Traveling wave detection method and system based on high-frequency current sensor

Abstract

The invention discloses a traveling wave detection method and system based on a high-frequency current sensor. The high-frequency current sensor comprises a main coil and an auxiliary coil. The traveling wave detection method comprises the following steps: synchronously obtaining a power frequency current signal output by a main coil and a high frequency traveling wave signal output by an auxiliary coil; fusing the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal; carrying out notch filtering on the fused signal, and enabling a notch center frequency to dynamically track a background noise peak value of the power distribution network; and performing Teager energy operator transformation on the filtered signal to capture a fault traveling wave head sudden change point, and determining a traveling wave arrival time. According to the traveling wave detection method and system based on the high-frequency current sensor provided by the invention, the problem of signal connection distortion in a traditional scheme can be eliminated through a three-stage processing architecture of fusion, filtering and TEO detection.

Description

Technical Field

[0001] The present invention relates to the technical field of traveling wave detection, and in particular to a traveling wave detection method and system based on a high-frequency current sensor. Background Art

[0002] In the process of intelligent upgrading of the distribution network, the integrated primary and secondary pole-mounted circuit breaker, as a core device, plays an important role in line fault isolation. However, in the face of high-resistance grounding faults, its recognition accuracy is relatively low. The root cause lies in the performance bottleneck of the existing traveling wave detection technology restricted by the current sensor, which is mainly manifested in the following defects:

[0003] (1) The bandwidth of traditional electromagnetic CT is insufficient, and high-frequency signals are severely attenuated.

[0004] The traveling wave signals in the distribution network contain rich high-frequency components in the MHz range (such as the rising edge frequency of lightning shock waves from 1 MHz to 10 MHz, and the high-frequency oscillation of fault arcs from 2 MHz to 20 MHz). However, the working frequency band of traditional electromagnetic CT is usually limited to 50 Hz to 10 kHz, and the signal attenuation exceeds -20 dB at >100 kHz. For example: the amplitude of the 1 MHz traveling wave signal is attenuated to 10% of the theoretical value, and the 10 MHz component is almost completely lost, resulting in the inability to effectively extract the characteristics of the traveling wave head and a high false negative rate for high-resistance fault detection.

[0005] (2) Inherent defects of the magnetic core material cause signal distortion.

[0006] Traditional CT generally uses silicon steel sheets or ordinary ferrite cores, which have two major drawbacks: when a short-circuit fault occurs in the line, the magnetic core saturation causes the output waveform to be clipped and distorted; at high frequencies, the phase error is large, resulting in large detection deviations and positioning errors not only at the moment of the traveling wave head.

[0007] (3) Lack of anti-interference mechanism and low signal-to-noise ratio.

[0008] There are various MHz-level electromagnetic interference sources in the distribution network, such as switch operation oscillations and arc discharge noises. Traditional CT only relies on a single-layer metal shell for shielding and is not optimized for high-frequency interference. The measured signal-to-noise ratio is small, resulting in the traveling wave signal being submerged by noise and an increase in the fault misjudgment rate.

[0009] (4) Rigid installation structure and poor scene adaptability.

[0010] Existing sensors are mostly of a closed-loop structure and cannot adapt to the complex scenarios of the distribution network. As a result, monitoring devices cannot be deployed at some key nodes.

[0011] In order to solve at least one of the above technical problems, the present invention provides a traveling wave detection method and system based on a high-frequency current sensor. Summary of the Invention

[0012] The object of the present invention is to provide a traveling wave detection method and system based on a high-frequency current sensor, which can eliminate the problem of signal connection distortion in the traditional scheme through a three-level processing architecture of fusion, filtering, and TEO detection.

[0013] The object of the present invention is achieved by the following technical solutions:

[0014] On the one hand, the present invention provides a traveling wave detection method based on a high-frequency current sensor, characterized in that the high-frequency current sensor includes a main coil and an auxiliary coil;

[0015] The traveling wave detection method includes the following steps:

[0016] Synchronously acquire the power frequency current signal output by the main coil and the high-frequency traveling wave signal output by the auxiliary coil;

[0017] Fuse the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal;

[0018] Perform notch filtering on the fused signal and make the notch center frequency dynamically track the peak value of the distribution network background noise;

[0019] Perform Teager energy operator transformation on the filtered signal to capture the mutation point of the fault traveling wave head and determine the arrival time of the traveling wave.

[0020] The beneficial effects of the above solution are: The present invention ensures the stability of power frequency current measurement through the main coil (thick wire diameter), and the auxiliary coil (thin wire diameter closely wound) specializes in capturing high-frequency traveling waves, solves the problem of limited bandwidth of traditional sensors, expands the effective bandwidth from the traditional 5 MHz to 20 MHz, and reduces the amplitude error of traveling wave signals. In addition, through a three-level processing architecture of fusion, filtering, and TEO detection, the problem of signal connection distortion in the traditional scheme is eliminated.

[0021] Further, the fusing the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal includes:

[0022] Perform wavelet packet decomposition on the high-frequency traveling wave signal, extract the sub-signal energy of the first frequency band, and calculate the dynamic weight coefficient according to the power frequency current signal and the sub-signal energy;

[0023] Based on the dynamic weight coefficient, determine the fused signal by using the power frequency current signal and the high-frequency traveling wave signal;

[0024] Wherein, the first frequency band is determined according to the fault characteristic frequency band.

[0025] The beneficial effects of the above solution are: The present invention accurately extracts the energy of the traveling wave characteristic frequency band in the first frequency band through wavelet packet decomposition, and the signal-to-noise ratio is improved under the interference of switching operations. In addition, the dynamic weight can avoid misoperation caused by high-frequency noise under normal operating conditions.

[0026] Further, the high-frequency current sensor is deployed at a preset node of the 35 kV line;

[0027] The dynamic weight coefficient includes:

[0028] When a short-circuit fault occurs on the line, the energy of the sub-signal suddenly increases, and the dynamic weight coefficient is 1;

[0029] When the line is operating normally, the energy of the sub-signal approaches zero, and the dynamic weight coefficient is 0.

[0030] The beneficial effect of the above solution is that: by means of the dynamic weight, the present invention strengthens the ability to capture traveling wave characteristics during a fault, ensures the non-attenuated transmission of traveling wave signals, and improves the protection speed; during normal operation, it maintains the power frequency measurement accuracy, completely isolates high-frequency channel interference, and reduces the power frequency measurement error.

[0031] Further, an auxiliary induction coil is provided inside the shielding layer of the high-frequency current sensor;

[0032] The notch filtering of the fusion signal and the dynamic tracking of the notch center frequency to the peak value of the background noise of the distribution network include:

[0033] Collect the bus voltage signal through a voltage sensor and extract the power frequency noise component;

[0034] Collect the spatial electromagnetic noise signal through the auxiliary induction coil;

[0035] Perform FFT calculation on the spatial electromagnetic noise within the first sliding time window to identify the main peak frequency of the background noise;

[0036] When the main peak frequency is the power frequency noise component or greater than the first preset frequency band and the same-frequency verification is established, update the notch center frequency to the main peak frequency;

[0037] Wherein, the first preset frequency band is determined according to high-frequency interference and effective traveling wave signals.

[0038] The beneficial effect of the above solution is that: the present invention can suppress both power frequency harmonics and high-frequency interference. In addition, through a dual verification mechanism (main peak detection and same-frequency verification), it prevents mis-tracking of the traveling wave signal itself, and can reduce the mis-tracking rate in the arc fault scenario.

[0039] Further, the same-frequency verification includes:

[0040] Detect whether there is a spectral peak greater than the first preset peak value at the main peak frequency of the background noise of the fusion signal:

[0041] If it exists, the same-frequency verification is established;

[0042] Among them, the first preset peak value is determined according to the interference amplitude.

[0043] The beneficial effects of the above solution are as follows: By setting the first preset peak value, the present invention ensures that only significant interference is suppressed, avoiding excessive filtering and reducing the waveform distortion rate of the traveling wave head. In addition, different voltage levels can be compatible through the first preset peak value, reducing the on-site calibration workload.

[0044] Further, performing the Teager energy operator transform on the filtered signal to capture the mutation point of the fault traveling wave head and determine the arrival time of the traveling wave includes the following steps:

[0045] Discretize the filtered signal to calculate the instantaneous energy;

[0046] According to the instantaneous energy, calculate the energy mean value and energy standard within the second sliding time window, and determine the traveling wave trigger threshold according to the energy mean value and energy standard;

[0047] When the instantaneous energy is greater than the first preset condition, it is the arrival time of the traveling wave, where the first preset condition includes the traveling wave trigger threshold.

[0048] The beneficial effects of the above solution are as follows: With the real-time dynamic traveling wave trigger threshold of the present invention, it can adapt to the noise environment, solve problems such as false operation during thunderstorms and missed detection of high-resistance faults caused by fixed thresholds, and improve the detection rate of high-resistance grounding.

[0049] Further, the first preset condition includes:

[0050] Within the first duration, the instantaneous energy is greater than the traveling wave trigger threshold, and the energy rising rate is greater than the rising threshold;

[0051] Among them, the first duration is determined according to the duration characteristics of the traveling wave head; the rising threshold is determined according to impulse interference and oscillating decay waves.

[0052] The beneficial effects of the above solution are as follows: By verifying the energy mutation detection through the first duration, the present invention can not only filter out most of the impulse interference but also exclude oscillating decay waves. In addition, through the verification of the steepness of the traveling wave head, the real traveling wave head can be identified.

[0053] On the other hand, the present invention provides a traveling wave detection system based on a high-frequency current sensor, including:

[0054] An edge computing unit, the edge computing unit is built with an FPGA chip, and the FPGA chip is used to execute the above-mentioned traveling wave detection method and output the arrival time of the traveling wave and the fault direction identifier.

[0055] The beneficial effects of the above solution are as follows: The present invention can identify the fault direction based on the traveling wave polarity, and solve the problem of fault line selection in multi-branch distribution networks.

[0056] Furthermore, the traveling wave detection system based on the high-frequency current sensor further includes:

[0057] A fault diagnosis cloud platform for receiving the data of the arrival time of traveling waves at multiple nodes and calculating the fault distance by means of the double-ended traveling wave location method.

[0058] The beneficial effects of the above solution are as follows: Through double-ended traveling wave location, the present invention can dynamically calibrate the traveling wave speed and shorten the fault search time. In addition, the fusion of multi-node data to identify complex faults can improve the accuracy of fault diagnosis.

[0059] Furthermore, the edge computing unit integrates a noise learning module, and the noise learning module is used to periodically collect the background noise spectrum and update the parameters of the notch filter.

[0060] The beneficial effects of the above solution are as follows: By optimizing the notch center frequency, the present invention maintains a high noise suppression ability; by optimizing the spectrum feature library, it can improve the seasonal adaptability.

[0061] Compared with the prior art, the beneficial effects of the present invention at least include:

[0062] The present invention ensures the stability of power frequency current measurement through the main coil, and the auxiliary coil specializes in capturing high-frequency traveling waves, solves the problem of limited bandwidth of traditional sensors, expands the effective bandwidth from the traditional 5 MHz to 20 MHz, and reduces the amplitude error of traveling wave signals. In addition, through the three-level processing architecture of fusion, filtering, and TEO detection, the problem of signal connection distortion in the traditional solution is eliminated. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic flow chart of the traveling wave detection method based on the high-frequency current sensor according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this invention will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art. Identical reference numerals in the figures denote identical or similar structures, and thus their repetitive description will be omitted.

[0065] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first", "second", etc. may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise stated, the meaning of "a plurality" is two or more.

[0066] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "connected", "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood through specific circumstances.

[0067] The high-frequency current sensor of the present invention can be used for traveling wave detection of 35 kV lines. Specifically, the high-frequency current sensor is deployed at a preset node of the 35 kV line.

[0068] In some embodiments, the high-frequency current sensor is deployed at the outgoing line end, branch point, or cable joint of the 35 kV line.

[0069] The high-frequency current sensor of the present invention includes: a composite magnetic core module, a distributed signal acquisition module, a double-layer electromagnetic shielding structure, and a detachable installation structure.

[0070] The composite magnetic core module of the present invention is composed of an inner layer nanocrystalline alloy ring and an outer layer ferrite ring coaxially nested, and the effective bandwidth is extended from less than 5 MHz of traditional sensors to 20 MHz, covering the full frequency band requirements of traveling wave signals.

[0071] Specifically, the nanocrystalline alloy ring accounts for 70% to 80% of the total thickness of the magnetic core. In addition, the nanocrystalline alloy ring is a Fe-Si-B-based nanocrystalline strip, the strip thickness is 0.02 mm to 0.03 mm, and the initial magnetic permeability is greater than or equal to 5×10 4 @1 kHz, significantly improving the signal sensitivity of medium and low frequencies from 0.1 Hz to 10 MHz.

[0072] The ferrite ring accounts for 20% to 30% of the total thickness of the core. In addition, the ferrite ring is nickel-zinc ferrite Ni-Zn, with a magnetic permeability greater than or equal to 5000@1MHz, which compensates for the high-frequency attenuation of nanocrystals.

[0073] The distributed signal acquisition module of the present invention comprises a main coil and an auxiliary coil coaxially wound on a composite magnetic core module. The main coil is wound with a thick wire diameter to reduce resistance and reduce thermal noise caused by power frequency current; the auxiliary coil is wound with a thin wire diameter and densely wound, and the high number of turns improves the coupling efficiency of high-frequency signals and enhances high-frequency response. In addition, the main coil and the auxiliary coil are coaxially wound and output is fused through an impedance matching circuit.

[0074] Specifically, the axial spacing between the main coil and the auxiliary coil is 0.5mm to 1.0mm, and the output ends of the two are merged into a single signal output end through an impedance matching circuit. Furthermore, the main coil is wound with 8 to 12 turns of enameled copper wire with a wire diameter of 1.0mm to 1.5mm, and the auxiliary coil is wound with 90 to 110 turns of high-frequency enameled wire with a wire diameter of 0.1mm to 0.2mm.

[0075] In some embodiments, an auxiliary induction coil is provided inside the shielding layer of the high-frequency current sensor. The double-layer electromagnetic shielding structure of the present invention includes an inner permalloy magnetic shielding layer and an outer copper electric shielding layer that are close to the sensor housing. The permalloy magnetic shielding layer suppresses low-frequency magnetic field interference; the copper electric shielding layer reflects high-frequency electromagnetic waves. The double-layer shielding improves the noise suppression capability by 15dB to 20dB in the 1MHz to 20MHz frequency band.

[0076] Specifically, the thickness of the Permalloy magnetic shielding layer is 0.1mm to 0.3mm, and the thickness of the copper electric shielding layer is 0.5mm to 1.0mm. An insulating buffer layer with a thickness of 0.2mm to 0.5mm is filled between the Permalloy magnetic shielding layer and the copper electric shielding layer, and the material of the insulating buffer layer is silicone rubber or epoxy resin. The detachable installation structure of the present invention includes a split open magnetic core and a matching spring buckle, and a self-locking mechanism is provided at the opening of the magnetic core.

[0077] The spring buckle has a built-in pressure sensor to detect the closing status of the magnetic core in real time and provide feedback on the installation sealing through an LED indicator.

[0078] In some embodiments, the detachable mounting structure also includes a standardized mounting interface at the bottom of the shell, and the mounting interface includes a guide rail groove and a bolt hole, which is compatible with the guide rail installation in the DTU cabinet and the outdoor pole tower fixing.

[0079] The traveling wave detection method of the present invention can be implemented based on a high-frequency current sensor. Specifically, through the three-level processing architecture of fusion, filtering and TEO detection, the problem of signal connection distortion in the traditional solution is eliminated. In the actual measurement of 35kV cable fault, the wave head detection delay is reduced from 5μs to 0.8μs.

[0080] Reference Figure 1 , the traveling wave detection method of the present invention includes: step SS1 to step SS4.

[0081] Step SS1: Synchronously acquire the power frequency current signal Imain(t) output by the main coil and the high-frequency traveling wave signal Iaux(t) output by the auxiliary coil.

[0082] In application, the power frequency current signal Imain(t) output by the main coil and the high-frequency traveling wave signal Iaux(t) output by the auxiliary coil are synchronously acquired through the distributed signal acquisition module.

[0083] Step SS2: Fuse the power frequency current signal Imain(t) and the high-frequency traveling wave signal Iaux(t) to obtain the fused signal Ifuse(t).

[0084] In order to improve the signal-to-noise ratio and avoid misoperation caused by high-frequency noise under normal conditions, step SS2 of the present invention includes: step SS21 to step SS22.

[0085] Step SS21: Perform wavelet packet decomposition on the high-frequency traveling wave signal Iaux(t), extract the sub-signal energy Ehigh of the first frequency band, and calculate the dynamic weight coefficient α according to the power frequency current signal and the sub-signal energy.

[0086] In application, the first frequency band is determined according to the fault characteristic frequency band to ensure that the wavelet packet decomposition accurately extracts the fault characteristic frequency band and improves the signal-to-noise ratio. The high-frequency components of the fault traveling wave are concentrated in the range of 1 MHz to 20 MHz, and the effective bandwidth of the high-frequency current sensor is extended to 20 MHz. Therefore, the first frequency band can be the frequency band of 1 MHz to 20 MHz.

[0087] In practical application, the dynamic weight coefficient α includes the following formula:

[0088]

[0089] In the formula, α is the dynamic weight coefficient, Ehigh is the sub-signal energy, Imain(t) is the power frequency current signal, K1 is the dynamic adjustment coefficient, and K1 of the present invention is 0.1 - 0.5.

[0090] In some embodiments, in order to enhance the ability to capture traveling wave characteristics during a fault, ensure the non-attenuated transmission of the traveling wave signal, and improve the protection speed: when a short-circuit fault occurs on the line, the sub-signal energy Ehigh suddenly increases, and the dynamic weight coefficient α is 1. In some embodiments, in order to maintain the power frequency measurement accuracy during normal operation, completely isolate the high-frequency channel interference, and reduce the power frequency measurement error: when the line is operating normally, the sub-signal energy Ehigh approaches zero, and the dynamic weight coefficient α is 0.

[0091] Step SS22: Determine the fusion signal Ifuse(t) based on the dynamic weight coefficient, using the power frequency current signal Imain(t) and the high-frequency traveling wave signal Iaux(t).

[0092] When applied, the fusion signal Ifuse(t) includes the following formula:

[0093] Ifuse(t)=α·Iaux(t)+(1-α)·Imain(t)

[0094] In the formula, Ifuse(t) is the fusion signal, Iaux(t) is the high-frequency traveling wave signal, Imain(t) is the power frequency current signal, and α is the dynamic weight coefficient.

[0095] Step SS3: Perform notch filtering on the fusion signal Ifuse(t), and make the notch center frequency dynamically track the noise peak value of the distribution network background.

[0096] When applied, the noise peak value can be 50Hz, 150Hz or 250Hz.

[0097] In order to be able to suppress power frequency harmonics and high-frequency interference at the same time; in order to prevent mis-tracking of the traveling wave signal itself, that is, in the arc fault scenario, the mis-tracking rate can be reduced. Step SS3 of the present invention includes: Step SS31 to Step SS34.

[0098] Step SS31: Collect the bus voltage signal through a voltage sensor and extract the power frequency noise component.

[0099] When applied, the power frequency noise component includes: the power frequency noise fundamental wave and its main harmonic components. Among them, the power frequency noise fundamental wave is 50Hz, and the main harmonic components of the power frequency noise fundamental wave are 150Hz or 250Hz.

[0100] Step SS32: Collect the spatial electromagnetic noise signal Nref(t) through an auxiliary induction coil.

[0101] When applied, the spatial electromagnetic noise signal Nref(t) covers the frequency band from 100kHz to 20MHz.

[0102] Step SS33: Perform FFT calculation on the spatial electromagnetic noise Nref(t) within the first sliding time window to identify the main peak frequency fnoise of the background noise.

[0103] When applied, the window length of the first sliding time window can be 10ms.

[0104] Step SS34: When the main peak frequency fnoise is the power frequency noise component or greater than the first preset frequency band, and the same frequency verification is established, update the notch center frequency to the main peak frequency fnoise.

[0105] During application, a first preset frequency band is determined according to high-frequency interference and effective traveling wave signals, ensuring that the notch filter only tracks significant noise above this threshold and avoiding mis-filtering the effective components of traveling waves. The effective bandwidth of the high-frequency current sensor is extended to 20 MHz, so the effective traveling wave signals cover from 1 MHz to 20 MHz. High-frequency interference in the background noise of the distribution network, such as switching operation oscillations and arc noise, is mainly concentrated in the frequency band greater than 1 MHz. Therefore, the first preset frequency band can be 1 MHz. The power frequency noise components include: the power frequency noise fundamental wave and its main harmonic components. Among them, the power frequency noise fundamental wave is 50 Hz, and the main harmonic components of the power frequency noise fundamental wave are 150 Hz or 250 Hz.

[0106] The in-frequency verification of the present invention includes: detecting whether there is a spectral peak greater than the first preset peak at the main peak frequency fnoise of the background noise in the fused signal Ifuse(t); if it exists, the in-frequency verification is established; if it does not exist, the in-frequency verification is not established.

[0107] During application, the signal-to-noise ratio of the traditional scheme is low. In order to suppress significant noises such as power frequency harmonics and high-frequency interference while avoiding wavefront distortion of traveling waves caused by over-filtering, the first preset peak is determined according to the interference amplitude. Based on the measured comparison of the noise attenuation defect of the traditional scheme when the frequency is greater than 100 kHz and the noise suppression ability improvement of the high-frequency current sensor of the present application in the frequency band from 1 MHz to 20 MHz, the interference amplitude is greater than or equal to 10 dB. Therefore, the first preset peak can be 10 dB to ensure that only significant interference is suppressed to avoid over-filtering and reduce the wavefront waveform distortion rate of traveling waves. In some embodiments, different voltage levels can be made compatible by adjusting below the first preset peak, reducing the on-site calibration workload.

[0108] Step SS4: Perform Teager energy operator transformation on the filtered signal x(t) to capture the mutation point of the fault traveling wave head and determine the arrival time t0 of the traveling wave.

[0109] In some embodiments, the continuous form of the mutation point of the fault traveling wave head includes the following formula:

[0110]

[0111] In the formula, Ψ[x(t)] is the continuous form of the mutation point of the fault traveling wave head, and x(t) is the filtered signal.

[0112] In some embodiments, the discrete form of the mutation point of the fault traveling wave head includes the following formula:

[0113] Ψ[n]=x²[n]-x[n - 1]·x[n + 1]

[0114] Wherein, Ψ[n] is the mutation point of the wavefront of the fault traveling wave in discrete form, and n is the discrete time sequence.

[0115] In some embodiments, in order to adapt to the noise environment, solve problems such as false operation during thunderstorms caused by fixed thresholds and missed detection of high-resistance faults, and improve the detection rate of high-resistance grounding, step SS4 of the present invention includes: step SS41 to step SS43.

[0116] Step SS41: Discretize the filtered signal x(t) to calculate the instantaneous energy Ψ(t).

[0117] In application, the sampling rate fs can be greater than or equal to 40 MHz.

[0118] In actual application, the instantaneous energy Ψ(t) includes the following formula:

[0119] Ψ(t) = x(t)^2 - x(t - 1)·x(t + 1)

[0120] Wherein, Ψ(t) is the instantaneous energy, x(t) is the filtered signal, and t is the time.

[0121] Step SS42: According to the instantaneous energy Ψ(t), calculate the energy mean μ and the energy standard σ within the second sliding time window, and determine the traveling wave trigger threshold Vth according to the energy mean μ and the energy standard σ.

[0122] In application, the window length of the second sliding time window is 10 ms.

[0123] In actual application, the traveling wave trigger threshold Vth includes the following formula:

[0124] Vth = μ + K2·σ

[0125] μ = mean(Ψ[x(t)])

[0126] σ = std(Ψ[x(t)])

[0127] Wherein, Vth is the traveling wave trigger threshold, μ is the energy mean, σ is the energy standard, K2 is a dynamic coefficient, and K2 can be 3.0 to 4.0.

[0128] Step SS43: When the instantaneous energy Ψ(t) is greater than the first preset condition, it is the arrival time of the traveling wave, where the first preset condition includes the traveling wave trigger threshold Vth.

[0129] In some embodiments, the first preset condition of the present invention includes: within the first duration Δt, the instantaneous energy Ψ(t) is greater than the traveling wave trigger threshold Vth, and the energy rising rate Greater than the rising threshold. By verifying the energy mutation detection with the first duration, the present invention can not only filter out most of the pulse interferences, but also exclude the oscillating decaying waves. In addition, by verifying the steepness of the traveling wave front, the true traveling wave front can be identified.

[0130] In application, the first duration is determined according to the duration characteristic of the traveling wave front to ensure that the instantaneous energy mutation lasts long enough to distinguish from the transient interference. The mutation duration of the fault traveling wave is usually in the order of nanoseconds to microseconds. Therefore, the first duration Δt can be greater than or equal to 200 ns.

[0131] In practical application, the rising threshold can be determined according to the voltage change and time change of the traveling wave front in the time domain; it can also be determined according to the pulse interference and the oscillating decaying wave; it can also be determined according to the voltage change and time change of the traveling wave front in the time domain, as well as the pulse interference and the oscillating decaying wave. In some embodiments, the rising threshold can be 0.5×10 9 (V 2 / s).

[0132] The traveling wave detection system of the present invention is based on the application of high-frequency current sensors.

[0133] The traveling wave detection system of the present invention includes an edge computing unit. Further, it may also include a fault diagnosis cloud platform.

[0134] In order to be able to identify the fault direction based on the traveling wave polarity and solve the problem of fault line selection in multi-branch distribution networks, the edge computing unit of the present invention is built-in with an FPGA chip, and the FPGA chip is used to execute the above-mentioned traveling wave detection method to output the traveling wave arrival time t0 and the fault direction identifier.

[0135] In application, in order to maintain a high noise suppression ability and improve the seasonal adaptability, the edge computing unit of the present invention integrates a noise learning module, and the noise learning module is used to periodically collect the background noise spectrum and update the notch filter parameters to suppress the interference at specific frequency points, such as 5 MHz switching noise.

[0136] In practical application, the noise learning module collects the background noise spectrum once every 24 hours.

[0137] In order to be able to shorten the fault finding time and improve the fault diagnosis accuracy, the fault diagnosis cloud platform of the present invention is used to receive the traveling wave arrival time t0 data of multiple nodes and calculate the fault distance Lfault by the double-ended traveling wave location method.

[0138] In application, the fault distance Lfault includes the following formula:

[0139]

[0140] Wherein, Lfault is the fault distance, v is the traveling wave propagation speed, L is the total length of the line, and t01 and t02 are the detection times of the sensors at both ends.

[0141] In some embodiments, the fault diagnosis cloud platform is built-in with a high-resistance fault identification model. The traveling wave head steepness S and the first duration Δt are input into the high-resistance fault identification model. When the traveling wave head steepness S is less than the preset steepness Sset and the first duration Δt is greater than the preset duration, it is determined as a high-resistance grounding fault.

[0142] In application, the traveling wave head of a high-resistance grounding fault has a low steepness and a long duration. The duration of an arc fault can reach dozens of microseconds or more. According to the predicted data, the preset duration can be greater than 50 μs.

[0143] In practical application, the traveling wave head steepness S includes the following formula:

[0144]

[0145] Wherein, S is the traveling wave head steepness, Ψ[x(t)] is the mutation point of the fault traveling wave head, and x(t) is the filtered signal.

[0146] In some embodiments, the present invention can also provide another traveling wave detection system and is used based on a high-frequency current sensor.

[0147] The traveling wave detection system of the present invention includes: an acquisition module, a fusion module, a filtering module, and a determination module.

[0148] In application, the acquisition module of the present invention is used to synchronously acquire the power frequency current signal output by the main coil and the high-frequency traveling wave signal output by the auxiliary coil; the fusion module is used to fuse the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal; the filtering module is used to perform notch filtering on the fused signal and make the notch center frequency dynamically track the peak value of the distribution network background noise; the determination module is used to perform Teager energy operator transformation on the filtered signal to capture the mutation point of the fault traveling wave head and determine the arrival time of the traveling wave.

[0149] In some embodiments, the fusion module can also be configured to perform wavelet packet decomposition on the high-frequency traveling wave signal, extract the sub-signal energy of the first frequency band, and calculate the dynamic weight coefficient according to the power frequency current signal and the sub-signal energy; based on the dynamic weight coefficient, use the power frequency current signal and the high-frequency traveling wave signal to determine the fused signal.

[0150] In some embodiments, the filtering module may further be configured to collect a bus voltage signal through a voltage sensor and extract a power frequency noise component; collect a spatial electromagnetic noise signal through an auxiliary induction coil; perform FFT calculation on the spatial electromagnetic noise within a first sliding time window to identify the main peak frequency of the background noise; when the main peak frequency is the power frequency noise component or greater than a first preset frequency band and the same frequency verification is established, update the notch center frequency to the main peak frequency.

[0151] In some embodiments, the determination module may further be configured to discretize the filtered signal to calculate the instantaneous energy; calculate the energy mean and energy standard within a second sliding time window according to the instantaneous energy, and determine a traveling wave trigger threshold according to the energy mean and energy standard; when the instantaneous energy is greater than a first preset condition, it is the arrival time of the traveling wave, where the first preset condition includes the traveling wave trigger threshold.

[0152] As can be seen from the above embodiments, the present invention ensures the stability of power frequency current measurement through the main coil, and the auxiliary coil specializes in capturing high-frequency traveling waves, solving the problem of limited bandwidth of traditional sensors. The effective bandwidth is extended from the traditional 5 MHz to 20 MHz, and the amplitude error of the traveling wave signal is reduced. In addition, through a three-level processing architecture of fusion, filtering, and TEO detection, the problem of signal connection distortion in traditional solutions is eliminated. Further, the present invention accurately extracts the energy of the traveling wave characteristic frequency band in the first frequency band through wavelet packet decomposition, and the signal-to-noise ratio is improved under the interference of switch operations. In addition, the dynamic weight can avoid misoperation caused by high-frequency noise under normal conditions. Further, the present invention realizes strengthening the traveling wave characteristic capture ability during a fault through dynamic weight, ensuring the non-attenuated transmission of the traveling wave signal, and improving the protection speed; under normal conditions, maintaining the power frequency measurement accuracy, completely isolating the high-frequency channel interference, and reducing the power frequency measurement error. Further, the present invention can suppress both power frequency harmonics and high-frequency interference. In addition, through a dual verification mechanism (main peak detection and same frequency verification), mis-tracking of the traveling wave signal itself is prevented, and the mis-tracking rate can be reduced in the arc fault scenario. Further, the present invention ensures that only significant interference is suppressed by setting a first preset peak value to avoid over-filtering and reducing the waveform distortion rate of the traveling wave head. In addition, different voltage levels can be compatible through the first preset peak value, reducing the on-site calibration workload. Further, the present invention can adapt to the noise environment through a real-time dynamic traveling wave trigger threshold, solving problems such as misoperation during thunderstorms and missed detection of high-resistance faults caused by fixed thresholds, and improving the detection rate of high-resistance grounding. Furthermore, the present invention filters most of the pulse interference and excludes oscillatory decaying waves through the first duration verification of energy mutation detection. In addition, the true traveling wave head can be identified through the traveling wave head steepness verification.

[0153] In addition, the present invention can identify the fault direction based on the traveling wave polarity, solving the problem of fault line selection in multi-branch distribution networks. Through double-ended traveling wave positioning, the traveling wave speed can be dynamically calibrated, shortening the fault search time. In addition, multi-node data fusion is used to identify complex faults, which can improve the accuracy of fault diagnosis. By optimizing the notch center frequency, a high noise suppression ability can be maintained; by optimizing the spectral feature library, the seasonal adaptability can be improved.

[0154] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0155] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0156] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0157] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0158] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims. All of these are within the protection scope of the present invention.

Claims

1. A traveling wave detection method based on a high-frequency current sensor, characterized in that, The high-frequency current sensor includes a main coil and an auxiliary coil; The traveling wave detection method includes the following steps: Synchronously acquire the power frequency current signal output by the main coil and the high-frequency traveling wave signal output by the auxiliary coil; Fuse the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal; Perform notch filtering on the fused signal, and make the notch center frequency dynamically track the peak value of the background noise of the distribution network; Perform Teager energy operator transformation on the filtered signal to capture the mutation point of the fault traveling wave head and determine the arrival time of the traveling wave.

2. The traveling wave detection method based on a high-frequency current sensor according to claim 1, characterized in that The fusing the power frequency current signal and the high-frequency traveling wave signal to obtain a fused signal includes: Perform wavelet packet decomposition on the high-frequency traveling wave signal, extract the sub-signal energy of the first frequency band, and calculate the dynamic weight coefficient according to the power frequency current signal and the sub-signal energy; Based on the dynamic weight coefficient, determine the fused signal by using the power frequency current signal and the high-frequency traveling wave signal; Wherein, the first frequency band is determined according to the fault characteristic frequency band.

3. The traveling wave detection method based on a high-frequency current sensor according to claim 2, wherein The high-frequency current sensor is deployed at a preset node of a 35 kV line; The dynamic weight coefficient includes: When a short circuit fault occurs on the line, the sub-signal energy suddenly increases, and the dynamic weight coefficient is 1; When the line is operating normally, the sub-signal energy approaches zero, and the dynamic weight coefficient is 0.

4. The traveling wave detection method based on a high-frequency current sensor according to claim 1, wherein An auxiliary induction coil is provided inside the shielding layer of the high-frequency current sensor; The performing notch filtering on the fused signal and making the notch center frequency dynamically track the peak value of the background noise of the distribution network includes: Collect the bus voltage signal through a voltage sensor and extract the power frequency noise component; Collect the spatial electromagnetic noise signal through the auxiliary induction coil; Perform FFT calculation on the spatial electromagnetic noise within the first sliding time window to identify the main peak frequency of the background noise; When the main peak frequency is the power frequency noise component or greater than the first preset frequency band and the same frequency verification is established, update the notch center frequency to the main peak frequency; Wherein, the first preset frequency band is determined according to the high-frequency interference and the effective traveling wave signal.

5. The traveling wave detection method based on a high-frequency current sensor according to claim 4, wherein The same frequency verification includes: Detect whether there is a spectral peak greater than the first preset peak value at the main peak frequency of the background noise of the fused signal: If it exists, the same frequency verification is established; Wherein, the first preset peak value is determined according to the interference amplitude.

6. The traveling wave detection method based on a high-frequency current sensor according to claim 1, characterized in that The performing Teager energy operator transformation on the filtered signal to capture the mutation point of the fault traveling wave head and determine the arrival time of the traveling wave includes the following steps: Perform discretization processing on the filtered signal to calculate the instantaneous energy; According to the instantaneous energy, calculate the energy mean and the energy standard within the second sliding time window, and determine the traveling wave trigger threshold according to the energy mean and the energy standard; When the instantaneous energy is greater than the first preset condition, it is the arrival time of the traveling wave, where the first preset condition includes the traveling wave trigger threshold.

7. The traveling wave detection method based on a high-frequency current sensor according to claim 6, characterized in that, The first preset condition includes: Within the first duration, the instantaneous energy is greater than the traveling wave trigger threshold, and the energy rising rate is greater than the rising threshold; Wherein, the first duration is determined according to the duration characteristic of the traveling wave head; the rising threshold is determined according to the pulse interference and the oscillating decay wave.

8. A traveling wave detection system based on a high-frequency current sensor, characterized in that, Includes: An edge computing unit, which is built with an FPGA chip. The FPGA chip is used to execute the traveling wave detection method according to any one of claims 1 to 7, and output the arrival time of the traveling wave and the fault direction identifier.

9. The traveling wave detection system based on a high-frequency current sensor according to claim 8, characterized in that, It further includes: A fault diagnosis cloud platform, which is used to receive the traveling wave arrival time data of multiple nodes and calculate the fault distance by means of the double-end traveling wave location method.

10. The traveling wave detection system based on a high-frequency current sensor according to claim 8, characterized in that, The edge computing unit integrates a noise learning module, and the noise learning module is used to periodically collect the background noise spectrum and update the notch filter parameters.

Images