A method for adaptive isolation protection of excitation circuit ground fault positioning of a hydroelectric generating set

Abstract

The application discloses a kind of hydroelectric generating set excitation loop grounding fault positioning adaptive isolation protection methods, it is related to the technical field of relay protection of hydroelectric generating set, including: the electrical signal of each segmented monitoring point of excitation loop is collected, and detection signal is injected between excitation loop and ground to obtain ground resistance value;According to the electrical signal and ground resistance value, the fault characteristic quantity for representing the nature of fault is extracted;According to the ground resistance value and fault characteristic quantity, the type of ground fault is identified;When the type of ground fault is identified, the fault location is determined by a variety of positioning means according to the electrical signal;According to the type of ground fault and the fault position, adaptive isolation protection action is performed.The application realizes the fine identification and differential treatment of the type of excitation loop ground fault, and issues an effective early warning when insulation deterioration is still in the early stage.

Description

Technical Field

[0001] This invention relates to the field of relay protection technology for hydroelectric generator sets, specifically a method for adaptive isolation protection of ground fault location in the excitation circuit of a hydroelectric generator set. Background Technology

[0002] As a crucial power source in the power system, the safe and stable operation of hydropower units directly impacts the reliability of the power grid. The excitation circuit is a core component of the hydropower generator unit, responsible for providing excitation current to the generator rotor and regulating terminal voltage and reactive power. With the continuous increase in the capacity of individual hydropower units and the sustained improvement in excitation voltage levels, the insulation of the excitation circuit to ground faces higher electrical and thermal stresses, leading to a higher probability of grounding faults. Currently, excitation circuit grounding protection mainly adopts a protection principle based on switching logic. It determines whether a grounding fault has occurred by detecting changes in the excitation circuit's voltage to ground and combining this with switching tests. Conventional protection schemes include two types: bridge-type grounding protection and injection-type grounding protection. Bridge-type protection detects grounding resistance using an unbalanced bridge, while injection-type protection determines the insulation condition by injecting a low-frequency signal between the excitation circuit and ground and measuring leakage current. These protection schemes can identify single-point and two-point grounding faults and execute tripping or alarm actions upon fault detection.

[0003] However, the existing technical solutions still have the following significant drawbacks: First, the existing solutions are relatively crude in their identification of fault types, only able to distinguish between "single-point grounding" and "two-point grounding," failing to finely identify the physical properties of the fault. That is, they cannot distinguish between fault types with different hazard mechanisms and development trends, such as metallic grounding, high-resistance grounding, and arcing grounding. This leads to a simplistic protection strategy, using a uniform tripping or alarm handling method regardless of the fault type. Second, due to the lack of fine-grained fault type identification, the existing protection strategies cannot take differentiated measures based on the specific nature of the fault. For metallic grounding faults with large current and urgent hazards, the protection action may not be fast enough, leading to further equipment damage. For high-resistance grounding faults with weak current but potentially long-term existence, direct tripping may cause unnecessary unplanned shutdowns, affecting power generation efficiency. Third, the existing protection schemes lack the ability to predict the evolution trend of faults and cannot issue effective warnings when insulation degradation is still in its early stages. They can only respond after the fault actually occurs, failing to achieve the transformation from "post-event action" to "pre-event warning." Therefore, how to overcome the shortcomings of existing technologies, such as crude fault type identification, simplistic protection strategies, and lack of fault evolution trend prediction capabilities, and achieve refined identification and differentiated processing of excitation circuit grounding fault types, and issue effective early warnings when insulation degradation is still in its early stages, is a problem we need to solve. To this end, we now provide an adaptive isolation protection method for locating grounding faults in the excitation circuit of hydropower units. Summary of the Invention

[0004] The purpose of this invention is to provide an adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator.

[0005] The objective of this invention can be achieved through the following technical solution: A method for adaptive isolation protection of ground fault location in the excitation circuit of a hydroelectric generator, comprising:

[0006] The electrical signals of each monitoring point in the excitation circuit are collected, and a detection signal is injected between the excitation circuit and the ground. The grounding resistance value is obtained based on the leakage response of the detection signal.

[0007] Based on the electrical signal and the grounding resistance value, extract fault characteristic quantities to characterize the nature of the fault;

[0008] The grounding fault type is identified based on the grounding resistance value and the fault characteristic quantity;

[0009] When the ground fault type is identified, the fault location is determined based on the electrical signal;

[0010] Based on the ground fault type and the fault location, an adaptive isolation protection action is performed.

[0011] Furthermore, the electrical signals include the voltage signal to ground and the branch current signal at each monitoring point.

[0012] Furthermore, the process of injecting a detection signal between the excitation circuit and ground, and obtaining the grounding resistance value based on the leakage response of the detection signal, includes:

[0013] A low-frequency AC signal is injected between the excitation circuit and ground as the detection signal;

[0014] The leakage current of the detection signal leaking to ground through the excitation circuit is collected;

[0015] The time-domain signal of the leakage current is sampled to obtain a discrete sampling sequence. The discrete sampling sequence is then subjected to a fast Fourier transform to obtain the effective value of the fundamental component of the leakage current, the amplitude of the fundamental component of the leakage current, and the amplitude of each harmonic component.

[0016] The effective voltage value of the detection signal is obtained, and the grounding resistance value is obtained by combining the effective voltage value of the detection signal with the effective value of the fundamental component of the leakage current according to Ohm's law.

[0017] Furthermore, the process of extracting fault characteristic quantities to characterize the nature of the fault based on the electrical signal and the grounding resistance value includes:

[0018] The time-domain signal of the leakage current is sampled to obtain a discrete sampling sequence. A fast Fourier transform is performed on the discrete sampling sequence to obtain the amplitude of the fundamental component and each harmonic component of the leakage current. The total harmonic distortion rate is obtained based on the amplitude of the fundamental component and each harmonic component.

[0019] The waveform of the leakage current is analyzed in the time domain to obtain the waveform distortion coefficient;

[0020] Within a preset time window, the grounding resistance value is collected by an equal-interval sampling method to obtain the grounding resistance value of multiple sampling points. The average value and standard deviation of the grounding resistance value of the multiple sampling points are obtained, and the grounding resistance fluctuation rate is obtained through the average value and standard deviation.

[0021] The ultra-high frequency electromagnetic wave signal generated by the insulation degradation of the excitation circuit is collected.

[0022] Furthermore, the process of identifying the ground fault type based on the grounding resistance value and the fault characteristic quantity is as follows:

[0023] The grounding resistance value, total harmonic distortion rate, waveform distortion coefficient, grounding resistance fluctuation rate, and ultra-high frequency partial discharge signal are comprehensively judged according to the preset judgment logic to identify the grounding fault type.

[0024] The grounding fault types include metallic grounding faults, high-resistance grounding faults, arcing grounding faults, and insulation degradation warnings.

[0025] Furthermore, the process of determining the fault location using multiple positioning methods based on the electrical signals includes:

[0026] Based on the branch current signal and the ground voltage signal, the fault location is calculated using the injection signal method to obtain the location result of the injection signal method.

[0027] Based on the branch current signal, the fault location is calculated using differential current analysis, and the location result is obtained by differential current analysis.

[0028] The position result of the injection signal method is compared with the position result of the differential current analysis method. If the deviation between the two is less than or equal to a preset deviation threshold, the position result of the injection signal method or the position result of the differential current analysis method is confirmed as the final fault location.

[0029] If the deviation between the two is greater than the preset deviation threshold, then the position result of the injection signal method and the position result of the differential flow analysis method are weighted and averaged, and the weighted average value is taken as the final fault location.

[0030] Furthermore, the process of performing adaptive isolation protection actions based on the ground fault type and the fault location includes:

[0031] When the ground fault type is a metallic ground fault, disconnect the branch where the fault location is located;

[0032] When the ground fault type is the high-resistance ground fault, a current-limiting resistor is connected to the branch where the fault location is located, and the leakage current is limited to below a preset safe current threshold by the current-limiting resistor.

[0033] When the ground fault type is an arcing ground fault, after detecting the arc characteristics, observe for a preset number of power frequency cycles. If the arc extinguishes itself within the preset number of power frequency cycles and does not reignite, then no isolation action is performed. If the arc continues to burn for more than the preset number of power frequency cycles, or if the number of reignitions within the preset number of power frequency cycles exceeds a preset threshold, then the arc extinguishing device is activated or the branch where the fault is located is disconnected.

[0034] When the ground fault type is the insulation degradation warning, only a warning signal is issued and no isolation action is performed.

[0035] Furthermore, a method for adaptive isolation protection of ground fault location in the excitation circuit of a hydroelectric generator is characterized by further comprising:

[0036] When two grounding points exist simultaneously, the branch where the fault location is located is cut off and a two-point grounding shutdown signal is issued.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] This invention extracts multiple fault characteristic quantities to characterize the nature of the fault and combines them with the grounding impedance parameter and the multiple fault characteristic quantities according to a preset judgment logic to achieve refined classification and diagnosis of metallic grounding, high-resistance grounding, arcing grounding, and insulation degradation early warning. It overcomes the shortcomings of existing technologies that can only distinguish between "single-point grounding" and "two-point grounding" but cannot identify the physical nature of the fault, filling the technical gap in the refined identification of excitation circuit grounding fault types. By executing corresponding adaptive isolation protection actions according to the identified grounding fault type, it achieves rapid disconnection of metallic grounding faults, current-limiting operation of high-resistance grounding faults, intelligent arc suppression of arcing grounding faults, and early warning of insulation degradation. It changes the existing technology's "one-size-fits-all" uniform tripping or alarm handling mode, minimizing unplanned downtime while ensuring equipment safety. At the same time, the insulation degradation early warning function extends the protection starting point from "fault-after-action" to "fault-before-warning", providing maintenance personnel with a time window for scheduled maintenance and realizing a shift from passive response to proactive management. This invention realizes the transformation from coarse response to refined diagnosis of grounding faults in excitation circuits, the extension from post-fault action to pre-fault early warning, and the upgrade from single trip protection to hierarchical adaptive protection. It solves the problems of coarse fault type identification, single protection strategy, and lack of fault evolution trend prediction capability in the prior art. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0040] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0041] like Figure 1 As shown, an adaptive isolation protection method for locating ground faults in the excitation circuit of a hydroelectric generator unit includes:

[0042] The electrical signals of each monitoring point in the excitation circuit are collected, and a detection signal is injected between the excitation circuit and the ground. The grounding resistance value is obtained based on the leakage response of the detection signal.

[0043] Based on the electrical signal and the grounding resistance value, extract fault characteristic quantities to characterize the nature of the fault;

[0044] The grounding fault type is identified based on the grounding resistance value and the fault characteristic quantity;

[0045] When the ground fault type is identified, the fault location is determined based on the electrical signal;

[0046] Based on the ground fault type and the fault location, execute the adaptive isolation protection action corresponding to the ground fault type.

[0047] It should be further explained that, in the specific implementation process, the process of collecting electrical signals from each monitoring point of the excitation circuit includes:

[0048] In this embodiment, monitoring points are set up at the segment nodes of the excitation circuit. These segment nodes include the low-voltage side of the excitation transformer, the positive end of the rotor slip ring, the negative end of the rotor slip ring, and each tap of the excitation winding. Voltage transformers and current transformers are installed at each monitoring point to collect the voltage-to-ground signal and branch current signal at each monitoring point. The voltage-to-ground signal at each monitoring point is collected through the voltage transformers. ,in Indicates the monitoring point number. , The total number of monitoring points; the branch current signal of each monitoring point is collected through a current transformer. .

[0049] The ground voltage signal The sampling rate is denoted as the instantaneous voltage value of each monitoring point relative to the ground, collected at a frequency of no less than 256 points per cycle. Hz; the branch current signal The instantaneous current values ​​flowing through the branches where each segment monitoring point is located are obtained using a synchronous sampling method with the ground voltage signal.

[0050] It should be further explained that, in the specific implementation process, the process of injecting a detection signal between the excitation circuit and ground, and obtaining the grounding resistance value based on the leakage response of the detection signal, includes:

[0051] In this embodiment, a 20Hz low-frequency AC signal is injected between the excitation circuit and ground via a low-frequency signal injection device as a detection signal. The effective voltage value of the low-frequency AC signal is denoted as... Its waveform expression is ,in The frequency is a time variable; in addition, the choice of the 20Hz frequency takes into account two aspects: on the one hand, the frequency is much lower than the power frequency of 50Hz, which can effectively avoid aliasing interference with the power frequency signal; on the other hand, the frequency is high enough to make the coupling capacitor required by the signal injection device of moderate size, which is convenient for engineering implementation.

[0052] Furthermore, the leakage current of the detection signal leaking to ground through the excitation circuit is collected by a low-frequency signal injection device, and the leakage current is denoted as... The time-domain signal of the leakage current is sampled to obtain a discrete sampling sequence. ,in , To determine the number of points for the Fast Fourier Transform, this embodiment takes... The discrete sequence of the leakage current is subjected to a Fast Fourier Transform (FFT) to obtain the full-frequency spectrum; the output of the Fast Fourier Transform is a complex sequence. ( ),in This is the frequency index, and the corresponding frequency is... ,in For the sampling rate, in this embodiment... Hz; Extract the amplitude of the 20Hz frequency component from the full-frequency spectrum and divide by Hz. The effective value of the fundamental component of the leakage current was then obtained, denoted as . .

[0053] Furthermore, according to Ohm's law, the effective value of the detected signal voltage is... Divide by the effective value of the fundamental component of the leakage current The grounding resistance value is obtained. Its calculation expression is:

[0054] ;

[0055] Since the injected signal is a low-frequency AC signal, the capacitive reactance of the excitation circuit to ground at a frequency of 20Hz is... Larger, of which The capacitance to ground is negligible compared to the grounding resistance; therefore, the value calculated above can be considered a pure resistance value, i.e., the grounding resistance value. In this embodiment, the effective voltage value of the detection signal is acquired in real time by a voltage measurement unit inside the low-frequency signal injection device. The voltage measurement unit adopts a resistive voltage divider sensor. After the injection signal is emitted, the voltage measurement unit synchronously acquires the instantaneous voltage value at the injection point and obtains it after effective value calculation (RMS).

[0056] It should be further explained that, in the specific implementation process, the process of extracting fault characteristic quantities to characterize the nature of the fault based on the electrical signal and the grounding resistance value includes:

[0057] The total harmonic distortion rate is calculated based on the amplitude of the fundamental component and the amplitudes of each harmonic component of the leakage current. Specifically, the amplitude of the fundamental component, i.e., the corresponding frequency, is extracted from the same full-frequency spectrum mentioned above. The amplitude of Hz is denoted as Extract the amplitude of each harmonic component and record it as follows: The frequencies corresponding to each harmonic component are as follows: Hz, , The highest harmonic order is taken in this embodiment. ;

[0058] The total harmonic distortion rate is the ratio of the square root of the sum of the squares of the amplitudes of all harmonic components to the amplitude of the fundamental component, and its calculation expression is as follows:

[0059] ;

[0060] The total harmonic distortion (THD) is used to characterize the content of harmonic components in the leakage current waveform. When the ground fault is purely resistive, the leakage current waveform is close to a sine wave, and the THD value is small. When there are nonlinear phenomena such as electric arcs, the leakage current waveform contains abundant harmonic components, and the THD value increases significantly.

[0061] Furthermore, the process of performing time-domain analysis on the waveform of the leakage current to obtain the waveform distortion coefficient includes:

[0062] In this embodiment, leakage current Peak detection is performed on the time-domain waveform to obtain the maximum instantaneous value of the leakage current within one power frequency cycle, which is denoted as the peak value. Simultaneously, the effective value of the leakage current is calculated based on the sampling sequence, and denoted as... Its calculation expression is:

[0063] ;

[0064] in The number of sampling points within one power frequency cycle;

[0065] For a standard sine wave, the peak value and the RMS value satisfy the following relationship: The identity relationship; the peak value of the actual leakage current. Compared with the theoretical peak value of the sine wave Compare the waveforms and calculate the waveform distortion coefficients, denoted as... Its calculation expression is:

[0066] ;

[0067] When the leakage current waveform is a standard sine wave ,at this time When the leakage current waveform is distorted, such as by a flattened top, the appearance of spikes, or discontinuities, the actual peak value deviates from the theoretical peak value. The absolute value increases; in this embodiment, we take The absolute value is used for subsequent judgment.

[0068] Furthermore, within a preset time window, the grounding resistance value is collected using an equally spaced sampling method to obtain the grounding resistance values ​​at multiple sampling points. The process of obtaining the average and standard deviation of the grounding resistance values ​​at these multiple sampling points, and then obtaining the grounding resistance fluctuation rate using the average and standard deviation, includes:

[0069] In this embodiment, the length of the preset time window is set. , One power frequency cycle; one power frequency cycle is 20ms, therefore The selection of this time window length is based on the fact that the alternation process of arcing and extinguishing in arcing ground faults usually occurs within a time scale of tens to hundreds of milliseconds. A 2-second time window can fully capture the intermittent characteristics of the arc without reducing the response speed of the protection due to an excessively long window.

[0070] Within the preset time window Within, the grounding resistance value is continuously collected according to the equal-interval sampling method. The sampling interval is denoted as... In this embodiment, we take ms means sampling twice within each power frequency cycle, thus obtaining the number of sampling points. The obtained grounding resistance value sequence is denoted as _ ... ;

[0071] First, calculate the average value of the grounding resistance within this time window, denoted as . Its calculation expression is:

[0072] ;

[0073] in For the first The grounding resistance value at each sampling point;

[0074] Then calculate the standard deviation of the grounding resistance value within this time window, denoted as . Its calculation expression is:

[0075] ;

[0076] Standard deviation Divide by average The fluctuation rate of the grounding resistance value within a preset time window is obtained, i.e., the grounding resistance fluctuation rate. ;

[0077] When the grounding resistance value remains stable within the time window, each sampling point All are close to the average value Standard deviation Volatility approaches 0 The standard deviation approaches 0; when the grounding resistance value fluctuates drastically, the standard deviation... Increase, volatility It increases accordingly.

[0078] Furthermore, the process of acquiring the ultra-high frequency electromagnetic wave signal generated by the insulation degradation of the excitation circuit includes:

[0079] In this embodiment, an ultra-high frequency (UHF) partial discharge sensor is used to collect UHF electromagnetic wave signals generated by insulation degradation in the excitation circuit. The UHF partial discharge sensor is installed on the casing of the excitation transformer and the grounding lead. The UHF partial discharge sensor operates in the frequency band of 300MHz to 1500MHz and can effectively receive UHF electromagnetic wave signals radiated when partial discharge occurs inside the insulating material.

[0080] When the insulation material of the excitation circuit develops local defects due to aging, moisture, or mechanical damage, a weak partial discharge will occur at the defect location under the influence of an electric field, radiating ultra-high frequency electromagnetic waves. The ultra-high frequency partial discharge sensor converts the received ultra-high frequency electromagnetic wave signal into a voltage signal, which is then amplified, filtered, and output to the central processing unit. When the signal amplitude output by the ultra-high frequency partial discharge sensor exceeds the preset partial discharge detection threshold... When the voltage reaches mV, it is determined that an ultra-high frequency partial discharge signal has been detected, denoted as . Otherwise, record as .

[0081] It should be further explained that, in the specific implementation process, the process of identifying the ground fault type based on the grounding resistance value and the fault characteristic quantity includes:

[0082] In this embodiment, the central processing unit obtains the grounding resistance value. Total Harmonic Distortion (THD) and Waveform Distortion Coefficient Volatility and UHF partial discharge signal The above five characteristic quantities are used as inputs, and a comprehensive judgment is made according to the preset judgment logic to identify the grounding fault type; the grounding fault type includes metallic grounding fault, high resistance grounding fault, arcing grounding fault and insulation degradation warning.

[0083] The specific judgment logic is as follows:

[0084] If the grounding resistance value And the total harmonic distortion rate And the waveform distortion coefficient And the grounding resistance fluctuation rate If so, it is determined to be a metallic grounding fault;

[0085] If the grounding resistance value And the total harmonic distortion rate The waveform distortion coefficients are both in to If the values ​​are between these values, it is determined to be a high-resistance grounding fault;

[0086] If the total harmonic distortion rate And the waveform distortion coefficient And the grounding resistance fluctuation rate Or the ultra-high frequency partial discharge signal is detected. If so, it is determined to be an arc-induced grounding fault;

[0087] If the grounding resistance value If the value shows a continuous downward trend, it is determined to be an insulation degradation warning; the continuous downward trend is the grounding resistance value. The weekly decline exceeded 5% over seven consecutive days.

[0088] The specific values ​​of the above thresholds are based on statistical analysis of the physical characteristics of different types of grounding faults; metallic grounding faults are characterized by low resistance, low distortion, and low fluctuation; high resistance grounding faults are characterized by medium to high resistance and medium distortion; arcing grounding faults are characterized by high distortion and high fluctuation; and insulation degradation is characterized by ultra-high resistance and a continuous downward trend.

[0089] It should be further explained that, in the specific implementation process, when the ground fault type is identified, the process of determining the fault location based on the electrical signal using multiple positioning methods includes:

[0090] In this embodiment, once the ground fault type is identified, i.e., it is determined to be any one of a metallic ground fault, a high-resistance ground fault, or an arcing ground fault, the fault location process is initiated; in particular, insulation degradation warning does not trigger the location process.

[0091] Fault location was determined by combining two independent methods: the injection signal method and the differential current analysis method.

[0092] Furthermore, the process of calculating the fault location using the injection signal method based on the branch current signal and the voltage signal to ground includes:

[0093] A low-frequency AC signal with a frequency of 20Hz is injected between the excitation circuit and the ground through a low-frequency signal injection device; the injected signal and the above-mentioned detection signal are from the same signal source, and are continuously injected during the fault location stage.

[0094] The amplitude attenuation and phase shift of the low-frequency AC signal at each segment monitoring point are measured. Specifically, the amplitude and phase of the injected signal at each monitoring point are obtained by performing Fourier analysis on the ground voltage signal collected at that monitoring point: the amplitude of the 20Hz frequency component in the ground voltage signal is extracted as the amplitude of the injected signal at that monitoring point. The phase angle of the 20Hz frequency component is extracted and used as the phase of the injected signal at the monitoring point. ; based on the amplitude at the injection point and phase For reference, the first Amplitude attenuation at each monitoring point Phase offset .

[0095] According to transmission line theory, during signal propagation on a conductor, both amplitude attenuation and phase shift are proportional to the propagation distance. Let the amplitude attenuation coefficient per unit length be... The phase offset coefficient per unit length is Then the first Distance from each monitoring point to the fault point satisfy:

[0096] ;

[0097] In this embodiment, and The data is pre-determined through calibration tests before leaving the factory and stored in the central processing unit.

[0098] By combining the distance calculation results of each segmented monitoring point, the position result of the injection signal method is obtained. Specifically, using the spatial position of each monitoring point as a known quantity, the calculated distances of each monitoring point are... Cross-comparison is performed to determine the segmented interval where the fault point is located, and this is recorded as the location result of the injection signal method. .

[0099] Furthermore, the process of calculating the fault location using differential current analysis based on the branch current signal includes:

[0100] For each branch containing a segment monitoring point, obtain the first current value at the beginning of the branch and the second current value at the end of the branch. For the first... The current at the beginning of each branch is denoted as . The terminal current is denoted as The first-end current and the last-end current are both acquired by current transformers installed at both ends of the branch.

[0101] Calculate the difference between the first current value and the second current value to obtain the differential current value of this branch. Its calculation formula is .

[0102] According to Kirchhoff's Current Law, under normal operating conditions, the current at the beginning of a branch is equal to the current at the end, and the differential current value is... The current approaches 0; when a ground fault occurs in a branch, some current leaks to ground through the fault point, causing the current at the beginning and the current at the end to be unequal, resulting in a difference in current value. Significantly increased.

[0103] Differential value Exceeding the preset differential current threshold The branch is identified as a faulty branch. The preset differential current threshold... Based on the measurement error of the current transformer and the normal leakage current level of the excitation circuit, this embodiment takes... ,in This is the rated current of the branch circuit.

[0104] Based on the location of the faulty branch, the location result obtained by differential current analysis is denoted as... Specifically, the spatial interval where the faulty branch is located is used as the location result of the differential flow analysis method.

[0105] Furthermore, the position results obtained using the injection signal method are obtained. Location results of sum and difference flow analysis Then, the two are compared. If the deviation between the two is less than or equal to a preset deviation threshold... If the location result obtained by the injection signal method or the differential current analysis method is confirmed as the final fault location, then the fault location is determined. ,Right now or In this embodiment The deviation is defined as ,in This is the total length of the excitation circuit;

[0106] If the deviation between the two is greater than the preset deviation threshold Then, a weighted average is calculated between the location results obtained by the injection signal method and the location results obtained by the differential current analysis method, and the weighted average is taken as the final fault location. The expression for the weighted average is:

[0107] ;

[0108] Wherein, 0.6 is the weight of the position result of the injection signal method and 0.4 is the weight of the position result of the differential current analysis method. The basis for setting this weight allocation is: the injection signal method directly measures the attenuation and phase shift of the signal on the propagation path, and is less affected by changes in the operating conditions of the excitation circuit, so the positioning accuracy is relatively high; the differential current analysis method depends on the measurement accuracy of the current transformer, and the measurement error may increase when the fault current is small, such as when the high resistance grounding is in the case of high resistance grounding, so it is given a lower weight.

[0109] It should be further explained that, in the specific implementation process, the process of executing adaptive isolation protection actions based on the ground fault type and the fault location includes:

[0110] In this embodiment, the central processing unit determines the ground fault type and the final fault location. It retrieves the corresponding protection action instructions from the preset protection strategy library and executes the corresponding isolation protection action through controllable switches and current-limiting resistor arrays.

[0111] Specifically, when the ground fault type is a metallic ground fault, the central processing unit sends a signal to the final fault location. The high-speed bypass switch of the branch circuit sends a tripping command, disconnecting the branch circuit within 20 milliseconds or less. The high-speed bypass switch uses a fast vacuum circuit breaker or power electronic switching device (such as IGBT), with a mechanical tripping time of no more than 5 milliseconds and a contact separation time of no more than 15 milliseconds from receiving the command. At the same time, the central processing unit issues an audible and visual alarm signal of "Metallic grounding fault - emergency isolation" to notify the operation and maintenance personnel.

[0112] When the ground fault type is a high-resistance ground fault, the central processing unit will send a signal to the final fault location within 100 milliseconds. A current-limiting resistor is connected in the branch circuit. The resistance value of the current-limiting resistor is denoted as... Based on the current grounding resistance value Dynamic selection, The current-limiting resistor is connected in parallel between this branch and ground to limit the fault current below a safe threshold, which is defined as follows: Ensure that the fault current does not exceed the withstand capability of the equipment insulation.

[0113] Meanwhile, the central processing unit continuously monitors the grounding resistance value. The changing trend. If it is found during the monitoring process. Continue to decrease until it is less than or equal to the first impedance threshold , If the fault is determined to have worsened into a metallic grounding fault, the protection strategy will be automatically upgraded to the isolation protection action corresponding to a metallic grounding fault, that is, the fault branch will be immediately disconnected; the central processing unit will issue a "high resistance grounding fault - current limiting isolation" alarm signal and recommend that maintenance be arranged in the near future.

[0114] When the ground fault type is an arcing ground fault, the central processing unit observes for 2 to 5 power frequency cycles (40ms to 100ms) after detecting arc characteristics. If the arc extinguishes itself without reigniting during the observation period, it is determined to be a transient arc fault, and only the event log is recorded without performing isolation actions. If the arc continues to burn beyond the observation window, or reignites more than 3 times within the observation window, it is determined to be a persistent arc fault, and the central processing unit activates the arc suppression device or disconnects the branch where the fault is located. The arc suppression device is a neutral point reactor. By connecting the neutral point reactor between the neutral point of the excitation circuit and ground, the impedance characteristics of the circuit are changed, suppressing the reignition of the arc. The central processing unit issues an "Arcing Ground Fault - Arc Suppression / Isolation" alarm signal.

[0115] When an insulation degradation warning is detected, the central processing unit only issues an "insulation degradation warning" signal, suggesting scheduled maintenance, without performing any isolation actions. The warning signal includes the fault location. Current grounding resistance value And its downward trend information.

[0116] Furthermore, when the central processing unit detects two grounding points simultaneously, regardless of the fault type at each point, it immediately performs full-loop rapid isolation: the central processing unit disconnects all faulty branches within 15 milliseconds and issues a "two-point grounding - emergency shutdown" signal. The criterion for determining two-point grounding is: after locating a fault location... Based on this, if another independent fault location is detected using the same location method... ,and and If the distance between them is greater than the preset minimum spacing threshold, which is 10% of the total length of the excitation circuit, then it is determined to be a two-point grounding fault.

[0117] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications or equivalent substitutions 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 scope of the present invention.

Claims

1. A method for adaptive isolation protection of grounding fault location in the excitation circuit of a hydroelectric generator, characterized in that, include: The electrical signals of each monitoring point in the excitation circuit are collected, and a detection signal is injected between the excitation circuit and the ground. The grounding resistance value is obtained based on the leakage response of the detection signal. Based on the electrical signal and the grounding resistance value, extract fault characteristic quantities to characterize the nature of the fault; The grounding fault type is identified based on the grounding resistance value and the fault characteristic quantity; When the ground fault type is identified, the fault location is determined based on the electrical signal; Based on the ground fault type and the fault location, an adaptive isolation protection action is performed.

2. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator unit according to claim 1, characterized in that, The electrical signals include the ground voltage signal and branch current signal at each monitoring point.

3. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator as described in claim 1, characterized in that, The process of injecting a detection signal between the excitation circuit and ground, and obtaining the grounding resistance value based on the leakage response of the detection signal, includes: A low-frequency AC signal is injected between the excitation circuit and ground as the detection signal; The leakage current of the detection signal leaking to ground through the excitation circuit is collected; The time-domain signal of the leakage current is sampled to obtain a discrete sampling sequence. The discrete sampling sequence is then subjected to a fast Fourier transform to obtain the effective value of the fundamental component of the leakage current, the amplitude of the fundamental component of the leakage current, and the amplitude of each harmonic component. The effective voltage value of the detection signal is obtained, and the grounding resistance value is obtained by combining the effective voltage value of the detection signal with the effective value of the fundamental component of the leakage current according to Ohm's law.

4. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator as described in claim 3, characterized in that, The process of extracting fault characteristic quantities to characterize the nature of the fault based on the electrical signal and the grounding resistance value includes: The total harmonic distortion rate is obtained based on the amplitude of the fundamental component and the amplitude of each harmonic component of the leakage current. The waveform of the leakage current is analyzed in the time domain to obtain the waveform distortion coefficient; Within a preset time window, the grounding resistance value is collected by an equal-interval sampling method to obtain the grounding resistance value of multiple sampling points. The average value and standard deviation of the grounding resistance value of the multiple sampling points are obtained, and the grounding resistance fluctuation rate is obtained through the average value and standard deviation. The ultra-high frequency electromagnetic wave signal generated by the insulation degradation of the excitation circuit is collected.

5. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydropower unit according to claim 4, characterized in that, The process of identifying the ground fault type based on the grounding resistance value and the fault characteristic quantity is as follows: The grounding resistance value, total harmonic distortion rate, waveform distortion coefficient, grounding resistance fluctuation rate, and ultra-high frequency partial discharge signal are comprehensively judged according to the preset judgment logic to identify the grounding fault type. The grounding fault types include metallic grounding faults, high-resistance grounding faults, arcing grounding faults, and insulation degradation warnings.

6. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator unit according to claim 2, characterized in that, The process of determining the fault location based on the electrical signals includes: Based on the branch current signal and the ground voltage signal, the fault location is calculated using the injection signal method to obtain the location result of the injection signal method. Based on the branch current signal, the fault location is calculated using differential current analysis, and the location result is obtained by differential current analysis. The position result of the injection signal method is compared with the position result of the differential current analysis method. If the deviation between the two is less than or equal to a preset deviation threshold, the position result of the injection signal method or the position result of the differential current analysis method is confirmed as the final fault location. If the deviation between the two is greater than the preset deviation threshold, then the position result of the injection signal method and the position result of the differential flow analysis method are weighted and averaged, and the weighted average value is taken as the final fault location.

7. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator as described in claim 5, characterized in that, The process of performing adaptive isolation protection actions based on the ground fault type and the fault location includes: When the ground fault type is a metallic ground fault, disconnect the branch where the fault location is located; When the ground fault type is the high-resistance ground fault, a current-limiting resistor is connected to the branch where the fault location is located, and the leakage current is limited to below a preset safe current threshold by the current-limiting resistor. When the ground fault type is an arcing ground fault, after detecting the arc characteristics, observe for a preset number of power frequency cycles. If the arc extinguishes itself within the preset number of power frequency cycles and does not reignite, then no isolation action is performed. If the arc continues to burn for more than the preset number of power frequency cycles, or if the number of reignitions within the preset number of power frequency cycles exceeds a preset threshold, then the arc extinguishing device is activated or the branch where the fault is located is disconnected. When the ground fault type is the insulation degradation warning, only a warning signal is issued and no isolation action is performed.

8. The adaptive isolation protection method for locating grounding faults in the excitation circuit of a hydroelectric generator unit according to claim 7, characterized in that, Also includes: When two grounding points exist simultaneously, the branch where the fault location is located is cut off and a two-point grounding shutdown signal is issued.

Images