Lightning stroke fault positioning method and system based on starting time of action current of lightning arrester

Abstract

The invention belongs to the field of lightning stroke fault positioning, and provides a lightning stroke fault positioning method and system based on lightning arrester action current starting time, and the method comprises the steps: obtaining a lightning arrester action current signal and positioning information; performing wavelet transformation and denoising on the lightning arrester action current signal, and calculating the current change rate of the denoised lightning arrester action current signal; based on the current change rate, abrupt change features are extracted, candidate starting time is obtained, a self-adaptive double-threshold verification mechanism is introduced, the candidate starting time is screened from the two aspects of amplitude features and space-time constraints, and final action current starting time is determined; the time measurement error, the wave velocity fluctuation and the positioning synchronization error are comprehensively considered, the positioning information is compensated, and the corrected positioning information is obtained; and based on the determined final action current starting time and the corrected positioning information, carrying out distance calculation and direction and phase logical judgment to realize lightning stroke point positioning. The lightning stroke fault point positioning capability is high.

Description

Technical Field

[0001] This invention belongs to the field of lightning fault location, specifically relating to a method and system for lightning fault location based on the start time of the surge arrester's operating current. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In power systems, lightning strikes are a major cause of power outages and equipment damage. Severe lightning overvoltages can lead to serious accidents such as flashover of line insulators, breakdown of transformer windings, and explosions of switchgear, resulting not only in huge equipment repair costs but also posing a significant threat to the continuity of power supply and the safety of users. As a core component of power system lightning protection, zinc oxide surge arresters effectively reduce the impact damage of lightning strikes on electrical equipment by discharging lightning current and limiting overvoltage amplitude.

[0004] However, the operating current waveform of the zinc oxide surge arrester during a lightning strike also records key fault information. Its current always rises rapidly from the initial zero value and is not affected by the superposition of the power grid frequency component on the transient waveform, so it can more accurately reflect the arrival time of the lightning current waveform.

[0005] Existing methods for locating lightning strike faults mostly rely on line traveling wave signal analysis or tower grounding resistance measurement, but they often face difficulties such as power frequency voltage and current interference and waveform superposition distortion, and are also limited by factors such as complex distribution network topology and many line branches.

[0006] Traditional traveling wave location technology requires rapid capture of transient signals after a fault. However, the continuous presence of power frequency components can interfere with the true waveform characteristics of lightning current, leading to time difference calculation errors in the location algorithm. The grounding resistance method is easily affected by soil resistivity and terrain environment, making it difficult to achieve accurate fault location. Summary of the Invention

[0007] To address the aforementioned problems, this invention proposes a lightning fault location method and system based on the start time of the surge arrester's operating current. This invention achieves strong lightning fault location capability, adapts to different types of power grid topologies, and is unaffected by power frequency current and load fluctuations, thereby improving the reliability of lightning fault identification and location.

[0008] According to some embodiments, the present invention adopts the following technical solution: A lightning fault location method based on the start time of the surge arrester's operating current includes the following steps: Acquire surge arrester operating current signal and location information; Perform wavelet transform and denoising on the surge arrester operating current signal, and calculate the current change rate of the denoised surge arrester operating current signal; Based on the rate of change of current, abrupt change features are extracted to obtain candidate start times. An adaptive dual threshold verification mechanism is introduced to screen candidate start times from both amplitude features and spatiotemporal constraints to determine the final action current start time. Taking into account time measurement error, wave velocity fluctuation and positioning synchronization error, the positioning information is compensated to obtain the corrected positioning information; Based on the determined start time of the final operating current and the corrected positioning information, distance calculation and logical judgment of direction and phase are performed to achieve lightning strike point location.

[0009] As an alternative implementation method, the process of performing wavelet transform on the surge arrester operating current signal includes: Time-frequency localization analysis is performed by wavelet transform. The inner product operation between the wavelet basis function and the signal is used to perform scaling and translation operations, decomposing the signal into components with different time-frequency windows. This retains the transient characteristics of the current starting from zero while filtering out noise components.

[0010] As a further implementation, the process of performing scaling and translation operations using the inner product of wavelet basis functions and signals includes: applying the db4 wavelet basis ψ(t) to the original current signal. i (t) Perform continuous wavelet transform:

[0011] In the formula, the scaling factor a Take 2 j j=1,2,...,5, achieving a 5-level decomposition through small-scale... a =2,4 Capture high-frequency noise, large scale a =16,32 preserves the trend of low-frequency signals; translation factor b Slide the sampling interval Δt to ensure that the time resolution is consistent with the original signal.

[0012] As an alternative implementation method, the process of denoising the surge arrester operating current signal includes: denoising the high-frequency coefficient... j Layers 1, 2, and 3 use soft thresholding to suppress noise.

[0013] The noise standard deviation is estimated by the median of the high-frequency layer coefficients: σ = median(∣ W T |) / 0.6745, where N is the signal length, is used to reconstruct the signal using inverse wavelet transform, resulting in an improved and denoised signal. i' The signal-to-noise ratio of (t) is improved, and the distortion rate of the steepness of the starting edge is reduced.

[0014] As an alternative implementation, the process of calculating the rate of change of the current in the denoised arrester operating current signal includes: processing the current signal after wavelet denoising. i' (t) The first derivative is used to calculate the rate of change of current:

[0015] Δt is the time interval.

[0016] As an alternative implementation, the process of extracting abrupt change features based on the rate of change of current includes: characterizing the abrupt change features of the rate of change using the second derivative:

[0017] Δt is the time interval, and the second derivative passes through the zero-crossing point where it changes from negative to positive, which is the theoretical position of the starting time.

[0018] As an alternative implementation, an adaptive dual-threshold verification mechanism is introduced to screen candidate start times from both amplitude characteristics and spatiotemporal constraints. The process of determining the final operating current start time includes: in terms of amplitude threshold verification, based on the noise baseline of the signal before denoising... I noise Take the maximum peak value of the no-signal period of the previous set duration. Based on engineering experience, set the amplitude threshold as follows:

[0019] Combined with the propagation speed of lightning current v and maximum monitoring distance l The time threshold is:

[0020] in, t l The inherent time of lightning intrusion. T max This is the upper limit of the positioning time; The earliest candidate time that passes the dual-threshold verification is the valid start time, which determines the final operating current start time. : .

[0021] As an alternative implementation method, the process of compensating for positioning information, taking into account time measurement errors, wave velocity fluctuations, and positioning synchronization errors, includes: The corrected distance is obtained by quantifying and compensating using an error correction formula. s * Interval:

[0022] In the formula, time measurement error It is mainly related to the sampling rate of the high-frequency current sensor. Select sensor time resolution; wave velocity error It is mainly related to changes in line parameters, such as high line temperature and icing, which cause wave velocity fluctuations. The specific situation must be considered to determine the cause. The positioning synchronization error δ is related to the performance of the positioning equipment.

[0023] As an alternative implementation method, a Type I distribution network line requires at least 2 surge arrester data points; a Type T distribution network line requires a total of 3 data points, including a surge arrester at the center point and one surge arrester on each of the two branches; and a cross-shaped distribution network line requires a total of 4 data points, including a surge arrester at the center point and one surge arrester on each of the three branches. Furthermore, multiple surge arrester data exist, and the intersection of the positioning intervals is taken to further improve the accuracy of the positioning information.

[0024] A lightning fault location system based on the start time of the surge arrester's operating current includes: The data acquisition module is configured to acquire the surge arrester's operating current signal and location information; The current change rate module is configured to perform wavelet transform and denoising on the surge arrester operating current signal, and calculate the current change rate of the denoised surge arrester operating current signal. The start time determination module is configured to extract abrupt change features based on the current change rate to obtain candidate start times. An adaptive dual threshold verification mechanism is introduced to screen candidate start times from both amplitude features and spatiotemporal constraints to determine the final action current start time. The distance correction module is configured to comprehensively consider time measurement errors, wave speed fluctuations, and positioning synchronization errors to compensate for the positioning information and obtain corrected positioning information. The lightning strike location module is configured to perform distance calculations and logical judgments of direction and phase based on the determined start time of the final operating current and the corrected location information, thereby achieving lightning strike location.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves accurate location in complex distribution network topologies by extracting the start time and amplitude of the operating current and inverting the electrical distance and phase of the lightning fault point. Utilizing the inherent characteristic of the zinc oxide surge arrester's current "starting from zero", it can effectively avoid interference from power grid frequency voltage and current and load fluctuations. The attenuation of the lightning current amplitude and energy loss do not affect the extraction of the start time, ensuring reliability under complex operating conditions. Simulation results show that the location error is less than 1%.

[0026] This invention achieves accurate extraction of the start time of the action current in engineering through a multi-step processing mechanism of wavelet threshold denoising, derivative sequence mutation point detection, and dual threshold verification. Combined with a distance error correction algorithm, it compensates for factors such as time measurement and wave velocity fluctuation, effectively improving the anti-interference capability of the start time extraction and the reliability of distance calculation.

[0027] This invention does not rely on additional traveling wave detection devices. It can directly utilize the monitoring data of zinc oxide surge arresters already installed in the distribution network, combined with high-frequency sensors and GPS positioning technology, to provide an economical and efficient solution for the rapid investigation and accurate location of lightning strike faults in the distribution network.

[0028] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0029] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0030] Figure 1 One embodiment of the volt-ampere characteristic curve of a zinc oxide surge arrester; Figure 2 This is an example of a lightning current model and waveform; Figure 3 This is a diagram showing the dimensions and parameters of a tower according to one embodiment; Figure 4 This describes the surge arrester voltage and current under power frequency voltage in one embodiment. Figure 5 This describes the voltage and current of a surge arrester under lightning current impulse in one embodiment. Figure 6 This is a type "one" circuit model of one embodiment; Figure 7 This is one embodiment of a "T" type circuit model; Figure 8 This is a cross-shaped circuit model of one embodiment; Figure 9 This is an embodiment of the surge arrester current of phase A of each tower when lightning strikes phase A of tower No. 4; Figure 10 This is an embodiment of the surge arrester current for phases A and B of tower No. 4 when lightning strikes both phases A and B of tower No. 4; Figure 11 This is an embodiment of the surge arrester current of phase A of each tower when lightning strikes phase A of tower No. 2; Figure 12 This is an embodiment of the surge arrester current of each tower's phase A when lightning strikes tower No. 4 ("T-type line"). Figure 13 This is an embodiment of the surge arrester current of each tower phase A when lightning strikes tower No. 4 ("+" type line); Figure 14 This is an embodiment of the surge arrester current of each tower phase A when lightning strikes tower No. 1 ("+" type line extension). Figure 15 This is a schematic diagram of a lightning strike point location method based on the current information of a zinc oxide surge arrester, according to one embodiment. Detailed Implementation

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0032] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0034] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0035] Example 1 A method for locating lightning faults based on the start time of the surge arrester's operating current, such as... Figure 15 As shown, the location of a lightning strike fault is determined by monitoring the start time of the operating current of a zinc oxide surge arrester. This embodiment also establishes a typical 10kV distribution network topology simulation circuit using EMTP software, analyzes the current waveforms of each phase surge arrester on each tower under lightning strike, and inverts the electrical distance between the lightning strike point and the surge arrester, as well as the fault phase, by analyzing the current start time and peak current. This enables rapid location of the lightning strike fault. Furthermore, an optimized method for data extraction and processing is proposed from an engineering analysis perspective. Simulation results show that this method has strong lightning strike fault location capability, adapts to different types of power grid topologies, and is unaffected by power frequency current and load fluctuations, thus improving the reliability of lightning strike fault identification and location.

[0036] The following is a detailed description, starting with the principle of lightning fault location based on the characteristics of zinc oxide surge arresters.

[0037] As an overvoltage protection device, zinc oxide surge arresters have excellent volt-ampere characteristics, such as... Figure 1 As shown, under the power frequency voltage of a 10kV distribution network, the surge arrester exhibits a high resistance state, and the current flowing through it is almost zero; under lightning overvoltage, the surge arrester exhibits a low resistance state, quickly conducts to discharge the lightning current, and clamps the voltage at the residual voltage level.

[0038] When a tower in a power distribution network is struck by lightning, the lightning wave will travel at a speed of v Propagation in the conductor: (1) in, L The unit inductance of the line, C This is the unit capacitance of the circuit.

[0039] In engineering analysis, it is generally assumed that the L and C values ​​of the same line are the same, meaning the lightning wave propagation speed remains constant. The location of the lightning strike point can be deduced by the difference in the arrival time of the lightning wave at each surge arrester. Since 10kV distribution network overhead lines generally do not have lightning protection wires installed, when lightning strikes a tower, it needs to discharge through insulators or surge arresters before the lightning wave propagates to the next tower. Surge arresters only dissipate the lightning wave energy, resulting in a reduction in peak amplitude, but do not affect the wave's arrival time.

[0040] In the simulated power supply design, the power frequency power supply is implemented using AC power. The lightning current waveform, a standard 2.6 / 50μs waveform, is generated using the Heidler model. Its steep rising edge and post-peak decay characteristics more closely resemble the actual physical process of lightning, resulting in simulation accuracy superior to the double exponential model. The lightning model and waveform are as follows: Figure 2 As shown.

[0041] In the simulation design of towers and surge arresters, the towers are modeled using a 10kV distribution network. Specific parameters and dimensions are as follows: Figure 3 As shown.

[0042] The parameters of the zinc oxide surge arrester are shown in Table 1. The MOV model was selected and the volt-ampere characteristic parameters were fitted.

[0043] Table 1 Parameters of Zinc Oxide Surge Arresters

[0044] Operating under power frequency voltage conditions, such as Figure 4 As shown, under power frequency voltage, its current is almost zero. Under lightning current impulse conditions, as... Figure 5 As shown, the voltage rises from the power frequency voltage at that moment and is clamped below 55kV, while the current rises from zero and is not affected by the power frequency current at that moment. The waveform is similar to the lightning current waveform.

[0045] Typical topology modeling of transmission lines In the simulation of power distribution line design, three typical line topologies are modeled. Figure 6 The diagram shows a Type I line, also known as a straight line. Seven towers, numbered sequentially, are installed along this line, spaced 50m apart. The power supply and end loads are positioned at distances greater than 50m from each tower. Surge arresters are installed on each tower and grounded via a 30Ω grounding resistor. Lightning strikes can enter the grid from any phase of any tower and propagate in both directions along the conductor. Figure 6 The image shows a lightning strike on tower No. 4 (phase A), with the lightning wave propagating to towers No. 3 and No. 5 respectively.

[0046] Based on this, a branch line is drawn from tower number four, forming a "T" shaped line, such as... Figure 7 As shown, this line has a total of 10 towers, and the lightning waves propagate in three directions. If two branch lines are drawn from tower number four, a cross-shaped line is formed, as shown below. Figure 8 As shown, the line has a total of 13 towers, and lightning waves propagate in four directions. By modeling the topologies of typical "I", "T", and "+" type transmission lines, most complex distribution network structures can be covered.

[0047] Waveform characteristic analysis and lightning strike location of zinc oxide surge arresters Analysis of current waveform characteristics of zinc oxide surge arresters In a "Type I" line, phase A of tower No. 4 was selected as the lightning strike point. A simulation model was run to obtain the operating current waveform of the surge arrester for phase A of each tower, as follows: Figure 9 As shown, after the lightning strike, within 0.405 μs, the current in the A-phase surge arrester on pole #4 rapidly increased from 0, beginning to discharge current, with a waveform approximating the lightning current. Subsequently, the lightning current wave propagated to towers #3 and #5. Due to the wave velocity within the same line... v At a constant value, at 0.572 μs, the surge arresters of phase A on towers 3 and 5 simultaneously began discharging, with the current rapidly increasing from 0. The waveform approximates a lightning current, but the amplitude decreased significantly. This is because the surge arrester of phase A on tower 4 consumed the lightning wave energy, and there was also some loss during wave propagation on the line. Subsequently, the lightning wave propagated to towers 2 and 6, where the arresters discharged at 0.739 μs, with the amplitude decreasing further. Finally, the lightning wave propagated to towers 1 and 7, where the arresters discharged at 0.906 μs, with the amplitude decreasing and the waveform distorted, demonstrating that the lightning current energy was discharged in stages.

[0048] Introducing the formula for calculating the distance to the lightning strike point: (2) In the formula, t c The start time and initial time of the surge arrester current abruptly changing from 0. t lThe intrinsic time of lightning intrusion is given, which is related to the properties of the lightning wave; in this paper, it is taken as 0.405 μs. v This is the propagation speed of lightning waves, which is related to the line parameters; in this paper, we take 300 m / μs.

[0049] extract t c The electrical distance between the surge arrester and the lightning strike point can be calculated using the distance calculation formula, achieving inversion and location. As shown in Table 2, the error between the calculated distance and the actual distance is less than 1%.

[0050] Table 2 Calculation of the initiation time and distance of the surge arrester current for phase A of each tower (lightning strike No. 4)

[0051] When lightning strikes both phase A and phase B simultaneously, data from the first four towers is used to obtain the phase A and phase B currents for each tower, as shown below. Figure 10 As shown, when lightning strikes both phases A and B simultaneously, both phase surge arresters operate, and the current waveforms are almost identical and approximate the lightning current waveform, indicating that the lightning wave propagates in phases A and B.

[0052] Compared to lightning striking only phase A, the current wave shape changes slightly and the amplitude decreases. This is because the two-phase surge arresters share the lightning wave energy, but the current initiation time remains unchanged, and the two phases are relatively independent. The calculated distance to the lightning strike point is consistent with the previous one.

[0053] Based on the above analysis, this method can not only calculate the location of the lightning strike point by the current initiation time, but also determine the phase of the lightning strike based on the amplitude and propagation characteristics of the lightning current wave in each phase.

[0054] Analysis of Lightning Strike Location in Typical Network Topology (a) Analysis of Type I Line In a "Type I" line, phase A of tower No. 2 is selected as the lightning strike point. A simulation model is run to obtain the operating current waveform of the surge arrester for phase A of each tower, as follows: Figure 11 As shown, the lightning current wave arrives at towers 1 and 3 at the same time, while the arrival times at the other towers are different, exhibiting more varied time differences. It is noteworthy that the surge arrester current waveforms of towers 1 and 3 are identical before 0.972μs, but show a significant deviation after 0.972μs. This is related to tower 1's proximity to the power source, but it does not affect the current initiation time.

[0055] The current initiation time was extracted and the lightning strike location was calculated, as shown in Table 3. Comparing with Table 1, it was found that the calculated distance for tower 3 was 50.1m. If only the surge arrester current waveform data from tower 3 was available in the actual calculation, only the distance could be calculated, and the direction could not be determined. However, by comparing the data with that from the adjacent tower 4, the abrupt change time on tower 4 was greater than that on tower 3, indicating that the lightning current wave traveled from tower 3 to tower 4, thus determining the direction and uniquely identifying the lightning strike location. If the lightning strike point was located between two selected surge arresters, the calculated distance could also be compared to uniquely identify the lightning strike location.

[0056] Therefore, for a "Type I" line, at least the current data of two surge arresters on the line are needed to locate the lightning strike fault point.

[0057] Table 3 Calculation of the start time and distance of the surge arrester current for phase A of each tower (lightning strike No. 2)

[0058] (b) Analysis of “T” type lines In a "T"-shaped line, phase A of tower No. 4 was selected as the lightning strike point. A simulation model was run to obtain the operating current waveform of the surge arrester for phase A of each tower, as follows: Figure 12 As shown, the time characteristics of the current waveform are the same as those analyzed in 3.1, but the current amplitude is lower than that of the "Type I" line, and the energy absorbed by each surge arrester is also reduced. The current start time and the location of the lightning strike point are extracted and calculated, as shown in Table 4.

[0059] Table 4 Calculation of the start time and distance of the A-phase surge arrester current for each tower ("T" type line, lightning strike No. 4)

[0060] The lightning strike point was changed to phase A of tower No. 2. The operating current waveform of the surge arrester of phase A of each tower was obtained, the current start time was extracted and the lightning strike point location was calculated, as shown in Table 5.

[0061] Table 5 Calculation of the start time and distance of the A-phase surge arrester current for each tower ("T" type line, lightning strike No. 2)

[0062] The lightning strike point was changed to phase A of tower No. 6. The operating current waveform of the surge arrester of phase A of each tower was obtained, the current start time was extracted and the lightning strike point location was calculated, as shown in Table 6.

[0063] Table 6 Calculation of the time and distance of sudden change in current of phase A surge arrester on each tower ("T" type line, lightning strike No. 6)

[0064] Analyzing the data in Tables 3, 4, and 5, in a "T"-shaped circuit, if only the surge arrester current data of the center point (tower 4) is taken, it is impossible to determine which line the lightning strike point is on unless the lightning strike point is located at the center point. If the data of a surge arrester adjacent to the center tower on another branch line is taken, such as tower 8, and the magnitude of the abrupt change time of towers 4 and 8 is compared, the direction of lightning current wave propagation on that branch line can be determined. Unless the lightning strike point is located on that line, it is impossible to determine which of the remaining two lines the lightning strike point is on. If the data of a surge arrester adjacent to the center tower on another branch line is taken, such as tower 5, and the magnitude of the abrupt change time of towers 4, 5, and 8 is compared, the direction of lightning current wave propagation on the three lines can be determined. Combined with distance calculation, the location of the lightning strike point can be uniquely determined.

[0065] Therefore, for a "T" type line, if a center point surge arrester is selected, at least the current data of three surge arresters (one on each of the two branches) are needed to locate the lightning strike fault point.

[0066] (c) Analysis of the "T"-shaped route In the "+" type line, phase A of tower No. 4 was selected as the lightning strike point. A simulation model was run to obtain the operating current waveform of the surge arrester for phase A of each tower, as follows: Figure 13 As shown, the time characteristics of the current waveform are the same as those analyzed in 3.1, but the current amplitude is lower than that of the "I" and "T" type lines, and each surge arrester absorbs less energy. The current initiation time and the location of the lightning strike point are extracted and calculated, as shown in Table 7.

[0067] Table 7 Calculation of the start time and distance of the A-phase surge arrester current for each tower ("+" type line, lightning strike No. 4)

[0068] The lightning strike point was changed to phase A of tower No. 2. The operating current waveform of the surge arrester of phase A of each tower was obtained, the current start time was extracted and the lightning strike point location was calculated, as shown in Table 8.

[0069] Table 8 Calculation of the start time and distance of the A-phase surge arrester current for each tower ("+" type line, lightning strike No. 2)

[0070] The lightning strike point was changed to phase A of tower No. 6. The operating current waveform of the surge arrester of phase A of each tower was obtained, the current start time was extracted and the lightning strike point location was calculated, as shown in Table 9.

[0071] Table 9 Calculation of the starting time and distance of the A-phase surge arrester current for each tower ("+" type line, lightning strike No. 6)

[0072] Change the lightning strike point to phase A of tower No. 9, obtain the current waveform of the surge arrester of phase A of each tower, extract the time when the current changes abruptly from 0 and calculate the location of the lightning strike point, as shown in Table 10.

[0073] Table 10 Calculation of the starting time and distance of the A-phase surge arrester current for each tower ("+" type line, lightning strike No. 9)

[0074] Analyzing the data in Tables 7, 8, 9, and 10, and based on the analysis of the "T" type line, it can be seen that the "+" type line is equivalent to four branches extending from the center point. With the center point surge arrester selected, at least the current data of four surge arresters (one on each of the three branches) are needed to locate the lightning strike fault point.

[0075] Analysis of the lightning strike location range of zinc oxide surge arresters Because surge arresters discharge lightning current energy in stages, the operating current of surge arresters outside a certain range becomes severely distorted and loses its regularity, making it impossible to determine and locate lightning strike faults based on current information.

[0076] Based on the above analysis, it can be seen that under the same lightning strike, the current amplitude of the same surge arrester is: Type I > Type T > Type X. From the topology analysis of the largest current drop, the following should be selected: Figure 8 The line is shown in a "+" shape, and extended after tower 13 on the upper branch, with six identical towers (14-19) erected to extend the line distance. The A-phase of the furthest tower (1) is selected as the lightning strike point, and the current waveform of the A-phase arrester on each tower is obtained, as shown below. Figure 14 As shown, starting from tower 18, the waveform becomes severely distorted and no longer follows a regular pattern. The current at tower 19 experiences a significant time deviation from its initial jump from 0, and the jump amplitude is too small to be of any value for analysis. Therefore, the positioning range using this method is approximately 10 towers' distance, or 500m.

[0077] Design of lightning strike fault location schemes for complex power grid topologies, such as Figure 15 As shown, it includes the following steps: 1. Extraction of operating current start time (1) Wavelet thresholding for noise reduction The acquisition of surge arrester operating current signals is affected by high-frequency electromagnetic interference, manifesting as random oscillations superimposed on the initial waveform. Time-frequency localization analysis is performed using wavelet transform, employing the wavelet basis function ψ(t) and the signal... i The inner product operation of (t) is used to perform scaling and translation operations to decompose the signal into components of different time-frequency windows, thereby achieving accurate capture of transient change signals. It can retain the transient characteristics of the current starting from zero while adaptively filtering out noise components.

[0078] The original current signal is processed using the db4 wavelet basis ψ(t). i (t) Perform continuous wavelet transform (CWT): (3) In the formula, the scaling factor a Take 2 j Achieving a 5-level decomposition using (j=1,2,...,5) and utilizing small-scale ( a =2,4) Capture high-frequency noise, large-scale ( a =16,32) Preserve the trend of low-frequency signals. Translation factor b Slide the sampling interval Δt = 0.01 μs to ensure that the time resolution is consistent with the original signal.

[0079] For high frequency coefficients ( j (For layers 1, 2, and 3) noise is suppressed using soft thresholding. (4) The noise standard deviation is estimated by the median of the high-frequency layer coefficients: σ = median(∣ W T |) / 0.6745, signal length N=1024 (corresponding to a 10.24μs data window). After reconstruction using inverse wavelet transform, the denoised signal can be improved. i' The signal-to-noise ratio of (t) is reduced, and the kurtosis distortion rate at the start edge is decreased.

[0080] (2) Detection of abrupt change points in derivative sequences The surge arrester's operating current exhibits a typical characteristic of rapidly rising from zero, with the initial moment corresponding to the critical point where the current slope abruptly changes from zero. This is relevant to the current signal after wavelet denoising. i' (t) By performing the first derivative calculation, the rate of change of current can be obtained: (5) However, the first derivative is susceptible to noise interference, and the second derivative is needed to further characterize the abrupt changes in the rate of change. (6) This is a measure of signal curvature. When a current signal affected by noise enters a rapid rising phase from a stable state, the second derivative will undergo a zero-crossing process from negative to positive, which is the theoretical position of the starting time.

[0081] (3) Dual threshold verification mechanism Although zero-crossing points of the second derivative can effectively locate signal abrupt changes, electromagnetic interference in complex power grid environments, such as power frequency harmonics and switching operation pulses, can also generate spurious abrupt changes. Therefore, an adaptive dual-threshold verification mechanism is introduced to evaluate candidate start times from both amplitude characteristics and spatiotemporal constraints. t candidatePerform the screening.

[0082] In amplitude threshold verification, it should be ensured that the detected signal is operating current and not noise. The noise baseline should be based on the signal before denoising. I noise The maximum peak value of the first 100μs of signal-free period is taken. Based on engineering experience, the amplitude threshold is set as follows: (7) Combined with the propagation speed of lightning current v With a maximum monitoring distance of 500m, the time threshold is limited to: (8) in, t l The inherent time of lightning intrusion. T max This is the upper limit for positioning time.

[0083] The earliest candidate time that passes the dual threshold verification described above is the effective start time, and the final operating current start time is determined accordingly. : (9) 2. Distance error correction In practical applications, positioning distance s The distance can be corrected by quantifying and compensating for errors such as time measurement errors, wave speed fluctuations, and GPS synchronization errors using error correction formulas. s * Interval: (10) In the formula, time measurement error It is mainly related to the sampling rate of the high-frequency current sensor. Select sensor time resolution; wave velocity error It is mainly related to changes in line parameters, such as high line temperature and icing, which cause fluctuations in wave velocity. The specific determination can be made based on the specific circumstances. The GPS synchronization error δ is related to the performance of the positioning device.

[0084] By adjusting the distance and taking into account both theoretical model calculations and engineering practice, the error can be controlled to be less than 2% for a positioning range of 500m. Furthermore, for the data of n surge arresters in the distribution network, the intersection of the positioning intervals can be taken, further improving the reliability of the method.

[0085] Positioning scheme design This method is applied to lightning strike location. The current of the zinc oxide surge arrester is detected by a magnetic field induction high-frequency sensor. After processing by a microcontroller, the start time and amplitude of the operating current are accurately extracted. The data is transmitted to the monitoring terminal via wireless transmission. The data is then compared to make a logical judgment on the direction and phase. Combined with the GPS positioning system installed on the surge arrester, the lightning strike location can be realized.

[0086] Example 2 A lightning fault location system based on the start time of the surge arrester's operating current includes: The data acquisition module is configured to acquire the surge arrester's operating current signal and location information; The current change rate module is configured to perform wavelet transform and denoising on the surge arrester operating current signal, and calculate the current change rate of the denoised surge arrester operating current signal. The start time determination module is configured to extract abrupt change features based on the current change rate to obtain candidate start times. An adaptive dual threshold verification mechanism is introduced to screen candidate start times from both amplitude features and spatiotemporal constraints to determine the final action current start time. The distance correction module is configured to comprehensively consider time measurement errors, wave speed fluctuations, and positioning synchronization errors to compensate for the positioning information and obtain corrected positioning information. The lightning strike location module is configured to perform distance calculations and logical judgments of direction and phase based on the determined start time of the final operating current and the corrected location information, thereby achieving lightning strike location.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for locating lightning faults based on the start time of the surge arrester's operating current, characterized in that, Includes the following steps: Acquire surge arrester operating current signal and location information; Perform wavelet transform and denoising on the surge arrester operating current signal, and calculate the current change rate of the denoised surge arrester operating current signal; Based on the rate of change of current, abrupt change features are extracted to obtain candidate start times. An adaptive dual threshold verification mechanism is introduced to screen candidate start times from both amplitude features and spatiotemporal constraints to determine the final action current start time. Taking into account time measurement error, wave velocity fluctuation and positioning synchronization error, the positioning information is compensated to obtain the corrected positioning information; Based on the determined start time of the final operating current and the corrected positioning information, distance calculation and logical judgment of direction and phase are performed to achieve lightning strike point location.

2. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, The process of performing wavelet transform on the surge arrester operating current signal includes: Time-frequency localization analysis is performed by wavelet transform. The inner product operation between the wavelet basis function and the signal is used to perform scaling and translation operations, decomposing the signal into components with different time-frequency windows. This retains the transient characteristics of the current starting from zero while filtering out noise components.

3. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 2, characterized in that, The process of scaling and translation operations using the inner product of wavelet basis functions and signals includes: applying the db4 wavelet basis ψ(t) to the original current signal. i (t) Perform continuous wavelet transform: In the formula, the scaling factor a Take 2 j j=1,2,...,5, achieving a 5-level decomposition through small-scale... a =2,4 Capture high-frequency noise, large scale a =16,32 preserves the trend of low-frequency signals; translation factor b Slide the sampling interval Δt to ensure that the time resolution is consistent with the original signal.

4. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, The process of denoising the surge arrester operating current signal includes: denoising the high-frequency coefficients. j Layers 1, 2, and 3 use soft thresholding to suppress noise. The noise standard deviation is estimated by the median of the high-frequency layer coefficients: σ = median(∣ W T |) / 0.6745, where N is the signal length, is used to reconstruct the signal using inverse wavelet transform, resulting in an improved and denoised signal. i' The signal-to-noise ratio of (t) is improved, and the distortion rate of the steepness of the starting edge is reduced.

5. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, The process of calculating the rate of change of the current in the denoised surge arrester operating current signal includes: processing the current signal after wavelet denoising. i' (t) The first derivative is used to calculate the rate of change of current: Δt is the time interval.

6. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 5, characterized in that, The process of extracting abrupt change features based on the rate of change of current includes: characterizing the abrupt change features of the rate of change using the second derivative. Δt is the time interval, and the second derivative passes through the zero-crossing point where it changes from negative to positive, which is the theoretical position of the starting time.

7. A lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, it introduces... The adaptive dual-threshold verification mechanism filters candidate start times based on both amplitude characteristics and spatiotemporal constraints to determine the final operating current start time. The process includes: in amplitude threshold verification, based on the noise baseline of the pre-denoised signal... I noise Take the maximum peak value of the no-signal period of the previous set duration. Based on engineering experience, set the amplitude threshold as follows: Combined with the propagation speed of lightning current v and maximum monitoring distance l The time threshold is: in, t l The inherent time of lightning intrusion. T max This is the upper limit of the positioning time; The earliest candidate time that passes the dual-threshold verification is the valid start time, which determines the final operating current start time. : 。 8. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, Taking into account time measurement errors, wave velocity fluctuations, and positioning synchronization errors, the process of compensating for positioning information includes: The corrected distance is obtained by quantifying and compensating using an error correction formula. s * Interval: In the formula, time measurement error It is mainly related to the sampling rate of the high-frequency current sensor. Select sensor time resolution; wave velocity error It is mainly related to changes in line parameters, such as high line temperature and icing, which cause wave velocity fluctuations. The specific situation must be considered to determine the cause. The positioning synchronization error δ is related to the performance of the positioning equipment.

9. The lightning fault location method based on the start time of the surge arrester's operating current as described in claim 1, characterized in that, For a Type I distribution network line, at least 2 surge arrester data points are required; for a Type T distribution network line, 3 data points are required, including one surge arrester at the center point and one surge arrester on each of the two branch lines; for a Type C distribution network line, 4 data points are required, including one surge arrester at the center point and one surge arrester on each of the three branch lines. Furthermore, multiple surge arrester data exist, and the intersection of the positioning intervals is taken to further improve the accuracy of the positioning information.

10. A lightning fault location system based on the start time of the surge arrester's operating current, characterized in that, include: The data acquisition module is configured to acquire the surge arrester's operating current signal and location information; The current change rate module is configured to perform wavelet transform and denoising on the surge arrester operating current signal, and calculate the current change rate of the denoised surge arrester operating current signal. The start time determination module is configured to extract abrupt change features based on the current change rate to obtain candidate start times. An adaptive dual threshold verification mechanism is introduced to screen candidate start times from both amplitude features and spatiotemporal constraints to determine the final action current start time. The distance correction module is configured to comprehensively consider time measurement errors, wave speed fluctuations, and positioning synchronization errors to compensate for the positioning information and obtain corrected positioning information. The lightning strike location module is configured to perform distance calculations and logical judgments of direction and phase based on the determined start time of the final operating current and the corrected location information, thereby achieving lightning strike location.

Images