A power equipment weather disaster failure analysis method and system

Abstract

The present application relates to the technical field of fault prediction, in particular to a kind of meteorological disaster fault analysis method and system of electric power equipment, comprising the following steps: obtaining the real-time wind speed, wind direction and air pressure data acting on transmission line and insulator string, calculating the tension vector and spatial displacement of transmission line under wind load, generating the real-time swing coordinates of the line, and substituting the X-axis, Y-axis and Z-axis components of the real-time swing coordinates of the line into the spatial position analytical matrix.In the present application, the air breakdown voltage threshold under the current environment is calculated by combining real-time air pressure data and applying the Bashen law ionization formula, the grid operating phase voltage is compared with the dynamic physical threshold by difference, the multi-physical field coupling diagnosis logic is constructed by fusing meteorology, mechanics, electromagnetism and insulation physics, the real-time and accuracy of transmission line windage discharge fault diagnosis under complex environment such as strong wind are improved, and the limitation that single electrical quantity monitoring is difficult to represent environment-induced fault is overcome.

Description

Technical Field

[0001] This invention relates to the field of fault prediction technology, and in particular to a method and system for analyzing meteorological disaster faults in power equipment. Background Technology

[0002] Fault prediction technology utilizes sensor networks, condition monitoring systems, and historical operation and maintenance records to collect multi-dimensional time-series data during equipment operation. It combines signal processing technology, statistical analysis, physical modeling theory, and artificial intelligence algorithms to conduct real-time assessment of the equipment's health status and predict its evolution trend.

[0003] Current technologies for power equipment fault prediction primarily rely on collecting electrical quantities or simple environmental parameters using general-purpose sensors and on mining statistical patterns from historical big data or training black-box artificial intelligence models. When dealing with specific faults caused by meteorological disasters, meteorological data is often treated merely as an external environmental label rather than an internal variable, lacking modeling of the physical mechanisms by which environmental mechanical effects translate into changes in electrical characteristics. In dynamic and complex environments such as strong winds, static safety distance criteria or fixed electrical thresholds are often used for monitoring, ignoring the nonlinear spatial sway of conductors driven by wind loads, which leads to drastic real-time changes in phase-to-phase distance and insulation strength, thus reducing the accuracy of fault prediction. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and to propose a method and system for analyzing meteorological disaster faults in power equipment.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for analyzing meteorological disaster faults in power equipment, comprising the following steps: Real-time wind speed, wind direction and air pressure data acting on transmission conductors and insulator strings are acquired, the tension vector and spatial displacement of the transmission conductors under wind load are calculated, the real-time swing coordinates of the conductors are generated, and the X-axis, Y-axis and Z-axis components of the real-time swing coordinates of the conductors are substituted into the spatial position analysis matrix to calculate and generate the phase-to-phase geometric mean distance. Based on the phase-to-phase geometric mean distance, calculate the flux linkage change and electric field coupling strength per unit length, generate instantaneous mutual inductance and capacitance parameters, perform calculations with the power grid angular frequency, calculate the distribution value of the imaginary part in the total impedance in the time domain, and generate the line dynamic reactance value. The reactance change rate index is calculated based on the dynamic reactance value of the line. The reactance change rate index is compared with the standard sinusoidal coupling characteristic value to generate the reactance fluctuation entropy criterion. Based on the minimum spatial gap associated with the reactance fluctuation entropy criterion, the dielectric strength limit under the current environment is calculated in combination with air pressure data to generate an air breakdown voltage threshold. The voltage amplitude of the grid operating phase is compared with the air breakdown voltage threshold to generate a discharge fault diagnosis result.

[0006] Preferably, the step of obtaining the geometric mean distance between phases is as follows: Based on real-time wind speed, wind direction and air pressure data, the wind pressure surface density is calculated and air pressure correction is performed. The insulator string swing angle is extracted and the wind angle of attack is corrected. The tension vector and spatial displacement are solved. The three-dimensional coordinate system is mapped in the order from mid-span to tension point to generate the real-time swing coordinates of the conductor. Based on the real-time swing coordinates of the conductor, the X-axis, Y-axis and Z-axis components are extracted and filled into the spatial position analysis matrix. The Euclidean distance between the two phases is calculated by enumerating each phase and combining them at the same index position. The Euclidean distance between the phases is corrected according to the phase sequence to form the equivalent geometric radius between the three-phase transmission conductors. Based on the equivalent geometric radius between the three-phase transmission conductors, phase distance combinations are constructed according to the pairing relationship of the phase Euclidean distances. Logarithmic scaling is performed on each phase distance combination according to the phase sequence, and side span distortion segments are removed. The combined geometric average value of the entire span is calculated to generate the phase mutual geometric average distance.

[0007] Preferably, the steps for obtaining the instantaneous mutual inductance and capacitance parameters are as follows: Based on the phase-to-phase geometric mean distance, a relative position sequence is established according to the phase sequence and the diagonal and off-diagonal terms of the potential coefficient matrix are filled. The conductor path is divided into segments according to the length ratio and the link direction is marked. The flux contribution is accumulated segment by segment and the coupling contribution is accumulated synchronously according to the item strength to generate the flux change and electric field coupling strength. Based on the flux linkage change and electric field coupling strength, the phase current amplitude sequence and phase voltage amplitude sequence are called to perform time-division ratio calculation and the results are recorded according to the phase sequence. Jump segments are eliminated moment by moment, and the effective range is limited by the geometric mean distance between phases as the screening condition. The results of each phase are merged to generate instantaneous mutual inductance and capacitance parameters.

[0008] Preferably, the step of obtaining the dynamic reactance value of the line is as follows: Based on the instantaneous mutual inductance and capacitance parameters, the parameters are multiplied one by one with the power grid angular frequency according to the time index to form a complex impedance vector. Phasor superposition and phase sign constraint are performed, and the distribution value of the imaginary part of the total impedance in the time domain is extracted to generate the dynamic reactance value of the line.

[0009] Preferably, the step of obtaining the reactance change rate index is as follows: Based on the line dynamic reactance value, all samples are rearranged in ascending order of timestamp within the sampling period, the time interval difference between adjacent samples is calculated, and linear interpolation is performed on the time interval with gaps to generate a line dynamic reactance value sequence with uniform sampling interval, thus obtaining the reconstructed reactance sequence. Based on the reconstructed reactance sequence, the difference between adjacent sample points is calculated, a first-order difference sequence is constructed, and abnormal mutation points are corrected to form the oscillation amplitude characteristics of the electrical parameters. The reactance change rate index is calculated based on the oscillation amplitude characteristics of the electrical parameters.

[0010] Preferably, the step of obtaining the reactance fluctuation entropy criterion is as follows: Based on the reactance change rate index, a sampling sequence with uniform time step is established according to the time sampling order. Fourier transform operation is performed on the sampling sequence to extract the fundamental amplitude and harmonic amplitude in the frequency domain, and the phase and energy information of each frequency component is recorded to generate a spectral feature set. Calculate the degree of waveform distortion based on the spectral feature set; Based on the spectral feature set, harmonic components with distortion levels higher than the threshold are extracted, the frequency of amplitude occurrence is statistically analyzed, and the information entropy value is calculated. The information entropy value is compared with the set reactance stability judgment threshold, and a reactance fluctuation entropy criterion is generated based on the comparison result.

[0011] Preferably, the step of obtaining the air breakdown voltage threshold is as follows: The minimum spatial gap corresponding to the reactance fluctuation entropy criterion is queried, the collected air pressure data is called to complete the air pressure correction, the minimum spatial gap is defined as the electrode spacing parameter according to the shortest path of the conductor to the ground, the corrected air pressure data is defined as the gas pressure parameter, the dielectric strength limit is calculated and converted into the potential difference threshold, and the air breakdown voltage threshold is generated.

[0012] Preferably, the step of obtaining the discharge fault diagnosis result is as follows: Based on the air breakdown voltage threshold, the voltage amplitude sequence of the operating phase of the power grid is extracted according to the sampling time index. The relationship between the voltage amplitude of the operating phase of the power grid and the air breakdown voltage threshold is compared sample by sample. Sample points that exceed the air breakdown voltage threshold are marked and the number of samples is counted. The proportion of the number of samples that exceed the air breakdown voltage threshold to the total number of samples is calculated to form the intersection probability. The system statistically analyzes the time periods during which the voltage continuously exceeds the air breakdown voltage threshold and calculates the maximum continuous duration. It then combines the maximum continuous duration with the intersection probability to determine the interval, classifies the severity according to the discharge warning threshold, and generates a discharge fault diagnosis result.

[0013] The present invention also provides a system comprising: The data acquisition module is used to acquire real-time wind speed, wind direction and air pressure data acting on the transmission conductor and insulator string, calculate the tension vector and spatial displacement of the transmission conductor under wind load, generate the real-time swing coordinates of the conductor, and substitute the X-axis, Y-axis and Z-axis components of the real-time swing coordinates of the conductor into the spatial position analysis matrix to calculate and generate the phase-to-phase geometric mean distance. The electromagnetic parameter calculation module is used to calculate the change in magnetic flux and electric field coupling strength per unit length based on the geometric mean distance between phases, generate instantaneous mutual inductance and capacitance parameters, perform calculations with the power grid angular frequency, calculate the distribution value of the imaginary part in the total impedance in the time domain, and generate the dynamic reactance value of the line. The feature analysis module is used to calculate the reactance change rate index based on the dynamic reactance value of the line, compare the reactance change rate index with the standard sinusoidal coupling feature value, and generate a reactance fluctuation entropy criterion. The fault diagnosis module is used to calculate the dielectric strength limit under the current environment based on the minimum spatial gap associated with the reactance fluctuation entropy criterion and combined with air pressure data, generate an air breakdown voltage threshold, compare the voltage amplitude of the grid operating phase with the air breakdown voltage threshold, and generate a discharge fault diagnosis result.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, real-time wind speed, direction, and pressure data are substituted into the catenary force balance model to solve for the tension vector and spatial displacement of the transmission conductor under wind load. Real-time conductor swing coordinates are generated and mapped to a spatial position analytical matrix. The phase-to-phase geometric mean distance is calculated, achieving deep coupling between the meteorological and mechanical environment and the conductor's geometry. Based on the phase-to-phase geometric mean distance, the change in flux linkage per unit length and the electric field coupling strength are derived, generating instantaneous mutual inductance and capacitance parameters and synthesizing the line's dynamic reactance value, revealing the dynamic modulation law of conductor mechanical motion on electrical parameters. The rate of change and waveform nonlinear distortion analysis are performed on the line's dynamic reactance value. Entropy statistical calculations are used to quantify the uncertainty of reactance fluctuations, constructing a reactance fluctuation entropy criterion to effectively capture weak fault precursor signals. Based on the minimum spatial gap associated with the fluctuation entropy criterion, combined with real-time air pressure data and applying Paschen's law ionization formula, the air breakdown voltage threshold under the current environment is calculated. The difference between the operating phase voltage of the power grid and this dynamic physical threshold is compared to construct a multi-physical field coupled diagnostic logic that integrates meteorology, mechanics, electromagnetics and insulation physics. This improves the real-time performance and accuracy of wind-induced discharge fault diagnosis of transmission lines under complex environments such as strong winds, and overcomes the limitation of single electrical quantity monitoring in characterizing environment-induced faults. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] Please see Figure 1 This invention provides a technical solution, a method for analyzing meteorological disaster faults in power equipment, comprising the following steps: The system acquires real-time wind speed, wind direction, and air pressure data acting on transmission conductors and insulator strings, calculates the tension vector and spatial displacement of transmission conductors under wind load, generates real-time oscillation coordinates of conductors, substitutes the X-axis, Y-axis, and Z-axis components of the real-time oscillation coordinates of conductors into the spatial position analysis matrix, and calculates and generates the phase-to-phase geometric mean distance. The flux linkage change and electric field coupling strength per unit length are calculated based on the phase-to-phase geometric mean distance. Instantaneous mutual inductance and capacitance parameters are generated. The instantaneous mutual inductance and capacitance parameters are then calculated with the power grid angular frequency to calculate the distribution of the imaginary part of the total impedance in the time domain, thereby generating the dynamic reactance value of the line. The reactance change rate index is calculated based on the dynamic reactance value of the line. The reactance change rate index is compared with the standard sinusoidal coupling characteristic value to generate the reactance fluctuation entropy criterion. Based on the minimum spatial gap associated with the reactance fluctuation entropy criterion, the dielectric strength limit under the current environment is calculated in combination with air pressure data to generate the air breakdown voltage threshold. The voltage amplitude of the grid operating phase is compared with the air breakdown voltage threshold to generate the discharge fault diagnosis result.

[0018] The steps to obtain the geometric mean distance between phases are as follows: Based on real-time wind speed, wind direction and air pressure data, the wind pressure surface density is calculated and air pressure correction is performed. The insulator string swing angle is extracted and the wind angle of attack is corrected. The tension vector and spatial displacement are solved. The three-dimensional coordinate system is mapped in the order from mid-span to tension point to generate the real-time swing coordinates of the conductor. Based on the real-time swing coordinates of the conductor, the X-axis, Y-axis and Z-axis components are extracted and filled into the spatial position analysis matrix. The Euclidean distance between the two phases is calculated by enumerating each phase combination at the same index position. The span scale is corrected for the Euclidean distance between the phases according to the phase sequence to form the equivalent geometric radius between the three-phase transmission conductors. Based on the equivalent geometric radius between the three-phase transmission conductors, phase distance combinations are constructed according to the pairing relationship of the phase Euclidean distances. Logarithmic scaling is performed on each phase distance combination according to the phase sequence, and the side span distortion segments are removed. The combined geometric mean of the entire span is calculated to generate the phase mutual geometric mean distance.

[0019] Specifically, based on real-time wind speed, wind direction, and air pressure data, the standard atmospheric pressure reference value is set to 101.325 kPa. The current ambient air pressure value is read and its ratio to the standard atmospheric pressure reference value is calculated to obtain the air pressure correction coefficient. The real-time collected wind speed value is squared and divided by 1600 to obtain the standard wind pressure value. The standard wind pressure value is multiplied by the air pressure correction coefficient to obtain the corrected wind pressure surface density. The transmission line design database is called to read the conductor diameter and unit length mass parameters of the current line segment. The corrected wind pressure surface density is multiplied by the conductor diameter to calculate the horizontal wind load per unit length. The vertical gravity load per unit length is calculated. The horizontal wind load and gravity load are vector-synthesized using the Pythagorean theorem to obtain the comprehensive specific load of the conductor. Simultaneously, the real-time tilt angles of the insulator strings in the lateral and longitudinal directions are obtained using dual-axis tilt sensors installed at the ends of the insulator strings. The wind direction angle is subtracted from the azimuth angle of the line alignment, and further subtracted from the lateral tilt angle of the insulator strings to obtain the corrected angle of attack. The square of the sine of the angle of attack is calculated as a wind load reduction factor, updating the conductor's force balance equation. The catenary theory model is used to solve for the tension vector and spatial displacement of the conductor under wind deflection conditions. The calculation formula is as follows: ,in, This is the horizontal displacement of the midpoint of the conductor span. The horizontal tension component of the conductor. The corrected conductor composite load ratio, This refers to the conductor span length. To represent a hyperbolic cosine function, a three-dimensional coordinate system is established with the tension tower anchorage point as the origin, the track alignment as the X-axis, the vertical ground direction as the Z-axis, and the vertical track direction as the Y-axis. The span length is then... Divide the data into equal parts in 1-meter increments to generate a discrete node sequence from mid-span to tension point. Then, superimpose the calculated horizontal displacement and vertical sag change onto the static coordinates of each discrete node to generate the real-time oscillation coordinates of the conductor.

[0020] Based on the real-time oscillation coordinates of the conductor, a three-dimensional array is initialized as a spatial position analysis matrix. The number of rows in the array corresponds to the number of discrete nodes of the conductor, the number of columns corresponds to phases A, B, and C, and the depth corresponds to the X, Y, and Z coordinate components. The three-dimensional coordinate values ​​of each node generated in the previous steps are filled into the corresponding positions in the matrix. The loop traversal program is started, and for each discrete node index, the phase A coordinates are extracted. Phase coordinates of B The Euclidean distance between two phases is calculated using the principles of spatial geometry. The formula is as follows: ,in, This represents the straight-line distance between phase A and phase B at this node. Let A be the coordinate components of the conductor at the current node. Given the coordinate components of phase B conductor at the current node, calculate the distance between phase B and phase C using the same logic. and the distance between phase C and phase A Considering the differences in electric field distribution characteristics of transmission conductors at different locations within the span, a span-scale correction coefficient is introduced. This coefficient is dynamically assigned based on the radius of curvature of the conductor's catenary. For example, the correction coefficient at the mid-span is set to 1.0, and as it extends towards the two end suspension points, the coefficient decreases parabolically to 0.95. The specific decrease function is obtained by fitting the electric field non-uniformity curve in the insulation coordination design specification. The calculated Euclidean distance between each phase is multiplied by the corresponding span-scale correction coefficient to obtain the corrected phase-to-phase distance. , and The geometric mean algorithm is used to fuse the three corrected interphase distances on the same node section. The calculation formula is as follows: ,in, Let be the equivalent geometric radius of the three-phase transmission conductor at this node. , , These represent the corrected distances between phase A and phase B, phase B and phase C, and phase C and phase A, respectively. The cube root operator stores the calculation results in the result sequence according to the node order, forming the equivalent geometric radius between the three-phase transmission conductors.

[0021] Based on the equivalent geometric radius between the three-phase transmission conductors, a numerical sequence of radii is constructed across the entire span. The sequence length is determined by the total number of discrete nodes. To eliminate the end-effect influence of fittings and insulator strings near the tower on the electric field distribution, a side-span distortion removal threshold is set. This threshold is based on the effective shielding radius of the tower's equipotential ring. For example, the area within 5% of the span length from each end suspension point is defined as the distortion segment. The number of sample points to be removed is then calculated. ,in, To exclude points from one end, This is the floor function. The total number of sampling points across the entire range is calculated by including the first and last points of the radius numerical sequence. Data points are marked as invalid and discarded, retaining the data from the middle 90% region as the valid sample set. A natural logarithmic transformation is performed on each equivalent geometric radius value in the valid sample set, converting the product-like geometric parameters into additive parameters in the logarithmic domain. The average of the transformed logarithmic sequence is calculated, and then an inverse exponential transformation is performed on this average to calculate the combined geometric mean across the entire range. The calculation formula is as follows: ,in, To obtain the final interphase geometric mean distance, For the natural constant An exponential function with base 0. This represents the total number of valid samples after removing distorted segments. It is the natural logarithm function. For the first in the valid sample set The equivalent geometric radius of each point is used to smooth out the random fluctuations in local distance caused by wind sway through the integral averaging process, generating the geometric mean distance between phases.

[0022] The steps for obtaining instantaneous mutual inductance and capacitance parameters are as follows: Based on the phase-to-phase geometric mean distance, a relative position sequence is established according to the phase sequence and the diagonal and off-diagonal terms of the potential coefficient matrix are filled. The conductor path is divided into segments according to the length ratio and the link direction is marked. The flux contribution is accumulated segment by segment and the coupling contribution is accumulated synchronously according to the item strength to generate the flux change and electric field coupling strength. Based on the change in flux linkage and the electric field coupling strength, the phase current amplitude sequence and phase voltage amplitude sequence are called to perform time-division ratio calculation and the results are recorded according to the phase sequence. Jump segments are eliminated moment by moment, and the effective range is limited by the geometric mean distance between phases. The results of each phase are merged to generate instantaneous mutual inductance and capacitance parameters.

[0023] Specifically, based on the phase-to-phase geometric mean distance, the structural parameters of the transmission line are extracted, and the vacuum permittivity is set as... and vacuum permeability , build dimensional potential coefficient matrix The self-potential coefficient and mutual potential coefficient in the matrix are calculated using the method of images. For any element in the matrix... The calculation formula is: ,in, For the first Phase conductor and the first The potential coefficient between phase conductors, For the first Phase conductor and the first The straight-line spatial distance between phase conductors (the value here is taken as the geometric mean distance between phases) ), For the first Phase conductor and the first The distance between the mirror images of phase conductors needs to be calculated in conjunction with the average height of the conductors above the ground. The calculation expression is: The calculated coefficients are then filled into the diagonal and off-diagonal positions of the potential coefficient matrix. A path integral model is then established along the conductor's direction, representing the entire length... According to the preset step size Divided into Each infinitesimal element is labeled with a current flow vector. For each infinitesimal element, the magnetic flux generated is calculated using the integral form of the Biot-Savart law, and then accumulated over the entire path. The calculation formula is as follows: ,in, The total flux linkage contribution after accumulation. For unit reference current, Let be the equivalent geometric radius of the conductor. For fragment index, The total number of segments is determined synchronously based on the potential coefficient matrix. The electric field coupling strength is calculated by solving the capacitance coefficient matrix through matrix inverse operations. The elements in the capacitance coefficient matrix are used as weighting factors to sum the unit potential differences distributed along the line, generating the flux linkage change and electric field coupling strength.

[0024] Based on the flux linkage change and electric field coupling strength, real-time data collected by a synchronous phasor measurement unit (PMU) is used to obtain a sampling frequency of [missing value]. Phase current amplitude sequence Phase voltage amplitude sequence Perform a ratio calculation on the data at each sampling time using the formula. Calculate the instantaneous mutual inductance parameters, where for The instantaneous mutual inductance value at a given moment. For the change in magnetic flux at the corresponding time, similarly, the formula is used. Calculate the instantaneous capacitance parameters, where To calculate the charge based on the electric field coupling strength, the calculated parameters are stored in a buffer queue in chronological order. The data cleaning process is then initiated, and a first-order difference operation is performed on the instantaneous mutual inductance sequence to obtain the difference sequence. Calculate the average value of the difference sequence. and standard deviation Set the transition detection threshold as Check the difference values ​​item by item, if If the data at that moment is deemed to have an abnormal jump and is removed, it is filled by linear interpolation of adjacent valid points. Then, physical constraints are introduced for verification, using the formula... Inversely calculate the conductor spacing, where To calculate the spacing, Let the radius be the conductor. For the vacuum permeability, the effective range is set as follows: ,in Given the interphase geometric mean distance obtained from the previous steps, if If the value falls outside this range, the corresponding mutual inductance and capacitance parameters are marked as invalid and set to zero. Finally, the effective parameters of the three phases are averaged and merged to generate instantaneous mutual inductance and capacitance parameters.

[0025] The steps for obtaining the dynamic reactance value of the line are as follows: Based on the instantaneous mutual inductance and capacitance parameters, the parameters are multiplied one by one with the power grid angular frequency according to the time index to form a complex impedance vector. Phasor superposition and phase sign constraint are performed, and the distribution value of the imaginary part of the total impedance in the time domain is extracted to generate the dynamic reactance value of the line.

[0026] Specifically, based on the instantaneous mutual inductance and capacitance parameters, the standard power frequency of the power grid is read. Calculate the power frequency angular frequency In time To iterate through all sampling points using an index, a complex impedance model incorporating resistance, inductance, and capacitance effects is constructed. The calculation formula is as follows: ,in, for The complex impedance vector at time t. The AC resistance constant of the conductor (with a value of...) ), The imaginary unit, These are instantaneous mutual inductance parameters. For instantaneous capacitance parameters, perform phasor superposition and phase sign constraints, and check the sign of the imaginary part of the complex impedance. This indicates an abnormal state dominated by capacitive forces, and it needs to be corrected to a positive value closer to the current time or set to a very small positive number. Extract the imaginary part of the corrected complex impedance vector; the calculation formula is as follows: ,in This is the dynamic reactance value of the line at that moment, calculated from all moments. The waveforms are rearranged according to the time axis to form a continuous time sequence, thereby generating the dynamic reactance value of the line.

[0027] The steps for obtaining the reactance change rate index are as follows: Based on the line dynamic reactance value, all samples are rearranged in ascending order of timestamp within the sampling period, the time interval difference between adjacent samples is calculated, and linear interpolation is performed on the time interval with gaps to generate a line dynamic reactance value sequence with uniform sampling interval, thus obtaining the reconstructed reactance sequence. Based on the reconstructed reactance sequence, the difference between adjacent sample points is calculated, a first-order difference sequence is constructed, and abnormal mutation points are corrected to form the oscillation amplitude characteristics of the electrical parameters. Based on the oscillation amplitude characteristics of the electrical parameters, the reactance change rate index is calculated using the following formula: ; in, The reactance change rate index For the first time in the sampling period The dynamic reactance value of each line, For the first The dynamic reactance value of each line, For the first The dynamic reactance value of each line, The time interval between adjacent sampling points. This represents the total number of sampling points within the sampling period. It is a dynamic sensitivity weighting factor used to characterize the coupling ratio between the first-order difference term and the second-order difference term.

[0028] Specifically, based on the dynamic reactance value of the line, the timestamp tag and corresponding reactance value of each sample point are read. All sample points are placed in the buffer queue to be processed, and a sorting operation based on the timestamp value is performed. For example, the quicksort algorithm is used to rearrange the samples in chronological order to ensure the monotonically increasing characteristic of the data in time. The sorted sample sequence is traversed, and the time difference between two adjacent sample points is calculated one by one. A standard sampling interval benchmark is set, which is based on the reciprocal of the power grid frequency cycle. For example, for a 50Hz power grid frequency, the standard sampling interval is set to 0.02 seconds. Each calculated time difference is compared with this standard sampling interval. If the time difference between two adjacent samples is found to be greater than 1.5 times the standard sampling interval, it is determined that there is a data gap in that time period, and the reactance value at the start of the gap is extracted. Reactance value at the end time and the corresponding timestamp and The interpolation step size is calculated based on the number of missing sampling points, and the estimated value at the missing time step is calculated using the principle of linear interpolation. The calculation logic is as follows: for the missing time step... Its reactance value is equal to Add the slope multiplied by the time increment, i.e. The calculated interpolated data is inserted into the original sequence according to the corresponding time position to fill all detected time breaks. For redundant data points with a time difference less than 0.5 times the standard sampling interval, average merging is performed. Finally, all processed data points are reorganized according to a fixed standard sampling interval to form a discrete data string with a continuous time axis and consistent step size, thus obtaining the reconstructed reactance sequence.

[0029] Based on the reconstructed reactance sequence, a first-order difference array matching the length of the original sequence is constructed. Starting from the second data point of the sequence, the array is traversed to extract the reconstructed reactance value at the current time step. Reconstruction reactance value compared to the previous moment The difference value is obtained by performing subtraction. The difference value is stored in the corresponding index position of the difference group. After completing the full sequence traversal, the original first-order difference sequence is obtained. In order to eliminate non-physical abrupt changes caused by measurement noise or instantaneous interference, statistical analysis of the difference sequence is required to calculate the average value of all values ​​in the difference sequence. and standard deviation Based on statistical principles, an abnormal mutation judgment interval is set. The upper limit of this interval is set as the mean plus three standard deviations, and the lower limit is set as the mean minus three standard deviations. Each value in the difference array is checked item by item. If a value is found to be outside the set range, it is marked as an abnormal mutation point. The abnormal point is corrected by a local smoothing strategy. For example, the average of the two normal difference values ​​before and after the abnormal point is taken to replace the abnormal value, or the previous valid difference value is directly used to replace it. The corrected difference value is updated back into the difference array. The fluctuation component reflecting the dynamic characteristics of the line is retained while high-frequency noise is removed. The final processed difference data is combined into basic data to describe the degree of change of electrical parameters in the time domain, forming the oscillation amplitude characteristics of electrical parameters.

[0030] In the formula for calculating the reactance change rate index, the first term quantifies the rate of change of the reactance value, and the second term quantifies the increment of the rate of change of the reactance value (i.e., the difference between adjacent rates). The two are then weighted and fused under the same dimension to construct a comprehensive index that reflects the dynamic instability of the line. The steps for obtaining the parameters are as follows: directly read the total number of data points in the reconstructed reactance sequence within the sampling period. This value is determined by dividing the total sampling duration by the standard sampling interval. For example, within a 100-millisecond microscopic analysis window, resampling is performed at a frequency of 50Hz, meaning one point is collected every 20 milliseconds. At this time, the sequence contains 5 data points. This value is obtained through the array length counting function. In this example, it is set as follows: ; , , The parameters are obtained by directly extracting them from the reconstructed reactance sequence generated in the previous step according to the time index, which respectively represent the first time step. The, the The and the first The dynamic reactance of the line at each moment, in ohms. These values ​​reflect the imaginary impedance characteristics of the line at each moment during wind-induced swaying. In this example, a representative set of fluctuation data is selected, and the following settings are made: , , , , ; The parameter acquisition steps are as follows: directly read the uniform sampling time interval set in the reconstructed reactance sequence generation step, in seconds. This parameter determines the time base for the differential operation; in this example, it is set according to the power frequency cycle. ; The steps for obtaining the parameter are as follows: This parameter serves as a dynamic sensitivity weighting factor, used to balance the proportions of the first-order rate of change (velocity) and the second-order increment of change (velocity difference) in the index. It is dimensionless. The setting method is as follows: collect historical reactive power fluctuation data under strong wind conditions, and calculate the squared mean of the first-order difference term. The square mean of the second difference term To ensure that both contribute equally to the sensitivity of the indicators, the following settings are made: In this example, based on historical data statistics, the following settings are defined: ; Calculations based on parameters: According to the settings, The range of summation is from arrive .

[0031] when When (using) ): First-order term (rate): ; Second-order term (rate difference): ; Calculate the unit value: ; when When (using) ): First-order term (rate): ; Second-order term (rate difference): ; Calculate the unit value: ; when When (using) ): First-order term (rate): ; Second-order term (rate difference): ; Calculate the unit value: ; Sum of total values: ; Calculate the mean: ; Square root operation: ; The results indicate that within the current monitoring window, the comprehensive fluctuation index of line reactance is 19.89. This value objectively reflects the instability of the electrical parameters of the conductor under wind load. The larger the value, the more drastic the rate of reactance change and the degree of abrupt change in the rate. When this index exceeds the preset safety benchmark, it indicates that there may be a high-risk galloping or discharge hazard.

[0032] The steps for obtaining the reactance fluctuation entropy criterion are as follows: Based on the reactance change rate index, a sampling sequence with uniform time step is established according to the time sampling order. Fourier transform operation is performed on the sampling sequence to extract the fundamental amplitude and harmonic amplitude in the frequency domain, and the phase and energy information of each frequency component is recorded to generate a spectral feature set. Based on the spectral feature set, the degree of waveform distortion is calculated using the following formula: ; in, The degree of distortion is used to characterize the overall degree of distortion in a waveform. The fundamental amplitude is the value within the spectral feature set. For the first in the spectral feature set One harmonic amplitude, The highest harmonic order used in the calculation. It is the natural logarithm function. The normalized energy probability after stabilization correction is calculated as follows: , The stabilization factor is a very small positive number. Based on the spectral feature set, harmonic components with distortion levels higher than the threshold are extracted, the frequency of amplitude occurrence is statistically analyzed, and the information entropy value is calculated. The information entropy value is compared with the set reactance stability judgment threshold, and a reactance fluctuation entropy criterion is generated based on the comparison result.

[0033] Specifically, based on the reactance change rate index, a pre-allocated circular buffer in memory is invoked, and the continuously calculated reactance change rate values ​​are sequentially filled in according to their generation time, with the sampling frequency set to [value missing]. That is, 100 data points are collected per second. When the amount of data in the buffer reaches the preset time window length, for example, the window length is set to 100. Seconds, including The system reads all data sequences within a given time window using sampling points. To suppress spectral leakage caused by signal truncation, the original data sequences are windowed using the Hanning window as the window function. Each point in the data sequence is multiplied by the corresponding window function coefficient to obtain a weighted time sequence. The Fast Fourier Transform (FFT) algorithm library is then called to perform a transform operation on the weighted sequence, converting the time-domain signal into a frequency-domain complex sequence. The magnitude of each frequency point in the complex sequence is calculated, which is the square root of the sum of the squares of the real and imaginary parts, resulting in the amplitude spectrum. The amplitude spectrum is then iterated to find the main peak position of the energy concentration, which is defined as the fundamental frequency component, and the corresponding amplitude is the fundamental amplitude. Simultaneously, based on integer multiples of the fundamental frequency, the amplitude data of the second harmonic, third harmonic, and so on up to the higher harmonics are extracted sequentially. The phase angle information of each frequency component is calculated using the inverse tangent function, and the amplitude is squared to obtain the corresponding energy value. The extracted fundamental frequency, fundamental amplitude, harmonic amplitude, phase, and energy data are structured and encapsulated according to the frequency index to generate a spectral feature set.

[0034] In the formula for calculating the degree of distortion, the concept of information entropy is introduced to correct the traditional method of calculating the total harmonic distortion rate. It not only considers the energy proportion of the harmonic amplitude, but also reflects the degree of dispersion of the harmonic energy in the spectrum distribution through the weighting of the entropy term, thus capturing the complex distortion characteristics of the waveform more sensitively. The steps for obtaining the parameter are as follows: This parameter represents the highest harmonic order involved in the calculation and is used to limit the effective bandwidth range of the frequency domain analysis. Its value setting needs to be determined based on the sampling theorem and the spectral characteristics of the actual line oscillation. It is usually taken as 5 to 10 times the fundamental frequency to cover the main nonlinear distortion components. In this example, considering that the main energy of reactance fluctuation is concentrated in the low frequency band, the value is set to... That is, only the 5th harmonic is calculated; The parameter acquisition steps are as follows: This parameter represents the fundamental amplitude in the spectral feature set, and is the amplitude data corresponding to the main oscillation frequency directly extracted from the spectral feature set generated in the previous steps. Its dimensions are... The physical meaning lies in characterizing the dominant energy intensity of the dynamic changes of the line. In this example, the fundamental amplitude is obtained by reading the spectrum analysis results. ; The steps for obtaining the parameter are as follows: this parameter represents the first element in the spectral feature set. The amplitude of the second harmonic, of which The value range is from 2 to These values ​​are also directly derived from the FFT-transformed spectral data, corresponding to amplitudes at 2 times, 3 times, etc., of the fundamental frequency, with the same dimensions. This reflects the nonlinear disturbance component present in the waveform. In this example, the amplitudes of each harmonic are read sequentially: Second harmonic Third harmonic Fourth harmonic Fifth harmonic ; The steps to obtain the parameter are as follows: This parameter is a stabilization factor, which is a very small positive number. Set... ; The steps for obtaining the parameter are as follows: the parameter is a normalized energy probability after stabilization correction, used to characterize the first... The relative proportion of subharmonic energy in the total energy of all higher harmonics depends on the amplitude of each harmonic. and stabilizing factors The calculation formula is: ; Calculations based on parameters: First, calculate the sum of squares of the amplitudes of each harmonic (ignore the rest). (Minor effect on the summation value) ; Calculate the first part (distortion rate term, result is dimensionless): ; Calculate the normalized energy probability of each harmonic. (denominator) ): ; ; ; ; Calculate the entropy term : ; ; ; ; Summation: ; Calculate the second part (entropy weights): ; Calculate the total distortion : ; This result indicates that the overall distortion level of the current waveform is... This value combines the proportion of harmonic energy with the degree of disorder in its distribution. The higher the value, the more severe the nonlinear fluctuations of the line reactance and the more complex the frequency components, suggesting the possible presence of unstable airflow disturbances or abnormal structural parameters.

[0035] Based on the spectral feature set, a significant threshold for harmonic distortion is set. This threshold is calculated according to the provisions of the national power quality standards for power systems regarding harmonic voltage limits for public power grids, for example, set as the fundamental amplitude. It iterates through all harmonic components in the spectral feature set, comparing the amplitude of each harmonic with the fundamental amplitude. If the amplitude of a certain harmonic exceeds a certain threshold... If it is significantly distorted, it is marked as a significant distortion component, and a sliding time window is established to statistically analyze the past... Within each sampling period, the number of times each harmonic is marked as a significant distortion component is calculated. The frequency probability distribution of each harmonic is calculated, and the information entropy value of this distribution is calculated based on the Shannon information entropy principle. The calculation formula is expressed as the negative summation of probability multiplied by the logarithm of the probability, resulting in the entropy value reflecting the dispersion of the harmonic distribution. A threshold for reactance stability is set, which is based on the statistical characteristics in the historical transmission line galloping accident database. By analyzing the entropy level before the galloping event, a critical value is taken. Multiples can be used as a safety boundary, for example, setting a threshold of 100%. The calculated real-time information entropy value is compared with the threshold. If the real-time entropy value is lower than the threshold, it indicates that the energy is concentrated on a few specific harmonic frequencies, which may indicate that regular low-frequency dancing is forming. If it is higher than the threshold, it reflects random vibration. Based on this logic, an electro-reactive wave entropy criterion with "stable", "warning", or "dangerous" state labels is generated.

[0036] The steps for obtaining the air breakdown voltage threshold are as follows: The minimum spatial gap corresponding to the reactance fluctuation entropy criterion is queried, the collected air pressure data is called to complete the air pressure correction, the minimum spatial gap is defined as the electrode spacing parameter according to the shortest path of the conductor to the ground, the corrected air pressure data is defined as the gas pressure parameter, the dielectric strength limit is calculated and converted into the potential difference threshold, and the air breakdown voltage threshold is generated.

[0037] Specifically, the minimum spatial gap corresponding to the reactance fluctuation entropy criterion is queried, the minimum Euclidean distance data generated during the calculation of the phase-to-phase geometric mean distance is retrieved, and this data is directly assigned as the electrode spacing parameter. For example, if the obtained value is 2.5 meters, it will simultaneously retrieve real-time collected air pressure data. Set standard atmospheric pressure The gas pressure is corrected for temperature using the gas law, and the correction formula is as follows: ,in The ambient temperature (in degrees Celsius) is used as the reference, and the corrected result is the gas pressure parameter. The breakdown field strength of the air medium is calculated according to Paschen's law. Considering the non-uniform electric field characteristics of the actual environment of the transmission line, a field strength non-uniformity coefficient is introduced. This coefficient is set based on the surface roughness of the conductor and the structure of the split conductor; for example, a value of 0.85 is used to calculate the dielectric strength limit. ,in Units are Multiply the calculated field strength limit by the electrode spacing parameter, i.e. The units are uniformly converted to kilovolts (kV). This value represents the highest voltage that the air insulation layer can withstand under the current weather and gap conditions, generating the air breakdown voltage threshold.

[0038] The steps for obtaining discharge fault diagnosis results are as follows: Based on the air breakdown voltage threshold, the voltage amplitude sequence of the operating phase of the power grid is extracted according to the sampling time index. The relationship between the voltage amplitude of the operating phase of the power grid and the air breakdown voltage threshold is compared sample by sample. Sample points that exceed the air breakdown voltage threshold are marked and the number of samples is counted. The proportion of the number of samples that exceed the air breakdown voltage threshold to the total number of samples is calculated to form the intersection probability. The system statistically analyzes the time periods during which the voltage continuously exceeds the air breakdown voltage threshold and calculates the maximum continuous duration. It then combines the maximum continuous duration with the intersection probability to determine the interval, classifies the severity according to the discharge warning threshold, and generates discharge fault diagnosis results.

[0039] Specifically, based on the air breakdown voltage threshold, the synchronously recorded power grid operating phase voltage amplitude sequence is read. Set the sequence length to Initialize the counter and marker array The loop iterates through the voltage amplitude samples at each time point, and in each loop step, it records the phase voltage amplitude at the current moment. Compared with the air breakdown voltage threshold generated in the previous step Perform numerical comparisons, if If the condition is met, the counter is incremented by 1, and the corresponding position in the marker array is set to 1; otherwise, it is set to 0. After the traversal is complete, the formula is used to... The percentage of samples exceeding the standard is calculated. For example, if 50 out of 1000 sampling points exceed the threshold, the result is 5%. This percentage reflects the frequency of voltage waveforms intruding into the insulation danger zone, forming the intersection probability.

[0040] Count the time periods during which the voltage continuously exceeds the air breakdown voltage threshold and calculate the maximum continuous duration, then iterate through the marker array. Identify subsequence segments with consecutive 1s, record the length of each consecutive segment (i.e., the number of consecutive out-of-range sampling points), and multiply these numbers by the sampling interval. (For example, 0.01 seconds), obtain the duration of each consecutive exceedance, find the maximum value among all recorded durations, and define it as the maximum consecutive duration. The severity level of discharge warnings is categorized based on insulator flashover characteristic curves. For example, the threshold for Level 1 warnings is set as follows: and The level 2 warning threshold is and , calculate Intersection probability If the combined input into the hierarchical judgment logic, and If it is determined to be a "high-risk discharge", then it is considered a "high-risk discharge". and If the discharge is deemed "medium-risk discharge", it is classified as "low-risk fluctuation". The classification label is used as the final diagnostic conclusion to generate a discharge fault diagnosis result.

[0041] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for analyzing meteorological disaster faults in power equipment, characterized in that, Includes the following steps: Real-time wind speed, wind direction and air pressure data acting on transmission conductors and insulator strings are acquired, the tension vector and spatial displacement of the transmission conductors under wind load are calculated, the real-time swing coordinates of the conductors are generated, and the X-axis, Y-axis and Z-axis components of the real-time swing coordinates of the conductors are substituted into the spatial position analysis matrix to calculate and generate the phase-to-phase geometric mean distance. Based on the phase-to-phase geometric mean distance, calculate the flux linkage change and electric field coupling strength per unit length, generate instantaneous mutual inductance and capacitance parameters, perform calculations with the power grid angular frequency, calculate the distribution value of the imaginary part in the total impedance in the time domain, and generate the line dynamic reactance value. The reactance change rate index is calculated based on the dynamic reactance value of the line. The reactance change rate index is compared with the standard sinusoidal coupling characteristic value to generate the reactance fluctuation entropy criterion. Based on the minimum spatial gap associated with the reactance fluctuation entropy criterion, the dielectric strength limit under the current environment is calculated in combination with air pressure data to generate an air breakdown voltage threshold. The voltage amplitude of the grid operating phase is compared with the air breakdown voltage threshold to generate a discharge fault diagnosis result.

2. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the geometric mean distance between the phases are as follows: Based on real-time wind speed, wind direction and air pressure data, the wind pressure surface density is calculated and air pressure correction is performed. The insulator string swing angle is extracted and the wind angle of attack is corrected. The tension vector and spatial displacement are solved. The three-dimensional coordinate system is mapped in the order from mid-span to tension point to generate the real-time swing coordinates of the conductor. Based on the real-time swing coordinates of the conductor, the X-axis, Y-axis and Z-axis components are extracted and filled into the spatial position analysis matrix. The Euclidean distance between the two phases is calculated by enumerating each phase and combining them at the same index position. The Euclidean distance between the phases is corrected according to the phase sequence to form the equivalent geometric radius between the three-phase transmission conductors. Based on the equivalent geometric radius between the three-phase transmission conductors, phase distance combinations are constructed according to the pairing relationship of the phase Euclidean distances. Logarithmic scaling is performed on each phase distance combination according to the phase sequence, and side span distortion segments are removed. The combined geometric average value of the entire span is calculated to generate the phase mutual geometric average distance.

3. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the instantaneous mutual inductance and capacitance parameters are as follows: Based on the phase-to-phase geometric mean distance, a relative position sequence is established according to the phase sequence and the diagonal and off-diagonal terms of the potential coefficient matrix are filled. The conductor path is divided into segments according to the length ratio and the link direction is marked. The flux contribution is accumulated segment by segment and the coupling contribution is accumulated synchronously according to the item strength to generate the flux change and electric field coupling strength. Based on the flux linkage change and electric field coupling strength, the phase current amplitude sequence and phase voltage amplitude sequence are called to perform time-division ratio calculation and the results are recorded according to the phase sequence. Jump segments are eliminated moment by moment, and the effective range is limited by the geometric mean distance between phases as the screening condition. The results of each phase are merged to generate instantaneous mutual inductance and capacitance parameters.

4. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the dynamic reactance value of the line are as follows: Based on the instantaneous mutual inductance and capacitance parameters, the parameters are multiplied one by one with the power grid angular frequency according to the time index to form a complex impedance vector. Phasor superposition and phase sign constraint are performed, and the distribution value of the imaginary part of the total impedance in the time domain is extracted to generate the dynamic reactance value of the line.

5. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the reactance change rate index are as follows: Based on the line dynamic reactance value, all samples are rearranged in ascending order of timestamp within the sampling period, the time interval difference between adjacent samples is calculated, and linear interpolation is performed on the time interval with gaps to generate a line dynamic reactance value sequence with uniform sampling interval, thus obtaining the reconstructed reactance sequence. Based on the reconstructed reactance sequence, the difference between adjacent sample points is calculated, a first-order difference sequence is constructed, and abnormal mutation points are corrected to form the oscillation amplitude characteristics of the electrical parameters. The reactance change rate index is calculated based on the oscillation amplitude characteristics of the electrical parameters.

6. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the reactance fluctuation entropy criterion are as follows: Based on the reactance change rate index, a sampling sequence with uniform time step is established according to the time sampling order. Fourier transform operation is performed on the sampling sequence to extract the fundamental amplitude and harmonic amplitude in the frequency domain, and the phase and energy information of each frequency component is recorded to generate a spectral feature set. Calculate the degree of waveform distortion based on the spectral feature set; Based on the spectral feature set, harmonic components with distortion levels higher than the threshold are extracted, the frequency of amplitude occurrence is statistically analyzed, and the information entropy value is calculated. The information entropy value is compared with the set reactance stability judgment threshold, and a reactance fluctuation entropy criterion is generated based on the comparison result.

7. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the air breakdown voltage threshold are as follows: The minimum spatial gap corresponding to the reactance fluctuation entropy criterion is queried, the collected air pressure data is called to complete the air pressure correction, the minimum spatial gap is defined as the electrode spacing parameter according to the shortest path of the conductor to the ground, the corrected air pressure data is defined as the gas pressure parameter, the dielectric strength limit is calculated and converted into the potential difference threshold, and the air breakdown voltage threshold is generated.

8. The method for analyzing meteorological disaster faults in power equipment according to claim 1, characterized in that, The steps for obtaining the discharge fault diagnosis results are as follows: Based on the air breakdown voltage threshold, the voltage amplitude sequence of the operating phase of the power grid is extracted according to the sampling time index. The relationship between the voltage amplitude of the operating phase of the power grid and the air breakdown voltage threshold is compared sample by sample. Sample points that exceed the air breakdown voltage threshold are marked and the number of samples is counted. The proportion of the number of samples that exceed the air breakdown voltage threshold to the total number of samples is calculated to form the intersection probability. The system statistically analyzes the time periods during which the voltage continuously exceeds the air breakdown voltage threshold and calculates the maximum continuous duration. It then combines the maximum continuous duration with the intersection probability to determine the interval, classifies the severity according to the discharge warning threshold, and generates a discharge fault diagnosis result.

9. The system of the method for analyzing meteorological disaster faults in power equipment according to any one of claims 1-8, characterized in that, include: The data acquisition module is used to acquire real-time wind speed, wind direction and air pressure data acting on the transmission conductor and insulator string, calculate the tension vector and spatial displacement of the transmission conductor under wind load, generate the real-time swing coordinates of the conductor, and substitute the X-axis, Y-axis and Z-axis components of the real-time swing coordinates of the conductor into the spatial position analysis matrix to calculate and generate the phase-to-phase geometric mean distance. The electromagnetic parameter calculation module is used to calculate the change in magnetic flux and electric field coupling strength per unit length based on the geometric mean distance between phases, generate instantaneous mutual inductance and capacitance parameters, perform calculations with the power grid angular frequency, calculate the distribution value of the imaginary part in the total impedance in the time domain, and generate the dynamic reactance value of the line. The feature analysis module is used to calculate the reactance change rate index based on the dynamic reactance value of the line, compare the reactance change rate index with the standard sinusoidal coupling feature value, and generate a reactance fluctuation entropy criterion. The fault diagnosis module is used to calculate the dielectric strength limit under the current environment based on the minimum spatial gap associated with the reactance fluctuation entropy criterion and combined with air pressure data, generate an air breakdown voltage threshold, compare the voltage amplitude of the grid operating phase with the air breakdown voltage threshold, and generate a discharge fault diagnosis result.

Images