Arc grounding overvoltage suppression method based on rapid transfer of fault grounding point

Abstract

The arc grounding overvoltage suppression method based on the quick transfer of the fault grounding point solves the problem of how to quickly and accurately suppress the arc grounding overvoltage, and belongs to the single-phase grounding fault detection field of the distribution network.The present application comprises: real-time monitoring of the three-phase voltage, three-phase current and neutral point signal of the distribution network, and extraction of high-frequency transient characteristics; if the arc grounding fault and the fault phase are determined according to the high-frequency transient characteristics, the fault type is further judged to be transient fault or permanent fault; if it is transient fault, the quick grounding switch of the fault phase is closed, the fault point is converted from unstable arc grounding to stable metallic grounding, and is opened after the adaptive time; if the arc reignites and exceeds the set number of times, it is determined to be permanent fault; if it is permanent fault, an alarm is given. The present application quickly converts the grounding fault into stable metallic grounding.

Description

Technical Field

[0001] This invention relates to a method for suppressing arc grounding overvoltage based on rapid transfer of fault grounding point, belonging to the field of single-phase grounding fault detection in distribution networks. Background Technology

[0002] Arcing grounding overvoltage is one of the most dangerous phenomena among single-phase grounding faults in distribution networks. With continuous economic and social development and ever-increasing electricity load, the safe and stable operation of the power system faces severe challenges. As a crucial link connecting the transmission network and users, the distribution network plays a vital role in power distribution. It is large in scale, has complex topology, and a high penetration rate of distributed power sources, resulting in a significantly higher frequency of faults compared to the transmission system. Statistics show that approximately 90% of power system faults occur at the distribution network level, with single-phase grounding faults accounting for over 80%. These faults are mostly caused by damage to line insulation, specifically including lightning strikes, external force damage, cable aging, equipment insulation deterioration, and construction errors.

[0003] In power distribution network operation, the grounding current generated by a single-phase grounding fault is usually small, and regulations allow the system to continue operating for 1-2 hours after the grounding. However, when the grounding current is small, unstable intermittent arcs are easily generated at the fault point, exhibiting periodic burning, extinguishing, and reignition phenomena, thus triggering arcing grounding overvoltages. Studies have shown that the temporary extinguishing time of the arc is about half a cycle of the power frequency. During this time, the insulation recovery rate at the fault point is often lower than the recovery voltage rise rate, leading to dielectric re-breakdown. Each arc reignition will excite strong oscillations of the system's electromagnetic energy, generating significant overvoltages in the non-faulty phases, the neutral point, and even the faulty phase itself. Operational data shows that arcing grounding overvoltages can reach up to 3.5 times the phase voltage, seriously threatening the insulation safety of equipment and the stable operation of the system.

[0004] To suppress arc-to-ground overvoltage, distribution networks widely adopt neutral point grounding via arc-suppression coils. This technology, by compensating for ground capacitance current and reducing the rate of rise of recovery voltage at the fault point, theoretically promotes arc extinction and reduces reignition. However, in practical applications, this method has significant limitations: First, the compensation effect of the arc-suppression coil is highly dependent on the matching degree between its reactance parameters and the system's ground capacitance. Changes in system operation and grid expansion can alter ground capacitance parameters, requiring continuous monitoring and adjustment; otherwise, insufficient compensation or even overcompensation may occur. Second, the arc-suppression coil prolongs the time from arc extinction to the recovery voltage reaching its peak, increasing the deionization time of the original arc channel, which may induce re-breakdown at the peak of the recovery voltage, thus increasing the overvoltage amplitude. Third, during normal operation, due to three-phase line asymmetry and incomplete transposition, the arc-suppression coil may raise the system neutral point potential, affecting power quality. Fourth, in abnormal situations such as circuit breaker non-full-phase operation or line breakage, the arc-suppression coil can exacerbate neutral point voltage shift, further threatening equipment insulation.

[0005] Furthermore, existing technologies cannot effectively distinguish between transient and permanent faults, lack adaptive control strategies for different fault types, and often delay fault handling due to repeated attempts to extinguish the arc, or even expand the scope of the accident. Summary of the Invention

[0006] To address the problem of how to quickly and accurately suppress arc grounding overvoltage, this invention provides a method for suppressing arc grounding overvoltage based on rapid transfer of the fault grounding point.

[0007] The present invention provides a method for suppressing arc grounding overvoltage based on rapid transfer of fault grounding point, comprising:

[0008] S1. Real-time monitoring of three-phase voltage, three-phase current and neutral point signal of distribution network, and extraction of high-frequency transient characteristics;

[0009] S2. Determine whether an arcing ground fault and the faulty phase have occurred based on the high-frequency transient characteristics. If an arcing ground fault is determined to have occurred, continue to determine whether the fault type is transient or permanent. If it is a transient fault, proceed to S3; if it is a permanent fault, proceed to S5.

[0010] S3. The fast grounding switch of the faulty phase is closed, changing the fault point from unstable arc grounding to stable metallic grounding, and then disconnecting after the adaptive time.

[0011] S4. Monitor whether the arc grounding fault has been eliminated. If the arc reignites, switch to S2 and count the number of times the fast grounding switch of the corresponding faulty phase is closed. If the number of times exceeds the set number, it is determined to be a permanent fault, an alarm is triggered, and fault isolation is initiated.

[0012] S5. Issue an alarm.

[0013] The beneficial effects of this invention are that the method of this invention can quickly transform a ground fault into a stable metallic ground, eliminate the conditions for arc grounding overvoltage when a single-phase ground fault occurs in the distribution network, thereby fundamentally eliminating arc grounding overvoltage, reducing the overvoltage multiple from a maximum of 3.5 times to less than 2 times, and thus eliminating the currently used arc suppression coil, completely solving the problem of arc grounding overvoltage and other problems caused by the use of arc suppression coil. Attached Figure Description

[0014] Figure 1 The waveform of the three-phase voltage during arc grounding overvoltage;

[0015] Figure 2 This is a schematic flowchart of the method of the present invention;

[0016] Figure 3 To convert the arc grounding overvoltage into a three-phase voltage waveform after metallic grounding. Detailed Implementation

[0017] 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.

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0019] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0020] When a phase of a line experiences a ground fault due to insulation failure, the fault point is often not stably conductive to ground, but rather generates an unstable, intermittent electric arc. The arc may temporarily extinguish when the fault current naturally crosses zero, but the charge already fully charged on the three-phase-to-ground capacitance, which cannot be discharged in time, will cause a significant DC offset potential at the neutral point. After about half a power frequency cycle, the recovery voltage on both sides of the fault point reaches its maximum due to the superposition of power supply voltage changes and neutral point potential offset. This is sufficient to break down the partially restored insulation air gap again, causing the arc to reignite. This repeated "extinguishing-reigniting" process triggers violent high-frequency electromagnetic energy oscillations between the line inductance and the ground capacitance. Simultaneously, each cycle further accumulates residual charge and neutral point offset potential, thereby generating persistent overvoltages with amplitudes far exceeding the normal operating voltage on both the healthy and faulty phases. (See attached...) Figure 1 It is known that after 1 / 4 of the power frequency cycle, the power supply charges the non-faulty phase to ground capacitance through the leakage reactance, generating high-frequency oscillations, which can cause the non-faulty phase to reach 2.5 times the overvoltage; after half a power frequency cycle, the ground current of the faulty phase crosses zero, the arc is extinguished, and the voltage of the faulty phase to ground gradually recovers; after another half a power frequency cycle, the recovered voltage of the faulty phase is as high as 2 times the overvoltage, the arc reignites, and high-frequency oscillations are formed again. During the transition process, the non-faulty phase can reach up to 3.5 times the overvoltage, which can cause serious damage to the personal safety of power grid operation and maintenance personnel.

[0021] The main process of this embodiment for suppressing arc grounding overvoltage faults in power distribution systems is as follows: It detects and identifies whether an arc grounding overvoltage fault has occurred in the power distribution network and the phase in which the fault occurs. An operation command is sent to the drive unit of the fast grounding switch. Driven by the drive unit, the fast grounding switch closes, completing the transfer of the single-phase grounding fault current from the fault point to the fast grounding switch device, thereby completely extinguishing the arc at the fault point and eliminating the conditions for generating arc grounding overvoltage. Furthermore, this embodiment also has a certain degree of automated logic judgment function. The fast grounding switch will open after being closed for a period of time. If the fault is not completely eliminated and the arc reignites, the fault signal will be detected again and the system will automatically activate, repeating the above process. If the fault is still not completely eliminated after a set number of repetitions, an alarm will be triggered and relevant personnel will be notified for handling. Specifically, the arc grounding overvoltage suppression method based on rapid transfer of the fault grounding point in this embodiment includes:

[0022] Step 1: Monitor the three-phase voltage, three-phase current and neutral point signal of the distribution network in real time, and extract high-frequency transient characteristics;

[0023] The three-phase voltage in this step is The three-phase current is The neutral point voltage is ;

[0024] High-frequency transient characteristics include high-frequency transient energy spectrum characteristics and arc nonlinear distortion characteristics. Zero-sequence voltage abrupt gradient and current zero-crossing decay characteristics; high-frequency transient energy spectrum characteristics include arc high-frequency energy. and three-phase voltage high-frequency energy asymmetry The current zero-crossing decay characteristic includes the decay time constant. and oscillation frequency ; Subscript , indicating three phases; specifically obtained in the following way:

[0025] (1) Voltage A 6-level wavelet packet decomposition was performed, resulting in 64 frequency bands (numbered 0-63). A set of frequency bands reflecting high-frequency arc oscillations was selected. (Corresponds to 5-50 kHz);

[0026] Phase voltage in frequency band The wavelet packet coefficients on are Calculate its energy:

[0027]

[0028] The high-frequency transient energy spectrum characteristics of each phase voltage are as follows:

[0029]

[0030] Three-phase voltage high-frequency energy asymmetry:

[0031]

[0032] (2) Pressure Perform FFT to obtain the spectrum Calculate the nonlinear distortion characteristics of the electric arc:

[0033]

[0034] in, , This is the effective value of the voltage;

[0035] (3) Calculate the zero-sequence voltage gradient:

[0036]

[0037] (4) For each phase current Detected past midnight Extract the high-frequency oscillation component within 1 ms after the zero crossing. (5-50 kHz bandpass filter), fitted exponential decay function:

[0038]

[0039] Obtain the decay time constant and oscillation frequency .

[0040] 14-dimensional feature vector of high-frequency transient features:

[0041] .

[0042] This step integrates 14-dimensional features, including high-frequency transient energy, distortion, zero-sequence mutation, and current zero-crossing decay, through multi-scale spatiotemporal feature fusion. This multi-feature approach reduces the risk of misjudgment based on a single feature.

[0043] Step 2: Determine whether an arcing ground fault has occurred and the faulty phase based on the high-frequency transient characteristics. If an arcing ground fault is confirmed, further determine whether the fault type is transient or permanent. If it is a transient fault, proceed to Step 3; if it is a permanent fault, proceed to Step 5. The method for determining whether an arcing ground fault has occurred based on the high-frequency transient characteristics in this step includes:

[0044] Construct a fault state time-series trajectory matrix based on the sequence of features in each dimension of the high-frequency transient features. And perform dynamic time-corrected distance matching with the preset arc grounding template; the dynamic time-corrected distance is:

[0045]

[0046]

[0047] in, This serves as a reference trajectory template for arc grounding. This is a normal trajectory template;

[0048] For each feature sequence of the trajectory matrix S, an approximate entropy is calculated, resulting in 14 approximate entropy values. Mean:

[0049]

[0050] Calculate the confidence level of arc grounding:

[0051]

[0052] in, Represents the intercept term. , , , , , For coefficients;

[0053] Fault determination based on thresholds:

[0054] like If the fault persists for three consecutive cycles, an arc grounding fault is determined to have occurred.

[0055] like Proceed to the initial review step;

[0056] Review steps:

[0057] If zero-sequence voltage voltage of the fault phase Inverted, and If the fault occurs within the range of 10-30 kHz, an arcing ground fault is determined to have occurred.

[0058] This step employs Dynamic Time Warping (DTW) matching and approximate entropy analysis to calculate the confidence level of arc grounding, and sets up a verification mechanism to improve accuracy. The confidence level mechanism improves the reliability of the judgment; it can complete fault identification within milliseconds, meeting the requirements of rapid action; when the confidence level is moderate, phase and frequency verification are used to further reduce misjudgment.

[0059] Furthermore, this step also includes: dynamically adjusting based on system status. , , , , , , Specifically, it includes:

[0060] Get current load ground capacitance current and distributed power penetration , as input variables of the fuzzy logic system;

[0061] Divide the 14-dimensional feature vector into 5 groups, with each group corresponding to an adjustment factor:

[0062] Corresponding adjustment factor ;

[0063] Corresponding adjustment factor ;

[0064] Corresponding adjustment factor ;

[0065] Corresponding adjustment factor ;

[0066] Corresponding adjustment factor ;

[0067] , , , and As an output variable of the fuzzy logic system;

[0068] Group weights , To adjust the number of times;

[0069] Fuzzy logic systems convert input variables into fuzzy sets; specifically, fuzzy sets such as load level can be used. {Low, Medium, High}, Capacitive Current {Small, Medium, Large}, Distributed Power Penetration Rate {Low, Medium, High}. Using fuzzy rules and the membership degrees of input variables, fuzzy rules can be formulated based on expert experience or data analysis. For example: if For high, For the great, If it is low, then For high, For low, For high, It is low.

[0070] Calculate the fuzzy set for each output variable and then convert it to an exact value, for example, using the centroid method, to obtain the adjustment factor. , , , and ;

[0071] Using adjustment factors , , , and Adjust group weights separately And update the adjusted group weights. ;

[0072] right Normalize and assign feature weights ;

[0073]

[0074] Based on the value, For the first The baseline scaling factor for each feature. This is the sensitivity function for the intercept term.

[0075] This step introduces a fuzzy logic system to dynamically adjust feature weights based on three system state variables: load level, ground capacitance current, and distributed power source penetration rate. By grouping adjustment factors and normalizing weight allocation, the algorithm automatically optimizes feature importance under different operating conditions. This step can automatically adjust according to the actual system state, adapting to different power grid structures and operating modes, avoiding performance degradation under different operating conditions with fixed weights, simulating expert experience, and achieving adaptive judgment similar to humans.

[0076] Furthermore, the sensitivity function for the intercept term in this step is:

[0077]

[0078] in, Represents the capacitance current to ground The membership degree of the corresponding fuzzy set is high. Indicates load It belongs to the corresponding fuzzy set with a low membership degree. Indicates the penetration rate of distributed power sources The membership degree of the corresponding fuzzy set can be set according to actual needs.

[0079] This step's intercept item The system is dynamically adjusted based on its state using a fuzzy membership function, reflecting the overall fault tendency of the system. Sensitivity is increased under high capacitive current and high DG penetration, and decreased under low load to prevent false triggering. This step makes the fault detection threshold adaptively change with the system state, making it more sensitive under fault-prone conditions and more conservative under stable conditions, avoiding false triggering or missed detection caused by a fixed threshold.

[0080] This embodiment of the method for determining whether a fault is a transient or permanent fault includes:

[0081] Obtaining the arc combustion continuity index Arc energy attenuation coefficient Voltage recovery rate ratio Equivalent insulation resistance Oscillation frequency drift Construct the fault type feature vector:

[0082]

[0083] The above parameters are obtained in the following manner:

[0084] (1) Observation period Internal statistics Time proportion:

[0085]

[0086] in This is an indicator function. Arc combustion continuity index. for:

[0087]

[0088] (2) For the faulty phase (assuming it is phase p) By performing exponential fitting, the arc energy attenuation coefficient is obtained. :

[0089]

[0090] (3) Fault phase voltage The ratio of the recovery rate after zero crossing to the recovery rate before the fault:

[0091]

[0092] The ratio of fault phase voltage to leakage current:

[0093]

[0094] The equivalent insulation resistance was calculated. .

[0095] The oscillation frequency drift is obtained based on the frequency changes of multiple zero-crossing events:

[0096]

[0097] This step, based on the arc evolution law and transient recovery characteristics, extracts features such as arc continuity, energy attenuation, and insulation recovery. The fault type feature vector is then input into a trained classifier, which outputs the fault type.

[0098] .

[0099] This step accurately identifies the nature of the fault, providing a basis for subsequent control strategies. Transient faults are quickly recovered from, and permanent faults are promptly alerted, improving processing efficiency. It also avoids repeated attempts to resolve permanent faults, shortening fault handling time.

[0100] Step 3: Close the fast grounding switch of the faulty phase to change the fault point from unstable arc grounding to stable metallic grounding, and disconnect it after waiting for the adaptive time.

[0101] In this step, the method of converting the fault point from an unstable arc grounding to a stable metallic grounding is achieved using a hybrid fast switching unit. Specifically, the hybrid fast switching unit includes a thyristor (SCR) valve group connected in reverse parallel, a high-speed mechanical switch (UFV), a triggering unit, and a synchronization monitoring unit;

[0102] The thyristor (SCR) valve assembly and the high-speed mechanical switch (UFV) are connected in parallel;

[0103] The triggering unit receives a grounding control command and generates a high-voltage pulse to trigger the reverse-parallel thyristor (SCR) valve group; after receiving the high-voltage pulse trigger, the reverse-parallel thyristor (SCR) valve group is forced to conduct, grounding the fault phase, and within 1-20 milliseconds, sends a closing command to the high-speed mechanical switch; causing the mechanical switch to close under near-zero voltage conditions and bear the subsequent steady-state fault current;

[0104] The synchronous monitoring unit monitors the voltage and current passing through the high-speed mechanical switch (UFV) in real time.

[0105] When the trigger unit receives a disconnect control command, it uses the voltage and current zero-crossing point prediction to issue a disconnect command to the high-speed mechanical switch at the predicted current zero-crossing point.

[0106] In this step, the hybrid fast switching unit is a composite switching structure combining power electronics and high-speed mechanics. Its core lies in utilizing the complementary characteristics of the two switching devices to achieve rapid forced switching of fault grounding and reliable long-term use. Its working principle can form a precise timing control closed loop:

[0107] First, microsecond-level forced conduction is achieved through a thyristor valve group. Upon detecting an arcing ground fault and issuing a grounding control command, the trigger unit immediately generates a high-voltage pulse, triggering the reverse-parallel thyristor (SCR) valve group. As a fully controllable semiconductor device, the SCR can switch from cutoff to full conduction within 10-100 microseconds after gate triggering. Once conduction occurs, a near-zero impedance path to ground is established for the faulty phase. The arc at the fault point is instantly extinguished by being short-circuited by the SCR path, and the unstable arcing ground is forcibly converted into a stable metallic ground. This process occurs within half a power frequency cycle (10ms), fundamentally eliminating the conditions for arc reignition and electromagnetic oscillation, thereby immediately suppressing arcing overvoltage.

[0108] Then, millisecond-level arc-free commutation is achieved by a high-speed mechanical switch. Within 1-20 milliseconds after the SCR turns on, the system issues a closing command to the high-speed mechanical switch (UFV). Since the fault current has already flowed through the SCR at this time, the voltage difference across the high-speed mechanical switch is only the on-state voltage drop of the SCR (usually only a few volts), which is close to zero voltage. Under this condition, closing the mechanical switch will not generate a pre-breakdown arc and will not cause electrical wear on the contacts. After the mechanical switch closes, because its on-state resistance is much lower than that of the SCR, the fault current will automatically and smoothly transfer from the SCR path to the mechanical switch path. Subsequently, the SCR will naturally turn off when the current crosses zero and stop working. At this point, the mechanical switch becomes the only path to carry the steady-state fault current.

[0109] Finally, the synchronous monitoring unit and the control unit work together to achieve intelligent zero-crossing disconnection. The synchronous monitoring unit continuously and accurately measures the current flowing through the mechanical switch and the voltage across its terminals. When the adaptive waiting time ends and a disconnection command is issued, the synchronous monitoring unit uses the monitored real-time current waveform and algorithms (such as linear extrapolation or model prediction) to accurately predict the next natural zero-crossing point of the current. The synchronous monitoring unit issues a disconnection command one inherent opening time of the mechanical switch before the predicted zero-crossing point, ensuring that the contacts separate precisely at the current zero-crossing point. This minimizes arc energy and even achieves arc-free disconnection, greatly improving the electrical life and disconnection reliability of the switch.

[0110] Furthermore, the hybrid fast switching unit of this application also includes an energy absorption unit, which includes an RC snubber circuit connected in parallel across the mechanical switch to suppress overvoltage during mechanical switch operation and dampen high-frequency oscillations.

[0111] The RC snubber circuit is mainly connected in parallel across the two ends of a high-speed mechanical switch (vacuum circuit breaker, UFV). Its function is to suppress and absorb harmful electromagnetic energy generated during switch operation, ensuring switch safety.

[0112] The method for determining the adaptive time in this step is as follows:

[0113] First, establish a historical database to obtain the ground capacitance current of the power grid fault system. The waiting time and whether the database was successfully restored within the waiting time;

[0114] Then, the system's ground capacitance current is acquired in real time. Using the formula:

[0115]

[0116] Determine the initial waiting time , This indicates the configurable safety factor (initial value can be set to 3~5). This represents a quantity related to the discharge time constant of the system's capacitance to ground, and... Inversely proportional; Represents the system phase voltage. It represents the power frequency angular frequency.

[0117] Then search the database for the same capacitance current to ground. The waiting time during a failure is adjusted by adjusting the configurable safety factor K to obtain the adjusted waiting time. ; as the final adaptive time.

[0118] In this embodiment, the waiting time mainly depends on the discharge rate of residual charge at the fault point, and this rate is related to the system's capacitance current to ground. Strong correlation The larger the size, the faster the discharge, and the shorter the waiting time.

[0119] Query the historical database to search for information related to the current situation. Fault records of similar or identical phases.

[0120] The value of K in the formula is dynamically adjusted based on the success / failure results of the most recent N operations:

[0121] If the current If a failure occurs in the three most recent records of similar and identical fault records, the K value is increased (by a relatively large margin, but not exceeding the upper limit of 8.0).

[0122] If the current Among similar and identical fault records, if the most recent 3 records are all successful and the overall success rate is greater than 95%, a decision is made based on the comparison between the current basic waiting time and the historical average waiting time to either significantly reduce the K value or slightly reduce the K value (but not below the lower limit of 1.5).

[0123] If the current If, among similar and identical fault records, the three most recent records are all successful but the overall success rate does not reach 95%, then the K value should be conservatively reduced (the reduction should be ΔK / 2, but not less than 2.0); ΔK represents the adjustment step size, which is determined based on the actual situation. If the number of successful failures is less than 3 in similar or identical fault records, the current K value will be maintained.

[0124] Step 4: Monitor whether the arc grounding fault has been eliminated. If the arc reignites, proceed to step 2 and count the number of times the fast grounding switch of the corresponding faulty phase is closed. If the number of times exceeds the set number, it is determined to be a permanent fault, an alarm is triggered, and fault isolation is initiated.

[0125] Step 5: Set an alarm.

[0126] Specific example: When an arcing ground fault occurs at a certain point in phase A of an ungrounded neutral system, the fault is determined to be in phase A based on high-frequency transient characteristics. According to internal logic, if the number of closing operations of the phase A hybrid fast switching unit is less than 3 (the initial value of the closing operation count is 0), a closing command is sent to the phase A hybrid fast switching unit. Simultaneously, the closing operation count is incremented by 1. Upon receiving the command, the phase A hybrid fast switching unit closes for the first time, converting the phase A arcing ground fault into a single-phase ground fault. This reduces the maximum overvoltage multiple of the non-faulty phases B and C from 3.5 times to 2 times, completely suppressing the arcing ground fault. The hybrid fast switching unit waits for the adaptive time to disconnect. If the arc does not reignite, the suppression is successful, and the internal logic counter is automatically reset to zero.

[0127] If the arc reignites, it indicates that the arc grounding overvoltage fault still exists. Based on the high-frequency transient characteristics, it is determined that a fault has occurred in phase A. According to the internal logic, the number of closing times of the phase A hybrid fast switching unit is less than 3 (at this time, the closing time value is 1). A closing command is sent to the phase A hybrid fast switching unit, and the closing time record is incremented by 1. After receiving the command, the phase A hybrid fast switching unit closes for the second time, converting the phase A arc grounding overvoltage fault into a single-phase grounding fault. The maximum overvoltage multiple of the non-faulty phases B and C is reduced from 3.5 times to 2 times, completely suppressing the generation of arc grounding overvoltage. The hybrid fast switching unit waits for the adaptive time to disconnect. If the arc does not reignite, it means that the suppression is successful, and the internal logic counter is automatically cleared to zero.

[0128] If the arc reignites, it indicates that the arc grounding overvoltage fault still exists. Based on the high-frequency transient characteristics, it is determined that a fault has occurred in phase A. According to the internal logic, the number of closing times of the phase A hybrid fast switch unit is less than 3 (at this time, the number of closing times is 2). A closing command is sent to the phase A hybrid fast switch unit, and the closing count record is incremented by 1. After receiving the command, the phase A hybrid fast switch unit closes for the third time, converting the phase A arc grounding overvoltage fault into a single-phase grounding fault. The maximum overvoltage multiple of the non-faulty phases B and C is reduced from 3.5 times to 2 times, completely suppressing the generation of arc grounding overvoltage. If the arc does not reignite, it means that the suppression is successful, and the internal logic counter is automatically cleared to zero.

[0129] If the arc reignites, it indicates that the arc grounding overvoltage fault still exists. Based on the high-frequency transient characteristics, a fault is determined to occur in phase A. According to the internal logic, the phase A hybrid fast switching unit has closed 3 times, indicating that the phase A arc grounding overvoltage fault is relatively severe and this application has failed to suppress it. An alarm mechanism is used to notify personnel that a phase A arc grounding overvoltage fault has occurred, requiring human intervention to prevent the phase A arc grounding overvoltage fault from escalating and causing more serious damage. Furthermore, it can alert relevant personnel in the distribution network to the presence of a phase A arc grounding overvoltage fault before the fault is resolved, preventing electric shock to relevant personnel. After the alarm is triggered, the internal logic counter is automatically reset to zero. In summary, this application can convert a phase A arc grounding overvoltage fault into a single-phase fault when it is not very severe, while reducing the maximum overvoltage multiple of the non-faulty phases B and C. It can also provide alarm processing when the phase A arc grounding overvoltage fault is severe, which has strong practical value and significance.

[0130] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for suppressing arc grounding overvoltage based on rapid transfer of fault grounding point, characterized in that, include: S1. Real-time monitoring of three-phase voltage, three-phase current and neutral point signal of distribution network, and extraction of high-frequency transient characteristics; S2. Determine whether an arcing ground fault and the faulty phase have occurred based on the high-frequency transient characteristics. If an arcing ground fault is determined to have occurred, continue to determine whether the fault type is transient or permanent. If it is a transient fault, proceed to S3; if it is a permanent fault, proceed to S5. S3. The fast grounding switch of the faulty phase is closed, changing the fault point from unstable arc grounding to stable metallic grounding, and then disconnecting after the adaptive time. S4. Monitor whether the arc grounding fault has been eliminated. If the arc reignites, switch to S2 and count the number of times the fast grounding switch of the corresponding faulty phase is closed. If the number of times exceeds the set number, it is determined to be a permanent fault, an alarm is triggered, and fault isolation is initiated. S5. Issue an alarm; High-frequency transient characteristics include high-frequency transient energy spectrum characteristics and arc nonlinear distortion characteristics. Zero-sequence voltage abrupt gradient and current zero-crossing decay characteristics; high-frequency transient energy spectrum characteristics include arc high-frequency energy. and three-phase voltage high-frequency energy asymmetry The current zero-crossing decay characteristic includes the decay time constant. and oscillation frequency ; Subscript , indicating three phases; 14-dimensional feature vector of high-frequency transient features: ； Methods for determining whether an arcing ground fault has occurred based on high-frequency transient characteristics include: Construct a fault state time-series trajectory matrix based on the sequence of features in each dimension of the high-frequency transient features. And perform dynamic time-corrected distance matching with the preset arc grounding template; the dynamic time-corrected distance is: ； ； in, This serves as a reference trajectory template for arc grounding. This is a normal trajectory template; For each feature sequence of the trajectory matrix S, an approximate entropy is calculated, resulting in 14 approximate entropy values. Mean: ； Calculate the confidence level of arc grounding: ； ； in, Represents the intercept term. , , , , , For coefficients; Fault determination based on thresholds: like If the fault persists for three consecutive cycles, an arc grounding fault is determined to have occurred. like Proceed to the initial review step; Review steps: If zero-sequence voltage voltage of the fault phase Inverted, and If the frequency is within the range of 10-30kHz, an arcing ground fault is determined to have occurred.

2. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 1, characterized in that, The method further includes: dynamically adjusting according to the system status. , , , , , , Specifically, it includes: Get current load ground capacitance current and distributed power penetration As the input variable of the fuzzy logic system, the 14-dimensional feature vector is divided into 5 groups, each group corresponding to an adjustment factor: Corresponding adjustment factor ; Corresponding adjustment factor ; Corresponding adjustment factor ; Corresponding adjustment factor ; Corresponding adjustment factor ; , , , and As an output variable of the fuzzy logic system; Group weights , To adjust the number of times; The fuzzy logic system converts input variables into fuzzy sets, uses fuzzy rules and the membership degrees of the input variables to calculate the fuzzy set of each output variable, and then converts it back into precise values ​​to obtain the adjustment factor. , , , and ; Using adjustment factors , , , and Adjust group weights separately And update the adjusted group weights. ; right Normalize and assign feature weights ; ； Based on the value, For the first The baseline scaling factor for each feature. This is the sensitivity function for the intercept term.

3. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 2, characterized in that, The sensitivity function for the intercept term is: ； in, Represents the capacitance current to ground The membership degree of the corresponding fuzzy set is high. Indicates load It belongs to the corresponding fuzzy set with a low membership degree. Indicates the penetration rate of distributed power sources The degree of membership of the corresponding fuzzy set.

4. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 1, characterized in that, Methods for determining whether a fault is transient or permanent include: Obtaining the arc combustion continuity index Arc energy attenuation coefficient Voltage recovery rate ratio Equivalent insulation resistance Oscillation frequency drift Construct the fault type feature vector: ； The fault type feature vector is input into the trained classifier, and the classifier outputs the fault type: 。 5. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 1, characterized in that, In step S3, the method to change the fault point from unstable arc grounding to stable metallic grounding is to use a hybrid fast switching unit.

6. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 5, characterized in that, The hybrid fast switching unit includes a thyristor valve group connected in reverse parallel, a high-speed mechanical switch, a triggering unit, and a synchronous monitoring unit; The thyristor valve group and the high-speed mechanical switch are connected in parallel; The triggering unit receives a grounding control command and generates a high-voltage pulse to trigger the reverse-parallel thyristor valve group; after receiving the high-voltage pulse trigger, the reverse-parallel thyristor valve group is forced to conduct, grounding the faulty phase, and within 1-20 milliseconds, sends a closing command to the high-speed mechanical switch. This allows the mechanical switch to close under near-zero voltage conditions, thus absorbing the subsequent steady-state fault current. The synchronous monitoring unit monitors the voltage and current passing through the high-speed mechanical switch in real time. When the trigger unit receives the disconnect control command, it uses the voltage and current zero-crossing point prediction to issue a disconnect command to the high-speed mechanical switch at the predicted current zero-crossing point.

7. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 6, characterized in that, The hybrid fast switching unit also includes an energy absorption unit, which includes an RC snubber circuit connected in parallel across the mechanical switch to suppress overvoltage during mechanical switch operation and dampen high-frequency oscillations.

8. The arc grounding overvoltage suppression method based on rapid transfer of fault grounding point according to claim 1, characterized in that, In S3, the method for determining the adaptive time is as follows; First, establish a historical database to obtain the ground capacitance current of the power grid fault system. The waiting time and whether the database was successfully restored within the waiting time; Then, the system's ground capacitance current is acquired in real time. Using the formula: ； Determine the initial waiting time , This indicates the configurable safety factor. This represents a quantity related to the discharge time constant of the system's capacitance to ground, and... Inversely proportional; Represents the system phase voltage. Indicates the power frequency angular frequency; Then search the database for the same capacitance current to ground. The waiting time during a failure is adjusted by adjusting the configurable safety factor K to obtain the adjusted waiting time. ; as the final adaptive time.

Images