Rapid current protection method and system for power distribution network

Abstract

The invention discloses a power distribution network rapid current protection method and system, and the method comprises the steps: determining the real-time frequency of a current break variable signal based on the moment corresponding to the peak value of the current break variable signal and the moment corresponding to a signal zero crossing point after the current break variable signal is monitored; performing uniform harmonic filtering and half-cycle sampling value adjustment preprocessing on the current break variable signal, obtaining a distribution condition of the current break variable signal in a time-frequency domain by adopting discrete orthogonalization Stockwell transform, extracting a real part and an imaginary part of a fundamental wave of the current break variable signal, and calculating an effective value of the current break variable signal; when a climbing trend protection starting criterion and a sampling value out-of-limit protection action criterion of multi-point increment comparison are met at the same time, a current protection action is carried out. The action time of power distribution system protection can be effectively shortened, and the problem that the current protection cooperation stage difference time is too long under the multi-stage trend of the power distribution network is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of power distribution line protection, and relates to a fast current protection method and system for a power distribution network. Background Art

[0002] The distribution system is directly connected to the user, and is a key link in ensuring the quality of power supply and user service quality and improving the operating efficiency of the power system. At present, domestic urban distribution networks usually use radial or open-loop networks with low resistance grounding for power supply. When a fault occurs, it is easy to cause power outages to users. In addition, compared with the transmission network, the distribution network is more prone to failure. Statistics show that 88% of the total power outage time of users is caused by faults in medium and low voltage distribution networks. Therefore, the study of urban distribution network protection is of great significance to ensure the safe operation of urban distribution networks, avoid or reduce power outages caused by faults, and promote urban economic development.

[0003] As the trend of multi-level distribution network becomes increasingly obvious, shortening the differential time has become an urgent problem to be solved in urban distribution network protection, and it is also the basis for expanding the multi-level protection configuration. The differential time generally includes the inherent action time of the circuit breaker and the protection action time. The magnetically controlled circuit breaker has a fast action time, which greatly shortens the action time of the primary equipment. However, the overcurrent protection still uses the Fourier algorithm to calculate the effective value of the current, and realizes the protection judgment based on the comparison between the effective value and the fixed value. The inherent time of the protection action is difficult to break through within 40ms.

[0004] Prior art document 1 (CN114944647A - fast current protection method based on comparison of the first half-wave sampling value of a fault) adopts a full-wave Fourier transform FFT algorithm to calculate the effective value of the current mutation. The full-wave Fourier transform FFT algorithm requires a cycle of sampling points, and when faced with the actual current mutation signal with frequency fluctuations, noise and harmonics, the accuracy of the algorithm is poor. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the present invention provides a distribution network fast current protection method and system, which starts from the phasor perspective of current, takes into account the influence factors of frequency fluctuation and harmonics, and uses half-cycle current sampling values ​​to calculate the effective value of current. Only half-cycle sampling points are required, which effectively shortens the action time of primary equipment and the time difference of distribution system protection, and solves the problem of long current protection coordination difference time under the multi-level trend of distribution network.

[0006] The present invention adopts the following technical solution.

[0007] A first aspect of the present invention provides a distribution network fast current protection method, comprising:

[0008] Monitor the mutation signal of the current signal in real time. When the mutation signal of the current is detected, determine the real-time frequency of the mutation signal of the current based on the time corresponding to the peak value of the mutation signal of the current and the time corresponding to the zero crossing point of the signal.

[0009] Based on the real-time frequency of the current mutation signal, the current mutation signal is preprocessed by uniform harmonic filtering and half-cycle sampling value adjustment, and the discrete orthogonalized Stockwell transform is used to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real and imaginary parts of the fundamental wave of the current mutation signal. The effective value of the current mutation signal is calculated according to the real and imaginary parts of the fundamental wave.

[0010] When the effective value of the current mutation signal satisfies both the ramp trend protection start-up criterion of the multi-point incremental comparison and the sampling value over-limit protection action criterion, the current protection action is performed.

[0011] Preferably, the real-time monitoring of the mutation amount signal of the current signal, after the current mutation amount signal is monitored, determines the real-time frequency of the current mutation amount signal based on the time corresponding to the peak value of the current mutation amount signal and the time corresponding to the zero crossing point of the signal, comprising the following steps:

[0012] Real-time monitoring of the sudden change signal i of the current signal m (n), when the current mutation signal i is detected m (n), the current mutation signal i m (n) performing cosine sliding filtering;

[0013] The filtered current mutation signal is subjected to piecewise spline interpolation and peak estimation to obtain the time corresponding to the signal peak, and linear interpolation and zero-crossing analysis are performed to obtain the time corresponding to the signal zero-crossing point;

[0014] Calculate the real-time frequency F of the current mutation signal according to the time corresponding to the signal peak and the time corresponding to the signal zero point. c ;

[0015] Determine the calculated real-time frequency F c Does the absolute difference between the frequency and the standard frequency F0 meet the error requirements? If so, set F c Real-time frequency as a current mutation signal Otherwise, the standard frequency F0 is used as the real-time frequency of the current mutation signal.

[0016] Preferably, the piecewise spline interpolation and peak estimation are performed to obtain the time corresponding to the signal peak, specifically as follows:

[0017] The filtered current mutation signal i m(n) Perform half-cycle sampling data analysis and obtain the approximate peak value i′ of the current mutation signal based on the amplitude of the current mutation signal. p (n);

[0018] The piecewise spline polynomial is used to approximate the peak value i′ p (n) The amplitude of the previous sampling point i′ p (n-1), the amplitude of the next sampling point i′ p The current mutation signal i at (n+1) m (n) Interpolation, based on the piecewise spline interpolation method, calculate the actual peak value i of the current mutation signal p ,i p The corresponding time is the moment t corresponding to the actual peak value of the current mutation signal p .

[0019] Preferably, the linear interpolation and zero-crossing analysis are performed to obtain the time corresponding to the signal zero-crossing point, and the specific formula is:

[0020]

[0021] Where, t zc is the moment corresponding to the zero crossing point;

[0022] i′ zc (n)<0 and i′ zc (n+1)>0, or i′ zc (n)>0 and i′ zc (n+1)<0, or i′ zc (n) = 0;

[0023] t′ zc (n) and t′ zc (n+1) are i′ zc (n) and i′ zc (n+1) corresponding moment;

[0024] i′ zc (n) and i′ zc (n+1) are the amplitudes of the filtered current mutation signal at the nth sampling point and the n+1th sampling point respectively.

[0025] Preferably, the real-time frequency F of the current mutation signal c The calculation formula is:

[0026]

[0027] Among them, t zc is the time corresponding to the signal zero crossing point;

[0028] t pis the time corresponding to the signal peak.

[0029] Preferably, based on the real-time frequency of the current mutation signal, the current mutation signal is preprocessed by uniform harmonic filtering and half-cycle sampling value adjustment, and the current mutation signal after preprocessing is subjected to discrete orthogonalized Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real part and imaginary part of the fundamental wave of the current mutation signal, and the effective value of the current mutation signal is calculated according to the real part and imaginary part of the fundamental wave, including the following steps:

[0030] Real-time frequency based on current mutation signal For the current mutation signal i m (n) performing uniform harmonic filtering and half-cycle sampling value adjustment preprocessing;

[0031] The discrete current mutation signal i after preprocessing m (n) Perform fast discrete orthogonal Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain, select the frequency component information corresponding to the fundamental frequency, and decompose the frequency component information output into orthogonal components to obtain the real and imaginary parts of the fundamental wave of the current mutation signal, and then calculate the effective value of the current mutation signal.

[0032] Preferably, the real-time frequency based on the current mutation signal For the current mutation signal i m (n) The uniform harmonic filtering function for uniform harmonic filtering is as follows:

[0033]

[0034] Where H(z) and Z are the transfer function and Z transformation factor respectively; F s is the sampling frequency.

[0035] Preferably, the half-period sampling value adjustment preprocessing includes:

[0036] The filtered current mutation signal i m (n) Adjust the half-cycle data sampling value as follows:

[0037] 1) Determine the displacement factor α:

[0038]

[0039] Where F0 is the standard frequency;

[0040] 2) Assuming that parameter γ = |α|, during the sampling value adjustment process, when γ is less than or equal to 0.5, the displacement factor α is used to determine the adjusted sampling value, and the formula is as follows:

[0041] im (n) s =c0i m (n)+c1i m (n+1)+c -1 i m (n-1) (5)

[0042] In the formula, coefficient c0 = 1-α 2 , c1=0.5(1+α 2 ), c -1 =-1+α 2 ;

[0043] i m (n) s is the current sampling value adjusted at the nth sampling point; i m (n), i m (n+1), i m (n-1) are the current sampling values ​​at the nth, n+1th, and n-1th sampling points respectively;

[0044] When γ is greater than 0.5, the adjusted sampling value is determined using the new displacement factor β and the sample reference point m, as follows:

[0045] i m (n) S The calculation formula is

[0046] i m (n) S =c′0i m (m)+c′1i m (m+1)+c′ -1 i m (m-1) (7)

[0047] In the formula, coefficient c′0=1-β 2 , c′1=0.5(1+β 2 ), c′ -1 =-1+β 2 ;

[0048] i m (m), i m (m+1), i m (m-1) are the current sampling values ​​at the mth, m+1th, and m-1th sampling points respectively;

[0049] m and β are based on γ, F0, Sure.

[0050] Preferably, according to γ, F0, The way to determine m and β is:

[0051] When γ≤0.5, β=n△f n ; m = n;

[0052] γ∈(0.5,1.5], When β=-1+α; m=n-1;

[0053] γ∈(0.5,1.5], When β=1-α; m=n+1;

[0054] γ∈(1.5,2.5], When β=-2+α; m=n-2;

[0055] γ∈(1.5,2.5], When β = 2-α, m = n + 2;

[0056] And so on, until γ∈(r-1,r], When β=-r+α;m=nr;

[0057] γ∈(r-0.5,r+0.5], When β=r-α;m=n+r;

[0058] in r is the rounded integer of the γ value.

[0059] Preferably, the pre-processed discrete current mutation signal i m (n) Perform a fast discrete orthogonal Stockwell transform as follows:

[0060]

[0061] In the formula, S[i m (n)] is i m Discrete orthogonalized Stockwell transform of (n);

[0062] is the basis function of the discrete orthogonalized Stockwell transform;

[0063] n=0, 1, 2, ..., N-1; N is the number of samples per cycle;

[0064] i is the imaginary unit; v c is a frequency variable; β w is the bandwidth; t l For time.

[0065] Preferably, the calculation formula of the effective value of the current mutation signal is:

[0066]

[0067] in, are respectively the real part and imaginary part of the fundamental wave of the current mutation signal at the nth sampling point;

[0068] Estimated amplitude of the current mutation signal at the nth sampling point;

[0069] I m (n) is the effective value of the current mutation signal at the nth sampling point.

[0070] Preferably, the starting criterion of the climbing trend protection of the multi-point incremental comparison is:

[0071]

[0072] Where st(n) represents the transformation trend value from the nth sampling point to the nn+1th sampling point, which is 1 if the transformation trend increases and 0 if the transformation trend does not increase.

[0073] I m (n+1) is the effective value of the current mutation at the n+1th sampling point;

[0074] M1 is the number of sampling points in the data window for slope determination; M2 is the anti-interference threshold for trend determination.

[0075] Preferably, the sampling value over-limit protection action criterion is:

[0076]

[0077] Among them, I set is the start threshold;

[0078] I m (n), I m (n+1), I m (n+2) is the effective value of the current mutation signal at the nth, n+1th, and n+2nd sampling points.

[0079] A second aspect of the present invention provides a distribution network fast current protection system, comprising:

[0080] The frequency calculation module is used to monitor the mutation signal of the current signal in real time. When the current mutation signal is monitored, the real-time frequency of the current mutation signal is determined based on the time corresponding to the peak value of the current mutation signal and the time corresponding to the zero crossing point of the signal;

[0081] The effective value calculation module is used to perform uniform harmonic filtering and half-cycle sampling value adjustment preprocessing on the current mutation signal based on the real-time frequency of the current mutation signal, and to use discrete orthogonalized Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real part and imaginary part of the fundamental wave of the current mutation signal, and calculate the effective value of the current mutation signal according to the real part and imaginary part of the fundamental wave.

[0082] The protection action module is used to perform current protection action when the effective value of the current mutation signal simultaneously meets the climbing trend protection start criterion of multi-point incremental comparison and the sampling value over-limit protection action criterion.

[0083] A third aspect of the present invention provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0084] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0085] Compared with the prior art, the beneficial effects of the present invention include at least:

[0086] The present invention takes into account the problems of frequency fluctuation, noise and harmonics in the actual current signal, calculates the real-time frequency of the current based on the peak value of the current mutation signal and the time corresponding to the zero-crossing point, uses uniform harmonic filtering technology to process the current mutation signal within a half-cycle, eliminates the influence of harmonics in the current mutation signal, and uses half-cycle sampling value adjustment to solve the problem of signal discontinuity when the frequency fluctuates. It has strong robustness, can cope with non-standard frequency transient problems, is not affected by noise, and has a fast response speed;

[0087] The scheme for extracting the fundamental phasor (real and imaginary parts) of the signal based on discrete orthogonalized Stockwell transform proposed in the present invention only requires half-cycle data points, thus reducing the calculation time required for phasor calculation in current protection of distribution network;

[0088] The present invention can effectively improve the speed and reliability of distribution system protection, shorten the time difference of distribution system protection, solve the problem of long current protection coordination difference time under the multi-level trend of distribution network, and improve the power supply reliability of distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 This is a flowchart of the method for fast current protection of a power distribution network according to the present invention;

[0090] Figure 2 This is a schematic diagram of the fast current protection method for the distribution network of the present invention;

[0091] Figure 3 It is a sine wave signal diagram;

[0092] Figure 4 A flow chart of hybrid frequency estimation based on peak and zero-crossing values;

[0093] Figure 5 Flowchart for selecting displacement factors and sample reference points. DETAILED DESCRIPTION

[0094] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only embodiments of a part of the present invention, rather than all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present invention.

[0095] like Figure 1-2 As shown, Embodiment 1 of the present invention provides a method for fast current protection of a distribution network, which calculates the fundamental phasor of the current, obtains the effective value of the current, and then performs fast current protection, which specifically includes:

[0096] S1: Real-time monitoring of the sudden change signal i of the current signal m (n), when the current mutation signal i is detected m (n), the real-time frequency of the current mutation signal is determined based on the time corresponding to the peak value of the current mutation signal and the time corresponding to the zero crossing point of the signal.

[0097] Further preferably, the present invention introduces the concepts of peak value and zero-crossing point, proposes a new frequency estimation algorithm based on peak value and zero-crossing point, realizes real-time frequency estimation of the signal, obtains the real-time frequency and feeds it to the pre-filtering stage of S2, and the overall process of the frequency estimation algorithm is as follows: Figure 4 As shown in the figure, the algorithm can reduce the number of samples required for frequency estimation, introduce the concepts of peak value and zero crossing point, use piecewise spline polynomial to interpolate the real-valued sine function, calculate the value at the local maximum, and use linear interpolation to calculate the moment when the zero crossing condition is met, and then calculate the final frequency, which includes:

[0098] S1.1: Real-time monitoring of the sudden change signal i of the current signal m (n), when the current mutation signal i is detected m (n), the current mutation signal i m (n) performing cosine sliding filtering;

[0099] Real-time monitoring of current mutations. As the fault develops, the effective value will show a climbing trend. The formula for the current fault component is:

[0100] i m (n) = i(n) - i(n-2N)

[0101] Where N represents the number of sampling points in a cycle; i m (n) is the current fault component corresponding to time n, that is, the current mutation corresponding to the nth sampling point.

[0102] In order to avoid the interference of noise and harmonics on the frequency estimation of the current mutation signal, the present invention first inputs the current mutation signal into a cosine sliding Goertzel filter before applying the frequency estimation algorithm based on peak and zero-crossing values, and uses a cosine sliding Goertzel filter (CSGT) to remove noise and harmonics in the signal. The filtered signal is used as an input for frequency estimation.

[0103] S1.2: Perform segmented spline interpolation and peak estimation on the filtered current mutation signal to obtain the time corresponding to the signal peak, and perform linear interpolation and zero-crossing analysis to obtain the time corresponding to the signal zero-crossing point;

[0104] (1) Use the piecewise spline interpolation method to estimate the basic sine wave between sampling points, use the piecewise spline polynomial to interpolate the real-valued sine function, and calculate the value at the local maximum:

[0105] Figure 3 An example of a sine wave discrete current mutation signal i m (n), the approximate peak value i′ of the sinusoidal current mutation signal can be obtained by analyzing the sampling data of the half cycle p (n) (The approximate peak value is the extreme value in the signal, which can be calculated based on the amplitude of the signal.) Since the position of the extreme value depends on the sampling frequency used to construct the discrete signal, the approximate peak value of the sinusoidal current mutation signal may not be the exact peak value. Therefore, the piecewise spline interpolation method is used to estimate the sinusoidal current mutation signal between the sampling points.

[0106] The piecewise spline polynomial is used to approximate the peak value i′ p (n) The amplitude of the previous sampling point i′ p (n-1), the amplitude of the next sampling point i′ p The discrete current mutation signal i at (n+1) m (n) Interpolation, based on the piecewise spline interpolation method, calculate the approximate peak value i of the sinusoidal current mutation signal p ,i p The corresponding time is the moment t corresponding to the actual peak value of the current mutation signal p .

[0107] (2) At the same time, linear interpolation is used to calculate the time when the zero-crossing condition is met;

[0108] Find the zero-crossing moment that satisfies the conditions, that is, i′ zc (n)<0 and i′ zc (n+1)>0 or i′ zc (n)>0 and i′ zc (n+1)<0 or i′ zc (n) = 0. Time t zc The linear interpolation method is used for calculation, and the calculation formula is as follows:

[0109]

[0110] Where t′ zc (n) and t′ zc (n+1) are i′ zc (n) and i′ zc (n+1) corresponding moment;

[0111] i′ zc (n) and i′ zc (n+1) are the amplitudes of the current mutation signal after S1.1 filtering at the nth sampling point and the n+1th sampling point.

[0112] S1.3: Calculate the real-time frequency F of the signal based on the time corresponding to the signal peak and the time corresponding to the signal zero crossing point c ;

[0113] After finding the time corresponding to the maximum value and zero-crossing value of the current mutation signal, the frequency of the current mutation signal is calculated. The frequency calculation method is as follows:

[0114]

[0115] S1.4: Determine the real-time frequency F calculated by S1.3 c Does the absolute difference between the frequency and the standard frequency F0 meet the error requirements? If so, set F c Real-time frequency as a current mutation signal Otherwise, the standard frequency F0 is used as the real-time frequency of the current mutation signal.

[0116] When the exact peak value is equal to the approximate peak estimate i′ p (n), the spline interpolation method may deviate slightly from the actual peak. Therefore, an error check should be performed before initializing the spline interpolation. If the estimated frequency F obtained by S1.3 c The absolute difference between the standard frequency F0 (the standard frequency is the rated frequency specified in the power system, which is 50Hz) is less than 10 -4 Hz, then the estimated frequency F c Real-time frequency as a current mutation signal Otherwise, the real-time frequency of the current mutation signal Use the rated frequency specified by the power system.

[0117] S2: Real-time frequency based on current mutation signal The uniform harmonic filtering technology and half-cycle sampling value adjustment are used to solve the problem of signal discontinuity during harmonic interference and frequency fluctuation. The discrete orthogonalized Stockwell transform is used to obtain the distribution of the current mutation signal in the time-frequency domain after the processing, and the real and imaginary parts of the fundamental wave of the current mutation signal are extracted to obtain the current mutation signal i m (n) effective value I m (n);

[0118] Further preferably, the current mutation signal needs to go through a pre-filtering stage, a half-cycle sampling value adjustment and a half-cycle DOST to calculate the fundamental phasor of the current mutation signal.

[0119] Specifically, based on the real-time frequency in S1 The uniform harmonic filtering technology is used to process the current mutation signal within a half cycle, the influence of harmonics in the current mutation signal is eliminated, and the half-cycle sampling value is used to adjust the signal discontinuity when the frequency fluctuates; after the input current mutation signal is preprocessed, the discrete orthogonal Stockwell transform is used to extract the fundamental phasor of the current mutation signal, and then the effective value of the fundamental wave of the current mutation signal is obtained, including the following steps:

[0120] S2.1: Real-time frequency based on current mutation signal The signal is preprocessed using uniform harmonic filtering (EHF) and sample value adjustment (SVA).

[0121] (1) The uniform harmonic filtering technology is used for the current mutation signal, and the adaptive filter is used as a function of the input frequency for preprocessing:

[0122] Since relay protection requires a faster response time for the phasor measurement algorithm, the phasor extraction of the discrete orthogonalized Stockwell transform adopts a half-cycle discrete Fourier transform (HC-DFT), while DOST usually adopts a full-cycle discrete Fourier transform (FC-DFT). This in turn limits the size of the data window to half a cycle. In addition, the even-order harmonic rejection capability of HC-DFT is not as good as that of FC-DFT, which illustrates the contradiction between accuracy and speed. Therefore, in order to regain accuracy, the present invention adopts EHF and utilizes an adaptive filter as a function of the input frequency, according to the following transfer function, based on the real-time frequency of the current mutation signal For the original current mutation signal i m (n) Perform uniform harmonic filtering preprocessing:

[0123]

[0124] Where H(z) and Z are the transfer function and Z transformation factor respectively, and F s is the sampling frequency, It is the real-time frequency of the current mutation signal estimated by S1.

[0125] (2) To address the problem of signal discontinuity during spectrum leakage and frequency fluctuation, the Taylor series expansion and central difference method are used to calculate the sampling value of the signal, and the sampling value of the half-cycle current data is adjusted:

[0126] As mentioned above, FDOST is based on DFT, which has spectrum leakage problems and signal discontinuity problems when the frequency fluctuates. To solve this problem, the following will briefly introduce the sampling value adjustment method (SVA) used for half-cycle current data, and it is called half-cycle sampling value adjustment (HC-SVA). SVA is an interpolation technology that converts current mutation signals of any frequency into standard frequency. The sampling value adjustment method must know the frequency information of the current mutation signal. The real-time estimated frequency of the current mutation signal has been obtained in S1.

[0127] Consider a discrete current mutation signal containing harmonics, whose model is

[0128]

[0129] In the formula, n Indicates the sampling point number of the current mutation signal; n = 0, 1, 2, ..., N-1, N is the number of sampling times per cycle. h and φ h are the amplitude and phase angle of the hth harmonic component respectively. H is the harmonic order, and h=1 corresponds to the fundamental component.

[0130] Calculate i using Taylor series expansion and central difference method m (n) The adjusted sampling value i m (n) S .i m (n) S The output value is

[0131] i m (n) s =c0i m (n)+c1i m (n+1)+c -1 i m (n-1) (5)

[0132] In the formula, coefficient c0 = 1-α 2, c1=0.5(1+α 2 ), c -1 =-1+α 2 α is the displacement factor, because i m (n) and i m (n) S The sample values ​​at the nth sampling are not equal. Instead, i m (n) S =i m (n±α) depends on the negative frequency deviation or the positive frequency deviation respectively. The method for determining α is as follows:

[0133]

[0134] In the formula, F0 is the standard frequency. As shown in the above formula, α As the number of samples n increases, it increases and reduces the interpolation effect of old samples. In order to avoid this problem, the selection of the displacement factor is improved. The selection process is shown in Figure 5 .

[0135] In order to determine the improved displacement factor β, assuming γ = |α|, during the sample value adjustment process, if γ is higher than 0.5, the adjusted sample value can be determined more accurately using the new displacement factor β and the sample reference point m. m (n) S The calculation formula is

[0136] i m (n) S =c′0i m (m)+c′1i m (m+1)+c′ -1 i m (m-1) (7)

[0137] In the formula, c′0, c′1 and c′ -1 is to use something like c0, c1, and c -1 The coefficients of β are calculated from m and β. m and β are determined according to a regular procedure for different values ​​of γ.

[0138] S2.2: Preprocessed discrete current mutation signal i m (n) Perform fast discrete orthogonal Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain, select the frequency component information corresponding to the fundamental frequency, and decompose the frequency component information output into orthogonal components to obtain the real and imaginary parts of the fundamental wave of the current mutation signal, and then calculate the effective value of the current mutation signal.

[0139] The basis of discrete orthogonal Stockwell transform is Stockwell transform (S transform), which is the representation of time series signal in time-frequency domain. Stockwell transform uses a frequency-dependent Gaussian window to analyze the signal and is considered to be an improvement of short-time Fourier transform and continuous wavelet transform. It has the advantages of multi-scale analysis and low complexity, and retains the phase information of the signal.

[0140] For a signal x(t), its S transform can be obtained by multiplying the wavelet transform by a phase factor:

[0141] S x (τ,f)=e i2πft W x (τ,f) (8)

[0142]

[0143] Where W x (τ,f) represents the wavelet transform of the signal x(t); f is the frequency; w(t-τ,d) is the Gaussian window, and the parameters of the Gaussian window on the time t axis are τ .

[0144] The selection of wavelet mother function w(t,f) is as follows:

[0145]

[0146] Substituting equations (9) and (10) into equation (8), the complete expression of S transformation can be obtained as follows:

[0147]

[0148] The main feature of ST is that it has absolute reference phase information, which is not available in wavelet transform. m The DFT of (n) is

[0149]

[0150] Based on the above formula, let τ→j△T, then the S transformation of the discrete current mutation signal can be obtained as

[0151]

[0152] Where p, k and j range from 0 to N-1; △T is the sampling time interval of the system.

[0153] Discrete orthogonalized Stockwell transform is a non-redundant and efficient form of S transform, which can solve the storage and calculation problems of S transform.

[0154] Each basis function of DOST is represented by a frequency variable v at the center of the band c , bandwidth β w and time t l DOST is expressed as follows:

[0155]

[0156] In the formula, S[i m (n)] is i m Discrete orthogonalized Stockwell transform of (n);

[0157] is the basis function of the discrete orthogonalized Stockwell transform;

[0158] i is the imaginary unit;

[0159] The following rules are used to determine the parameters of the time-frequency space (tf space) to ensure orthogonality.

[0160] 1)t l =0,1,...,β w -1.

[0161] 2)v c and β w The selection must be done in a way that each Fourier frequency is used simultaneously.

[0162] In order to make the algorithm faster, a half-cycle discrete Fourier transform (HC-DFT) is used, which only requires N / 2 sampling points per cycle. The HC-DFT of equation (12) is expressed as:

[0163]

[0164] In this case, the signal octave sampling with a buffer length of N / 2=32 sampling points becomes {1, 1, 2, 4, 8, 1, 8, 4, 2, 1}.

[0165] After the original signal is subjected to discrete orthogonal Stockwell transform, the distribution of the signal in the time-frequency domain can be obtained, the frequency component information corresponding to the fundamental frequency is selected, and the frequency component information output is decomposed into orthogonal components, that is, the real part of the fundamental wave of the current mutation signal and the imaginary part The effective value of the current can be obtained based on the real and imaginary parts of the fundamental wave. The calculation formula is:

[0166]

[0167] in, are respectively the real part and imaginary part of the fundamental wave of the current mutation signal at the nth sampling point (i.e., time n);

[0168] Estimated amplitude of the current mutation signal at the nth sampling point;

[0169] I m (n) is the effective value of the current mutation signal at the nth sampling point.

[0170] HC-DOST performs Fourier transformation on the discrete current mutation signal and then multiplies it by a window function, the shape and size of which can be adjusted by parameters. By changing the parameters of the window function, the time-frequency analysis results of the current mutation signal at different scales can be obtained. Compared with other time-frequency analysis methods, HC-DOST has better resolution and stability when analyzing the current mutation signal.

[0171] S3: The effective value of the current mutation is combined with the climbing trend protection start-up criterion of the multi-point increment comparison and the sampling value over-limit protection action criterion to perform current protection action.

[0172] Further preferably, S1 is used to monitor the sudden change of the current signal in real time to decide whether to continue to perform the fault judgment, and the starting judgment criterion is composed of effective value calculation, climbing trend judgment, and sampling value over-limit judgment. Specifically, S1 and S2 are used to monitor the sudden change of the current in real time and calculate the effective value of the sudden change of the current. S3 is based on the climbing effect of the effective value of the first wave of the fault current and the phenomenon that the fault current sampling value quickly exceeds the threshold, and constructs the climbing trend protection starting judgment and protection action judgment of multi-point incremental comparison, including the following steps:

[0173] Determine the effective value I of the current mutation signal m (n) Whether the climbing trend protection start-up criterion and protection action criterion of the multi-point incremental comparison are met at the same time. If both are met, the current protection will be activated; otherwise, the current protection will not be activated.

[0174] In this step, the start-up criterion is composed of effective value calculation, climbing trend judgment, and sampling value over-limit judgment.

[0175] First, multi-point incremental comparison is used to identify the effective value climbing trend. The protection start criteria are as follows:

[0176]

[0177] Wherein, M1 is the number of sampling points of the slope determination data window, and the length of the data window is T / 8; M2 is the anti-interference threshold of trend determination, which is (0.5~0.7)M1.

[0178] Secondly, the effective value climbing trend will also appear when the system is disturbed. It is difficult to meet the reliability requirements of protection by using this as the only criterion for protection. After a short circuit fault, the current surge is large, and the current sampling value will quickly rise to a large value and last for a certain period of time. Therefore, the sampling value over-limit judgment formula is set as follows:

[0179]

[0180] The protection can only be started when three consecutive current sampling values ​​are greater than the starting threshold, thereby limiting the frequent startup of protection under non-fault or interference conditions.

[0181] Embodiment 2 of the present invention provides a distribution network fast current protection system, including:

[0182] The frequency calculation module is used to monitor the mutation signal of the current signal in real time. When the current mutation signal is monitored, the real-time frequency of the current mutation signal is determined based on the time corresponding to the peak value of the current mutation signal and the time corresponding to the zero crossing point of the signal;

[0183] The effective value calculation module is used to perform uniform harmonic filtering and half-cycle sampling value adjustment preprocessing on the current mutation signal based on the real-time frequency of the current mutation signal, and to use discrete orthogonalized Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real part and imaginary part of the fundamental wave of the current mutation signal, and calculate the effective value of the current mutation signal according to the real part and imaginary part of the fundamental wave.

[0184] The protection action module is used to perform current protection action when the effective value of the current mutation signal simultaneously meets the climbing trend protection start criterion of multi-point incremental comparison and the sampling value over-limit protection action criterion.

[0185] Embodiment 3 of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0186] Embodiment 4 of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0187] Compared with the prior art, the beneficial effects of the present invention include at least:

[0188] The present invention takes into account the problems of frequency fluctuation, noise and harmonics in the actual current signal, calculates the real-time frequency of the current based on the peak value of the current mutation signal and the time corresponding to the zero-crossing point, uses uniform harmonic filtering technology to process the current mutation signal within a half-cycle, eliminates the influence of harmonics in the current mutation signal, and uses half-cycle sampling value adjustment to solve the problem of signal discontinuity when the frequency fluctuates. It has strong robustness, can cope with non-standard frequency transient problems, is not affected by noise, and has a fast response speed;

[0189] The present invention proposes an algorithm for extracting the fundamental phasor of a signal based on a half-cycle FDOST of an S-transform, which only requires a half-cycle data point, thus reducing the calculation time required for phasor calculation in current protection of distribution networks;

[0190] The present invention can effectively improve the speed and reliability of distribution system protection, shorten the time difference of distribution system protection, solve the problem of long current protection coordination difference time under the multi-level trend of distribution network, and improve the power supply reliability of distribution network.

[0191] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0192] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0193] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0194] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages-such as Smalltalk, C++, etc., and conventional procedural programming languages-such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network-including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA) may be personalized by utilizing the state information of a computer-readable program instruction, and the electronic circuit may execute a computer-readable program instruction, thereby realizing various aspects of the present disclosure.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A fast current protection method for a distribution network, characterized in that: include: Monitor the mutation signal of the current signal in real time. When the mutation signal of the current is detected, determine the real-time frequency of the mutation signal of the current based on the time corresponding to the peak value of the mutation signal of the current and the time corresponding to the zero crossing point of the signal. Based on the real-time frequency of the current mutation signal, the current mutation signal is preprocessed by uniform harmonic filtering and half-cycle sampling value adjustment, and the discrete orthogonalized Stockwell transform is used to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real and imaginary parts of the fundamental wave of the current mutation signal. The effective value of the current mutation signal is calculated according to the real and imaginary parts of the fundamental wave. When the effective value of the current mutation signal satisfies both the ramp trend protection start-up criterion of the multi-point incremental comparison and the sampling value over-limit protection action criterion, the current protection action is performed.

2. A method for fast current protection of a distribution network according to claim 1, characterized in that: The method of real-time monitoring the mutation signal of the current signal, when the current mutation signal is monitored, determining the real-time frequency of the current mutation signal based on the time corresponding to the peak value of the current mutation signal and the time corresponding to the zero crossing point of the signal, comprises the following steps: Real-time monitoring of the sudden change signal i of the current signal m (n), when the current mutation signal i is detected m (n), the current mutation signal i m (n) performing cosine sliding filtering; The filtered current mutation signal is subjected to piecewise spline interpolation and peak estimation to obtain the time corresponding to the signal peak, and linear interpolation and zero-crossing analysis are performed to obtain the time corresponding to the signal zero-crossing point; Calculate the real-time frequency F of the current mutation signal according to the time corresponding to the signal peak and the time corresponding to the signal zero point. c ; Determine the calculated real-time frequency F c Does the absolute difference between the frequency and the standard frequency F0 meet the error requirements? If so, set F c Real-time frequency as a current mutation signal Otherwise, the standard frequency F0 is used as the real-time frequency of the current mutation signal.

3. A fast current protection method for distribution network according to claim 2, characterized in that: The piecewise spline interpolation and peak estimation are performed to obtain the time corresponding to the signal peak, which is specifically as follows: The filtered current mutation signal i m (n) Perform half-cycle sampling data analysis and obtain the approximate peak value i′ of the current mutation signal based on the amplitude of the current mutation signal. p (n); The piecewise spline polynomial is used to approximate the peak value i′ p (n) The amplitude of the previous sampling point i′ p (n-1), the amplitude of the next sampling point i′ p The current mutation signal i at (n+1) m (n) Interpolation, based on the piecewise spline interpolation method, calculate the actual peak value i of the current mutation signal p ,i p The corresponding time is the moment t corresponding to the actual peak value of the current mutation signal p .

4. A distribution network fast current protection method according to claim 2, characterized in that: The linear interpolation and zero-crossing analysis are performed to obtain the time corresponding to the signal zero-crossing point. The specific formula is: Where, t zc is the moment corresponding to the zero crossing point; i' zc (n)<0 and i' zc (n+1)>0, or i′ zc (n)>0 and i' zc (n+1)<0, or i′ zc (n)=0; t′ zc (n) and t′ zc (n+1) are i′ zc (n) and i′ zc (n+1) corresponding moment; i′ zc (n) and i′ zc (n+1) are the amplitudes of the filtered current mutation signal at the nth sampling point and the n+1th sampling point respectively.

5. A fast current protection method for a distribution network according to claim 2, characterized in that: The real-time frequency F of the current mutation signal c The calculation formula is: Among them, t zc is the time corresponding to the signal zero crossing point; t p is the time corresponding to the signal peak.

6. A fast current protection method for a distribution network according to claim 1, characterized in that: The method is based on the real-time frequency of the current mutation signal, performs uniform harmonic filtering and half-cycle sampling value adjustment preprocessing on the current mutation signal, and uses discrete orthogonalized Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real part and imaginary part of the fundamental wave of the current mutation signal, and calculates the effective value of the current mutation signal according to the real part and imaginary part of the fundamental wave, including the following steps: Real-time frequency based on current mutation signal For the current mutation signal i m (n) performing uniform harmonic filtering and half-cycle sampling value adjustment preprocessing; The discrete current mutation signal i after preprocessing m (n) Perform discrete orthogonal Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain, select the frequency component information corresponding to the fundamental frequency, and decompose the frequency component information output into orthogonal components to obtain the real part and imaginary part of the fundamental wave of the current mutation signal, and then calculate the effective value of the current mutation signal.

7. A method for rapid current protection of a distribution network according to claim 6, characterized in that: The real-time frequency based on the current mutation signal For the current mutation signal i m (n) The uniform harmonic filtering function for uniform harmonic filtering is as follows: Where H(z) and Z are the transfer function and Z transformation factor respectively; F s is the sampling frequency.

8. A method for fast current protection of a distribution network according to claim 6, characterized in that: The half-period sampling value adjustment preprocessing includes: The filtered current mutation signal i m (n) Adjust the half-cycle data sampling value as follows: 1) Determine the displacement factor α: Where F0 is the standard frequency; 2) Assuming that parameter γ = |α|, during the sampling value adjustment process, when γ is less than or equal to 0.5, the displacement factor α is used to determine the adjusted sampling value, and the formula is as follows: i m (n) s =c0i m (n)+c1i m (n+1)+c -1 i m (n-1) (5) In the formula, coefficient c0 = 1-α 2 , c1=0.5(1+α 2 ), c -1 =-1+α 2 ; i m (n) s is the current sampling value adjusted at the nth sampling point; i m (n), i m (n+1), i m (n-1) are the current sampling values ​​at the nth, n+1th, and n-1th sampling points respectively; When γ is greater than 0.5, the adjusted sampling value is determined using the new displacement factor β and the sample reference point m, as follows: i m (n) S The calculation formula is i m (n) S =c′0i m (m)+c′1i m (m+1)+c′ -1 i m (m-1) (7) In the formula, coefficient c′0=1-β 2 , c′1=0.5(1+β 2 ), c′ -1 =-1+β 2 ; i m (m), i m (m+1), i m (m-1) are the current sampling values ​​at the mth, m+1th, and m-1th sampling points respectively; m and β are based on γ, F0, Sure.

9. A method for fast current protection of a distribution network according to claim 8, characterized in that: According to γ, F0, The way to determine m and β is: When γ≤0.5, β=n△f n ; m = n; γ∈(0.5,1.5], When β=-1+α;m=n-1; γ∈(0.5,1.5], When β=1-α; m=n+1; γ∈(1.5,2.5], When β=-2+α;m=n-2; γ∈(1.5,2.5], When β=2-α; m=n+2; And so on, until γ∈(r-1,r], When β=-r+α;m=nr; γ∈(r-0.5,r+0.5], When β=r-α;m=n+r; in r is the rounded integer of the γ value.

10. A fast current protection method for a distribution network according to claim 6, characterized in that: The discrete current mutation signal i after preprocessing m (n) Perform discrete orthogonal Stockwell transform, as follows: In the formula, S[i m (n)] is i m Discrete orthogonalized Stockwell transform of (n); is the basis function of the discrete orthogonalized Stockwell transform; n=0, 1, 2, ..., N-1; N is the number of samples per cycle; i is the imaginary unit; v c is a frequency variable; β w is the bandwidth; t l For time.

11. A fast current protection method for a distribution network according to claim 1, characterized in that: The calculation formula of the effective value of the current mutation signal is: in, are respectively the real part and imaginary part of the fundamental wave of the current mutation signal at the nth sampling point; Estimated amplitude of the current mutation signal at the nth sampling point; I m (n) is the effective value of the current mutation signal at the nth sampling point.

12. A method for fast current protection of a distribution network according to claim 11, characterized in that: The starting criterion of the climbing trend protection of the multi-point incremental comparison is: Where st(n) represents the transformation trend value from the nth sampling point to the nn+1th sampling point, which is 1 if the transformation trend increases and 0 if the transformation trend does not increase. I m (n+1) is the effective value of the current mutation at the n+1th sampling point; M1 is the number of sampling points in the data window for slope determination; M2 is the anti-interference threshold for trend determination.

13. A fast current protection method for a distribution network according to claim 11, characterized in that: The sampling value over-limit protection action criterion is: Among them, I set is the start threshold; I m (n), I m (n+1), I m (n+2) is the effective value of the current mutation signal at the nth, n+1th, and n+2nd sampling points.

14. A fast current protection system for a distribution network, using the method according to any one of claims 1 to 13, characterized in that: The system comprises: The frequency calculation module is used to monitor the mutation signal of the current signal in real time. When the current mutation signal is monitored, the real-time frequency of the current mutation signal is determined based on the time corresponding to the peak value of the current mutation signal and the time corresponding to the zero crossing point of the signal; The effective value calculation module is used to perform uniform harmonic filtering and half-cycle sampling value adjustment preprocessing on the current mutation signal based on the real-time frequency of the current mutation signal, and to use discrete orthogonalized Stockwell transform to obtain the distribution of the current mutation signal in the time-frequency domain and extract the real part and imaginary part of the fundamental wave of the current mutation signal, and calculate the effective value of the current mutation signal according to the real part and imaginary part of the fundamental wave. The protection action module is used to perform current protection action when the effective value of the current mutation signal simultaneously meets the climbing trend protection start criterion of multi-point incremental comparison and the sampling value over-limit protection action criterion.

15. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-13.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images