High resistance fault detection method, device and equipment based on arc current change trend

Abstract

The application discloses a high-resistance fault detection method, device and equipment based on arc current change trend, wherein the detection method first collects a feeder zero sequence current signal by using a signal detection device; secondly, the zero sequence current signal is processed by using a variational mode decomposition to obtain a component after filtering high-frequency modes; thirdly, the current instantaneous amplitude of the filtered component is extracted by using a windowed fast Fourier transform; then, the change trend of the arc current is quantitatively estimated by using a nonlinear least square method; finally, the high-resistance fault is detected by using the sign of the coefficient in the nonlinear function. The application uses the nonlinear least square method to quantitatively estimate the overall change trend of the arc current. Meanwhile, the irregularity and randomness of the arc current can be weakened. This stage can improve the robustness of the proposed HIF detection method.

Description

Technical Field

[0001] This invention belongs to the technical field of relay protection in power system distribution networks, specifically relating to a high-resistivity fault detection method, device, and equipment based on the trend of arc current variation. Background Technology

[0002] When a high-impedance fault occurs in an active distribution network, the arc current has weak characteristics, and its properties are determined by the system's geometry, environment, and electrical conditions. These non-fixed characteristics make detecting high-impedance faults (HIF) difficult and easily confused with capacitor switching (CS) or load switching (LS) currents. Simultaneously, the fault current is susceptible to the influence of harmonics and noise from distributed generators, potentially leading to low detection accuracy of existing HIF detection methods. Furthermore, when an HIF occurs, the grounding surface typically limits the fault current to less than 10% of the load current. According to data from the Power System Relay Protection Committee Working Group, the success rate of HIF detection using traditional protection strategies is less than 20%. Once protection fails, the energized conductor poses a hazard to the public and can cause fires and personal injury; therefore, accurate detection of high-impedance arc faults in active distribution networks is of paramount importance.

[0003] To address the problem of detecting high-resistance arc faults in active distribution networks, scholars both at home and abroad have conducted extensive research and proposed various types of detection methods. However, in summary, the accuracy and reliability of existing methods for detecting high-resistance arc faults in active distribution networks are still relatively low. Summary of the Invention

[0004] This invention provides a method, apparatus, and device for detecting high-resistance faults based on the trend of arc current variation, which can accurately detect high-resistance arc faults.

[0005] To achieve the above objectives, the present invention provides a high-resistance fault detection method based on the trend of arc current variation, comprising the following steps:

[0006] Step 1: Collect the zero-sequence current of the feeder, and use the variational mode decomposition algorithm to decompose the zero-sequence current to obtain the filtered component after removing the high-frequency modes.

[0007] Step 2: Process the filtered components and extract the instantaneous amplitude of the zero-sequence current. This is done by detecting the instantaneous amplitude of the zero-sequence current at the [missing information - likely a specific point or time value]. j Calculate the transient disturbances occurring at each time point. n The maximum current amplitude of the zero-sequence current within each sampling point;

[0008] Step 3: Based on the instantaneous amplitude of the maximum zero-sequence current, use a nonlinear function to quantitatively estimate the changing trend of the zero-sequence current and estimate the coefficients of the nonlinear function;

[0009] Step 4: Construct a high-resistance fault detection criterion using the signs of the coefficients of the nonlinear function, and determine whether a high-resistance fault has occurred based on the high-resistance fault detection criterion.

[0010] Furthermore, in step 1, the zero-sequence current is decomposed using a variational mode decomposition algorithm.

[0011] Furthermore, in step 2, a windowed fast Fourier transform is used to process the filtered components.

[0012] Furthermore, step 2 includes the following steps:

[0013] Step 2.1: Process the filtered components using WFFT transform. I FC (t), to obtain the discrete frequency domain signal ;

[0014] Step 2.2: Using the sliding discrete frequency domain signal The window is used to calculate the instantaneous amplitude sequence. A FC (t);

[0015] Step 2.3: Calculate the current amplitude Compared to the previous value The difference between D FC ( j ), j The number of sampling points. j =1,2,…, n ;

[0016] Step 2.4, when the difference ,in It is an instantaneous amplitude sequence The maximum value in indicates the value in the first... j Amplitude at each time point A transient disturbance occurs. When the transient disturbance is detected, matrix H is obtained. ;

[0017] Step 2.5: Decompose matrix H into matrices with dimension equal to The matrix G, , This indicates taking the largest integer, and , The number of sampling points in one period is N, where N is the total number of sampling points;

[0018] Step 2.6: Calculate the maximum current amplitude in each column of matrix G: ;

[0019] In the formula, It is the first in matrix G mThe maximum current amplitude is listed. Adjacent amplitudes and The time interval between them is T is the sampling period.

[0020] Furthermore, in step 2.1, the discrete frequency domain signal The formula for calculation is:

[0021]

[0022] In the formula, k is a positive integer, i These are sampling points. It is the resolution in the frequency domain. ; It is the filtered component The sampling time interval; A and These are the start and end points of the sampling time window, respectively, and the length of the window is... , For the filtered components Number of sampling points in one cycle.

[0023] Furthermore, step 3 includes the following steps:

[0024] Step 3.1: Determine the timeframe within the two cycles. Maximum current amplitude The corresponding instant is Select point As a starting point;

[0025] Step 3.2, for Count the X current amplitudes in the corresponding time; Represented as:

[0026] , ;

[0027] Step 3.3: Based on the obtained time points and current amplitude The expression of the nonlinear function Rewritten as ;

[0028] Step 3.4, through settings The nonlinear equation is transformed into a linear equation, as shown in the following expression:

[0029]

[0030] Step 3.5: Calculate the minimum sum of squared differences between observed and predicted values, resulting in the following formula:

[0031]

[0032] Step 3.6, through and get , The calculation formula is as follows:

[0033]

[0034] In the formula: , .

[0035] Furthermore, in step 3.2, X equals 5.

[0036] Furthermore, in step 4, the high-resistance fault detection criterion is:

[0037] like The current amplitude shows an increasing trend, indicating a high-resistance fault.

[0038] like This indicates that the current amplitude does not show an increasing trend, and it is determined that load switching or capacitor switching has occurred;

[0039] These are the coefficients of the nonlinear function.

[0040] A high-resistance fault detection device based on the trend of arc current variation includes:

[0041] The acquisition module is used to acquire the zero-sequence current of the feeder and decompose the zero-sequence current using a variational mode decomposition algorithm to obtain the filtered component after removing high-frequency modes.

[0042] The data extraction module processes the filtered components to extract the instantaneous amplitude of the zero-sequence current. This is achieved by detecting the instantaneous amplitude of the zero-sequence current at the [missing information - likely a specific point or time value]. j Calculate the transient disturbances occurring at each time point. n The maximum current amplitude of the zero-sequence current within each sampling point;

[0043] The coefficient calculation module is used to quantitatively estimate the changing trend of the zero-sequence current and the coefficients of the nonlinear function based on the instantaneous amplitude of the maximum zero-sequence current.

[0044] The fault determination module is used to construct a high-resistance fault detection criterion using the sign of the coefficients of the nonlinear function, and to determine whether a high-resistance fault has occurred based on the high-resistance fault detection criterion.

[0045] A computer device includes an electrically connected memory and a processor, wherein the memory stores a computer program executable on the processor, and when the processor executes the computer program, it implements the steps of the fault detection method described above.

[0046] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0047] 1) Regarding trend estimation: Nonlinear least squares method is used to quantitatively estimate the overall trend of arc current variation. This also reduces the irregularity and randomness of the arc current. This stage improves the robustness of the proposed HIF detection method.

[0048] 2) In the stage of constructing the detection criteria, the accumulation characteristics of the electric arc were analyzed, and a nonlinear function was derived to represent the relationship between the current amplitude and the voltage amplitude. The sign of the coefficients of the nonlinear function was used to detect HIF. This method has advantages such as applicability to different power distribution systems and strong noise interference resistance, thus improving the reliability of the detection method.

[0049] 3) Signal processing: Variational mode decomposition (VMD) and windowed fast Fourier transform (WFFT) are used to process the measured arc current to reduce noise and harmonics in the arc current, providing an accurate basis for HIF detection. Attached Figure Description

[0050] Figure 1 This is a flowchart of the HIF fault detection method for power distribution networks described in this specification.

[0051] Figure 2 A schematic diagram of the module structure of the fault device provided by the present invention;

[0052] Figure 3 The power distribution network model described in this specification;

[0053] Figure 4 This is a schematic diagram of a diode-resistance-based model of the HIF fault used in an embodiment of the present invention.

[0054] Figure 5 When the HIF, LS, and CS events described in this embodiment of the invention occur on feeder 1, and each event has 20dB of noise, the zero-sequence current is as follows: Figure 5 As shown in (a). After VMD decomposition, the obtained filtered component IFC is as follows. Figure 5 As shown in (b), WFFT is then used to extract the amplitude of IFC, and finally NLS is used to calculate the coefficients b1 of the nonlinear function, as follows. Figure 5 As shown in (c). Detailed Implementation

[0055] To make the objectives and technical solutions of this invention clearer and easier to understand, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0056] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0057] A HIF (High-Intensity Fault) detection method based on arc current variation trends is proposed. Utilizing the transient characteristics of HIF current, CS current, and LS current, it is found that normal switching currents (such as CS and LS currents) exhibit damping characteristics, while arc current exhibits accumulation characteristics. This enables accurate identification of faulty circuits.

[0058] Reference Figure 1 A high-resistance fault detection method based on the trend of arc current variation includes the following steps:

[0059] Step 1: The signal detection device collects the zero-sequence current of the feeder. The variational mode decomposition (VMD) algorithm was used to analyze the zero-sequence current signal. Decomposition is performed to obtain the filtered components after removing high-frequency modes. I FC (t).

[0060] Step 2: Apply windowed Fast Fourier Transform (WFFT) to the filtered components. I FC(t) is processed to obtain the discrete frequency domain signal. From discrete frequency domain signals Extracting the instantaneous amplitude sequence of zero-sequence current A FC (t), by detecting the instantaneous amplitude of the zero-sequence current A FC ( j ) in the j Calculate the transient disturbances occurring at each time point. n The maximum current amplitude of zero-sequence current within each sampling point ;

[0061] Step 3: Based on the maximum current amplitude of the zero-sequence current within the sampling point The nonlinear least squares (NLS) method is used to quantitatively estimate the high-resistivity fault at time points. zero-sequence current at The changing trend of arc zero-sequence current Given a nonlinear function, calculate its coefficients.

[0062] Step 4: Construct a high-resistance fault detection criterion using the sign of the nonlinear function coefficients, and determine the fault type based on the criterion. The high-resistance fault detection criterion is as follows: if the sign of the nonlinear function coefficient is greater than 0, it is determined to be a high-resistance fault; if it is less than or equal to 0, it is determined to be a load switching or capacitor switching.

[0063] Step 2 includes the following steps:

[0064] Step 2.1: Process the filtered components using WFFT transform. I FC (t), to obtain the discrete frequency domain signal The calculation formula is:

[0065]

[0066] In the formula, k is a positive integer, i These are sampling points. It is the resolution in the frequency domain. ; It is the filtered component The sampling time interval, (T=0.02s); A and These are the start and end points of the sampling time window, respectively, and the length of the window is... , For the filtered components Number of sampling points in one cycle.

[0067] Step 2.2: Using the sliding discrete frequency domain signal The window is used to obtain the instantaneous amplitude sequence of zero-sequence current extracted by WFFT. A FC (t), A FC (t), including the amplitude of each sampling point .

[0068] Step 2.3: Calculate the instantaneous amplitude of the current. ( j The number of sampling points. j =1,2,…, n ) and the previous instantaneous current amplitude The difference between D FC ( j The calculation formula is as follows:

[0069]

[0070] Step 2.4, when Indicates that in the first j Amplitude at each time point A transient disturbance occurs; when the transient disturbance is detected, the matrix is ​​obtained. ,in yes The maximum value in.

[0071] Step 2.5: Decompose matrix H into matrices with dimension equal to , The matrix G, This indicates taking the largest integer, and , The number of sampling points in one period is N, and the total number of sampling points is N.

[0072] Step 2.6: Calculate the maximum current amplitude in each column of matrix G. The calculation formula is as follows:

[0073]

[0074] In the formula, It is the first in matrix G m The maximum current amplitude is listed. Adjacent amplitudes and The time interval between them is T=0.02s.

[0075] Step 3 includes the following steps:

[0076] Step 3.1: Determine the timeframe within the two cycles. Maximum current amplitude The corresponding instant is Select point As a starting point.

[0077] Step 3.2, for The five current amplitudes are counted. The corresponding time points... Represented as:

[0078] , .

[0079] Step 3.3: Based on the obtained time points and zero-sequence current amplitude ( ), to nonlinear functions Taking the logarithm gives Then use Represented as , b is a coefficient, and b0 is a constant.

[0080] Step 3.4, through settings The nonlinear equation is transformed into a linear equation, as shown in the following expression:

[0081]

[0082] Step 3.5: Calculate the minimum sum of the squared differences between the observed and predicted values ​​of the zero-sequence current amplitude. The calculation formula is:

[0083]

[0084] Step 3.6: By analyzing the min in step 3.5 S ( b 0, b 1) Find and ,get ,coefficient The calculation formula is as follows:

[0085]

[0086] In the formula: ,

[0087] The sign of the nonlinear function coefficients in step 4 is based on the coefficients calculated in step 3.6. ,according to The criteria for constructing symbols are as follows:

[0088] like The current amplitude showed an increasing trend, indicating a high-resistance fault.

[0089] like This indicates that the current amplitude does not show an increasing trend, and it is determined that load switching or capacitor switching has occurred.

[0090] Working principle of the invention

[0091] 1. Derivation of the formula for high-resistivity arc current

[0092] To study the characteristics of arc faults during the start-up and stable combustion stages, this invention introduces a HIF model that covers the arc current accumulation and shoulder stages.

[0093] First, according to the energy balance equation, the energy stored in the electric arc train... Q arc Power loss P loss Measured arc voltage u arc and current i arc The relationship between them can be described as follows:

[0094]

[0095] In the formula, the time constant is defined. , It is arc conductivity. Resistance during the stable combustion phase of the electric arc Calculated using the following formula:

[0096]

[0097] Due to arc current There is a short zero-crossing time interval, therefore It will be non-linear. Therefore, It is very large during the zero-crossing period and decays to a small value after the zero-crossing period.

[0098] To simulate the arc accumulation characteristics during the arc initiation phase, an arc accumulation characteristic resistor is introduced. , Represented as:

[0099]

[0100] Finally, arc resistance It can be represented as:

[0101]

[0102] Because an electric arc is almost resistive, the angle difference between the arc voltage and the arc current is very small. When the arc voltage reaches its amplitude, the arc current is also approximately equal to its amplitude. For a given arc voltage... Assuming arc voltage At any moment t j Reach its amplitude Therefore, during the arc initiation phase, we can obtain:

[0103]

[0104] in, for t j The resistance during the stable combustion phase of the electric arc at any given moment. for t j Arc voltage at time, for t j The zero-sequence current of the electric arc at that moment, for t j The characteristic resistance of arc accumulation at any given time.

[0105] Therefore, arc resistance Mainly by Decision, arc resistance It can be represented as:

[0106]

[0107] in At any moment t j The corresponding arc resistance at that location, and t j It is the arc voltage. Reach its amplitude At that moment.

[0108] The changing trend is similar to a decaying exponential function. In this invention, to easily estimate the changing trend of the electric arc, based on the Taylor series, It can be approximately rewritten as:

[0109]

[0110] Arc resistance The trend of change is similar to that of time. t j The decay exponential function. Based on arc resistance. The corresponding arc current The expression is:

[0111]

[0112] because , It is a constant coefficient. Arc current Trend of change at different times t j It is an increasing exponential function.

[0113] 2 Variational Mode Decomposition (VMD)

[0114] This invention employs the VMD method to decompose the original zero-sequence current signals of each feeder and selects the filtered components after removing high-frequency modes. The principle of VMD is explained below:

[0115] Variational mode decomposition (VM) is a signal decomposition and estimation method. This method determines the frequency center and bandwidth of each component by iteratively searching for the optimal solution of the variational model during the acquisition of decomposed components. This enables adaptive frequency domain partitioning of the signal and effective separation of its components. The intrinsic mode function (IMF) is defined as an amplitude-frequency modulated (AM-FM) signal, and can be expressed as:

[0116]

[0117] The VMD algorithm can be divided into two parts: constructing and solving the variational problem.

[0118] 1) Construction of variational problems

[0119] Assuming each "mode" is a finite bandwidth with a center frequency, the variational problem is described as seeking... k Modal functions The constraint is to minimize the sum of the estimated bandwidths of each mode, where the sum of the bandwidths of all modes equals the input signal. f The specific construction steps are as follows:

[0120] Step 1: Obtain each mode through Hilbert transform. The analytic signal is used to obtain its one-sided spectrum:

[0121]

[0122] Step 2: Estimate the center frequency by mixing the analytical signals of each mode. The spectrum of each mode is modulated to the corresponding fundamental frequency band:

[0123]

[0124] Step 3: Calculate the square of the gradient of the demodulated signal above. Norms are used to estimate the bandwidth of each mode signal, and a variational problem is constructed:

[0125]

[0126] in, , , .

[0127] 2) Solving variational problems

[0128] Step 1: Introduce a secondary penalty factor and Lagrange multiplication operators This transforms the constrained variational problem into an unconstrained variational problem. The quadratic penalty factor ensures the reconstruction accuracy of the signal even in the presence of Gaussian noise, and the Lagrange operator keeps the constraints strict. The extended Lagrange expression is as follows:

[0129]

[0130] Step 2: VMD employs the Alternating Direction Multiplication Method (ADMM) to solve the above variational problem, through alternating updates. , and Seeking "saddle points" for extended Lagrange expressions.

[0131] in The problem of the value of can be expressed as:

[0132]

[0133] In the formula: Equivalent to ; Equivalent to Using the Parseval / Plancherel Fourier isometric transform, the above equation is transformed into the frequency domain:

[0134]

[0135] 3. Windowed Fast Fourier Transform (WFFT)

[0136] Windowed Fast Fourier Transform (WFFT) is widely used to extract the amplitude and phase angle of periodic signals. However, due to the nonlinear and non-stationary characteristics of arc current, traditional FFT may not be able to accurately extract the instantaneous amplitude information of arc current. In this invention, a windowed Fast Fourier Transform (WFFT) is introduced to accurately estimate the instantaneous amplitude information of arc current.

[0137] For the original time-domain signal The corresponding discrete frequency domain signal Using WFFT, it can be represented as:

[0138]

[0139] In the formula, It is the resolution in the frequency domain. ; It is the original signal The sampling time interval, , It is the number of sampling points in one period. A and B These are the start and end points of the window, and the length of the window is... .

[0140] 4. High-resistivity arc model

[0141] The present invention uses a high-resistance arc model based on diode resistance, as shown in the schematic diagram. Figure 3 As shown.

[0142] In the diode-resistance-based model, there are two variable DC voltage sources. V p and V n Connect two diodes respectively D p and D n . V p and V n The voltages are unequal, changing randomly every 0.1 ms. A variable DC voltage source can simulate the intermittent, random, and asymmetrical nature of arc current. Two variable resistors... R p and R n respectively with diode D p and D n Series, R p and R n The resistance changes independently and randomly every 0.1 ms. Therefore, the randomness of the arc resistance can be modeled. (Variable resistance) R arc_b ( t When connected in series with the diode resistor model, due to the introduction of... R arc_b ( t To simulate the characteristics of arc accumulation, therefore at the start of HIF, R arc_b ( tIt is relatively large, and then decreases in the transient state.

[0143] This invention provides a high-resistance fault detection device based on the trend of arc current variation, such as... Figure 2 As shown, it includes:

[0144] The acquisition module is used to acquire the zero-sequence current of the feeder and decompose the zero-sequence current using a variational mode decomposition algorithm to obtain the filtered component after removing high-frequency modes.

[0145] The data extraction module processes the filtered components to extract the instantaneous amplitude of the zero-sequence current. This is achieved by detecting the instantaneous amplitude of the zero-sequence current at the [missing information - likely a specific point or time value]. j Calculate the transient disturbances occurring at each time point. n The maximum current amplitude of the zero-sequence current within each sampling point;

[0146] The coefficient calculation module is used to quantitatively estimate the changing trend of the zero-sequence current and the coefficients of the nonlinear function based on the instantaneous amplitude of the maximum zero-sequence current.

[0147] The fault determination module is used to construct a high-resistance fault detection criterion using the sign of the coefficients of the nonlinear function, and to determine whether a high-resistance fault has occurred based on the high-resistance fault detection criterion.

[0148] The present invention provides a computer device comprising an electrically connected memory and a processor, wherein the memory stores a computer program that can run on the processor, and when the processor executes the computer program, it implements the above-described high-resistance fault detection method based on the trend of arc current variation.

[0149] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.

[0150] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0151] The memory can be used to store the computer program and / or module. The processor implements various functions of the fault detection device / terminal equipment by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0152] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0153] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0154] Application Examples

[0155] Establish as Figure 3 The active distribution network model shown is as follows: Figure 4 The high-resistance arc model shown represents a distribution network with five feeders and three inverter-interfaced distributed generations (IIDGs). Transformer T is the main transformer with a turns ratio of 110kV / 10.5kV, and the transformers at the ends of each branch have turns ratios of 10.5kV / 0.4kV. All transformer windings use a delta / y connection. Overhead line parameters are shown in Table 1.

[0156] Table 1 Parameters of Overhead Lines

[0157]

[0158] The cable line parameters are shown in Table 2:

[0159] Table 2 Cable Line Parameters

[0160]

[0161] (Inverter interface distributed generation) IIDG parameters are shown in Table 3:

[0162] Table 3 IIDG Parameters

[0163]

[0164] The high-resistance fault parameters are shown in Table 4:

[0165] Table 4 High-resistance fault parameters

[0166]

[0167] When HIF, LS (load switching), and CS (capacitor switching) events occur in feeder 1, respectively, and noise with a signal-to-noise ratio of 20dB is added, the zero-sequence current under different conditions is as follows: Figure 5 As shown in (a). [The text abruptly ends here.] I HIF , I CS and I LS After VMD decomposition, the corresponding filtered components are obtained. I FC like Figure 5 As shown in (b). Then, WFFT is used to extract the filtered components. I FC The amplitude. Finally, for The five current amplitudes are counted, and the coefficient b1 of the nonlinear function is calculated using the NLS method. The fitted function and the calculated b1 are as follows: Figure 5 As shown in (c).

[0168] from Figure 5 As can be seen in (a), the measured high-resistivity arc current... I HIF LS current I LS and CS current I CS The current is contaminated by noise and harmonics. After VMD decomposition, the influence of harmonics and noise on the current is reduced, and the filtered components... I FC like Figure 5 As shown in (b). Based on the obtained I FC WFFT is used to calculate I FC The amplitude. For example... Figure 5As shown in (c), when CS and LS events occur, the coefficients b1 calculated using the nonlinear least squares method are -87.6 and -0.4, respectively, both less than 0. However, when HIF occurs, the coefficient b1 is 11.9, greater than 0. Based on the sign of b1, the HIF fault can be identified.

[0169] To verify the effectiveness of the proposed arc fault detection method, HIF, LS, and CS events were considered in different feeders, with varying noise levels (SNR). The zero-sequence currents under different conditions were obtained, and after VMD decomposition, the filtered components were obtained. I FC Then, use WFFT to extract. I FC The amplitude. Finally, for The five current amplitudes are counted, and the coefficients of the nonlinear function are calculated using the NLS method. b 1. HIF, LS, and CS events occur on different lines and under different signal-to-noise ratios (SNR). The coefficients of the nonlinear function are calculated from the five current amplitude values. b As shown in Table 5, this demonstrates that the method proposed in this invention can accurately detect arc faults.

[0170] Table 5. Detection Results of Radial Distribution Network

[0171]

[0172] Note: Y indicates yes, N indicates no.

[0173] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A high impedance fault detection method based on arc current change trend, characterized in that, Includes the following steps: Step 1: Collect the zero-sequence current of the feeder, and use the variational mode decomposition algorithm to decompose the zero-sequence current to obtain the filtered component after removing the high-frequency modes. Step 2, processing the filtered component, extracting the zero sequence current instantaneous amplitude, calculating the maximum current amplitude of the zero sequence current within the j sample points by detecting the transient disturbance of the zero sequence current instantaneous amplitude at the first n time point; Step 3: Based on the instantaneous amplitude of the maximum zero-sequence current, use a nonlinear function to quantitatively estimate the changing trend of the zero-sequence current and estimate the coefficients of the nonlinear function; Step 4: Construct a high-resistance fault detection criterion using the signs of the coefficients of the nonlinear function, and determine whether a high-resistance fault has occurred based on the high-resistance fault detection criterion. Step 3 includes the following steps: Step 3.1: Determine the timeframe within the two cycles. Maximum current amplitude The corresponding instant is Select point As a starting point; Step 3.2, for Count the X current amplitudes in the corresponding time; Represented as: , ； Step 3.3, based on the obtained time point and the current amplitude the expression of the non-linear function is rewritten as ; Step 3.4, by setting The non-linear equation is converted to a linear equation, expressed as follows: Step 3.5: Calculate the minimum sum of squared differences between observed and predicted values, resulting in the following formula: Step 3.6, by and giving , the calculation formula is as follows: In the formulae: , ; In step 4, the high-resistance fault detection criterion is: like The current amplitude shows an increasing trend, indicating a high-resistance fault. If , it indicates that the current amplitude does not show an increasing trend, and it is determined that load switching or capacitor switching occurs; are coefficients of a non-linear function.

2. The high impedance fault detection method based on arc current trend according to claim 1, wherein, In step 2, the filtered components are processed using a windowed fast Fourier transform.

3. The high impedance fault detection method based on arc current trend according to claim 1, wherein, Step 2 includes the following steps: Step 2.1: Process the filtered components using WFFT transform. I FC (t), to obtain the discrete frequency domain signal ; Step 2.

2. Compute the instantaneous amplitude sequence by sliding a window over the discrete frequency domain signal A FC (t);​ Step 2.3: Calculate the current amplitude Compared to the previous value The difference between D FC ( j ), j The number of sampling points. j =1,2,…, n ; Step 2.4, when the difference ,in It is an instantaneous amplitude sequence The maximum value in indicates the value in the first... j Amplitude at each time point A transient disturbance occurs. When the transient disturbance is detected, matrix H is obtained. ; Step 2.5: Decompose matrix H into matrices with dimension equal to The matrix G, , , The number of sampling points in one period is N, where N is the total number of sampling points; Step 2.

6. Calculate the maximum current amplitude of each column in the matrix G: ; In the formula, is the maximum current amplitude of the column m in the matrix G, ; the time interval between adjacent amplitudes and is , and T is the sampling period.

4. The high impedance fault detection method based on arc current trend according to claim 3, wherein, In step 2.1, the discrete frequency domain signal is calculated as In the formula, k is a positive integer, i These are sampling points. It is the resolution in the frequency domain. ; It is the filtered component The sampling time interval; A and These are the start and end points of the sampling time window, respectively, and the length of the window is... , For the filtered components Number of sampling points in one cycle.

5. The high impedance fault detection method based on arc current trend according to claim 1, wherein, In step 3.2, X equals 5.

6. A high impedance fault detection device based on arc current trend for implementing the method of any one of claims 1-5, characterized in that, include: The acquisition module is used to acquire the zero-sequence current of the feeder and decompose the zero-sequence current using a variational mode decomposition algorithm to obtain the filtered component after removing high-frequency modes. The data extraction module processes the filtered components to extract the instantaneous amplitude of the zero-sequence current. This is achieved by detecting the instantaneous amplitude of the zero-sequence current at the [missing information - likely a specific point or time value]. j Calculate the transient disturbances occurring at each time point. n The maximum current amplitude of the zero-sequence current within each sampling point; The coefficient calculation module is used to quantitatively estimate the changing trend of the zero-sequence current and the coefficients of the nonlinear function based on the instantaneous amplitude of the maximum zero-sequence current. The fault determination module is used to construct a high-resistance fault detection criterion using the sign of the coefficients of the nonlinear function, and to determine whether a high-resistance fault has occurred based on the high-resistance fault detection criterion.

7. A computer device, characterized by The method includes an electrically connected memory and a processor, wherein the memory stores a computer program that can run on the processor, and when the processor executes the computer program, it implements the steps of the method according to any one of claims 1-5.

Images