High-resistance fault detection method and device based on multi-level SVD decomposition

Through multi-level SVD decomposition technology, the zero-sequence voltage characteristics of the neutral point ungrounded system of the distribution network are analyzed, and the substance characteristics are generated and the modal component difference is calculated, which solves the problem of harmonics and noise interference in high-impedance fault detection. Improves the accuracy and adaptability of the detection.

CN118884104APending Publication Date: 2024-11-01NORTH CHINA UNIVERSITY OF TECHNOLOGY
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202410904321.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-05
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

The existing high-impedance fault detection methods are susceptible to harmonics and noise interference when the electrical amount of high-impedance ground fault is weak, and have poor noise resistance and are difficult to deal with nonlinear non-stationary signals, resulting in a reduced detection reliability.

Method used

Using a multi-level SVD decomposition method, the zero-sequence voltage of the neutral point ungrounded system of the distribution network is analyzed through transient characteristics and spectrum characteristics, and the substance characteristics are generated through singular value decomposition and variational modal decomposition, and the fault and normal are calculated. The difference in the mode component during operation is detected compared with the preset threshold.

Benefits of technology

It effectively expands the signal characteristics before and after the fault, improves the accuracy of identification and anti-interference ability of high-impedance faults, and enhances the reliability and adaptability of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118884104A_ABST
    Figure CN118884104A_ABST
Patent Text Reader

Abstract

The invention relates to a high-resistance fault detection method and device based on multi-level SVD decomposition. The method comprises the steps of performing singular value decomposition (SVD) processing of a preset number of times on a zero-sequence voltage signal based on transient characteristics and spectral characteristics of the zero-sequence voltage, and generating essential characteristics of the zero-sequence voltage signal; decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition (VMD), generating a mode component of a preset mode, and performing summation processing to obtain a mode component sum value of the preset mode; and calculating a difference value between the modal component sum value of the preset modal under the high-resistance fault and the modal component sum value of the preset modal during normal operation of the line, comparing the difference value with a preset threshold value, and completing high-resistance fault detection based on a comparison result. According to the invention, through multi-level SVD noise amplitude weakening and VMD decomposition, the signal characteristics before and after the fault and under different interference factors are effectively expanded, and the high-resistance fault can be effectively discriminated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of distribution network fault detection, and in particular to a high-resistance fault detection method and device based on multi-level SVD decomposition. Background Art

[0002] Distribution network lines are usually close to the ground, easy to contact with trees or disconnected, so they are prone to high-resistance grounding faults such as air breakdown and grounding through trees. Especially for neutral point ungrounded systems, when a high-resistance grounding fault occurs, the sudden change of electrical quantities will be extremely weak due to the high excess resistance, which brings certain difficulties to the reliable identification of high-resistance faults. If the high-resistance fault is not detected and the protection is not started in time, it may also evolve into a two-phase fault or a more serious fault, and may even cause a fire. It can be seen that high-resistance faults pose a great threat to life and engineering safety. Therefore, in the context of the accelerated construction of urban and rural distribution networks in my country, it is increasingly important to quickly and reliably detect high-resistance faults. However, fault detection methods often reduce detection reliability because the sudden change of electrical quantities in high-resistance faults is weak and easily interfered by factors such as harmonics and noise.

[0003] Nowadays, there are many studies on high-resistance grounding fault detection methods. According to the different analysis domains, they can be classified into time domain method, frequency domain method and time-frequency domain method. The time domain method processes problems in the time domain, mainly capturing the fault characteristics of zero-sequence current and zero-sequence voltage signals in the time domain, and using mathematical processing methods to determine the high-resistance fault based on the fault characteristics. The frequency domain method usually achieves high-resistance fault detection by decomposing the fault signal, extracting the harmonic components, and using the harmonic components to construct the judgment criteria. In actual engineering, the frequency domain method usually has poor noise resistance due to the inability to combine the time domain characteristics of the fault. With the development of signal processing technology, in order to improve the defects of the time domain method and frequency domain method that are susceptible to interference, it is proposed to use time-frequency domain analysis technology to extract feature quantities. The more typical method in the early years was the wavelet transform method, which was called a "mathematical magnifying glass" and could amplify the characteristics of local signals in the time-frequency domain.

[0004] However, in actual engineering, the problems of wavelet analysis have gradually been exposed, such as poor adaptability, susceptibility to the Heisenberg uncertainty principle, and difficulty in processing nonlinear and non-stationary signals in actual engineering, which has reduced the scope of application of wavelet analysis. Although the EMD (Empirical Mode Decomposition) method introduced later has solved some of the problems of wavelet analysis, it is prone to mode aliasing.

[0005] Therefore, one or more methods are needed to solve the above problems.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] The purpose of the present disclosure is to provide a high-resistance fault detection method and device based on multi-level SVD decomposition, thereby overcoming one or more problems caused by limitations and defects of related technologies at least to a certain extent.

[0008] According to one aspect of the present disclosure, a high-resistance fault detection method based on multi-level SVD decomposition is provided, comprising:

[0009] Analyze the zero-sequence voltage of the distribution network line system with ungrounded neutral point during high-resistance fault to obtain transient characteristics and spectrum characteristics of the zero-sequence voltage;

[0010] Based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, the zero-sequence voltage signal is subjected to a preset number of singular value decomposition (SVD) processes to generate essential characteristics of the zero-sequence voltage signal;

[0011] Decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition (VMD), generating modal components of a preset mode, and performing summation processing to obtain the modal components and values ​​of the preset mode;

[0012] The difference between the modal component and value of the preset mode under the high resistance fault and the modal component and value of the preset mode when the line is operating normally is calculated, and the difference is compared with the preset threshold value, and the high resistance fault detection is completed based on the comparison result.

[0013] In an exemplary embodiment of the present disclosure, the method further includes:

[0014] The transient characteristics of zero-sequence voltage in the distribution network with ungrounded neutral point during high-resistance fault are analyzed, and the amplitude characteristics of zero-sequence voltage during high-resistance fault are obtained.

[0015] The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point during high-resistance fault are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage during high-resistance fault are obtained.

[0016] In an exemplary embodiment of the present disclosure, the method further includes:

[0017] The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point under interference are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage under interference are obtained.

[0018] In an exemplary embodiment of the present disclosure, the method further includes:

[0019] Based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, performing singular value decomposition (SVD) processing on the zero-sequence voltage signal to generate a first SVD processing result;

[0020] Taking the first SVD processing result as input, performing singular value decomposition SVD processing to generate a second SVD processing result;

[0021] The second SVD processing result is used as the essential feature of the zero-sequence voltage signal.

[0022] In an exemplary embodiment of the present disclosure, the method further includes:

[0023] Extracting a zero-sequence voltage signal within 0.1s after a high-resistance fault occurs, performing a preset number of singular value decomposition (SVD) processes on the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs, and generating essential features of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs;

[0024] The essential characteristics of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs are decomposed based on variational mode decomposition (VMD) to generate modal components under 5 modes, and the modal components and values ​​of the preset modes under the high-resistance fault are calculated.

[0025] In an exemplary embodiment of the present disclosure, the method further includes:

[0026] Extracting a zero-sequence voltage signal when the line is operating normally, performing a preset number of singular value decomposition (SVD) processes on the zero-sequence voltage signal when the line is operating normally, and generating essential features of the zero-sequence voltage signal when the line is operating normally;

[0027] The essential characteristics of the zero-sequence voltage signal during normal operation of the line are decomposed based on variational mode decomposition (VMD) to generate modal components under five modes, and the modal components and values ​​of the preset modes under high-resistance faults are calculated.

[0028] In an exemplary embodiment of the present disclosure, the method further includes:

[0029] Calculate the difference between the modal component sum value of the preset mode under high resistance fault and the modal component sum value of the preset mode when the line is operating normally;

[0030] The maximum value of the difference is selected and compared with a preset threshold, and high-resistance fault detection is completed based on the comparison result.

[0031] In one aspect of the present disclosure, a high-resistance fault detection device based on multi-level SVD decomposition is provided, comprising:

[0032] A characteristic analysis module is used to analyze the zero-sequence voltage of a distribution network line system with an ungrounded neutral point during a high-resistance fault, and obtain transient characteristics and spectrum characteristics of the zero-sequence voltage;

[0033] An SVD processing module, configured to perform a preset number of singular value decomposition SVD processes on the zero-sequence voltage signal based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, so as to generate essential characteristics of the zero-sequence voltage signal;

[0034] A VMD processing module, used for decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition VMD, generating modal components of a preset mode and performing summation processing to obtain the modal components and values ​​of the preset mode;

[0035] The high-resistance fault detection module is used to calculate the difference between the modal component and value of the preset mode under high-resistance fault and the modal component and value of the preset mode when the line is operating normally, and compare the difference with a preset threshold value, and complete high-resistance fault detection based on the comparison result.

[0036] A high-resistance fault detection method based on multi-level SVD decomposition in an exemplary embodiment of the present disclosure analyzes the zero-sequence voltage during a high-resistance fault in a distribution network line system with an ungrounded neutral point to obtain the transient characteristics and spectral characteristics of the zero-sequence voltage; based on the transient characteristics and spectral characteristics of the zero-sequence voltage, the zero-sequence voltage signal is subjected to a preset number of singular value decomposition SVD processing to generate the essential characteristics of the zero-sequence voltage signal; the essential characteristics of the zero-sequence voltage signal are decomposed based on variational mode decomposition VMD to generate modal components of a preset mode and perform summation processing to obtain the modal component sum of the preset mode; the difference between the modal component sum of the preset mode under a high-resistance fault and the modal component sum of the preset mode when the line is operating normally is calculated, and the difference is compared with a preset threshold value, and high-resistance fault detection is completed based on the comparison result. The present invention uses multiple SVD to weaken the noise amplitude of the zero-sequence voltage signal 0.1s after the line fault, and then performs VMD decomposition on the signal to calculate the sum of the absolute values ​​of the signals in one power frequency cycle after each modal fault, thereby effectively expanding the signal characteristics before and after the fault and under different interference factors, and can effectively identify high-resistance faults.

[0037] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and other features and advantages of the present disclosure will become more apparent by describing in detail example embodiments thereof with reference to the attached drawings.

[0039] Figure 1A flowchart of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0040] Figure 2 A Cassie arc model equivalent circuit diagram of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0041] Figure 3 A high-resistance fault equivalent circuit diagram of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0042] Figure 4 A comparison diagram of electrical quantities of high-resistance fault and low-resistance fault according to a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0043] Figure 5 A schematic diagram of zero-sequence voltage affected during normal operation of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0044] Figure 6 A schematic diagram of zero-sequence voltage spectrum under different states of a line according to a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0045] Figure 7 A VMD algorithm flow chart of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0046] Figure 8 A schematic diagram of singular value decomposition of a noise signal of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0047] Fig. 9 A high-resistance fault determination flow chart of a high-resistance fault detection method based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown;

[0048] Fig.10 A schematic block diagram of a high-resistance fault detection device based on multi-level SVD decomposition according to an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0049] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0050] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, materials, devices, steps, etc. may be adopted. In other cases, known structures, methods and devices, implementations, materials or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0051] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.

[0052] In this exemplary embodiment, a high-resistance fault detection method based on multi-level SVD decomposition is first provided; Figure 1 As shown in , the high-resistance fault detection method based on multi-level SVD decomposition may include the following steps:

[0053] Step S110, analyzing the zero-sequence voltage of the distribution network line neutral point ungrounded system during a high-resistance fault to obtain transient characteristics and spectrum characteristics of the zero-sequence voltage;

[0054] Step S120, based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, performing singular value decomposition (SVD) processing on the zero-sequence voltage signal for a preset number of times to generate essential characteristics of the zero-sequence voltage signal;

[0055] Step S130, decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition (VMD), generating modal components of a preset mode, and performing summation processing to obtain the modal components and values ​​of the preset mode;

[0056] Step S140, calculating the difference between the modal component sum value of the preset mode under the high resistance fault and the modal component sum value of the preset mode when the line is operating normally, and comparing the difference with a preset threshold value, completing high resistance fault detection based on the comparison result.

[0057] A high-resistance fault detection method based on multi-level SVD decomposition in an exemplary embodiment of the present disclosure analyzes the zero-sequence voltage during a high-resistance fault in a distribution network line system with an ungrounded neutral point to obtain the transient characteristics and spectral characteristics of the zero-sequence voltage; based on the transient characteristics and spectral characteristics of the zero-sequence voltage, the zero-sequence voltage signal is subjected to a preset number of singular value decomposition SVD processing to generate the essential characteristics of the zero-sequence voltage signal; the essential characteristics of the zero-sequence voltage signal are decomposed based on variational mode decomposition VMD to generate modal components of a preset mode and perform summation processing to obtain the modal component sum of the preset mode; the difference between the modal component sum of the preset mode under a high-resistance fault and the modal component sum of the preset mode when the line is operating normally is calculated, and the difference is compared with a preset threshold value, and high-resistance fault detection is completed based on the comparison result. The present invention uses multiple SVD to weaken the noise amplitude of the zero-sequence voltage signal 0.1s after the line fault, and then performs VMD decomposition on the signal to calculate the sum of the absolute values ​​of the signals in one power frequency cycle after each modal fault, thereby effectively expanding the signal characteristics before and after the fault and under different interference factors, and can effectively identify high-resistance faults.

[0058] Next, a high-resistance fault detection method based on multi-level SVD decomposition in this exemplary embodiment will be further described.

[0059] Embodiment 1:

[0060] In step S110, the zero-sequence voltage of the distribution network line with an ungrounded neutral point during a high-resistance fault may be analyzed to obtain transient characteristics and frequency spectrum characteristics of the zero-sequence voltage.

[0061] In the embodiment of this example, the method further includes:

[0062] The transient characteristics of zero-sequence voltage in the distribution network with ungrounded neutral point during high-resistance fault are analyzed, and the amplitude characteristics of zero-sequence voltage during high-resistance fault are obtained.

[0063] The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point during high-resistance fault are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage during high-resistance fault are obtained.

[0064] In the embodiment of this example, the method further includes:

[0065] The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point under interference are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage under interference are obtained.

[0066] In step S120, singular value decomposition (SVD) processing may be performed on the zero-sequence voltage signal for a preset number of times based on the transient characteristics and spectrum characteristics of the zero-sequence voltage to generate essential features of the zero-sequence voltage signal.

[0067] In the embodiment of this example, the method further comprises:

[0068] Based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, performing singular value decomposition (SVD) processing on the zero-sequence voltage signal to generate a first SVD processing result;

[0069] Taking the first SVD processing result as input, performing singular value decomposition SVD processing to generate a second SVD processing result;

[0070] The second SVD processing result is used as the essential feature of the zero-sequence voltage signal.

[0071] In the embodiment of this example, a zero-sequence voltage signal within 0.1s after a high-resistance fault occurs is extracted, and singular value decomposition (SVD) processing is performed on the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs for a preset number of times to generate essential features of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs.

[0072] In the embodiment of this example, a zero-sequence voltage signal when the line is operating normally is extracted, and a singular value decomposition (SVD) process is performed on the zero-sequence voltage signal when the line is operating normally for a preset number of times to generate essential features of the zero-sequence voltage signal when the line is operating normally.

[0073] In step S130, the essential characteristics of the zero-sequence voltage signal may be decomposed based on variational mode decomposition (VMD) to generate modal components of a preset mode and perform summation processing to obtain the modal components and values ​​of the preset mode.

[0074] In the embodiment of this example, the essential characteristics of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs are decomposed based on variational mode decomposition VMD to generate modal components under 5 modes, and the modal components and values ​​of the preset modes under the high-resistance fault are calculated.

[0075] In the embodiment of this example, the essential characteristics of the zero-sequence voltage signal during normal operation of the line are decomposed based on variational mode decomposition (VMD) to generate modal components under 5 modes, and the modal components and values ​​of the preset modes under high-resistance faults are calculated.

[0076] In step S140, the difference between the modal component sum value of the preset mode under high resistance fault and the modal component sum value of the preset mode when the line is operating normally can be calculated, and the difference can be compared with a preset threshold value, and high resistance fault detection is completed based on the comparison result.

[0077] In the embodiment of this example, the method further includes:

[0078] Calculate the difference between the modal component sum value of the preset mode under high resistance fault and the modal component sum value of the preset mode when the line is operating normally;

[0079] The maximum value of the difference is selected and compared with a preset threshold, and high-resistance fault detection is completed based on the comparison result.

[0080] Embodiment 2:

[0081] In the embodiment of this example, the high-resistance grounding fault characteristic analysis includes:

[0082] When a high-resistance grounding fault occurs, the zero-sequence voltage and zero-sequence current change more obviously after the fault than the three-phase voltage and three-phase current. However, even if the zero-sequence voltage changes significantly, its electrical quantity is still very weak compared to other fault types, which can easily be confused with the identification of harmonics or noise, thereby affecting the action of the protection device. In response to the above problems, the present invention analyzes the transient characteristics of high-resistance grounding faults, compares the characteristics of high-resistance fault electrical quantities with the time domain and frequency domain waveforms of harmonics and noise, and lays a theoretical foundation for high-resistance fault detection methods.

[0083] In the embodiment of this example, the Cassie arc model assumes that the arc has a cylindrical gas channel, the conductance outside the channel is smaller than the conductance inside the channel, and the temperature of the channel cross section is uniformly distributed and constant, and the arc energy and energy dissipation rate are proportional to the cross-sectional area of ​​the arc column channel. The model equation is:

[0084]

[0085] In formula (1), R arc is the arc dynamic resistance, C is the arc time constant, u C is the arc voltage constant or breakdown voltage, and u is the arc instantaneous voltage.

[0086] The arc resistance of the Cassie arc model can be regarded as a variable resistor R affected by the instantaneous voltage, breakdown voltage and time constant. arc , its equivalent circuit is as follows Figure 2 shown.

[0087] Its arc resistance can be expressed as formula (2):

[0088]

[0089] In formula (2), u is the instantaneous voltage of the arc, E is the arc breakdown voltage, and τ is the time constant. When a high-resistance fault occurs, the arc resistance changes with the alternation of arcing and arc extinction. The arc resistance is close to 0 when the arc is on and is the maximum value when the arc is extinguished. The arc resistance data used in the present disclosure when studying high-resistance fault detection are all arc resistance values ​​when the arc is extinguished, that is, the maximum arc resistance value.

[0090] In the embodiment of this example, the zero-sequence voltage transient characteristic analysis includes:

[0091] like Figure 3 This is the high-resistance fault equivalent circuit of the neutral point ungrounded system. In order to simplify the model of the neutral point ungrounded system, the fault point can be equivalent to a virtual power supply. Among them U m is the phase voltage amplitude of the fault phase during normal operation, ω is the angular frequency of the power frequency, is the fault initial phase angle. Figure 2 In, R h is the neutral point to ground resistance; I 0n is the zero-sequence current flowing through the fault line outlet, I 0f is the zero-sequence current flowing through the fault point; C j (j=1, 2, ...n) is the zero-sequence distributed capacitance of the j-th feeder to the ground; I cj Indicates the zero-sequence distributed capacitance C flowing through the jth feeder to ground j The current u c is the equivalent distributed capacitance C n The voltage, u 0f is the zero-sequence voltage of the bus.

[0092] Zero-sequence current I flowing through the fault line n outlet 0n It is composed of the transient component of zero-sequence current I flowing through the fault point 0f and the transient current i of the zero-sequence capacitance of the fault line to ground cn Together. Figure 2 It is not difficult to derive the first-order differential equation expression of the virtual power supply from the equivalent circuit:

[0093] u f =u c +R(i cΣ +i 0z ) (3)

[0094] In formula (3), i c∑ is the sum of the currents of all feeder-to-ground capacitances, i 0z is the zero-sequence current flowing into the earth through the neutral point. To simplify the calculation, assume that the neutral point is not grounded and the neutral point to ground resistance R h is approximately equal to 0, then i 0z tends to 0, we can get:

[0095]

[0096] In formula (4), C∑ is the sum of the zero-sequence distributed capacitance of all feeders to ground. Therefore, formula (3) can be written as:

[0097]

[0098] By solving equation (5), we can get the bus zero-sequence voltage u c As shown in formula (6):

[0099]

[0100] The parameters θ and δ in formula (6) are:

[0101]

[0102] Depend on Figure 3 It can be concluded that the basic equation relationship of the zero-sequence current at the fault line outlet of the ungrounded system is:

[0103] i 0n =-i 0f +i cn (8)

[0104] In formula (8), i 0f It is equal to the sum of the currents flowing through the zero-sequence distributed capacitance of all j feeders to ground. c∑ ,i cn is the current flowing through the fault line-to-ground capacitance, so equation (8) can be expanded to:

[0105]

[0106] Zero sequence current i of non-fault line in ungrounded neutral point system cj It can be written as:

[0107]

[0108] From equations (6) to (10), it can be concluded that when a high-resistance fault occurs in a neutral point ungrounded system, its zero-sequence voltage is composed of a DC component and a steady-state sinusoidal component, while its zero-sequence current is composed of a DC component and a steady-state cosine component. In addition, as the transition resistance R increases, the steady-state voltage, transient voltage, and initial amplitudes of the current components will decrease accordingly.

[0109] Compared with low-resistance grounding fault, the electrical quantity characteristics of high-resistance grounding fault are much weaker than those of low-resistance grounding fault. The comparison of zero-sequence voltage and zero-sequence current under high-resistance fault and low-resistance fault is shown in the figure below. Figure 4 shown.

[0110] from Figure 4It can be seen that the amplitudes of zero-sequence voltage and zero-sequence current are smaller for high-resistance faults than for general low-resistance faults. Since the transition resistance of high-resistance faults tends to infinity, the sudden change of zero-sequence current during faults is not significantly different from that during normal operation. The fault characteristics of zero-sequence voltage are more obvious than those of zero-sequence current.

[0111] When the system is operating normally, it is often disturbed by harmonics or noise, which also causes the zero-sequence voltage at the disturbed point to change as follows: Figure 5 Mutations shown.

[0112] Depend on Figure 4 and Figure 5 It shows that if the zero-sequence voltage is affected by harmonics or noise during normal operation, it will also change suddenly, and its sudden change value is very different from the sudden change value of the zero-sequence voltage when a high-resistance fault occurs. Therefore, it is not reliable to judge whether a high-resistance fault occurs only based on the sudden change value of the zero-sequence voltage at a certain moment. In order to distinguish the influence of harmonics and noise, it is necessary to further analyze the characteristics of high-resistance faults.

[0113] In the embodiment of this example, the zero-sequence voltage spectrum analysis of the high-resistance fault includes:

[0114] In order to further distinguish the characteristics of high-resistance faults, noise interference and harmonic effects occurring at a certain point in the line, the fast Fourier transform can be used to compare the spectra of the three states, such as Figure 6 As shown:

[0115] pass Figure 6 It can be seen that when a high-frequency fault occurs, Figure 6 The frequency band in the zero-sequence voltage shown in (a) is mainly concentrated near the fundamental frequency, while the high-frequency components are relatively small. Figure 6 (b) Zero-sequence voltage under noise interference, the frequency is evenly distributed in each frequency band with low amplitude. When there is harmonic influence, Figure 6 The frequency of the zero-sequence voltage shown in (c) is mostly distributed in the high-frequency part, and the component near the fundamental frequency is almost 0. Based on the spectral characteristics of the three, the idea of ​​high-resistance fault detection can start from the distribution of zero-sequence voltage frequency, and then distinguish the zero-sequence voltage mutation caused by high-resistance fault, noise interference and harmonic influence, so as to identify the high-resistance fault.

[0116] In order to more accurately extract the spectrum characteristics of the zero-sequence voltage of a high-resistance fault, the present disclosure proposes a method of using VMD and SVD to refine the spectrum characteristics, further expanding the fault characteristics, thereby improving the accuracy of high-resistance fault identification.

[0117] In the embodiment of this example, the high resistance fault identification method based on VMD-SVD includes a VMD algorithm and SVD, wherein the VMD algorithm includes:

[0118] The VMD algorithm defines the intrinsic mode function as an AM-FM function with a limited bandwidth, which can effectively avoid the endpoint effect. The main function of the VMD algorithm is to decompose the original signal into K IMF (intrinsic mode function, IMF) components by constructing and solving the constrained variational problem through iterative calculation.

[0119] To realize the function of VMD algorithm, we must first construct the variational problem, as shown in formula (11):

[0120]

[0121] In formula (11), K is the total number of modes expected to be decomposed, {u k} is the kth component after decomposition, {ω k} is the center frequency of the kth component. Then, to solve equation (11), it is necessary to introduce the Lagrangian operator λ and transform it into an unconstrained variational problem, and we can get:

[0122]

[0123] Introducing the quadratic penalty factor α in equation (12) can reduce the Gaussian noise interference in the signal. k n+1 (t),u i (t), ω(t) and λ(t) are transformed by Fourier isometry, and then the optimal components and center frequencies of each mode are calculated by iterative algorithm. The solution process is shown in equations (13) and (14):

[0124]

[0125] In order to limit the distortion of the decomposed signal within a certain range, it is necessary to introduce the noise tolerance γ. The solution process of the Lagrangian operator λ is shown in formula (15):

[0126]

[0127] Then the calculation result is compared with formula (16). If the accuracy requirement of formula (16) is met, the iteration is terminated. If the accuracy requirement is not met, it returns to step (13) to continue the next iterative calculation.

[0128]

[0129] The specific process of VMD algorithm is as follows Figure 7 As shown:

[0130] In the embodiment of this example, the singular value decomposition VMD includes:

[0131] Assume that matrix A is an m×n matrix and define the SVD of matrix A as:

[0132] A=USV T (17)

[0133] Where U is an m×m matrix, S is an m×n matrix, all elements except the main diagonal are 0, each element on the main diagonal is called a singular value, and V is an n×n matrix. U and V are both unitary matrices, that is, they satisfy U H U=E,V H V=E. For n×n square matrix A T A can be decomposed by eigenvalues:

[0134] (A T A)v i =λ i v i (18)

[0135] A T The n eigenvectors of A are written as an n×n matrix, which is the matrix V in equation (17). The eigenvectors in V are called right singular vectors of the matrix A. T Performing feature decomposition yields:

[0136] (AA T ) i =λ i u i (19)

[0137] AA T The m eigenvectors of A are written as an m×m matrix, which is the matrix U in equation (17). The eigenvectors in U are called left singular vectors of the matrix A. i Arrange from large to small, take the positive singular values ​​and take the square root to form a diagonal matrix:

[0138]

[0139] Add "0" to the rows and columns of matrix D to form an m×n matrix:

[0140]

[0141] Matrix S is a singular value matrix, and the elements on the diagonal of matrix S are the singular values ​​of matrix A. So far, all the substitution matrices in equation (17) are obtained, that is, the singular value decomposition of matrix A is completed.

[0142] SVD constructs a matrix for signal data, and then finds the singular value of the matrix to obtain the essential characteristics of the signal and filter out the secondary characteristics of the signal. When the normal operation of the line is disturbed by noise, the zero-sequence voltage and noise generated by the incompletely symmetrical operation form a mixed signal. Since the zero-sequence voltage during normal operation is very small and the signal-to-noise ratio is high, SVD cannot extract features from the mixed signal, but it can reduce the amplitude of the noise signal. For high-resistance fault signals and harmonic signals, the results after SVD do not affect the essential characteristics of the signal, so there will be no significant impact on its amplitude. Since the calculation amount of SVD is relatively small for computers, the output of one SVD can be used as the input of the second SVD for multiple SVDs, thereby greatly reducing the noise amplitude.

[0143] like Figure 8 As shown in the figure, the amplitude of the mixed signal is reduced after multiple SVDs, so the characteristics of noise and high-resistance faults can be further amplified by this method, laying a foundation for improving the reliability and adaptability of subsequent methods.

[0144] In the embodiment of this example, the VMD-SVD based high resistance fault detection includes:

[0145] Based on the above theory, this paper proposes a high-resistance fault detection scheme based on VMD and SVD:

[0146] Firstly, the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs is extracted. After performing n SVD operations on it and using the VMD algorithm to decompose the zero-sequence voltage, K (K=5) different modal components can be obtained.

[0147] Secondly, in order to better distinguish the signal characteristics under high-resistance fault and normal operation, the absolute value sum of each modal signal after VMD decomposition within one power frequency cycle after the fault occurs is calculated to amplify the characteristics of the zero-sequence voltage signal when the high-resistance fault and the normal operation of the line are disturbed by harmonics or noise. The sum value F of the signal under mode k k The expression is shown in formula (22):

[0148]

[0149] Wherein, N is the total number of sampling points of the signal. The present disclosure uses 20kHz to sample the zero-sequence voltage signal, that is, the number of sampling points N in one power frequency cycle is 400. k [n] is the sampled value of the signal component of the kth mode at the nth sampling point.

[0150] Again, the F k F k Perform the difference to get F kΔ , select F kΔThe maximum value F kΔmax , F kΔmax Compared with the preset threshold Q, if F kΔmax >Q, it is judged as a high-resistance fault, otherwise it is normal operation.

[0151] In summary, the high resistance fault detection process designed in this disclosure is as follows: Fig. 9 shown.

[0152] In the embodiment of this example, the high-resistance fault identification method based on VMD-SVD proposed in the present disclosure can draw the following conclusions: by performing VMD decomposition on the zero-sequence voltage signal 0.1s after the line fault, the sum of the absolute values ​​of the signals of one power frequency cycle after each modal fault is calculated, which effectively expands the signal characteristics before and after the fault and under different interference factors, and can effectively identify the difference in zero-sequence voltage characteristics under high-resistance faults and different interference factors; the present disclosure has selected a suitable SVD number through multiple groups of simulation verifications, which effectively increases the F of the zero-sequence voltage under noise interference and high-resistance faults. 5Δ , which improves the adaptability of the method to situations with higher transition resistance; through adaptability analysis, the method proposed in the present disclosure is not affected by factors such as line fault location, higher transition resistance, and noise with different signal-to-noise ratios, and has certain advantages in reliability and adaptability.

[0153] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0154] In addition, in this exemplary embodiment, a high-resistance fault detection device based on multi-level SVD decomposition is also provided. Fig.10 As shown, the high resistance fault detection device 200 based on multi-level SVD decomposition may include: a characteristic analysis module 210, an SVD processing module 220, a VMD processing module 230 and a high resistance fault detection module 240. Among them:

[0155] The characteristic analysis module 210 is used to analyze the zero-sequence voltage of the distribution network line neutral point ungrounded system during a high-resistance fault, and obtain the transient characteristics and spectrum characteristics of the zero-sequence voltage;

[0156] The SVD processing module 220 is used to perform a preset number of singular value decomposition SVD processes on the zero-sequence voltage signal based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, so as to generate essential characteristics of the zero-sequence voltage signal;

[0157] A VMD processing module 230 is used to decompose the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition VMD, generate modal components of a preset mode, and perform summation processing to obtain the modal components and values ​​of the preset mode;

[0158] The high resistance fault detection module 240 is used to calculate the difference between the modal component sum value of the preset mode under the high resistance fault and the modal component sum value of the preset mode when the line is operating normally, and compare the difference with the preset threshold value, and complete the high resistance fault detection based on the comparison result.

[0159] The specific details of each of the above-mentioned high-resistance fault detection device modules based on multi-level SVD decomposition have been described in detail in the corresponding high-resistance fault detection method based on multi-level SVD decomposition, so they will not be repeated here.

[0160] It should be noted that, although several modules or units of a high-resistance fault detection device 200 based on multi-level SVD decomposition are mentioned in the above detailed description, such division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0161] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0162] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0163] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A high resistance fault detection method based on multi-level SVD decomposition, characterized in that: The method comprises: Analyze the zero-sequence voltage of the distribution network line system with ungrounded neutral point during high-resistance fault to obtain transient characteristics and spectrum characteristics of the zero-sequence voltage; Based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, the zero-sequence voltage signal is subjected to a preset number of singular value decomposition (SVD) processes to generate essential characteristics of the zero-sequence voltage signal; Decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition (VMD), generating modal components of a preset mode, and performing summation processing to obtain the modal components and values ​​of the preset mode; The difference between the modal component and value of the preset mode under the high resistance fault and the modal component and value of the preset mode when the line is operating normally is calculated, and the difference is compared with the preset threshold value, and the high resistance fault detection is completed based on the comparison result.

2. The method according to claim 1, characterized in that The method further comprises: The transient characteristics of zero-sequence voltage in the distribution network with ungrounded neutral point during high-resistance fault are analyzed, and the amplitude characteristics of zero-sequence voltage during high-resistance fault are obtained. The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point during high-resistance fault are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage during high-resistance fault are obtained.

3. The method according to claim 2, characterized in that The method further comprises: The frequency spectrum characteristics of the zero-sequence voltage in the distribution network system with ungrounded neutral point under interference are analyzed, and the frequency band distribution characteristics of the zero-sequence voltage under interference are obtained.

4. The method according to claim 1, characterized in that The method further comprises: Based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, performing singular value decomposition (SVD) processing on the zero-sequence voltage signal to generate a first SVD processing result; Taking the first SVD processing result as input, performing singular value decomposition SVD processing to generate a second SVD processing result; The second SVD processing result is used as the essential feature of the zero-sequence voltage signal.

5. The method according to claim 1, characterized in that The method further comprises: Extracting a zero-sequence voltage signal within 0.1s after a high-resistance fault occurs, performing a preset number of singular value decomposition (SVD) processes on the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs, and generating essential features of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs; The essential characteristics of the zero-sequence voltage signal within 0.1s after the high-resistance fault occurs are decomposed based on variational mode decomposition (VMD) to generate modal components under 5 modes, and the modal components and values ​​of the preset modes under the high-resistance fault are calculated.

6. The method according to claim 5, characterized in that The method further comprises: Extracting a zero-sequence voltage signal when the line is operating normally, performing a preset number of singular value decomposition (SVD) processes on the zero-sequence voltage signal when the line is operating normally, and generating essential features of the zero-sequence voltage signal when the line is operating normally; The essential characteristics of the zero-sequence voltage signal during normal operation of the line are decomposed based on variational mode decomposition (VMD) to generate modal components under five modes, and the modal components and values ​​of the preset modes under high-resistance faults are calculated.

7. The method according to claim 6, characterized in that The method further comprises: Calculate the difference between the modal component sum value of the preset mode under high resistance fault and the modal component sum value of the preset mode when the line is operating normally; The maximum value of the difference is selected and compared with a preset threshold, and high-resistance fault detection is completed based on the comparison result.

8. A high resistance fault detection device based on multi-level SVD decomposition, characterized in that: The device comprises: A characteristic analysis module is used to analyze the zero-sequence voltage of a distribution network line system with an ungrounded neutral point during a high-resistance fault, and obtain transient characteristics and spectrum characteristics of the zero-sequence voltage; An SVD processing module, configured to perform a preset number of singular value decomposition SVD processes on the zero-sequence voltage signal based on the transient characteristics and spectrum characteristics of the zero-sequence voltage, so as to generate essential characteristics of the zero-sequence voltage signal; A VMD processing module, used for decomposing the essential characteristics of the zero-sequence voltage signal based on variational mode decomposition VMD, generating modal components of a preset mode and performing summation processing to obtain the modal components and values ​​of the preset mode; The high-resistance fault detection module is used to calculate the difference between the modal component and value of the preset mode under high-resistance fault and the modal component and value of the preset mode when the line is operating normally, and compare the difference with a preset threshold value, and complete high-resistance fault detection based on the comparison result.

Citation Information

Cited By

  • Heterogeneous data access method and device

    CN119513574A

  • 10kV distribution line forest fire fault early warning method based on zero sequence voltage signal

    CN119780604A

  • Power distribution network fault detection method and device based on variational mode decomposition

    CN120064825A