Arc suppression coil grounding fault positioning method based on transformer substation line protection device

Abstract

The invention discloses an arc suppression coil grounding fault positioning method based on a transformer substation line protection device, particularly relates to the technical field of system fault measurement, and is used for solving the problem of insufficient positioning precision caused by neglecting the nonlinear characteristic of a fault point arc channel in an existing fault positioning method. A specific frequency current signal is injected into a fault phase, the phase relation between fundamental wave and harmonic wave components in current collected by a line protection device is analyzed to obtain a phase coherence coefficient, a fault line distribution parameter model is established to obtain theoretical reference information, and a spatial gradient sequence is generated by calculating actual measurement and theoretical spatial entropy values. Identifying a line section corresponding to the abnormal abrupt change point, and finally determining a grounding fault point according to the spatial position of the line section with significant difference. The method can effectively overcome the influence of nonlinear attenuation and frequency dispersion effect on the positioning precision, and achieves the precise measurement and positioning of a single-phase grounding fault.

Description

Technical Field

[0001] This invention relates to the field of system fault measurement technology, and more specifically, to a method for locating ground faults in arc suppression coils of substation line protection devices. Background Technology

[0002] In smart grid distribution network neutral point grounded by arc suppression coils, signal injection is a commonly used fault location technique for quickly locating and isolating the fault point after a single-phase ground fault occurs. A specific frequency current signal is injected between the faulty phase and ground. This signal flows along the faulty line and into the ground at the grounding point. By using a dedicated detector or protective devices installed along the line to detect the path and intensity of the injected signal, the fault section or point can be located. Existing technologies generally construct location models based on linear circuit theory, assuming that the signal transmission path from the injection point to the fault point, especially the arc path at the fault point, has stable linear impedance characteristics.

[0003] However, the arc channel at the fault point in actual operation exhibits strong nonlinear dynamic characteristics. When the injected signal passes through the arc channel, it will encounter significant nonlinear attenuation and dispersion effects, resulting in waveform distortion, abnormal amplitude attenuation, and frequency component changes in the injected signal. This causes the signal characteristics received and analyzed by the protection device to deviate significantly from the theoretical expectations based on the linear model, ultimately leading to fundamental errors in the fault location results and a decrease in location accuracy. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for locating ground faults of arc suppression coils based on substation line protection devices to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The method for locating ground faults using arc suppression coils in substation line protection devices includes: S1. After a single-phase ground fault is detected, a current signal of a specific frequency is injected between the faulty phase and the ground through a signal injection device. S2. Obtain the current signal collected by the line protection device, and calculate the phase coherence coefficient by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal. S3. Establish a distributed parameter model for the faulty line to obtain theoretical reference information including the distribution data of parameters in the direction of line length; S4. Calculate the measured spatial entropy value based on the distribution of the phase coherence coefficient along the line length, and compare and analyze the measured spatial entropy value with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of the measured spatial entropy value and the theoretical spatial entropy value. S5. Line segments identified as anomalous mutation points in the spatial gradient sequence are determined as line segments whose measured spatial entropy values ​​differ significantly from theoretical spatial entropy values. S6. Based on the spatial location of line sections with significant differences, determine the final location of the grounding fault point.

[0006] Furthermore, upon detecting a single-phase ground fault, a current signal of a specific frequency is injected between the faulty phase and ground via a signal injection device, including: Determine the injection parameters for a current signal at a specific frequency. The injection parameters include the injection frequency value and the injection current amplitude. According to the injection parameters, a current signal of a specific frequency is injected between the faulty phase and ground through a signal injection device; Monitor the current signal waveform during the injection process, and calculate the current amplitude stability and frequency purity based on the current signal waveform; When the current amplitude stability is lower than the preset stability threshold or the frequency purity is lower than the preset purity threshold, adjust the injection parameters and re-execute the injection step until the current amplitude stability is greater than or equal to the preset stability threshold and the frequency purity is greater than or equal to the preset purity threshold.

[0007] Furthermore, the current signal collected by the line protection device is acquired, and the phase coherence coefficient is calculated by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal, including: The fundamental component and main harmonic components are separated from the current signal collected by the line protection device; Extract the instantaneous phase of the fundamental component and the instantaneous phase of the main harmonic components; Calculate the instantaneous phase difference between the fundamental component and each major harmonic component; The phase coherence coefficient is calculated based on the instantaneous phase difference sequence, where the phase coherence coefficient reflects the stability of the phase relationship between the fundamental component and the main harmonic components.

[0008] Furthermore, the calculation of phase coherence coefficients based on the instantaneous phase difference sequence includes: calculating the statistical variance of the instantaneous phase difference sequence, and normalizing the reciprocal of the statistical variance to obtain the phase coherence coefficients.

[0009] Furthermore, a distributed parameter model of the faulty line is established to obtain theoretical reference information including the distribution data of parameters along the line length direction, including: The resistance, inductance, capacitance, and admittance to ground per unit length of the faulty line are determined based on its structural parameters and material properties. A distributed parameter model of the faulty line is established based on the resistance, inductance, capacitance and admittance to ground per unit length of the line. The theoretical phase coherence coefficient of an injected specific frequency current signal propagating along the line length is calculated based on a distributed parameter model. The distribution of the theoretical phase coherence coefficient along the line length is used as part of the theoretical reference information.

[0010] Furthermore, the measured spatial entropy value is calculated based on the distribution of the phase coherence coefficient along the line length, and then compared and analyzed with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of the measured and theoretical spatial entropy values, including: The measured spatial entropy sequence is calculated based on the distribution of phase coherence coefficients along the line length. The measured spatial entropy sequence is obtained by applying the Shannon entropy formula to the distribution of phase coherence coefficients along the line length. The theoretical spatial entropy sequence is calculated based on the distribution of the theoretical phase coherence coefficients along the line length in the theoretical reference information. The theoretical spatial entropy sequence is obtained by applying the same Shannon entropy formula to the distribution of the theoretical phase coherence coefficients along the line length. The first-order difference sequence of the measured spatial entropy value sequence along the line length direction is calculated as the spatial gradient sequence of the measured spatial entropy value; The first-order difference sequence of the theoretical spatial entropy value sequence along the line length direction is calculated as the spatial gradient sequence of the theoretical spatial entropy value.

[0011] Furthermore, calculating the first-order difference sequence of the measured spatial entropy value sequence along the line length direction as the spatial gradient sequence of the measured spatial entropy value includes: calculating the difference of the measured spatial entropy values ​​of adjacent sections in order along the line length direction, and arranging the differences in order to form a spatial gradient sequence.

[0012] Furthermore, line segments identified as anomalous abrupt change points in the spatial gradient sequence are determined to be line segments where the measured spatial entropy value differs significantly from the theoretical spatial entropy value, including: The gradient difference between corresponding points is calculated between the spatial gradient sequence of the measured spatial entropy value and the spatial gradient sequence of the theoretical spatial entropy value to form a spatial gradient difference sequence. Calculate the gradient difference magnitude at each point in the spatial gradient difference sequence; Identify candidate mutation points whose gradient difference magnitude exceeds a preset difference threshold; Select points from candidate mutation points that simultaneously satisfy the local extremum condition as anomalous mutation points; The smallest line segment containing abnormal mutation points is identified as a line segment where the measured spatial entropy value differs significantly from the theoretical spatial entropy value.

[0013] Furthermore, selecting points from candidate mutation points that simultaneously satisfy the local extremum condition as anomalous mutation points includes: within a preset neighborhood of the spatial gradient difference sequence, determining the candidate mutation point with the maximum gradient difference amplitude as satisfying the local extremum condition.

[0014] Furthermore, based on the spatial locations of the line sections with significant differences, the final location of the grounding fault is determined, including: Obtain spatial location information of line sections with significant differences; The coordinates of the geometric center point of the corresponding line segment are determined based on the spatial location information of the line segments with significant differences. The coordinates of the geometric center point are corrected based on the distribution characteristics of abnormal abrupt change points within the line section with significant differences. The spatial location corresponding to the corrected geometric center point coordinates is determined as the final location result of the grounding fault point.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By introducing phase coherence coefficient analysis and spatial entropy comparison mechanism, the influence of nonlinear characteristics of the arc channel at the fault point on positioning accuracy is effectively overcome. Traditional methods, based on linear model assumptions, cannot accurately handle waveform distortion and frequency component changes generated when the signal passes through the arc channel. This invention, by analyzing the phase relationship between the fundamental and harmonic components, extracts the phase coherence coefficient as a key feature quantity. This coefficient can sensitively reflect the changes in phase stability of the signal during transmission, thereby capturing the changes in signal characteristics caused by nonlinear attenuation and dispersion effects. Combined with the theoretical reference information provided by the distributed parameter model, accurate modeling of signal propagation characteristics over the entire length of the line is achieved. This makes the positioning process no longer dependent on simple linear impedance assumptions, but based on more realistic multi-parameter distributed data, significantly improving the positioning robustness under nonlinear conditions.

[0016] 2. By calculating spatial entropy and analyzing gradient sequences, the fault location problem is transformed into a spatial anomaly detection problem. Utilizing the sensitivity of entropy to signal distribution inhomogeneity, it effectively identifies abrupt changes in local features caused by the presence of fault points. The comparison between measured and theoretical spatial entropy values ​​filters out environmental interference in normal sections of the line, while the spatial gradient sequence amplifies the boundary features of the fault section, making the detection of abrupt changes more accurate and reliable. Finally, a geometric correction mechanism is used to determine the fault location, ensuring that the location results are both theoretically consistent and conform to actual distribution characteristics. This provides stable and repeatable high-precision location capabilities while avoiding fundamental errors, fully meeting the application needs of the power system measurement field. Attached Figure Description

[0017] Figure 1This is a flowchart of the method for locating grounding faults in arc suppression coils based on substation line protection devices according to the present invention; Figure 2 This is a schematic diagram of the structure of the arc suppression coil grounding fault location system based on the substation line protection device of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1 The present invention provides a method for locating ground faults in arc-suppression coils of substation line protection devices, comprising: S1. After a single-phase ground fault is detected, a current signal of a specific frequency is injected between the faulty phase and the ground through a signal injection device. S2. Obtain the current signal collected by the line protection device, and calculate the phase coherence coefficient by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal. S3. Establish a distributed parameter model for the faulty line to obtain theoretical reference information including the distribution data of parameters in the direction of line length; S4. Calculate the measured spatial entropy value based on the distribution of the phase coherence coefficient along the line length, and compare and analyze the measured spatial entropy value with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of the measured spatial entropy value and the theoretical spatial entropy value. S5. Line segments identified as anomalous mutation points in the spatial gradient sequence are determined as line segments whose measured spatial entropy values ​​differ significantly from theoretical spatial entropy values. S6. Based on the spatial location of line sections with significant differences, determine the final location of the grounding fault point.

[0020] S1. After detecting a single-phase ground fault, a current signal of a specific frequency is injected between the faulty phase and ground through a signal injection device. Specifically, this is implemented as follows: After detecting a single-phase ground fault, the process of injecting a current signal of a specific frequency between the faulty phase and ground using a signal injection device first involves determining the injection parameters for the specific frequency current signal. These parameters consist of the injection frequency and the injection current amplitude. The injection frequency is determined based on the principle of being distinct from the power grid frequency and avoiding system resonance points. The power grid frequency is typically 50 Hz or 60 Hz. The system resonance point is obtained by analyzing the impedance-frequency characteristic curve of the arc suppression coil grounding system. This curve is calculated from system parameters including the line-to-ground capacitance and the arc suppression coil inductance. The specific frequency, i.e., the injection frequency value, is selected as a fixed value within the range of 15 Hz to 300 Hz, such as 220 Hz, to ensure that the signal avoids excessive attenuation or amplification caused by resonance during propagation. The injection current amplitude is determined based on the electrical parameters of the faulty line, including the line impedance per unit length and the line-to-ground capacitance, as well as the maximum output capacity of the signal injection device. It is optimized through simulation experiments or historical operating data. For example, the injection current amplitude is set to a value between 5 amps and 20 amps to balance signal detection sensitivity and equipment safety constraints.

[0021] According to the injection parameters, a current signal of a specific frequency is injected between the faulty phase and ground through a signal injection device. The signal injection device is implemented using a standard power electronic converter. Its control circuit generates a corresponding pulse width modulation signal according to the injection frequency value and the injection current amplitude, drives the insulated gate bipolar transistor power switching device to generate the required current waveform, and injects it between the faulty phase and ground through a coupling transformer to ensure that the current signal propagates stably in the faulty line.

[0022] The injected current signal waveform is monitored in real time during the injection process. Monitoring is performed using a high-precision current sensor and data acquisition system, recording the current signal waveform at a sampling rate of 10,000 points per second. Current amplitude stability is calculated based on the current signal waveform. This stability is obtained by analyzing the fluctuations in the current waveform amplitude over multiple consecutive cycles. Specifically, the average peak value of the current waveform in each cycle is taken, and then the standard deviation of these average values ​​within a time window is calculated. Current amplitude stability is defined as the reciprocal of the standard deviation, normalized to the range of 0 to 1; the closer the value is to 1, the more stable the amplitude. Simultaneously, frequency purity is calculated based on the current signal waveform. Frequency purity is obtained by analyzing the current waveform using Fast Fourier Transform (FFT). The amplitude of a specific frequency component is extracted, and its ratio to the total signal energy is calculated. The total signal energy is the sum of the squares of the amplitudes of all frequency components. Frequency purity is defined as the percentage of energy of a specific frequency component relative to the total energy; the closer the value is to 100%, the purer the frequency component.

[0023] The preset stability threshold and preset purity threshold are set based on system operating requirements and historical fault data. The process for setting the preset stability threshold includes collecting current amplitude stability data under normal operating conditions and calculating its statistical distribution. A specific quantile of this distribution is used as the threshold benchmark; for example, the value corresponding to the 95th percentile is taken as the preset stability threshold. The process for setting the preset purity threshold includes analyzing the minimum frequency purity requirements under different fault conditions and determining it in conjunction with the minimum identifiable signal-to-noise ratio of the signal detection device. For example, by simulating different grounding resistance fault scenarios in the laboratory, the lower limit of frequency purity required for reliable signal detection is determined as the preset purity threshold.

[0024] When the calculated current amplitude stability is lower than the preset stability threshold or the frequency purity is lower than the preset purity threshold, the injection parameters are adjusted and the injection step is repeated. Adjusting the injection parameters includes modifying the injection frequency or the injection current amplitude. For example, if the frequency purity is low, the injection frequency is fine-tuned to avoid interference frequencies; if the current amplitude stability is low, the injection current amplitude is increased to enhance signal strength. The adjustment range is dynamically determined based on the deviation between the measured value and the threshold; for example, a larger deviation results in a larger adjustment range, and a smaller deviation results in a smaller adjustment range. After re-executing the injection step, the current signal waveform is monitored again, and the current amplitude stability and frequency purity are calculated. This process is repeated until the current amplitude stability is greater than or equal to the preset stability threshold and the frequency purity is greater than or equal to the preset purity threshold, ensuring reliable quality of the injected signal. The preferred frequency range of 15 Hz to 300 Hz is determined based on the signal propagation characteristics of the power system. Within this range, signal attenuation is low and it is easily distinguishable from power frequency harmonics. For example, a fixed frequency of 220 Hz has been experimentally verified to have low system impedance and minimal resonance risk at this frequency, thus ensuring effective signal transmission and detection. The entire injection process is automated, with parameter adjustments and retries performed by the embedded processor. It makes real-time decisions based on a preset algorithm to ensure rapid and stable signal injection even under fault conditions.

[0025] S2. Acquire the current signal collected by the line protection device, and calculate the phase coherence coefficient by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal. The specific implementation is as follows: The process of acquiring the current signal collected by the line protection device and calculating the phase coherence coefficient by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal first includes separating the fundamental component and the main harmonic components from the current signal collected by the line protection device. The current signal collected by the line protection device is time-domain waveform data, and the sampling rate is set according to the power system measurement requirements, such as 6400 points per second, to ensure coverage of the fundamental and harmonic frequency components. The separation of the fundamental component is achieved using a digital bandpass filter. The center frequency of this filter is set to the power grid frequency, such as 50 Hz, and the passband width is set according to the fundamental frequency fluctuation range, such as ±0.5 Hz, to effectively extract the fundamental component. The separation of major harmonic components employs multiple digital bandpass filters processed in parallel. These major harmonic components include the 2nd, 3rd, 5th, and 7th harmonics. The selection of these harmonics is based on the common harmonic characteristics of power systems. The center frequency of the filter corresponds to the frequency of each harmonic; for example, 100 Hz for the 2nd harmonic and 150 Hz for the 3rd harmonic. The passband width is set according to the harmonic frequency stability, for example, ±1 Hz. The separation process is executed by an embedded digital signal processor, using either an infinite impulse response (IER) filter or a finite impulse response (FIR) filter. The filter coefficients are calculated based on frequency response requirements, ensuring accurate separation of the fundamental component and major harmonic components from the original signal without mutual interference.

[0026] The instantaneous phases of the fundamental component and the principal harmonic components are extracted using the Hilbert transform method. For the separated fundamental component signal, the Hilbert transform is applied to calculate its analytic signal. The real part of the analytic signal is the original fundamental component signal, and the imaginary part is the signal after the Hilbert transform. The instantaneous phase of the fundamental component is obtained by calculating the phase angle of the analytic signal. The phase angle is calculated using the four-quadrant arctangent function, and the output is a radian value that changes continuously with time. Similarly, for each principal harmonic component signal, the Hilbert transform process is repeated to calculate its respective analytic signal and extract the instantaneous phase. The instantaneous phase sequence is recorded at the same time interval, for example, one sampling point every 1 millisecond, to ensure that the phase data is synchronized with the original signal. During the instantaneous phase extraction process, the phase angle is processed by modulo operation to ensure that the phase value is within a continuous range, for example, from negative π radians to positive π radians.

[0027] The instantaneous phase difference between the fundamental component and each major harmonic component is calculated by comparing the instantaneous phase of the fundamental component with the instantaneous phase of each major harmonic component point by point. For each major harmonic component, such as the 2nd harmonic, the instantaneous phase difference is calculated as the instantaneous phase of the fundamental component minus the instantaneous phase of the 2nd harmonic component, expressed in radians. The same method is applied to the 3rd, 5th, and 7th harmonics. The instantaneous phase difference sequence is arranged in chronological order, with each sequence containing phase difference values ​​from multiple sampling points; for example, 1000 phase difference data points are generated within 1 second. During the calculation, the range of phase difference values ​​is limited to between negative π radians and positive π radians. Phase jumps are handled by adding or subtracting integer multiples of 2π radians to ensure the continuity and comparability of the phase difference sequence.

[0028] Phase coherence coefficients are calculated based on instantaneous phase difference sequences, reflecting the stability of the phase relationship between the fundamental component and the major harmonic components. The calculation begins with the instantaneous phase difference sequence for each major harmonic component. For example, for the second harmonic, the statistical variance of this sequence is calculated. This variance is obtained by averaging all values ​​in the instantaneous phase difference sequence, then summing the squares of the differences between each value and the average, and dividing by the sequence length minus one. The variance value indicates the degree of phase difference fluctuation. Next, the reciprocal of the statistical variance is normalized to obtain the phase coherence coefficient. Normalization is achieved by dividing the reciprocal of the variance by a normalization factor. This factor is set based on the theoretical maximum variance, calculated based on the possible range of phase differences. For example, when the phase difference range is from negative π radians to positive π radians, the theoretical maximum variance is the square of π divided by 3. After normalization, the phase coherence coefficient ranges from 0 to 1, with values ​​closer to 1 indicating a more stable phase relationship. For multiple major harmonic components, the phase coherence coefficient corresponding to each harmonic is calculated separately, and the average value of these coefficients is taken as the overall phase coherence coefficient to comprehensively reflect phase stability.

[0029] The specific implementation of calculating phase coherence coefficients based on instantaneous phase difference sequences includes calculating the statistical variance of the instantaneous phase difference sequence and normalizing the reciprocal of the statistical variance to obtain the phase coherence coefficients. The statistical variance is calculated using the standard deviation formula. The input is all data points of the instantaneous phase difference sequence, and the output is a single variance value. A larger variance value indicates greater phase fluctuation. Normalization is achieved by dividing the reciprocal of the variance by a reference value. The reference value is determined based on the theoretical characteristics of the phase difference sequence, such as the minimum expected value of the phase difference variance under ideal stable conditions, or a reference variance derived from historical data statistics. After normalization, the coefficients are ensured to be interpretable between 0 and 1. The entire calculation process is implemented using a digital signal processor (DSP) program, employing floating-point operations to ensure precision; for example, 32-bit floating-point numbers are used to process the phase difference data to avoid accumulated errors. The stability of the phase coherence coefficient is judged based on its value. For example, when the phase coherence coefficient is lower than the preset coherence threshold, the phase relationship is considered unstable. The preset coherence threshold is determined by analyzing the distribution characteristics of the phase coherence coefficient in historical fault data. Specifically, the distribution of coefficient values ​​under normal and fault conditions is statistically analyzed, and a critical value that can effectively distinguish between the two conditions is selected as the preset coherence threshold.

[0030] S3. Establish a distributed parameter model for the faulty line to obtain theoretical reference information including the distribution data of parameters along the line length direction. The specific implementation is as follows: The process of establishing a distributed parameter model for a faulty line and obtaining theoretical reference information including data on the distribution of parameters along the line length first involves determining the resistance, inductance, capacitance, and admittance to ground per unit length of the line based on its structural parameters and material properties. Structural parameters include conductor cross-sectional area, conductor diameter, conductor spacing, and conductor height to ground; these parameters are obtained from line design drawings. Material properties include conductor resistivity, relative permeability, and dielectric constant of the insulating medium; these parameters are obtained from material handbooks or experimental measurements. The resistance per unit length of the line is calculated based on the conductor resistivity and cross-sectional area, for example, using the formula that resistance equals resistivity multiplied by length divided by cross-sectional area, where the length is taken as 1 meter. The inductance per unit length of the line is calculated based on the conductor size and arrangement, using an inductance formula that considers both internal and external magnetic flux linkages; for example, for a parallel multi-conductor system, the geometric mean distance and self-geometric mean distance are used to calculate the inductance value. The capacitance per unit length of the line is calculated based on the conductor size and its relationship to ground, using Maxwell's potential coefficient method, by constructing a potential coefficient matrix and inverting it to obtain the capacitance matrix. The ground admittance per unit length of the line is calculated based on the characteristics of the insulating medium and the operating frequency. It includes ground conductance and ground tolerance. Ground conductance is calculated using the loss tangent of the insulating medium and the capacitance value, while ground tolerance is calculated using the capacitance value and the angular frequency.

[0031] A distributed parameter model of the faulty line is established based on the resistance, inductance, capacitance, and admittance to ground per unit length of the line. The distributed parameter model adopts a uniform transmission line model, dividing the entire line into multiple small segments, each of which is represented by lumped parameters of resistance, inductance, capacitance, and admittance to ground.

[0032] In establishing a distributed parameter model of a faulty line, the model establishment includes determining the model's fundamental differential equations, namely the telegraph equations. The telegraph equations describe the relationship between voltage and current as a function of the line's location, and their general form is: ; Where V(x) represents the voltage at position x on the line, in volts (V); I(x) represents the current at position x on the line, in amperes (A); x represents the position coordinate along the length of the line, in meters (m); R represents the resistance per unit length, in ohms per meter (Ω / m); L represents the inductance per unit length, in henries per meter (H / m); G represents the conductance per unit length, in sieems per meter (S / m); C represents the capacitance per unit length, in farads per meter (F / m); ω represents the angular frequency, in radians per second (rad / s), and is related to the injected specific frequency f by ω = 2πf; j represents the imaginary unit, satisfying j 2 =−1.

[0033] The model parameters include characteristic impedance and propagation constant. The formula for calculating characteristic impedance is: Where Z0 represents characteristic impedance, and the unit is ohms (Ω).

[0034] The formula for calculating the propagation constant is: ; where γ represents the propagation constant, the unit is per meter, usually a complex number, the real part represents the attenuation constant, and the imaginary part represents the phase constant.

[0035] The formula, based on uniform transmission line theory, is used to calculate the propagation characteristics of signals in faulty transmission lines. Parameters R, L, G, and C are determined through line structural parameters and material properties, using methods described previously, such as calculations based on conductor cross-sectional area, spacing, and dielectric properties. ω is determined by the specific injection frequency; for example, when the injection frequency is 220 Hz, ω = 2π × 220 rad / s. The square root operation in the formula is performed in the complex domain to ensure the model accurately reflects the amplitude and phase characteristics of the line.

[0036] It is worth noting that in the process of solving the distributed parameter model, the propagation matrix is ​​constructed using the four-terminal network method. The characteristic impedance and propagation constant are converted into a transmission matrix form using a hyperbolic function, specifically including the calculation of forward and backward propagation parameters. A hierarchical iterative strategy is adopted for boundary condition handling. First, the initial boundary is established based on the measured current value at the injection point. Then, the terminal boundary equation is established using the line terminal impedance condition. Frequency sweep analysis is used to verify the model's stability at different frequencies. A sliding window mechanism is introduced for calculating the theoretical phase coherence coefficient. A fixed-length analysis window is set on the current distribution data, and the complex forms of the fundamental and harmonic waves are calculated point by point. Phase information is extracted using discrete Fourier transform, and the covariance matrix of the phase difference sequence is established to evaluate phase stability. During numerical iteration, the convergence criterion is set as the voltage distribution error between two adjacent iterations being less than 10. -6 Meanwhile, an adaptive step size adjustment strategy is adopted to optimize the computational efficiency of the Newton-Raphson method.

[0037] The theoretical phase coherence coefficient of an injected current signal at a specific frequency propagating along the line length is calculated based on a distributed parameter model. The calculation process first uses the distributed parameter model to solve for the current signal at each point on the line at a specific frequency, determined in the injection step (e.g., 220 Hz). For each calculation point on the line, the fundamental component and principal harmonic components of the current signal are extracted. The fundamental component is a specific frequency component, and the principal harmonic components include the 2nd, 3rd, 5th, and 7th harmonics, which are generated by the nonlinear characteristics of the line. The theoretical phase difference between each harmonic component and the fundamental component is calculated. This theoretical phase difference is obtained by subtracting the phase values ​​of each component at the calculation point, with the phase values ​​extracted from the complex form of the current signal. The theoretical phase coherence coefficient is calculated based on the theoretical phase difference sequence, using the same method as the measured phase coherence coefficient, including calculating the statistical variance of the phase difference sequence and normalizing the reciprocal of the statistical variance. The theoretical phase coherence coefficient reflects the ideal stability of the phase relationship under fault-free conditions.

[0038] The distribution of the theoretical phase coherence coefficient along the line length is used as a component of the theoretical reference information. This distribution is obtained by repeating the calculation at each calculation point, which is evenly distributed along the line, for example, one calculation point every 10 meters, covering the entire line length. The distribution data is stored in array form, containing the line location coordinates and the corresponding theoretical phase coherence coefficient values. The theoretical reference information also includes the distribution data of other line parameters, such as the distribution of characteristic impedance and propagation constant along the line. These data together constitute a complete theoretical reference benchmark. The theoretical reference information is used for subsequent comparison with measured data. By comparing the theoretical and measured phase coherence coefficient distributions, abnormal sections of the line can be identified.

[0039] The distributed parameter model is validated by comparing the model's calculation results with actual measurement data. For example, current distribution is measured on a section of line with known parameters, and compared with the model's predicted values, with the error range controlled within, for example, 5%. Model parameter adjustments are based on the validation results. For example, if the calculated resistance value deviates significantly from the measured value, the conductor resistivity parameter is adjusted. The adjustment range is determined according to the magnitude of the deviation; for example, if the deviation is 10%, adjust by 5%, until the model output matches the measured data. The calculation of the theoretical phase coherence coefficient considers the normal operating conditions of the line and does not include the influence of fault points, thus providing an ideal reference benchmark. The entire modeling process is implemented through computer simulation, using numerical calculation methods to solve the distributed parameter model, ensuring computational accuracy and efficiency.

[0040] S4. Calculate the measured spatial entropy value based on the distribution of the phase coherence coefficient along the line length, and compare the measured spatial entropy value with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of the measured and theoretical spatial entropy values. The specific implementation is as follows: The process of calculating the measured spatial entropy value based on the distribution of phase coherence coefficients along the line length, and comparing and analyzing the measured spatial entropy value with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of measured and theoretical spatial entropy values, firstly includes calculating the measured spatial entropy value sequence based on the distribution of phase coherence coefficients along the line length. The distribution of phase coherence coefficients along the line length comes from the phase coherence coefficient values ​​obtained by analyzing the current signal in the previous steps. These values ​​are arranged sequentially along the line length. For example, if the total line length is 1000 meters, it is divided into 100 continuous segments, each segment being 10 meters long, and each segment corresponding to one phase coherence coefficient. The measured spatial entropy value sequence is obtained by applying the Shannon entropy formula to the distribution of phase coherence coefficients along the line length. The Shannon entropy formula is calculated by first normalizing all phase coherence coefficient values. Normalization is achieved by dividing each phase coherence coefficient value by the sum of all phase coherence coefficient values, ensuring that the sum of the normalized values ​​is 1, thus forming a probability distribution. The Shannon entropy is then calculated by summing the values ​​of each normalized phase coherence coefficient multiplied by the negative natural logarithm of that coefficient. Specifically, for each segment, the negative normalized phase coherence coefficient is multiplied by its natural logarithm, and the results for all segments are summed to obtain the entropy value for that segment. The measured spatial entropy value sequence is obtained by repeating this calculation at each location along the length of the line, for example, calculating the entropy value of each segment sequentially from the starting point to the end point of the line. In the application of the Shannon entropy formula, the logarithm is calculated using the natural logarithm, with the unit being knights, to ensure that the entropy value reflects the degree of uncertainty in the distribution of the phase coherence coefficients.

[0041] The theoretical spatial entropy sequence is calculated based on the distribution of theoretical phase coherence coefficients along the line length from the theoretical reference information. This distribution originates from the theoretical phase coherence coefficient values ​​calculated using the distribution parameter model in the preceding steps. These values ​​are arranged along the line length in identical segments, for example, 100 segments, each corresponding to one theoretical phase coherence coefficient. The theoretical spatial entropy sequence is obtained by applying the same Shannon entropy formula to the distribution of theoretical phase coherence coefficients along the line length. The calculation process is identical to that of the measured spatial entropy sequence, including normalizing the theoretical phase coherence coefficient values ​​and then calculating the Shannon entropy. Normalization ensures that the sum of the theoretical phase coherence coefficient values ​​is 1. The Shannon entropy calculation uses the same natural logarithm method, outputting a theoretical spatial entropy sequence that represents the ideal entropy variation of the phase coherence coefficient distribution under fault-free conditions. The theoretical spatial entropy sequence and the measured spatial entropy sequence have the same length and segment correspondence, facilitating subsequent comparative analysis.

[0042] The first-order difference sequence of the measured spatial entropy value sequence along the line length is calculated as the spatial gradient sequence of the measured spatial entropy value. The calculation of the first-order difference sequence is achieved by calculating the difference between the measured spatial entropy values ​​of adjacent segments sequentially along the line length. For example, for the i-th segment and the (i+1)-th segment in the measured spatial entropy value sequence, the difference is calculated as the measured spatial entropy value of the (i+1)-th segment minus the measured spatial entropy value of the i-th segment, where i starts from 1 and represents the segment number (to the total number of segments minus 1). These differences are arranged sequentially along the line length to form the spatial gradient sequence, which is one less than the measured spatial entropy value sequence. Each value represents the rate of change of entropy values ​​between adjacent segments. During the calculation, the differences are stored numerically in the same unit as the entropy value, such as knights per meter, reflecting the spatial gradient of entropy value along the line length. The generation of the spatial gradient sequence ensures data continuity, for example, by handling boundary segments through linear interpolation, but the core calculation is based on direct difference.

[0043] The first-order difference sequence of the theoretical spatial entropy value sequence along the line length is calculated as the spatial gradient sequence of the theoretical spatial entropy value. The calculation method for the first-order difference sequence is the same as that for the measured spatial gradient sequence. The difference between the theoretical spatial entropy values ​​of adjacent segments is calculated sequentially along the line length direction. For example, the theoretical spatial entropy value of the (i+1)th segment is subtracted from the theoretical spatial entropy value of the ith segment. The differences are arranged in order to form the theoretical spatial gradient sequence. The theoretical spatial gradient sequence represents the ideal gradient of entropy value change under fault-free conditions and is used for comparison with the measured spatial gradient sequence. In the calculation, the segment division of the theoretical spatial entropy value sequence is consistent with the measured sequence to ensure that the gradient values ​​correspond to the same spatial locations.

[0044] The specific implementation of calculating the spatial gradient sequence of measured spatial entropy values ​​along the length of the line, using the first-order difference sequence, involves calculating the difference between the measured spatial entropy values ​​of adjacent segments sequentially along the line length, and arranging these differences in order to form the spatial gradient sequence. The definition of adjacent segments is based on the continuity along the line length; for example, segments are numbered sequentially by length, from the starting point 1 to the ending point 100, and the difference calculation is performed between consecutively numbered segments. The difference calculation uses arithmetic subtraction, retaining the sign of the result to indicate the direction of change: a positive difference indicates an increase in entropy, and a negative difference indicates a decrease. The spatial gradient sequence is arranged in the same order as the line length, for example, storing gradient values ​​sequentially from the starting point to the ending point. The entire calculation is executed by a digital processor, using floating-point operations to ensure precision; for example, 64-bit floating-point numbers are used to process entropy values ​​and differences to avoid accumulated errors. The spatial gradient sequence is used for subsequent identification of anomalous abrupt changes, providing spatial characteristics of entropy changes.

[0045] S5. The line segments identified as anomalous abrupt change points in the spatial gradient sequence are determined as those with significantly different measured and theoretical spatial entropy values. The specific implementation is as follows: The process of identifying line segments with anomalous abrupt changes in the spatial gradient sequence as those with significant differences between measured and theoretical spatial entropy values ​​first involves calculating the gradient differences between corresponding points in the spatial gradient sequences of measured and theoretical spatial entropy values, forming a spatial gradient difference sequence. The spatial gradient sequence of measured spatial entropy values ​​is derived from data obtained in the preceding steps by calculating the first-order difference sequence of the measured spatial entropy value sequence along the line length. The spatial gradient sequence of theoretical spatial entropy values ​​is derived from data obtained in the preceding steps by calculating the first-order difference sequence of the theoretical spatial entropy value sequence along the line length. Gradient difference calculation is performed for each corresponding location point. For example, for the u-th location point on the line, the gradient difference value is the gradient value of the spatial gradient sequence of measured spatial entropy values ​​at the u-th location point minus the gradient value of the spatial gradient sequence of theoretical spatial entropy values ​​at the u-th location point, where u ranges from 1 to the sequence length, and u is the location point number. These gradient difference values ​​are arranged sequentially along the line length to form the spatial gradient difference sequence, which has the same length as the spatial gradient sequence. Each value represents the degree of difference between the measured and theoretical gradients at that location point.

[0046] Calculate the gradient difference magnitude at each point in the spatial gradient difference sequence. The gradient difference magnitude is obtained by taking the absolute value of the gradient difference at each point in the spatial gradient difference sequence. For example, for the gradient difference value at a certain point in the spatial gradient difference sequence, its absolute value is taken as the gradient difference magnitude at that point. The calculation of the gradient difference magnitude ensures that the result is a non-negative value, reflecting the magnitude of the gradient difference without considering the direction. The gradient difference magnitude sequence has the same length as the spatial gradient difference sequence, and each magnitude corresponds to a specific location point on the line, representing the intensity of the gradient difference at that point.

[0047] Candidate abrupt change points are identified when the gradient difference amplitude exceeds a preset threshold. The preset threshold is set based on the statistical characteristics of the spatial gradient difference sequence, specifically determined by analyzing the historical data distribution of the spatial gradient difference sequence under normal operating conditions. For example, the preset threshold is calculated by adding twice the standard deviation to the average gradient difference amplitude in historical normal data. Candidate abrupt change points are identified by iterating through each point in the gradient difference amplitude sequence and marking points with gradient difference amplitudes greater than the preset threshold. For instance, when the preset threshold is set to 0.05 nits per meter, all points with gradient difference amplitudes exceeding 0.05 nits per meter are selected as candidate abrupt change points. The location information of the candidate abrupt change points includes the line section number and spatial coordinates; these points constitute a preliminary set of outliers.

[0048] Points that simultaneously satisfy the local extremum condition are selected from candidate mutation points as anomalous mutation points. The local extremum condition is determined by comparing the gradient difference magnitude within a preset neighborhood of the spatial gradient difference sequence. The preset neighborhood is set according to the length of the line segment; for example, it is defined as two segments extending before and after the candidate mutation point. Within this preset neighborhood, the candidate mutation point with the maximum gradient difference magnitude is identified as satisfying the local extremum condition. For example, for candidate mutation point k, within a range of five segments (two segments before and two segments after it, totaling five segments), if the gradient difference magnitude of point k is greater than the gradient difference magnitudes of the other four points, then point k satisfies the local extremum condition. Candidate mutation points that simultaneously satisfy the condition of having a gradient difference magnitude exceeding a preset difference threshold and possessing the maximum gradient difference magnitude within the preset neighborhood are ultimately identified as anomalous mutation points.

[0049] The smallest line segment containing an anomalous mutation point is identified as a line segment where the measured spatial entropy value differs significantly from the theoretical spatial entropy value. The determination of the smallest line segment is based on the spatial location of the anomalous mutation point; for example, if the anomalous mutation point is located in a specific section, that section is designated as the smallest line segment. If multiple adjacent segments contain anomalous mutation points, these consecutive segments are merged into a single smallest line segment. The merging condition is based on segment continuity; for example, merging occurs when the difference in adjacent segment numbers is less than or equal to 2. The start and end positions of line segments identified as having significant differences are recorded. These segments represent areas where the measured spatial entropy value differs greatly from the theoretical spatial entropy value, potentially indicating the location of a fault.

[0050] The specific implementation of selecting points that simultaneously satisfy the local extremum condition from candidate mutation points as anomalous mutation points involves identifying candidate mutation points with the maximum gradient difference amplitude within a preset neighborhood of the spatial gradient difference sequence as satisfying the local extremum condition. The size of the preset neighborhood is adjusted according to the line characteristics and detection requirements. For example, for scenarios with high accuracy requirements, the preset neighborhood is set to 3 segments; for scenarios with high anti-interference requirements, the preset neighborhood is set to 5 segments. The determination of the maximum gradient difference amplitude is achieved by comparing the gradient difference amplitudes of all points within the preset neighborhood using a numerical comparison algorithm, such as iterating through comparisons to find the maximum value. The entire selection process is executed automatically by a computer program, which iterates through each candidate mutation point and checks the extremum condition within its preset neighborhood to ensure that all points meeting the condition are identified as anomalous mutation points. The spatial location information of anomalous mutation points is used for subsequent fault location analysis, providing accurate indication of abnormal areas.

[0051] S6. Based on the spatial location of line sections with significant differences, determine the final location of the grounding fault point. The specific implementation is as follows: The process of determining the final location of a ground fault point based on the spatial location of line segments with significant differences first includes acquiring the spatial location information of these line segments. This spatial location information comes from the location data of line segments whose measured spatial entropy values ​​differ significantly from their theoretical spatial entropy values, as determined in the preceding steps. This data includes the coordinates of the starting and ending points of the line segments, expressed in a line coordinate system. For example, a Cartesian coordinate system is established along the line route with the substation outgoing line point as the origin. The spatial location information is acquired by reading stored line segment location records. These records contain the x and y coordinates of the starting and ending points of each segment in the coordinate system, such as the starting point coordinates (x1, y1) and the ending point coordinates (x2, y2). For multiple line segments with significant differences, the spatial location information of each segment is acquired separately and stored in order of segment number.

[0052] The geometric center coordinates of line segments with significant differences are determined based on their spatial location information. The calculation of the geometric center coordinates is based on the starting and ending coordinates of the line segment, using the midpoint formula. For example, for a line segment with significant differences, the x-coordinate of its geometric center point is equal to the sum of the starting and ending x-coordinates divided by 2, and the y-coordinate is equal to the sum of the starting and ending y-coordinates divided by 2. When multiple line segments with significant differences exist, the geometric center coordinates of each segment are calculated independently. These coordinates represent the center position of each anomalous segment. The calculated geometric center coordinates are stored in coordinate pairs, such as (x3, y3), and associated with the corresponding line segment number.

[0053] The geometric center point coordinates are corrected based on the distribution characteristics of anomalous mutation points within a line segment exhibiting significant differences. These characteristics include the locational density of anomalous mutation points within the line segment and the magnitude of their gradient differences, derived from the anomalous mutation point data identified in the preceding steps. The correction process first calculates weight coefficients for the anomalous mutation points, determined by their gradient difference magnitudes; for example, a higher gradient difference magnitude assigns a higher weight coefficient. Then, a weighted average position is calculated based on these weight coefficients. For instance, the spatial coordinates of each anomalous mutation point are multiplied by its weight coefficient, summed, and then divided by the sum of all weight coefficients to obtain the corrected geometric center point coordinates. The correction amount is adjusted based on the distribution density of the anomalous mutation points, obtained by calculating the number of anomalous mutation points per unit length. For example, a higher correction amount is applied when the distribution is dense, and a lower correction amount is applied when the distribution is sparse. The threshold for distribution density is determined by statistically analyzing historical distribution density data under normal and fault conditions; for example, the 80th percentile of the historical distribution density is used as the threshold.

[0054] The specific setting of the weight coefficients is achieved by analyzing the correlation between gradient difference amplitude and failure probability in historical failure data. Specifically, this includes collecting historical failure case data, extracting the gradient difference amplitude sequence of abnormal mutation points and the corresponding failure point location information in each case, and establishing a statistical relationship model between gradient difference amplitude and failure probability. For example, a linear regression method is used, with gradient difference amplitude as the independent variable and failure probability as the dependent variable, to fit the weight coefficient function. The normalization process uses the minimum-maximum normalization method to map the gradient difference amplitude to the range of 0 to 1. That is, the normalized weight is equal to the current amplitude minus the historical minimum amplitude and divided by the difference between the historical maximum and minimum amplitude, thereby ensuring that the weight coefficients are positively correlated with the failure probability and that the value range is standardized.

[0055] The spatial location corresponding to the corrected geometric center point coordinates is determined as the final location of the grounding fault. The corrected geometric center point coordinates are mapped to the actual line location through coordinate transformation. This transformation is based on the correspondence between the line coordinate system and the actual geographic coordinate system, for example, through preset coordinate transformation parameters obtained through on-site measurement and calibration. The final location result includes the line distance to the fault point and the actual geographic coordinates. The line distance is calculated as the straight-line distance from the corrected geometric center point coordinates to the origin, and the actual geographic coordinates are calculated using the coordinate transformation formula. The final location result is output in a standard format, containing the line number, fault point distance, and geographic coordinate information, for subsequent fault handling and maintenance location.

[0056] The specific implementation of correcting the geometric center point coordinates based on the distribution characteristics of anomalous mutation points within line sections with significant differences includes analyzing the spatial distribution patterns of these anomalous mutation points. The spatial distribution pattern is assessed by calculating the uniformity of the distribution of anomalous mutation points within the line section. The uniformity is calculated using the coefficient of variation in statistics, which is the standard deviation of the anomalous mutation point locations divided by the mean. The threshold for uniformity is determined by analyzing the statistical distribution of uniformity in historical fault data. For example, the mean of uniformity in historical data plus one standard deviation can be used as the threshold benchmark. When the uniformity is below the threshold, it indicates that the anomalous mutation points are concentrated, and the correction direction points towards the densely distributed areas. When the uniformity is above the threshold, it indicates that the anomalous mutation points are dispersed, and the correction amount is reduced accordingly. The correction coefficients are determined by ranking the gradient difference magnitudes of anomalous mutation points. For example, anomalous mutation points in the top 50 percentiles of gradient difference magnitude are assigned higher weights, while those in the bottom 50 percentiles are assigned lower weights. The weights are linearly distributed based on the magnitude of the gradient difference; for instance, the anomalous mutation point with the largest gradient difference magnitude has a weight of 1, and the one with the smallest gradient difference magnitude has a weight of 0.5. Intermediate values ​​are determined through linear interpolation. The entire correction process is implemented through iterative calculations. After each iteration, the stability of the correction results is checked. Iteration stops when the coordinate change between two consecutive iterations is less than a preset tolerance. The preset tolerance is set according to the positioning accuracy requirements; for example, if the positioning accuracy requirement is 1 meter, the preset tolerance is set to 0.1 meters to ensure the accuracy of the correction results. The corrected geometric center point coordinates serve as the core basis for the final fault location result, providing more accurate fault location information.

[0057] Example 2: Figure 2 A schematic diagram of the arc suppression coil grounding fault location system based on a substation line protection device is provided. The system includes: The signal injection module is used to inject a current signal of a specific frequency between the faulty phase and the ground through a signal injection device after a single-phase ground fault is detected. The coefficient calculation module is used to acquire the current signal collected by the line protection device, and calculate the phase coherence coefficient by analyzing the phase relationship between the fundamental component and the main harmonic components in the corresponding current signal. The reference acquisition module is used to establish a distributed parameter model of the faulty line and obtain theoretical reference information including the distribution data of the line length and direction parameters. The sequence generation module is used to calculate the measured spatial entropy value based on the distribution of the phase coherence coefficient along the line length, and compare and analyze the measured spatial entropy value with the theoretical spatial entropy value calculated based on theoretical reference information to generate a spatial gradient sequence of the measured spatial entropy value and the theoretical spatial entropy value. The segment determination module is used to determine the line segments identified as abnormal mutation points in the spatial gradient sequence as line segments whose measured spatial entropy values ​​differ significantly from the theoretical spatial entropy values. The result determination module is used to determine the final location of the grounding fault point based on the spatial location of line sections with significant differences.

[0058] The system's data interaction among modules adopts a layered architecture. The data bus uses the industrial Ethernet protocol, defining a unified data frame format including timestamps, data type identifiers, and check bits. The signal injection module is equipped with a signal quality monitoring unit, which monitors the amplitude stability and frequency accuracy of the injected signal in real time. When the signal quality is substandard, a re-injection mechanism is automatically triggered. The coefficient calculation module has a built-in digital filter bank, which uses multi-rate signal processing technology to perform graded filtering on the acquired current signal to ensure accurate separation of the fundamental and harmonic components. The reference acquisition module integrates a model verification unit, which automatically calibrates model parameters by comparing simulation results with historical normal data. The sequence generation module adopts a parallel computing architecture, processing both measured and theoretical data simultaneously, and ensuring data synchronization through a caching mechanism. The segment determination module sets up a multi-level threshold judgment mechanism, combining gradient amplitude and rate of change as dual indicators to identify abnormal mutation points. The result determination module is equipped with a positioning result verification unit, which evaluates the confidence level of the positioning results output by cross-validating different algorithms. The status monitoring units of all modules adopt a heartbeat mechanism, periodically sending operation status messages to the central controller. The central controller dynamically adjusts the task scheduling strategy according to the status of each module, ensuring that the system can still operate in a degraded manner when some modules are abnormal.

[0059] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0060] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0061] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0062] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0063] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0064] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0065] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0066] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0067] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0068] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for arc suppression coil grounding fault location based on substation line protection device, characterized in that, Comprise: S1, after detecting single-phase ground fault, injecting current signal of specific frequency between fault phase and ground through signal injection device; S2, obtaining current signal collected by line protection device, calculating phase coherence coefficient by analyzing phase relationship between fundamental component and main harmonic component in corresponding current signal; S3, establishing distribution parameter model of fault line, obtaining theoretical reference information containing line length direction parameter distribution data; S4, calculating measured spatial entropy value according to distribution of phase coherence coefficient along line length, comparing and analyzing measured spatial entropy value with theoretical spatial entropy value calculated based on theoretical reference information, generating spatial gradient sequence of measured spatial entropy value and theoretical spatial entropy value; S5, judging line section identified as abnormal mutation point in spatial gradient sequence as line section with significant difference between measured spatial entropy value and theoretical spatial entropy value; S6, determining final positioning result of ground fault point according to spatial position of line section with significant difference.

2. The method of arc suppression coil ground fault location based on substation line protection device according to claim 1, characterized in that, After detecting single-phase ground fault, injecting current signal of specific frequency between fault phase and ground through signal injection device, comprising: Determining injection parameters of current signal of specific frequency, injection parameters including injection frequency value and injection current amplitude; Injecting current signal of specific frequency between fault phase and ground through signal injection device according to injection parameters; Monitoring current signal waveform in injection process, and calculating current amplitude stability and frequency purity based on current signal waveform; When current amplitude stability is lower than preset stability threshold or frequency purity is lower than preset purity threshold, adjusting injection parameters and re-executing injection step until current amplitude stability is greater than or equal to preset stability threshold and frequency purity is greater than or equal to preset purity threshold.

3. The method of arc suppression coil ground fault location based on substation line protection device according to claim 1, characterized in that, Obtaining current signal collected by line protection device, calculating phase coherence coefficient by analyzing phase relationship between fundamental component and main harmonic component in corresponding current signal, comprising: Separating fundamental component and main harmonic component from current signal collected by line protection device; Extracting instantaneous phase of fundamental component and instantaneous phase of main harmonic component; Calculating instantaneous phase difference between fundamental component and each main harmonic component; Calculating phase coherence coefficient based on instantaneous phase difference sequence, wherein phase coherence coefficient reflects stability of phase relationship between fundamental component and main harmonic component.

4. The method of arc suppression coil ground fault location based on substation line protection device according to claim 3, characterized in that, Calculating phase coherence coefficient based on instantaneous phase difference sequence comprises: calculating statistical variance of instantaneous phase difference sequence, and normalizing inverse of statistical variance to obtain phase coherence coefficient.

5. The method of arc suppression coil ground fault location based on substation line protection device of claim 1, wherein, Establishing distribution parameter model of fault line, obtaining theoretical reference information containing line length direction parameter distribution data, comprising: Determining resistance, inductance, capacitance and ground admittance per unit length of line based on structural parameters and material characteristics of fault line; Establishing distribution parameter model of fault line based on resistance, inductance, capacitance and ground admittance per unit length of line; Calculating theoretical phase coherence coefficient of injected specific frequency current signal propagating along line length based on distribution parameter model; Taking distribution of theoretical phase coherence coefficient along line length as a part of theoretical reference information.

6. The method of arc suppression coil ground fault location based on substation line protection apparatus as claimed in claim 1 wherein, The measured spatial entropy value is calculated according to the distribution of the phase coherence coefficient along the length of the line, and the measured spatial entropy value is compared with the theoretical spatial entropy value calculated based on the theoretical reference information to generate a spatial gradient sequence of the measured spatial entropy value and the theoretical spatial entropy value, including: The measured spatial entropy value sequence is calculated based on the distribution of the phase coherence coefficient along the length of the line, and the measured spatial entropy value sequence is obtained by applying the Shannon entropy formula to the distribution of the phase coherence coefficient along the length of the line; The theoretical spatial entropy value sequence is calculated based on the distribution of the theoretical phase coherence coefficient along the length of the line in the theoretical reference information, and the theoretical spatial entropy value sequence is obtained by applying the same Shannon entropy formula to the distribution of the theoretical phase coherence coefficient along the length of the line; The first-order difference sequence of the measured spatial entropy value sequence along the length direction of the line is calculated as the spatial gradient sequence of the measured spatial entropy value; The first-order difference sequence of the theoretical spatial entropy value sequence along the length direction of the line is calculated as the spatial gradient sequence of the theoretical spatial entropy value.

7. The method of arc suppression coil ground fault location based on substation line protection device according to claim 6, characterized in that, The first-order difference sequence of the measured spatial entropy value sequence along the length direction of the line is calculated as the spatial gradient sequence of the measured spatial entropy value, including: calculating the difference value of the measured spatial entropy value of adjacent sections in the length direction of the line, and arranging the difference value in sequence to form the spatial gradient sequence.

8. The method of arc suppression coil ground fault location based on substation line protection apparatus as claimed in claim 1 wherein, The line section identified as an abnormal mutation point in the spatial gradient sequence is determined as a line section with a significant difference between the measured spatial entropy value and the theoretical spatial entropy value, including: The gradient difference between corresponding points in the spatial gradient sequence of the measured spatial entropy value and the spatial gradient sequence of the theoretical spatial entropy value is calculated to form a spatial gradient difference sequence; The gradient difference amplitude of each point in the spatial gradient difference sequence is calculated; Candidate mutation points with gradient difference amplitudes exceeding a preset difference threshold are identified; Points that meet the local extreme condition at the same time are selected from the candidate mutation points as abnormal mutation points; The smallest line section containing the abnormal mutation point is determined as a line section with a significant difference between the measured spatial entropy value and the theoretical spatial entropy value.

9. The method of arc suppression coil ground fault location based on substation line protection device according to claim 8, characterized in that, The points that meet the local extreme condition at the same time are selected from the candidate mutation points as abnormal mutation points, including: within a preset neighborhood range of the spatial gradient difference sequence, the candidate mutation point with the maximum gradient difference amplitude is determined as the one that meets the local extreme condition.

10. The method of arc suppression coil ground fault location based on substation line protection apparatus as claimed in claim 1 wherein, According to the spatial position of the line section with a significant difference, the final positioning result of the grounding fault point is determined, including: The spatial position information of the line section with a significant difference is obtained; The geometric center point coordinates of the corresponding line section are determined according to the spatial position information of the line section with a significant difference; The geometric center point coordinates are corrected based on the distribution characteristics of the abnormal mutation points in the line section with a significant difference; The spatial position corresponding to the corrected geometric center point coordinates is determined as the final positioning result of the grounding fault point.

Images