Capacitor compensation method and device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors

Abstract

The present application provides a capacitance compensation method and device for withstand voltage and partial discharge testing of ultra-high voltage shunt reactors, belonging to the field of reactor capacitance compensation. The method includes obtaining the inherent capacitance parameters of the winding, separating the high-frequency and low-frequency components, analyzing the spatial coupling path, calculating the temperature and humidity-induced capacitance offset, generating a collaborative capacitance parameter matrix after weighted fusion, introducing a nonlinear polarization attenuation factor and dynamically calibrating it, establishing a capacitance compensation benchmark model, and performing compensation through a multi-stage compensation device. Through dynamic modeling, the present application accurately compensates capacitance, suppresses voltage distortion, and improves the sensitivity of partial discharge detection, solving the problems of insufficient accuracy and poor adaptability of traditional compensation.

Description

Technical Field

[0001] The present application relates to the field of capacitor compensation for reactors, and in particular to a capacitor compensation method and device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors. Background Art

[0002] In the withstand voltage and partial discharge tests of UHV shunt reactors, it is crucial to accurately compensate the capacitor to suppress voltage waveform distortion and improve the sensitivity of partial discharge detection.

[0003] However, traditional methods face numerous challenges. First, UHV shunt reactor windings have complex inherent capacitance distributions, including interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and additional capacitance at the winding ends. These inherent capacitance parameters can significantly affect the stability of the voltage waveform during testing. Second, equivalent distributed capacitance also exists in the test circuit. This distributed capacitance, combined with the reactor's inherent capacitance, further exacerbates voltage waveform distortion. The polarization effect of the interlayer capacitance is particularly pronounced under high voltage gradients, easily leading to compensation deviations.

[0004] Furthermore, changes in ambient temperature and humidity can significantly affect the capacitance parameters of the test circuit. These fluctuations can cause changes in the dielectric constant and surface conductivity of the dielectric material, leading to dynamic capacitive reactance shifts in the distributed capacitance within the test circuit. Failure to accurately and promptly compensate for these capacitive reactance shifts caused by these changes in ambient temperature and humidity can severely impact the accuracy and reliability of the withstand voltage test.

[0005] Existing capacitance compensation methods often use a fixed-capacity compensation mode, which struggles to adapt to the dynamic changes in capacitance parameters caused by the nonlinear characteristics of reactors. Especially under high voltage gradients, the nonlinear polarization effect of the reactor windings and the capacitive resonant frequency vary with test conditions. Fixed-capacity compensation methods are unable to effectively track these dynamic changes, resulting in insufficient compensation accuracy.

[0006] In addition, when dealing with the spatial coupling effect of the distributed capacitance of the test circuit and the frequency domain characteristics of the partial discharge signal, traditional methods often ignore the modulation effect of the phase-capacitance coupling characteristics on the discharge signal spectrum, further reducing the accuracy and effect of compensation. Summary of the Invention

[0007] The purpose of this application is to overcome the defects in the above-mentioned prior art and provide a capacitance compensation method and device for ultra-high voltage shunt reactor withstand voltage and partial discharge tests.

[0008] This application provides a capacitance compensation method for ultra-high voltage shunt reactor withstand voltage and partial discharge tests, including:

[0009] Obtain the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit;

[0010] Based on the inherent capacitance parameter, a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component are separated;

[0011] Resolving the spatial coupling paths of the test loop;

[0012] When the ambient temperature and humidity change exceeds a preset change threshold, calculating the temperature and humidity induced capacitance offset resistance based on the spatial coupling path;

[0013] Inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component;

[0014] Generate the reactor-loop cooperative capacitance parameter matrix based on the weighted fusion results;

[0015] Introducing a reactor nonlinear polarization attenuation factor based on the capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix;

[0016] When the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, performing dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop collaborative capacitance parameter matrix;

[0017] When the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, a capacitance compensation reference model including the spatial coupling path characteristics is established, and the capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

[0018] Optionally, based on the inherent capacitance parameter, separating a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component includes:

[0019] Extracting the combined parameters of the interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and end additional capacitance of each preset segment of the reactor winding by segmented discrete sampling and multi-frequency impedance phase tracking method;

[0020] The combined parameters are separated to obtain the high-frequency capacitive coupling component in a frequency band greater than 100 kHz and the low-frequency dielectric loss fluctuation component in a frequency band less than 10 kHz.

[0021] Optionally, when the ambient temperature and humidity change exceeds a preset change threshold, calculating the temperature and humidity induced capacitance offset resistance based on the spatial coupling path includes:

[0022] Based on the three-dimensional electromagnetic field simulation topology of the test loop, the spatial coupling path of the distributed capacitance of the lead is analyzed;

[0023] The temperature and humidity induced varactor offset of the equivalent distributed capacitance is calculated according to the spatial coupling path and the ambient temperature and humidity.

[0024] Optionally, the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the temperature and humidity induced capacitance offset, and weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component is performed, including:

[0025] The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the calculation process of the temperature and humidity induced capacitance offset, and the weighted fusion is performed through frequency domain coupling coefficient weighted fusion.

[0026] Optionally, when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, dynamic baseline calibration of the polarization attenuation factor is performed to generate a calibrated reactor-loop cooperative capacitance parameter matrix, including:

[0027] Based on the capacitive coupling strength distribution characteristics of the reactor-loop cooperative capacitance parameter matrix, the nonlinear polarization attenuation factor is introduced to perform the dynamic baseline calibration.

[0028] Optionally, the capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device, including:

[0029] Based on the calibrated reactor-loop cooperative capacitance parameter matrix, continuously extracting frequency spectrum characteristic parameters of the partial discharge signal during the compensation process;

[0030] Feeding back the spectral characteristic parameters to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor. Optionally, based on the calibrated reactor-loop collaborative capacitance parameter matrix, continuously extracting the spectral characteristic parameters of the partial discharge signal during the compensation process includes:

[0031] collecting partial discharge signals of the test circuit by a broadband high-frequency sensor;

[0032] The partial discharge signal is subjected to time-frequency decomposition processing to obtain the pulse cluster energy density distribution and phase divergence factor.

[0033] Optionally, when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, a capacitance compensation reference model including the spatial coupling path characteristics is established, including:

[0034] Performing a complex domain conjugate point multiplication operation on the cross-modulation feature vector of the partial discharge signal and the voltage waveform distortion parameter collected in real time to generate a frequency domain correlation factor;

[0035] Constructing a dynamic weight allocation matrix based on the frequency domain correlation factors;

[0036] A compensation effect evaluation index is calculated according to the dynamic weight allocation matrix, and when the compensation effect evaluation index exceeds a capacitance compensation deviation threshold, the capacitance compensation reference model is established.

[0037] The present application also provides a capacitor compensation device for withstand voltage and partial discharge tests of an ultra-high voltage shunt reactor, comprising:

[0038] An acquisition module is used to obtain the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit;

[0039] A separation module, based on the inherent capacitance parameter, separates the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component;

[0040] An analysis module for analyzing a spatial coupling path of the test loop;

[0041] a calculation module, which calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path when the ambient temperature and humidity change exceeds a preset change threshold;

[0042] a fusion module, inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component;

[0043] Matrix module, which generates the reactor-loop collaborative capacitance parameter matrix based on the weighted fusion results;

[0044] An introduction module is provided for introducing a nonlinear polarization attenuation factor of the reactor based on a capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix;

[0045] a calibration module, which performs dynamic baseline calibration of the polarization attenuation factor when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, and generates a calibrated reactor-loop cooperative capacitance parameter matrix;

[0046] A model module is provided, which establishes a capacitance compensation reference model including the spatial coupling path characteristics when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold. The capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

[0047] Optionally, the separation module separates the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component based on the inherent capacitance parameter, including:

[0048] Extracting the combined parameters of the interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and end additional capacitance of each preset segment of the reactor winding by segmented discrete sampling and multi-frequency impedance phase tracking method;

[0049] The combined parameters are separated to obtain the high-frequency capacitive coupling component in a frequency band greater than 100 kHz and the low-frequency dielectric loss fluctuation component in a frequency band less than 10 kHz.

[0050] Optionally, when the ambient temperature and humidity change exceeds a preset change threshold, the calculation module calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path, including:

[0051] Based on the three-dimensional electromagnetic field simulation topology of the test loop, the spatial coupling path of the distributed capacitance of the lead is analyzed;

[0052] The temperature and humidity induced varactor offset of the equivalent distributed capacitance is calculated according to the spatial coupling path and the ambient temperature and humidity.

[0053] Optionally, the fusion module inputs the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performs weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component, including:

[0054] The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the calculation process of the temperature and humidity induced capacitance offset, and the weighted fusion is performed through frequency domain coupling coefficient weighted fusion.

[0055] Optionally, when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, the calibration module performs dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop cooperative capacitance parameter matrix, including:

[0056] Based on the capacitive coupling strength distribution characteristics of the reactor-loop cooperative capacitance parameter matrix, the nonlinear polarization attenuation factor is introduced to perform the dynamic baseline calibration.

[0057] Optionally, the capacitance compensation reference model in the model module performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device, including:

[0058] Based on the calibrated reactor-loop cooperative capacitance parameter matrix, continuously extracting frequency spectrum characteristic parameters of the partial discharge signal during the compensation process;

[0059] The spectral characteristic parameters are fed back to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor. Optionally, the model module continuously extracts the spectral characteristic parameters of the partial discharge signal during the compensation process based on the calibrated reactor-loop collaborative capacitance parameter matrix, including:

[0060] collecting partial discharge signals of the test circuit by a broadband high-frequency sensor;

[0061] The partial discharge signal is subjected to time-frequency decomposition processing to obtain the pulse cluster energy density distribution and phase divergence factor.

[0062] Optionally, when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, the model module establishes a capacitance compensation reference model including the spatial coupling path characteristics, including:

[0063] Performing a complex domain conjugate point multiplication operation on the cross-modulation feature vector of the partial discharge signal and the voltage waveform distortion parameter collected in real time to generate a frequency domain correlation factor;

[0064] Constructing a dynamic weight allocation matrix based on the frequency domain correlation factors;

[0065] A compensation effect evaluation index is calculated according to the dynamic weight allocation matrix, and when the compensation effect evaluation index exceeds a capacitance compensation deviation threshold, the capacitance compensation reference model is established.

[0066] The beneficial effects of this application are:

[0067] The present application provides a capacitance compensation method for a withstand voltage and partial discharge test of an ultra-high voltage shunt reactor, comprising: obtaining inherent capacitance parameters of a tested ultra-high voltage shunt reactor winding in a test circuit; separating a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component based on the inherent capacitance parameters; analyzing a spatial coupling path of the test circuit; calculating a temperature and humidity-induced capacitance offset based on the spatial coupling path when the ambient temperature and humidity change exceeds a preset change threshold; inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity-induced capacitance offset, performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component; and generating a reactor based on the weighted fusion result. -loop collaborative capacitance parameter matrix; introducing a nonlinear polarization attenuation factor of the reactor based on the capacitive coupling weight distribution of the reactor-loop collaborative capacitance parameter matrix; when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, performing dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop collaborative capacitance parameter matrix; when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, establishing a capacitance compensation reference model including the spatial coupling path characteristics, and the capacitance compensation reference model performs capacitance compensation operations through a multi-stage adjustable capacitance compensation device. This application uses dynamic modeling to accurately compensate for the test capacitance of the ultra-high voltage shunt reactor, effectively suppress voltage distortion, and improve the sensitivity of partial discharge detection, thereby solving the problems of insufficient accuracy and poor adaptability of traditional compensation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a schematic diagram of the capacitor compensation process of the UHV shunt reactor in this application;

[0069] Figure 2 This is a schematic diagram of the capacitor compensation device for the withstand voltage and partial discharge test of the ultra-high voltage shunt reactor in this application. DETAILED DESCRIPTION

[0070] In order to more clearly understand the above-mentioned objects, features and advantages of the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features therein can be combined with each other in the absence of conflict.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present application is not limited to the specific embodiments disclosed below.

[0072] like Figure 1 As shown, the present application discloses a capacitance compensation method for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors, comprising:

[0073] S101, obtaining the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit;

[0074] The winding is divided into 10 axial sections through segmented discrete sampling combined with multi-frequency impedance phase tracking method. The interlayer distributed capacitance (measured value 123-187pF), stray capacitance to ground (302-458pF), inter-turn coupling capacitance and end additional capacitance are measured section by section using a multi-frequency impedance analyzer with a frequency range of 1kHz-1MHz.

[0075] S102, separating a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component based on the inherent capacitance parameter;

[0076] The extracted inherent capacitance parameters are processed by wavelet transform to separate the high-frequency capacitive coupling component (equivalent capacitance value in the frequency band >100kHz, with an amplitude accounting for 18.3%) and the low-frequency dielectric loss fluctuation component (dielectric loss angle variation in the frequency band <10kHz, with a dielectric loss angle fluctuation of ±0.15°).

[0077] S103, analyzing the spatial coupling path of the test loop;

[0078] Based on the three-dimensional electromagnetic field simulation topology of the test loop, ANSYS Maxwell software was used to analyze the spatial parasitic coupling path of the lead distributed capacitance (simulation value 68pF / m) and calculate the electric field distribution under different lead spacing and directions.

[0079] S104, when the ambient temperature and humidity change exceeds a preset change threshold, calculating the temperature and humidity induced capacitance offset resistance based on the spatial coupling path;

[0080] When the ambient temperature and humidity change exceeds a threshold, the dynamic capacitive reactance offset of the equivalent distributed capacitance is calculated based on the spatial coupling path and combined with the temperature and humidity parameters (the capacitive reactance decreases by 2.7% for every 10°C increase in measured temperature). The calculation formula includes the temperature offset coefficient (a = 0.3% / °C) and the humidity correction factor (γ = 1.5% / 10% humidity).

[0081] The calculation of the dynamic capacitive reactance offset of the equivalent distributed capacitance requires a comprehensive calculation based on the temperature offset coefficient α and the humidity correction coefficient γ. When the ambient temperature and humidity change exceeds the preset threshold, the following calculation model is established based on the spatial coupling path data analyzed by the three-dimensional electromagnetic field simulation topology:

[0082] Capacitive reactance offset calculation formula:

[0083]

[0084] Where ΔZ is the total capacitive reactance offset, α is the temperature offset coefficient, ΔT is the actual temperature change, γ is the humidity correction coefficient, and ΔH is the actual humidity change.

[0085] S105, inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component;

[0086] The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are normalized and weighted fused using the frequency domain coupling coefficient. A specific example is to set the high-frequency weight coefficient to 0.7 and the low-frequency weight coefficient to 0.3.

[0087] Frequency Domain Coupling Coefficient The frequency domain coupling coefficient is a weight parameter used to coordinate the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component. Its core function is to resolve the physical characteristic differences of capacitance parameters in different frequency bands and achieve cross-band parameter fusion. In this application, the frequency domain coupling coefficient is calculated as follows:

[0088]

[0089] in, is the high frequency component weight, is the low-frequency component weight, is the correction factor for environment and nonlinearity.

[0090] S106. Generate a reactor-loop collaborative capacitance parameter matrix based on the weighted fusion result;

[0091] The weighted fusion results are subjected to matrix superposition operation with the capacitance distribution parameters of each segment of the reactor winding and the equivalent distributed capacitance of the test circuit to generate an equivalent capacitance parameter matrix (the mean value of the main diagonal elements is 243pF±5%), whose row and column elements correspond to the equivalent capacitance values ​​of different test nodes and their mutual coupling relationships.

[0092] S107, introducing a reactor nonlinear polarization attenuation factor based on the capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix;

[0093] A nonlinear polarization attenuation factor of the reactor is introduced (for example, nonlinear polarization attenuation factor β = 0.05 / min), which is calibrated through a polarization current attenuation experiment.

[0094] When constructing the reactor-loop cooperative capacitance parameter matrix, its structure reflects the electromagnetic coupling relationship between the winding's physical characteristics and the test loop. The matrix's row and column elements correspond to the winding's axial segments (e.g., 10 sections) and the test loop nodes, respectively. The main diagonal stores the measured equivalent capacitance values ​​(e.g., 243pF ± 5%), while the off-diagonal elements quantify the mutual influence between different segments / nodes. The capacitive coupling weight distribution is derived by normalizing these off-diagonal elements to create a spatial coupling strength map. The generation process is as follows: first, basic parameters such as the interlayer distributed capacitance and interturn coupling capacitance are obtained through segmented discrete sampling. This is then combined with the spatial coupling paths analyzed using 3D electromagnetic field simulation (e.g., the topological relationship of the lead distributed capacitance of 68pF / m). Finally, the off-diagonal elements of the matrix are row-vector normalized to form a weight distribution model that guides compensation priority allocation. This weight model directly determines the phase calibration focus areas and energy allocation ratios for subsequent compensation.

[0095] S108. When the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, performing dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop collaborative capacitance parameter matrix;

[0096] When the polarization attenuation factor matches the capacitive coupling weight distribution, dynamic baseline calibration is performed to reduce the capacitive coupling phase consistency error.

[0097] The physical essence of the nonlinear polarization attenuation factor (β) is to characterize the polarization loss characteristics of insulating materials under high voltage gradients. When the capacitive coupling weight distribution of the cooperative capacitance parameter matrix shows spatial inhomogeneity (i.e., the coupling strength of some sections is significantly higher than that of other areas), this factor needs to be introduced to correct the parameter drift under high electric field strength. The factor is calibrated by the polarization current decay experiment: after applying a step voltage to the test value, the polarization current decay curve is monitored and the equation is fitted. , extracting its nonlinear attenuation component β (typical value 0.05 / min). The essence of this calibration process is to establish a mapping relationship between the polarization characteristics of the material under high voltage and the capacitive parameters. Among them, I(t) in the fitting equation is the transient current intensity at time t; is the initial amplitude of the current; τ is the initial decay time constant of the current; β is the polarization decay factor; t is the time variable (unit: second).

[0098] Dynamic baseline calibration is triggered by a two-factor synergy: First, the capacitive coupling phase deviation must converge (decrease) after the factor is introduced; second, the current β value must be compatible with the spatial characteristics of the weight distribution (e.g., high β values ​​correspond to strong coupling sections). When both conditions are met, dynamic baseline calibration is performed. Using the capacitive coupling strength distribution of the matrix as a benchmark, an iterative algorithm adjusts the β value, minimizing the phase consistency error and outputting a calibrated cooperating capacitance matrix. This process essentially involves dynamically optimizing the β factor to achieve optimal matching between the spatial coupling strength distribution and the material's polarization properties. Specifically, using the capacitive coupling strength distribution of the equivalent capacitance parameter matrix as a benchmark, an initial value for the polarization attenuation factor β (e.g., β = 0.05) is set, and a convergence threshold for the phase consistency error is defined. Using gradient descent or Newton iteration, the β value is dynamically adjusted to gradually reduce the consistency error between the measured phase of the test loop and the target phase. During each iteration, the β step size is modified to account for the reactor's nonlinear characteristics (e.g., eddy current hysteresis). After each β adjustment, the matching degree between the phase divergence of the partial discharge signal and the capacitive coupling strength is measured. If the phase consistency error does not reach a threshold (e.g., 0.12 rad), the optimization process returns to continue. When the phase error meets the requirements, the optimal β value is locked, and the final collaborative capacitance matrix is ​​generated based on the capacitive coupling distribution at that time. This matrix is ​​used to guide the capacitance matching of the multi-stage compensation device.

[0099] S109. When the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, a capacitance compensation reference model including the spatial coupling path characteristics is established, and the capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

[0100] The preset phase deviation threshold is ±1.5°. When the deviation exceeds the threshold, a capacitance compensation benchmark model is established with the capacitive coupling phase consistency of the equivalent capacitance parameter matrix as a constraint condition.

[0101] The core feature of this model is to integrate the three-dimensional electromagnetic field topology characteristics of the spatial coupling path and perform precise compensation through a multi-level adjustable capacitance compensation device (such as the ABBMRS-CAP module).

[0102] Furthermore, in step S109, based on the capacitance compensation reference model and in combination with the real-time collected test voltage phase characteristics, a gradient adjustment interval of the compensation capacitance is dynamically generated, including:

[0103] Multi-scale phase fluctuation features are extracted from the real-time collected test voltage phase to separate its fundamental phase synchronization deviation and high-frequency harmonic phase distortion components. The fundamental phase synchronization deviation is matched with the capacitive coupling phase consistency constraint in the capacitor compensation benchmark model to generate the phase compensation dynamic coupling strength. The time-varying gradient step size of the compensation capacitance is determined by combining the capacitive resonant frequency identified in the high-frequency harmonic phase distortion component. Spatial coupling weights are assigned to the time-varying gradient step size using the capacitive coupling strength distribution characteristics of the test loop equivalent capacitance parameter matrix to form the gradient adjustment interval boundary. Based on the dynamic influence of the reactor's nonlinear polarization attenuation factor on the gradient adjustment interval boundary, a compensation sensitivity coefficient is introduced to adaptively scale the interval boundary to generate a gradient adjustment interval that includes phase-capacitance strong coupling correlation characteristics. The dynamic boundary parameters of the gradient adjustment interval are cross-domain correlated and mapped with the spectral characteristic parameters of the partial discharge signal to form the gradient adjustment interval output under the closed-loop feedback correction mechanism.

[0104] Specifically, the steps of generating a gradient adjustment interval including phase-capacitance strong coupling correlation characteristics include:

[0105] 1. Phase compensation dynamic coupling intensity generation:

[0106] The test voltage waveform is decomposed by FFT to separate the fundamental phase synchronization deviation and the high-frequency harmonic phase distortion components.

[0107] Compare the fundamental phase deviation Δθ with the capacitive coupling phase consistency threshold in the capacitor compensation benchmark model:

[0108] Dynamic coupling strength = k·Δθ·Wc

[0109] Where: k=0.8, Wc is the capacitive coupling weight of the current section.

[0110] For example, when Δθ=2.1° and Wc=0.6 is detected, the dynamic coupling strength = 0.8×2.1×0.6≈1.0, triggering priority compensation.

[0111] 2. Identification of capacitive resonance frequency and determination of time-varying gradient step size:

[0112] The partial discharge signal was collected using a Haefely PDS100 sensor and analyzed using Morlet wavelet time-frequency decomposition. The results showed a sudden increase in energy density (>20% of the baseline value) and a phase divergence of >30° in the 100-300kHz frequency band.

[0113] Gradient step size calculation:

[0114]

[0115] Among them, T=100ms is the basic step length; is the resonant frequency; β is the polarization attenuation factor.

[0116] For example: When , β=0.05, then Δt=100 / (150×0.05)≈13.3ms

[0117] Compensation sensitivity coefficient determination:

[0118]

[0119] Where, γ = 1.5 is the humidity correction factor; η = 0.03 / °C is the temperature attenuation coefficient; ΔH is the humidity change (%), and ΔT is the temperature change (°C).

[0120] For example: When ΔT=5℃, ΔH=15%, then .

[0121] Furthermore, the gradient adjustment interval is matched step by step by a multi-stage adjustable capacitance compensation device, and the capacitive deviation component generated by the nonlinear characteristics of the reactor in the test circuit is corrected synchronously, including:

[0122] The phase-capacitance coupling correlation characteristic parameters output from the gradient adjustment interval are converted into a dynamic impedance matching sequence for a multi-stage capacitor compensation device. The compensation stages are prioritized based on the inter-stage coupling attenuation characteristics of the dynamic impedance matching sequence. Based on the priority of the dynamic impedance matching sequence, the multi-stage adjustable capacitor compensation device is driven to switch on and off stage by stage, while simultaneously collecting the transient response spectrum of the capacitive deviation component stimulated by the nonlinear characteristics of the reactor in real time. The transient response spectrum is decomposed based on the nonlinear eddy current hysteresis effect, and the harmonic injection phase margin deviation for each compensation stage is extracted and mapped to the capacitive coupling strength distribution characteristics of the equivalent capacitance parameter matrix. The harmonic injection phase margin deviation is iteratively calculated using the eddy current-phase coupling coefficient with the dynamic boundary parameters of the gradient adjustment interval to generate a capacitance correction factor for each compensation capacitor stage. Based on the capacitance correction factor, the current switching stage of the multi-stage capacitor compensation device is reversely calibrated, and the dynamic compensation value of the reactor's nonlinear polarization attenuation factor is simultaneously superimposed to achieve synchronous correction of the capacitive deviation component of the test circuit.

[0123] The boundary scaling ratio of the adjustable gradient adjustment interval boundary is determined by the compensation sensitivity coefficient control:

[0124]

[0125] For example: original boundary [200pF, 300pF], , , then the new boundary = [200×3.1×0.99,300×3.1×0.99]≈[614pF,921pF].

[0126] Among them, the multi-stage capacitor compensation device includes ABBMRS-CAP modular capacitor group, SiemensSIVACON adjustable compensation unit, and GELV / MV capacitor compensation cabinet.

[0127] Furthermore, in step S109, the spectral characteristic parameters of the partial discharge signal are continuously extracted during the compensation process and fed back to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor: the partial discharge signal of the test circuit during the compensation process is collected by a wide-band high-frequency sensor, such as the Haefely PDS100 sensor (bandwidth 100kHz-30MHz), and the partial discharge signal is subjected to time-frequency decomposition processing, using the wavelet basis db10.

[0128] The pulse cluster energy density distribution and phase divergence factor are obtained, and a complex domain dot product operation is performed on the pulse cluster energy density distribution and the phase divergence factor to generate a cross-modulation feature vector containing the discharge intensity-phase coupling characteristics. Leveraging the capacitive coupling phase consistency constraint in the dynamic boundary parameters of the gradient adjustment interval, a multi-dimensional projection analysis is performed on the cross-modulation feature vector to extract the time-frequency aliasing interference factor between the partial discharge signal and the compensation threshold. When the time-frequency aliasing interference factor exceeds the preset capacitive adaptation threshold deviation limit, the correlation weight between the time-frequency aliasing interference factor and the capacitive coupling intensity distribution characteristic in the equivalent capacitance parameter matrix is ​​extracted, and the phase-capacitive intensity coupling correlation characteristic parameter of the gradient adjustment interval is reversely corrected based on the correlation weight. The corrected phase-capacitive intensity coupling correlation characteristic parameter is input into the frequency domain sliding mode iterator of the dynamic adaptation threshold to generate the optimized threshold interval for compensation capacitance and feed it back into the closed-loop feedback correction mechanism of the gradient adjustment interval.

[0129] Among them, the energy density Ed and phase divergence of the partial discharge pulse cluster are extracted ,include:

[0130] Calculate the interference factor:

[0131]

[0132] when (threshold), weight distribution matrix Reduce the compensation threshold of the corresponding section by 10%.

[0133] Furthermore, after step S109, the capacitor compensation effect is determined based on the correlation between the spectral characteristic parameters of the partial discharge signal and the voltage waveform distortion, and a closed-loop iterative compensation instruction is generated until the preset withstand voltage test accuracy standard is met, including:

[0134] A complex-domain conjugate dot product operation is performed between the cross-modulation feature vector of the partial discharge signal and the real-time collected voltage waveform distortion parameters (including total harmonic distortion (THD) and each subharmonic content) to generate a frequency-domain correlation factor representing the discharge-distortion coupling strength. Based on the frequency-domain correlation factor and the capacitive coupling strength distribution characteristics of the equivalent capacitance parameter matrix, a dynamic weight allocation matrix is ​​constructed to calculate the compensation effect evaluation index. When the compensation effect evaluation index exceeds a preset capacitive compensation deviation threshold, the nonlinear compensation hysteresis is extracted based on its mapping relationship with the phase-capacitance coupling correlation characteristic parameters of the gradient adjustment interval dynamic boundary parameters. This nonlinear compensation hysteresis is input into a frequency-domain sliding mode iterator and, combined with the dynamic compensation value of the reactor's nonlinear polarization attenuation factor, generates a phase-amplitude correction coefficient for the closed-loop iterative compensation instruction. Based on this phase-amplitude correction coefficient, the switching stages and capacitance of the multi-stage adjustable capacitor compensation device are adaptively and iteratively adjusted. The convergence of the voltage waveform distortion and partial discharge spectrum characteristic parameters is verified in real time until the withstand voltage test accuracy standard is met.

[0135] Spatial coupling weight assignment:

[0136] Weight distribution matrix generation

[0137]

[0138] in, is the coupling capacitance between the i-th segment and the j-th node in the cooperative capacitance matrix.

[0139] The weight is adjusted according to the temperature and humidity offset, and the weight increases by 5% for every 10% increase in humidity.

[0140] For example: the initial W10,j=0.15 in the winding end section (i=10), is corrected to 0.15×1.1=0.165 after the humidity increases by 20%.

[0141] In this application, the following steps are further included after step S109:

[0142] When the withstand voltage test is unexpectedly interrupted, the dynamic parameters of the gradient adjustment interval boundary, the switching level status of the multi-stage compensation device, and the instantaneous value of the reactor nonlinear polarization attenuation factor at the moment of test interruption are obtained as the historical benchmark data set.

[0143] The ambient temperature and humidity parameters after the interruption are collected in real time and the difference calculation is performed with the environmental parameters in the historical benchmark data set. The capacitive coupling drift caused by the environmental change is generated through the preset temperature, humidity and capacitive reactance offset correlation model. For example, the capacitive reactance offset coefficient a = 0.3% / °C for every 1°C change in temperature and the surface conductance correction coefficient γ = 1.5% for every 10% change in humidity.

[0144] The gradient adjustment interval boundaries in the historical baseline data set are vector-superimposed with the capacitive coupling drift to generate a preliminary corrected transition state interval boundary. The time accumulation effect of the polarization attenuation factor is also introduced to determine the compensation baseline attenuation based on the interruption duration.

[0145] Based on the capacitive coupling intensity distribution characteristics of the equivalent capacitance parameter matrix before interruption, the transition state interval boundary is nonlinearly scaled in combination with the compensation baseline attenuation, and the spatial weight allocation correction factor is generated using the priority state of the compensation device at the moment of interruption.

[0146] The scaled transition state interval boundary is subjected to a Hadamard product operation with the spatial weight distribution correction factor to generate the final reconstructed gradient adjustment interval boundary, which drives the multi-stage compensation device to recover to the working point before the interruption.

[0147] In this application, the following steps are further included after step S109:

[0148] The elements of the equivalent capacitance parameter matrix output by the capacitance compensation benchmark model are mapped to the initial synaptic weight values ​​of the input layer of the spiking neural network, where the row and column indices of the matrix correspond to the spatial topological connection relationship of the neurons.

[0149] Based on the dynamically generated compensation gradient adjustment interval boundary parameters, its upper and lower limits are converted into pulse emission rate thresholds. The pulse emission timing pattern is modulated by the phase-capacitance coupling correlation characteristic parameters to generate a pulse sequence with spatiotemporal characteristics.

[0150] The pulse sequence is input into the hidden layer neurons, the compensation lag is nonlinearly transformed through the membrane potential integral-leakage model, and the output layer neurons generate the pulse cluster decision code according to the synaptic weight iteration result.

[0151] The spatiotemporal patterns of pulse cluster decision encoding are analyzed, and the dynamic boundary parameters of the gradient regulation interval are assigned brain-like weights through the synaptic long-term potentiation factor to reconstruct the capacity matching strategy instruction flow.

[0152] After reconstructing the capacitance matching strategy instruction stream, it is converted into a control signal for a multi-stage adjustable capacitor compensation device, driving the dynamic adjustment of the switching levels and capacitance of the compensation capacitor. At the same time, the adjusted actual compensation parameters are fed back to the input layer of the pulse neural network to form a closed-loop adaptive optimization.

[0153] Please refer to Figure 2 As shown, the present application also provides a capacitor compensation device for ultra-high voltage shunt reactor withstand voltage and partial discharge tests, comprising:

[0154] An acquisition module 201 acquires the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit;

[0155] A separation module 202 is configured to separate a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component based on the inherent capacitance parameter;

[0156] An analysis module 203 analyzes the spatial coupling path of the test loop;

[0157] A calculation module 204 calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path when the ambient temperature and humidity change exceeds a preset change threshold;

[0158] A fusion module 205 inputs the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performs weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component;

[0159] A matrix module 206 generates a reactor-loop cooperative capacitance parameter matrix based on the weighted fusion result;

[0160] An introduction module 207 introduces a reactor nonlinear polarization attenuation factor based on a capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix;

[0161] The calibration module 208 performs a dynamic baseline calibration of the polarization attenuation factor when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, and generates a calibrated reactor-loop cooperative capacitance parameter matrix.

[0162] Model module 209, when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, establishes a capacitance compensation reference model including the spatial coupling path characteristics, and the capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

[0163] Furthermore, the separation module separates the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component based on the inherent capacitance parameter, including:

[0164] Extracting the combined parameters of the interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and end additional capacitance of each preset segment of the reactor winding by segmented discrete sampling and multi-frequency impedance phase tracking method;

[0165] The combined parameters are separated to obtain the high-frequency capacitive coupling component in a frequency band greater than 100 kHz and the low-frequency dielectric loss fluctuation component in a frequency band less than 10 kHz.

[0166] Furthermore, when the ambient temperature and humidity change exceeds a preset change threshold, the calculation module calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path, including:

[0167] Based on the three-dimensional electromagnetic field simulation topology of the test loop, the spatial coupling path of the distributed capacitance of the lead is analyzed;

[0168] The temperature and humidity induced varactor offset of the equivalent distributed capacitance is calculated according to the spatial coupling path and the ambient temperature and humidity.

[0169] Furthermore, the fusion module inputs the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performs weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component, including:

[0170] The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the calculation process of the temperature and humidity induced capacitance offset, and the weighted fusion is performed through frequency domain coupling coefficient weighted fusion.

[0171] Furthermore, when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, the calibration module performs dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop collaborative capacitance parameter matrix, including:

[0172] Based on the capacitive coupling strength distribution characteristics of the reactor-loop cooperative capacitance parameter matrix, the nonlinear polarization attenuation factor is introduced to perform the dynamic baseline calibration.

[0173] Furthermore, the capacitance compensation reference model in the model module performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device, including:

[0174] Based on the calibrated reactor-loop cooperative capacitance parameter matrix, continuously extracting frequency spectrum characteristic parameters of the partial discharge signal during the compensation process;

[0175] The spectral characteristic parameters are fed back to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor. Optionally, the model module continuously extracts the spectral characteristic parameters of the partial discharge signal during the compensation process based on the calibrated reactor-loop collaborative capacitance parameter matrix, including:

[0176] collecting partial discharge signals of the test circuit by a broadband high-frequency sensor;

[0177] The partial discharge signal is subjected to time-frequency decomposition processing to obtain the pulse cluster energy density distribution and phase divergence factor.

[0178] Furthermore, when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, the model module establishes a capacitance compensation reference model including the spatial coupling path characteristics, including:

[0179] Performing a complex domain conjugate point multiplication operation on the cross-modulation feature vector of the partial discharge signal and the voltage waveform distortion parameter collected in real time to generate a frequency domain correlation factor;

[0180] Constructing a dynamic weight allocation matrix based on the frequency domain correlation factors;

[0181] A compensation effect evaluation index is calculated according to the dynamic weight allocation matrix, and when the compensation effect evaluation index exceeds a capacitance compensation deviation threshold, the capacitance compensation reference model is established.

[0182] In the several embodiments provided in the application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical, or other forms.

[0183] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0184] In addition, all functional units in the embodiments of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the above-mentioned integrated units can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0185] Those skilled in the art will understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0186] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can be essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0187] The above description of the embodiments is intended to facilitate understanding and application of this application by those skilled in the art. It will be readily apparent to those skilled in the art that various modifications to the above embodiments can be made, and the general principles described herein can be applied to other embodiments without requiring creative effort. Therefore, this application is not limited to the above embodiments. Any improvements or modifications made to this application by those skilled in the art based on the disclosure of this application should fall within the scope of protection of this application.

Claims

1. A capacitance compensation method for ultra-high voltage shunt reactor withstand voltage and partial discharge tests, characterized in that: include: Obtain the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit; Based on the inherent capacitance parameter, a high-frequency capacitive coupling component and a low-frequency dielectric loss fluctuation component are separated; Resolving the spatial coupling paths of the test loop; When the ambient temperature and humidity change exceeds a preset change threshold, calculating the temperature and humidity induced capacitance offset resistance based on the spatial coupling path; Inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component; Generate the reactor-loop cooperative capacitance parameter matrix based on the weighted fusion results; Introducing a reactor nonlinear polarization attenuation factor based on the capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix; When the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, performing dynamic baseline calibration of the polarization attenuation factor to generate a calibrated reactor-loop collaborative capacitance parameter matrix; When the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, a capacitance compensation reference model including the spatial coupling path characteristics is established, and the capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

2. The capacitance compensation method for withstand voltage and partial discharge tests of a UHV shunt reactor according to claim 1, characterized in that: Based on the inherent capacitance parameters, the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are separated, including: Extracting the combined parameters of the interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and end additional capacitance of each preset segment of the reactor winding through segmented discrete sampling and multi-frequency impedance phase tracking method; The combined parameters are separated to obtain the high-frequency capacitive coupling component in a frequency band greater than 100 kHz and the low-frequency dielectric loss fluctuation component in a frequency band less than 10 kHz.

3. The capacitance compensation method for withstand voltage and partial discharge tests of a UHV shunt reactor according to claim 1, characterized in that: When the ambient temperature and humidity change exceeds a preset change threshold, calculating the temperature and humidity induced capacitance offset resistance based on the spatial coupling path includes: Based on the three-dimensional electromagnetic field simulation topology of the test loop, the spatial coupling path of the distributed capacitance of the lead is analyzed; The temperature and humidity induced varactor offset of the equivalent distributed capacitance is calculated according to the spatial coupling path and the ambient temperature and humidity.

4. The capacitance compensation method for withstand voltage and partial discharge tests of a UHV shunt reactor according to claim 1, characterized in that: Inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component, including: The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the calculation process of the temperature and humidity induced capacitance offset, and the weighted fusion is performed through frequency domain coupling coefficient weighted fusion.

5. The capacitance compensation method for ultra-high voltage shunt reactor withstand voltage and partial discharge tests according to claim 1, characterized in that: When the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, dynamic baseline calibration of the polarization attenuation factor is performed to generate a calibrated reactor-loop cooperative capacitance parameter matrix, including: Based on the capacitive coupling strength distribution characteristics of the reactor-loop cooperative capacitance parameter matrix, the nonlinear polarization attenuation factor is introduced to perform the dynamic baseline calibration.

6. The capacitance compensation method for ultra-high voltage shunt reactor withstand voltage and partial discharge tests according to claim 1, characterized in that: The capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device, including: Based on the calibrated reactor-loop cooperative capacitance parameter matrix, continuously extracting frequency spectrum characteristic parameters of the partial discharge signal during the compensation process; The spectrum characteristic parameters are fed back to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor.

7. The capacitance compensation method for withstand voltage and partial discharge tests of a UHV shunt reactor according to claim 6, characterized in that: Based on the calibrated reactor-loop cooperative capacitance parameter matrix, spectral characteristic parameters of the partial discharge signal are continuously extracted during the compensation process, including: collecting partial discharge signals of the test circuit by a broadband high-frequency sensor; The partial discharge signal is subjected to time-frequency decomposition processing to obtain the pulse cluster energy density distribution and phase divergence factor.

8. The capacitance compensation method for withstand voltage and partial discharge tests of a UHV shunt reactor according to claim 7, characterized in that: When the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, a capacitance compensation reference model including the spatial coupling path characteristics is established, including: Performing a complex domain conjugate point multiplication operation on the cross-modulation feature vector of the partial discharge signal and the voltage waveform distortion parameter collected in real time to generate a frequency domain correlation factor; Constructing a dynamic weight allocation matrix based on the frequency domain correlation factors; A compensation effect evaluation index is calculated according to the dynamic weight allocation matrix, and when the compensation effect evaluation index exceeds a capacitance compensation deviation threshold, the capacitance compensation reference model is established.

9. A capacitor compensation device for ultra-high voltage shunt reactor withstand voltage and partial discharge tests, characterized in that: include: An acquisition module is used to obtain the inherent capacitance parameters of the UHV shunt reactor winding under test in the test circuit; A separation module, based on the inherent capacitance parameter, separates the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component; An analysis module for analyzing a spatial coupling path of the test loop; a calculation module, which calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path when the ambient temperature and humidity change exceeds a preset change threshold; a fusion module, inputting the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performing weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component; Matrix module, which generates the reactor-loop collaborative capacitance parameter matrix based on the weighted fusion results; An introduction module is provided for introducing a nonlinear polarization attenuation factor of the reactor based on a capacitive coupling weight distribution of the reactor-loop cooperative capacitance parameter matrix; a calibration module, which performs dynamic baseline calibration of the polarization attenuation factor when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, and generates a calibrated reactor-loop cooperative capacitance parameter matrix; A model module is provided, which establishes a capacitance compensation reference model including the spatial coupling path characteristics when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold. The capacitance compensation reference model performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device.

10. A capacitor compensation device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors according to claim 9, characterized in that: The separation module separates the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component based on the inherent capacitance parameter, including: Extracting the combined parameters of the interlayer distributed capacitance, interturn coupling capacitance, stray capacitance to ground, and end additional capacitance of each preset segment of the reactor winding through segmented discrete sampling and multi-frequency impedance phase tracking method; The combined parameters are separated to obtain the high-frequency capacitive coupling component in a frequency band greater than 100 kHz and the low-frequency dielectric loss fluctuation component in a frequency band less than 10 kHz.

11. A capacitor compensation device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors according to claim 9, characterized in that: When the ambient temperature and humidity change exceeds a preset change threshold, the calculation module calculates the temperature and humidity induced capacitance offset resistance based on the spatial coupling path, including: Based on the three-dimensional electromagnetic field simulation topology of the test loop, the spatial coupling path of the distributed capacitance of the lead is analyzed; The temperature and humidity induced varactor offset of the equivalent distributed capacitance is calculated according to the spatial coupling path and the ambient temperature and humidity.

12. A capacitor compensation device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors according to claim 9, characterized in that: The fusion module inputs the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component into the temperature and humidity induced capacitance offset, and performs weighted fusion of the high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component, including: The high-frequency capacitive coupling component and the low-frequency dielectric loss fluctuation component are input into the calculation process of the temperature and humidity induced capacitance offset, and the weighted fusion is performed through frequency domain coupling coefficient weighted fusion.

13. The capacitor compensation device for withstand voltage and partial discharge tests of UHV shunt reactors according to claim 9, characterized in that: The calibration module performs dynamic baseline calibration of the polarization attenuation factor when the nonlinear polarization attenuation factor reduces the capacitive coupling phase deviation corresponding to the capacitive coupling weight distribution, and generates a calibrated reactor-loop cooperative capacitance parameter matrix, including: Based on the capacitive coupling strength distribution characteristics of the reactor-loop cooperative capacitance parameter matrix, the nonlinear polarization attenuation factor is introduced to perform the dynamic baseline calibration.

14. The capacitor compensation device for withstand voltage and partial discharge tests of UHV shunt reactors according to claim 9, characterized in that: The capacitance compensation reference model in the model module performs capacitance compensation operation through a multi-stage adjustable capacitance compensation device, including: Based on the calibrated reactor-loop cooperative capacitance parameter matrix, continuously extracting frequency spectrum characteristic parameters of the partial discharge signal during the compensation process; The spectrum characteristic parameters are fed back to the gradient adjustment interval to optimize the dynamic adaptation threshold of the compensation capacitor.

15. A capacitor compensation device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors according to claim 14, characterized in that: The model module continuously extracts the spectral characteristic parameters of the partial discharge signal during the compensation process based on the calibrated reactor-loop cooperative capacitance parameter matrix, including: collecting partial discharge signals of the test circuit by a broadband high-frequency sensor; The partial discharge signal is subjected to time-frequency decomposition processing to obtain the pulse cluster energy density distribution and phase divergence factor.

16. A capacitor compensation device for withstand voltage and partial discharge tests of ultra-high voltage shunt reactors according to claim 15, characterized in that: The model module establishes a capacitance compensation reference model including the spatial coupling path characteristics when the capacitive coupling phase deviation of the calibrated reactor-loop collaborative capacitance parameter matrix exceeds a preset phase deviation threshold, including: Performing a complex domain conjugate point multiplication operation on the cross-modulation feature vector of the partial discharge signal and the voltage waveform distortion parameter collected in real time to generate a frequency domain correlation factor; Constructing a dynamic weight allocation matrix based on the frequency domain correlation factors; A compensation effect evaluation index is calculated according to the dynamic weight allocation matrix, and when the compensation effect evaluation index exceeds a capacitance compensation deviation threshold, the capacitance compensation reference model is established.

Images