensing and detection module and data processing and analysis module into the circuit. Ground the low voltage end of the surge arrester. Step 2: Create controllable damage on the surface of the surge arrester using the artificial damage construction module; Step 3: Apply a strong electromagnetic pulse to the surge arrester with controllable damage through the strong electromagnetic pulse injection module; Step 4: Synchronously collect multi-dimensional physical quantity data related to damage of the surge arrester under the action of strong electromagnetic pulse through the multi-modal sensing and detection module, and transmit the multi-dimensional physical quantity data to the data processing and analysis module; Step 5: The received multidimensional physical quantity data is coupled through the data processing and analysis module, and the damage degree of the surge arrester is analyzed and evaluated based on the coupling processing results.

[0014] Preferably, step 2 specifically includes: Obtain at least two identical surge arresters, one of which is used for pollution degree measurement and the other is used to construct controlled damage; apply a pollution layer to the surge arresters by dip coating or spraying, and set up pollution discharge channels for the surge arresters other than the one used for pollution degree measurement.

[0015] Preferably, in step 4, multi-modal sensing and detection modules are used to synchronously collect multi-dimensional physical quantity data related to damage to the surge arrester under strong electromagnetic pulses, specifically: The multimodal sensing and detection module collects the voltage and current of the surge arrester through the electrical parameter sensing unit, the acoustic signal through the acoustic parameter sensing unit, the temperature rise change data through the temperature parameter sensing unit, and the pollution level parameter through the pollution level detection unit. Based on the electrical signal triggering time, the collected multimodal data is synchronously calibrated, and then the multimodal data is noise-reduced.

[0016] Preferably, in step 5, the received multidimensional physical quantity data is coupled through a data processing and analysis module, and the damage level of the surge arrester is analyzed and evaluated based on the coupling processing results. Specifically: Features related to the damage level of surge arresters are extracted from the noise-reduced multimodal data. The extracted features are normalized, and then a weighted fusion strategy is used to couple the normalized features to form fused features. A CNN-LSTM dual-branch coupled model is constructed, in which the fused features and the original temporal data of each modality are respectively input into the two branches. The CNN branch is responsible for extracting the spatial correlation features of the multimodal data, and the LSTM branch is responsible for learning the dynamic temporal features of the discharge process. The weights of the two branches are adaptively allocated through the attention mechanism, and the output layer uses the Softmax function to realize the classification and evaluation of discharge state and damage degree.

[0017] Preferably, the method of collecting dirtiness parameters through a dirtiness detection unit specifically includes: The waste is placed in a container with water, and the conductivity and temperature of the wastewater in the container are measured. Based on the conductivity and temperature, the equivalent salt density of the upper and lower surfaces of the surge arrester skirt is measured. The wastewater in the container is filtered and dried, and the remaining solid material is weighed and the ash density is calculated. The equivalent salt density and ash density constitute the pollution degree parameter.

[0018] The beneficial effects of this invention are as follows: This invention integrates modules for strong electromagnetic pulse injection, artificial damage construction, multimodal sensing and detection, and data processing and analysis to form an integrated experimental analysis device. It can realize the precise application of strong electromagnetic pulses, the construction of controllable damage on the surface of surge arresters, and the synchronous acquisition and coupled analysis of damage-related multidimensional physical quantity data. It effectively solves the problems of single sensing and detection dimensions, large randomness of damage scenarios, and one-sided damage assessment in the prior art. It can comprehensively characterize the damage mechanism of surge arresters under the action of strong electromagnetic pulses and accurately assess the damage degree of surge arresters.

[0019] The artificial damage construction module of this invention, through the cooperation of the contamination layer application unit and the discharge channel optimization unit, can standardize the construction of controllable damage scenarios such as surface breakdown of surge arresters, significantly improving the probability of discharge occurrence and the accuracy of discharge channel positioning. At the same time, it avoids the randomness problem of natural damage, improves the repeatability of the test and the comparability of the test data, and facilitates the accurate analysis of the characteristics of specific damage locations.

[0020] This invention relies on a multimodal sensing and detection module to achieve comprehensive acquisition of multi-dimensional data including electrical, acoustic, temperature, and pollution levels. Through a data processing and analysis module, it completes data preprocessing, feature extraction, and coupled modeling analysis. Based on the analysis results, it can also achieve quantitative classification and assessment of the damage degree of surge arresters. At the same time, it provides data support for building a database of surge arrester damage characteristics under strong electromagnetic pulse action, and provides a scientific basis for the reliability design, operation and maintenance protection, and intelligent damage detection of surge arresters in power distribution networks. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0022] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0023] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0024] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0025] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0026] Example: A surge arrester damage test and analysis device under strong electromagnetic pulse, comprising: A high electromagnetic pulse injection module is used to apply a high electromagnetic pulse to the surge arrester; Artificial damage construction module, used to construct controllable damage on the surface of surge arrester; The multimodal sensing and detection module is used to synchronously acquire multidimensional physical quantity data related to damage to the surge arrester under the action of strong electromagnetic pulse; The data processing and analysis module is used to couple and process multidimensional physical quantity data to analyze and evaluate the damage level of surge arresters.

[0027] In this embodiment, the high electromagnetic pulse injection module is used to apply a nanosecond-level high steep transient voltage pulse that meets the equivalent characteristics of HEMP (High Altitude Nuclear Electromagnetic Pulse) to the surge arrester under test. The module includes a fast leading edge pulse injection source, a current limiting and isolation protection device, the leading edge time of the pulse is ≤10ns, the peak voltage is adjustable in the range of 0-100kV, and it supports adjustable settings of injection polarity, amplitude and grounding method. The artificial damage construction module includes a contamination layer application unit and a discharge channel optimization unit. Kaolin (insoluble dust) is applied in a controlled amount according to the surface dust density (NSDD) of 0.01-0.1 mg / cm². The discharge channel optimization unit improves the probability of discharge occurrence and positioning accuracy by retaining a 1 cm wide contamination discharge channel (achieved by wiping after spraying a single sample) or by arranging conductive wires in the air gap arrester.

[0028] Under the influence of a strong electromagnetic pulse (ESP), one of the typical damage forms of surge arresters is surface discharge of the outer insulator skirts. According to experimental data, under the excitation of an ESP with a standard IEC waveform, surface flashover of surge arrester skirts typically requires a peak voltage of over 300kV. However, in actual field operating environments, contamination inevitably accumulates on the surface of the surge arrester skirts, which significantly reduces the surface breakdown field strength and thus significantly increases the probability of surface flashover.

[0029] Based on the above-mentioned field characteristics, the discharge channel optimization unit achieves defect simulation in two ways: First, for porcelain bushing or composite bushing surge arresters, after completing the dip-coating / spraying operation of the contamination layer application unit, the sample surface is locally wiped to retain a 1cm wide contamination strip as a discharge channel, thereby quantitatively and locally constructing surface breakdown defects; Second, for air gap surge arresters, by arranging conductive wires inside the air gap, the air gap spacing is effectively shortened, the local electric field concentration effect is strengthened, thereby increasing the probability of discharge and improving the accuracy of defect location.

[0030] The multimodal sensing and detection module includes an electrical parameter sensing unit, an acoustic parameter sensing unit, a temperature parameter sensing unit, and a pollution level detection unit. The electrical parameter sensing units are located at both ends of the surge arrester and in the grounding circuit to collect the voltage and current of the surge arrester. The acoustic parameter sensing units are evenly distributed around the surge arrester to collect acoustic signals. The temperature parameter sensing units are located on the surface of the surge arrester to collect temperature rise change data.

[0031] The electrical parameter sensing unit consists of a broadband high-voltage pulse voltage probe (bandwidth ≥ 1 GHz) and a broadband current probe (measurement range 0-1000 A), which are respectively arranged at both ends of the surge arrester and the grounding circuit to collect voltage and current waveforms. The acoustic parameter sensing unit specifically consists of at least two air-coupled ultrasonic sensors (frequency range 20kHz-1MHz), evenly distributed around the surge arrester, to collect acoustic signals during the discharge process. The temperature parameter sensing unit is specifically a temperature probe with an accuracy of ±0.1℃, which is attached to key positions on the surface of the surge arrester to collect temperature rise change data.

[0032] The data processing and analysis module includes a data preprocessing unit, a parameter extraction unit, a database construction unit, and a damage assessment unit. The data preprocessing unit is used to preprocess the multi-dimensional physical quantity data collected by the multi-modal sensing and detection module. The parameter extraction unit is used to extract feature parameters from the preprocessed multi-dimensional physical quantity data. The database construction unit is used to construct a coupled model through the feature parameters. The damage assessment unit is used to assess the damage degree of the surge arrester through the coupled model.

[0033] Existing surge arrester tests under strong electromagnetic pulses suffer from a limitation in sensing dimensions: they typically only collect electrical parameters such as voltage and current, neglecting crucial physical quantities like acoustic signals, equipment temperature rise, and surface contamination during the discharge process. This makes it difficult to comprehensively characterize the arrester's damage mechanism and achieve complementary verification of multi-dimensional damage information. It's important to note that strong electromagnetic pulses are rapid pulse waves on the order of 10ns. When applied to a surge arrester, relying solely on a single electrical quantity is insufficient to determine whether the discharge path is the internal zinc oxide varistor or the surface sheds. Therefore, multi-modal data coupling processing is needed to integrate cross-dimensional information, overcoming the limitations of single-modal data and achieving accurate characterization of the discharge state and damage path.

[0034] The output terminal of the strong electromagnetic pulse injection module is connected to the standard electrical connection port on the high-voltage side of the surge arrester. The strong electromagnetic pulse injection module applies a strong electromagnetic pulse to the surge arrester by directly injecting a fast-leading pulse into the standard electrical connection port on the high-voltage side of the surge arrester, and the low-voltage side of the surge arrester is grounded.

[0035] This embodiment also provides a method for analyzing surge arrester damage under strong electromagnetic pulse, such as... Figure 1 As shown, it includes the following steps: Step 1: Build a surge arrester damage test and analysis device under strong electromagnetic pulse action. Connect the strong electromagnetic pulse injection module, surge arrester, multi-mode sensing and detection module and data processing and analysis module into the circuit. Ground the low voltage end of the surge arrester. Step 2: Create controllable damage on the surface of the surge arrester using the artificial damage construction module; Step 3: Apply a strong electromagnetic pulse to the surge arrester with controllable damage through the strong electromagnetic pulse injection module; Step 4: Synchronously collect multi-dimensional physical quantity data related to damage of the surge arrester under the action of strong electromagnetic pulse through the multi-modal sensing and detection module, and transmit the multi-dimensional physical quantity data to the data processing and analysis module; Step 5: The received multidimensional physical quantity data is coupled through the data processing and analysis module, and the damage degree of the surge arrester is analyzed and evaluated based on the coupling processing results.

[0036] Step 2 is as follows: Two identical surge arresters were acquired, one for pollution measurement and the other for constructing controlled damage; a pollution layer was applied to the surge arresters by dip coating or spraying, with one of the surge arresters having a pollution discharge path.

[0037] In step 4, multi-dimensional physical quantity data related to damage to the surge arrester under strong electromagnetic pulse are synchronously collected through a multi-modal sensing and detection module. Specifically: The multimodal sensing and detection module acquires voltage u(t) and current i(t) signals (sampling frequency 1GHz, matched with a strong electromagnetic pulse with a 10ns rise time) via a high-voltage probe and Rogowski coil; it acquires acoustic signals s(t) generated by discharge using an ultrasonic sensor (center frequency 500kHz); it obtains surface temperature rise data T(x,y,t) of the umbrella skirt using an infrared thermal imager (frame rate ≥100fps); and it acquires surface dust density (NSDD) data ρ and equivalent salt density (ESDD) data using a contamination detector. Due to differences in time scale and dimensions among the modal data, synchronization calibration is required first. Using the electrical signal trigger time as a reference, the time alignment formula is: Where t is the original time series of the k-th mode, and Δt is the time offset of this mode relative to the electrical signal. This represents the synchronized time series of the k-th modality data after time calibration. This represents the original time series acquired from the data of the k-th modality. Noise reduction is then performed: wavelet thresholding is used for electrical signals, with the threshold formula being... (σ is the noise standard deviation, N is the data length); the ultrasonic signal is filtered using an adaptive LMS filter to eliminate environmental interference; the temperature rise data is filtered using a median filter to remove pixel noise.

[0038] In step 5, the received multidimensional physical quantity data is coupled through the data processing and analysis module. Based on the coupling processing results, the damage degree of the surge arrester is analyzed and evaluated. Specifically: Core features are extracted from the preprocessed data: Electrical dimensions are used to extract features of the volt-ampere characteristic curve (breakdown voltage U, volt-ampere slope k), and overshoot peak value. ( The parameters are: rated voltage of the surge arrester, residual voltage, and response time (time for the voltage to rise from 10% peak value to 90% peak value); acoustically, the peak value A and the center frequency f of the ultrasonic signal are extracted; thermally, the temperature rise amplitude is extracted. ( The initial temperature and the rate of temperature rise, v, are used; the contamination dimension is characterized by ρ. To eliminate dimensional differences, all features are normalized: (F represents the original feature, , (This refers to the sample extreme value of this feature). A weighted fusion strategy is used to achieve feature coupling. The weights are determined by calculating mutual information entropy, and the mutual information formula is... ( Features to be fused For reference features, (For probability density function), specifically: selecting the residual voltage feature in the electrical dimension that has the strongest correlation with the degree of damage to the surge arrester as the reference feature. Based on the mutual information formula, each normalized feature to be fused is calculated separately. With reference features Mutual information values ​​between ,Will Normalization is performed to obtain the weights of each feature. The formula is The final fusion feature is ( Let be the weight of the i-th feature, and ).

[0039] Construct a CNN-LSTM dual-branch coupled model to fuse features. The original time-series data for each modality are input into two branches: the CNN branch is responsible for extracting the spatial correlation features of the multimodal data (such as the correspondence between the ultrasonic signal spectrum distribution and the temperature rise region), and its input is the weighted fusion features. The system comprises two convolutional layers and one global average pooling layer, responsible for extracting spatial correlation features from multimodal data. The LSTM branch is responsible for learning the dynamic temporal features of the discharge process (such as the temporal synchronization between voltage peaks and ultrasonic signal peaks). Its input is the original temporal data of each modality, containing two bidirectional LSTM units responsible for learning the dynamic temporal features of the discharge process. An attention mechanism adaptively allocates the weights of the two branches. The output layer uses a Softmax function to classify and evaluate the discharge state (surface skirt discharge / internal valve plate discharge / normal state) and damage degree. The loss function is the cross-entropy function. ( For real labels, The model predicts the probability, where m represents the total number of samples.

[0040] The dirtiness parameters are collected through the dirtiness detection unit, specifically: The waste is placed in a container with water, and the conductivity and temperature of the wastewater in the container are measured. Based on the conductivity and temperature, the equivalent salt density of the upper and lower surfaces of the surge arrester skirt is measured. The wastewater in the container is filtered and dried, and the remaining solid material is weighed and the ash density is calculated. The equivalent salt density and ash density constitute the pollution degree parameter.

[0041] The specific implementation method is as follows: A1. Necessary equipment for measuring dirt levels; The equipment for measuring ESDD and NSDD includes: distilled or deionized water, conductivity meter, graduated cylinder, temperature probe, medical gloves, filter paper, tape, funnel, labeled wastewater storage container, desiccator or drying oven, washing container, balance, absorbent cotton, brush and sponge. A2. Methods for collecting sludge by measuring equivalent salt density and ash density; To avoid contamination damage, the insulating surface of the surge arrester skirt should not be touched when disassembling and handling it; containers, measuring cylinders, etc. should be cleaned before sampling surface contamination to ensure they are free of any contamination; clean medical gloves should be worn when sampling.

[0042] The procedure for wiping to collect soil samples is as follows: The distilled water volume used for surge arrester skirts is 300ml. For other surge arrester skirts, the water volume can be adjusted proportionally according to their surface area. If the area increases, it is recommended to select a water volume of ≤1500cm². 2 For 300ml, >1500~2000cm 2 For 400ml, >2000~2500cm 2 For 500ml, >2500~3000cm 2 For 600ml, >3000~4000cm 2 The dosage is 600ml. When cleaning dirt from the upper and lower surfaces separately, the amount of water should be allocated appropriately according to the surface area ratio.

[0043] Pour a measured amount of distilled water into a labeled container and immerse a sponge in the water (a brush or absorbent cotton can also be used). The conductivity of the water in which the sponge is immersed should be less than 10 μS / cm.

[0044] Use a sponge to scrub away dirt from both the top and bottom surfaces of the surge arrester skirt. Return the sponge with dirt to the container and dissolve the dirt in water by shaking and squeezing. Repeat scrubbing until no dirt remains on the surge arrester skirt surface. If any dirt remains after several scrubbings, scrape it off with a scraper and place it in the wastewater.

[0045] A3. Determination of equivalent salt density and ash density; A3.1 Calculation of equivalent salt density; The conductivity and temperature of wastewater should be measured after the wastewater has been thoroughly stirred. For highly soluble pollutants, the stirring time can be shorter, such as a few minutes; for low-soluble pollutants, a longer stirring time is generally required, such as 30 to 40 minutes.

[0046] Correct the conductivity according to formula A.1: (A.1); In the formula: θ represents the solution temperature; σ θ This represents the volumetric conductivity at a temperature θ℃. σ 20 This represents the volumetric conductivity at a temperature of 20°C. b represents a factor that depends on temperature θ, which can be calculated using formula A.2.

[0047] (A.2); The equivalent salt density (ESDD) of the surge arrester skirt surface is calculated according to formulas A.3 and A.4.

[0048] (A.3); ESDD (A.4); In the formula: This represents the equivalent amount of salt per unit area. This represents the volumetric conductivity at a temperature of 20°C. ESDD stands for Equivalent Salt Density; Indicates the volume of distilled water; This indicates the surface area of ​​the insulator of the surge arrester skirt.

[0049] If the equivalent salt density of the upper and lower surfaces of the surge arrester skirt is measured separately, the average value can be calculated according to formula A.5 (which can also be used to calculate the average value of the ash density): (A.5); In the formula: This indicates the ESDD on the upper surface of the surge arrester skirt; This indicates the ESDD on the lower surface of the surge arrester skirt; This indicates the area of ​​the upper surface of the surge arrester skirt; This indicates the area of ​​the lower surface of the surge arrester skirt; This indicates the total surface area of ​​the upper and lower surfaces of the surge arrester skirt.

[0050] If at 0.001 mg / cm 2 For precise measurement of equivalent salt density within a given range, water with very low conductivity, such as less than 1 μS / cm, is recommended. Generally, distilled water (deionized water) with a conductivity less than 10 μS / cm can also be used, but the equivalent salt content of the water must be subtracted from the equivalent salt content of the wastewater. The amount of distilled water (deionized water) used depends on the type and quantity of the contaminant. For very heavy contaminants or contaminants with low solubility, a larger volume is recommended. In practice, a volume of 0.2–1 ml / cm³ is acceptable. 2 Use water for washing. If the measured conductivity is very high, it may be due to insufficient water causing the contaminants to not dissolve completely. The stirring time before measuring conductivity depends on the type of contaminant; for contaminants with low solubility, measurements can be taken in several sessions with a maximum interval of 0.5 hours, and the conductivity value is taken when the measurement stabilizes.

[0051] A3.2 Calculation of gray density: First, weigh the filter paper (1.6μm or smaller). Then, filter the wastewater with the measured equivalent salinity using a funnel filter paper (vacuum filtration can be used if the time is too long). Dry the filter paper and residue together, and finally weigh them.

[0052] The density of the ash is calculated according to formula A.6: NSDD=1000(W f -W i ) / A(A.6); In the formula: NSDD represents the density of non-soluble sediments; W f This indicates the weight of the contaminated filter paper under dry conditions; W i This indicates the weight of the filter paper itself under dry conditions; A represents the surface area of ​​the surge arrester skirt.

[0053] A4. Chemical analysis of the filth: To understand the chemical composition of waste, quantitative chemical analysis should be performed. Chemical analysis of soluble salts can be conducted using solutions obtained after ESDD measurement, employing methods such as ion exchange chromatography (IC) or inductively coupled plasma optical emission spectrometry (ICP). The analytical results may show positive ions (such as Na+). + Ca 2+ K + Mg 2+ ) and negative ions (such as Cl) - , , ).

[0054] In this embodiment, the damage grading criteria are as follows: Level 1 (no damage): No shift in volt-ampere characteristics, residual pressure change rate ≤5%, no obvious ultrasonic signal, temperature rise ≤2℃, no significant changes in ESDD and NSDD; Level 2 (Slight Deterioration): Volt-ampere characteristic deviation ≤10%, residual voltage change rate 5%-10%, ultrasonic signal intensity ≤0.3V, temperature rise 2-5℃, ESDD and NSDD increase ≤20%; Level 3 (Moderate Damage): Volt-ampere characteristic shift 10%-20%, residual voltage change rate 10%-20%, ultrasonic signal intensity 0.3-0.8V, temperature rise 5-10℃, ESDD and NSDD increase 20%-50%; Level 4 (Severe Damage / Failure): Volt-ampere characteristic deviation ≥20%, residual voltage change rate ≥20%, ultrasonic signal intensity ≥0.8V, temperature rise ≥10℃, local thermal damage or structural deterioration, ESDD and NSDD increase ≥50%.

[0055] The following are specific applications of the technical solution in this embodiment: The collected voltage and current data are preprocessed to extract the volt-ampere characteristic curves, overshoot peak values, residual voltage, and response time for each pulse amplitude. Among them, at an amplitude of 80kV, the overshoot peak value is 38.5kV, the residual voltage is 26.8kV, and the response time is 80ns. Ultrasonic signal analysis showed that the discharge location completely coincided with the preset 1cm contamination channel, and the signal strength was 0.45V; temperature data showed that the temperature rise in the middle area of ​​the surge arrester was 6.2℃.

[0056] Compared with the parameters before the test (overshoot peak value 32.1kV, residual voltage 24.5kV, response time 65ns, no ultrasonic signal, temperature rise 1.1℃), the residual voltage change rate was calculated to be 9.4%, and the voltage-current characteristic deviation was 12.3%. Combined with the 33% increase in ESDD (0.032mg / cm²) and NSDD (0.024mg / cm²), according to the damage grading standard, the surge arrester was determined to be level 3 (moderate damage) under the action of an 80kV HEMP pulse.

[0057] The pulse parameters, artificial damage parameters, multimodal raw data, extracted feature parameters, and damage levels from this experiment will be stored in the HEMP damage feature database to provide a basis for comparison in subsequent experiments.

[0058] As can be seen from the above embodiments, the experimental analysis platform of the present invention can realize the synchronous acquisition and coupled analysis of multimodal data, controllable artificial damage scenarios, and accurate quantification of damage assessment, effectively solving the shortcomings of the prior art and having important engineering application value.

[0059] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0060] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A device for testing and analyzing damage to surge arresters under strong electromagnetic pulse, characterized in that, include: A high electromagnetic pulse injection module is used to apply a high electromagnetic pulse to the surge arrester; Artificial damage construction module, used to construct controllable damage on the surface of surge arrester; The multimodal sensing and detection module is used to synchronously acquire multidimensional physical quantity data related to damage to the surge arrester under the action of strong electromagnetic pulse; The data processing and analysis module is used to couple and process multidimensional physical quantity data to analyze and evaluate the damage level of surge arresters.

2. The surge arrester damage test and analysis device under strong electromagnetic pulse action according to claim 1, characterized in that, The artificial damage construction module includes a contamination layer application unit and a discharge channel optimization unit. The contamination layer application unit is used to apply contamination to the surface of the surge arrester, and the discharge channel optimization unit is used to set a discharge channel in the contamination layer or to arrange conductive wires in the surge arrester.

3. The surge arrester damage test and analysis device under strong electromagnetic pulse action according to claim 1, characterized in that, The multimodal sensing and detection module includes an electrical parameter sensing unit, an acoustic parameter sensing unit, a temperature parameter sensing unit, and a pollution level detection unit. The electrical parameter sensing units are located at both ends of the surge arrester and in the grounding circuit to collect the voltage and current of the surge arrester. The acoustic parameter sensing units are evenly distributed around the surge arrester to collect acoustic signals. The temperature parameter sensing units are located on the surface of the surge arrester to collect temperature rise change data.

4. The surge arrester damage test and analysis device under strong electromagnetic pulse action according to claim 3, characterized in that, The data processing and analysis module includes a data preprocessing unit, a parameter extraction unit, a database construction unit, and a damage assessment unit. The data preprocessing unit is used to preprocess the multi-dimensional physical quantity data collected by the multi-modal sensing and detection module. The parameter extraction unit is used to extract feature parameters from the preprocessed multi-dimensional physical quantity data. The database construction unit is used to construct a coupled model through the feature parameters. The damage assessment unit is used to assess the damage degree of the surge arrester through the coupled model.

5. The surge arrester damage test and analysis device under strong electromagnetic pulse action according to claim 1, characterized in that, The output terminal of the strong electromagnetic pulse injection module is connected to the standard electrical connection port on the high-voltage side of the surge arrester. The strong electromagnetic pulse injection module applies a strong electromagnetic pulse to the surge arrester by directly injecting a fast-leading pulse into the standard electrical connection port on the high-voltage side of the surge arrester, and the low-voltage end of the surge arrester is grounded.

6. A method for analyzing surge arrester damage under strong electromagnetic pulse, based on the surge arrester damage analysis device under strong electromagnetic pulse as described in any one of claims 1-5, characterized in that, Includes the following steps: Step 1: Build a surge arrester damage test and analysis device under strong electromagnetic pulse action. Connect the strong electromagnetic pulse injection module, surge arrester, multi-mode sensing and detection module and data processing and analysis module into the circuit. Ground the low voltage end of the surge arrester. Step 2: Create controllable damage on the surface of the surge arrester using the artificial damage construction module; Step 3: Apply a strong electromagnetic pulse to the surge arrester with controllable damage through the strong electromagnetic pulse injection module; Step 4: Synchronously collect multi-dimensional physical quantity data related to damage of the surge arrester under the action of strong electromagnetic pulse through the multi-modal sensing and detection module, and transmit the multi-dimensional physical quantity data to the data processing and analysis module; Step 5: The received multidimensional physical quantity data is coupled through the data processing and analysis module, and the damage degree of the surge arrester is analyzed and evaluated based on the coupling processing results.

7. The method for analyzing surge arrester damage under strong electromagnetic pulse as described in claim 6, characterized in that, Step 2 specifically includes: Obtain at least two identical surge arresters, one of which is used for pollution degree measurement and the other is used to construct controlled damage; apply a pollution layer to the surge arresters by dip coating or spraying, and set up pollution discharge channels for the surge arresters other than the one used for pollution degree measurement.

8. The method for analyzing surge arrester damage under strong electromagnetic pulse according to claim 6 or 7, characterized in that, In step 4, the multi-modal sensing and detection module synchronously collects multi-dimensional physical quantity data related to damage to the surge arrester under the action of a strong electromagnetic pulse, specifically: The multimodal sensing and detection module collects the voltage and current of the surge arrester through the electrical parameter sensing unit, the acoustic signal through the acoustic parameter sensing unit, the temperature rise change data through the temperature parameter sensing unit, and the pollution level parameter through the pollution level detection unit. Based on the electrical signal triggering time, the collected multimodal data is synchronously calibrated, and then the multimodal data is noise-reduced.

9. The method for analyzing surge arrester damage under strong electromagnetic pulse as described in claim 8, characterized in that, In step 5, the received multidimensional physical quantity data is coupled through a data processing and analysis module. Based on the coupling processing results, the damage level of the surge arrester is analyzed and evaluated. Specifically: Features related to the damage level of surge arresters are extracted from the noise-reduced multimodal data. The extracted features are normalized, and then a weighted fusion strategy is used to couple the normalized features to form fused features. A CNN-LSTM dual-branch coupled model is constructed, in which the fused features and the original temporal data of each modality are respectively input into the two branches. The CNN branch is responsible for extracting the spatial correlation features of the multimodal data, and the LSTM branch is responsible for learning the dynamic temporal features of the discharge process. The two branches are adaptively weighted using an attention mechanism, and the output layer uses a Softmax function to classify and evaluate the discharge state and damage level.

10. The method for analyzing surge arrester damage under strong electromagnetic pulse according to claim 8, characterized in that, The collection of dirtiness parameters through the dirtiness detection unit specifically refers to: The waste is placed in a container with water, and the conductivity and temperature of the wastewater in the container are measured. Based on the conductivity and temperature, the equivalent salt density of the upper and lower surfaces of the surge arrester skirt is measured. The wastewater in the container is filtered and dried, and the remaining solid material is weighed and the ash density is calculated. The equivalent salt density and ash density constitute the pollution degree parameter.

Images