A method and device for correcting electromagnetic transient frequency based on frequency domain information entropy

Abstract

The present invention provides an electromagnetic transient frequency correction method and device based on frequency domain information entropy, which belongs to the technical field of electromagnetic transient frequency correction of power system, and comprises the following steps: designing two bandpass filters whose amplitude-frequency response is symmetrical about 50Hz in the measurement frequency band; calculating the ratio R of the amplitude-frequency response of the two bandpass filters in the measurement frequency band; H , construct frequency about R H Function, establish frequency calculation model and corrected frequency calculation model; perform fast Fourier transform on the signal sampling value sequence, calculate the frequency domain information entropy and its mutation amount of the electrical quantity according to the spectrum; use two bandpass filters to filter and calculate the signal sampling value sequence to obtain two phasors; judge whether the frequency domain information entropy and its mutation amount exceed the preset threshold; perform frequency correction calculation or frequency calculation according to the judgment result. The present invention adopts the above-mentioned electromagnetic transient frequency correction method and device based on frequency domain information entropy to realize the correction of electromagnetic transient frequency measurement data of power system.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic transient frequency correction of power systems, and in particular to an electromagnetic transient frequency correction method and device based on frequency domain information entropy. Background Art

[0002] Synchronous phasor measurement unit (PMU) can measure and upload phasor and frequency data with a unified time scale in real time to realize dynamic monitoring of power systems. Therefore, it has been widely deployed in transmission networks and has begun to be deployed in distribution networks and new energy stations. It is of great significance to power system state estimation, stability control, fault identification and location.

[0003] The core of PMU lies in the PMU algorithm. Most existing PMU algorithms estimate synchronous phasors based on steady-state signal models, and then estimate the frequency based on the phase angle difference of the synchronous phasors. However, during the electromagnetic transient process of tens of milliseconds after the power system fault, the amplitude or phase angle of the electrical quantity will change suddenly. Model mismatch will cause the frequency calculation of the traditional PMU algorithm to have errors of several hertz or even tens of hertz. This erroneous measurement has no reference value for subsequent applications, and may even cause applications that do not take special treatment to run incorrectly. During the electromagnetic transient period, there are many electrical quantity mutation points, random time, low signal-to-noise ratio, and serious harmonic and interharmonic distortion. It is difficult to effectively improve the algorithm by establishing a transient signal model. Therefore, it is necessary to study an electromagnetic transient frequency correction method that does not rely on the signal model to avoid the harm of erroneous measurement to subsequent applications and explore the application potential of electromagnetic transient synchronous measurement data. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for correcting electromagnetic transient frequency based on frequency domain information entropy, which can realize the correction of electromagnetic transient frequency measurement data of power system.

[0005] To achieve the above object, the present invention provides an electromagnetic transient frequency correction method based on frequency domain information entropy, comprising the following steps:

[0006] S1. According to the requirements of control PMU, design two bandpass filters with amplitude-frequency response symmetrical about 50Hz in the measurement frequency band;

[0007] S2. Calculate the ratio R of the amplitude-frequency response of the two bandpass filters within the measurement frequency band. H , for frequency and R H Perform linear fitting and construct frequency about R H Function, establish frequency calculation model and correct frequency calculation model;

[0008] S3, obtaining a signal sampling value sequence with a time stamp through a signal sampling module, performing a fast Fourier transform on the signal sampling value sequence to obtain a spectrum, and calculating the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy according to the spectrum;

[0009] S4, using two bandpass filters to filter and calculate the signal sampling value sequence to obtain two phasors;

[0010] S5. According to the frequency domain information entropy and its mutation amount obtained in step S3, it is determined whether the frequency domain information entropy and the mutation amount of the frequency domain information entropy exceed a preset threshold;

[0011] S6. If the mutation amount of the frequency domain information entropy exceeds the preset threshold but the frequency domain information entropy does not fall below the preset threshold, it is determined that the current state is in a transient process, and the corrected frequency is calculated according to the corrected frequency calculation model of step S2 and the two phasors calculated in step S4;

[0012] S7. If the mutation amount of the frequency domain information entropy does not exceed the preset threshold or the mutation amount exceeds the preset threshold but the frequency domain information entropy falls below the preset threshold, it is determined that no transient process has occurred or the transient process has ended, and the uncorrected frequency is calculated according to the frequency calculation model of step S2 and the two phasors calculated in step S4.

[0013] Preferably, in step S1, the length of the bandpass filter is less than or equal to two cycles.

[0014] Preferably, in step S1, the width of the measurement frequency band is 48-52 Hz.

[0015] Preferably, in step S1, according to the requirements of the control type PMU, two bandpass filters with amplitude-frequency responses symmetrical about 50 Hz in the measurement frequency band are designed, and the specific operations are as follows:

[0016] Design a bandpass filter h1(m), m=1,2,…2N+1, with a passband center at 48Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points;

[0017] Wherein, m represents the mth coefficient in the bandpass filter coefficient, and N represents the number of sampling points in one cycle;

[0018] Design another bandpass filter h2(m), m=1,2,…2N+1, with a passband center at 52Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points.

[0019] Preferably, in step S2, the ratio of the amplitude-frequency response of the two bandpass filters within the measurement frequency band isH The expression is:

[0020]

[0021] Wherein, f0 represents frequency, ranging from 48 Hz to 52 Hz; H1(f0) represents the amplitude-frequency response of the bandpass filter h1(m) at f0; H2(f0) represents the amplitude-frequency response of the bandpass filter h2(m) at f0.

[0022] Preferably, in step S2, the frequency and R H Perform linear fitting and construct frequency about R H The function is:

[0023]

[0024] Among them, p i is the polynomial fitting coefficient; q is the fitting order; For R H to the power of .

[0025] Preferably, in step S2, a frequency calculation model is established and a frequency calculation model is corrected, and the specific operations are:

[0026] because:

[0027]

[0028] Among them, R X Represents the ratio of the amplitudes of two phase quantities; X1(t k ) and X2(t k ) represent the t calculated by bandpass filters h1(m) and h2(m) respectively. k The phasor at the moment; X m represents the magnitude of the real phasor;

[0029] Therefore, the frequency calculation model is:

[0030]

[0031] Further, the correction frequency calculation model is established as:

[0032]

[0033] Among them, R XE Represents the ratio of the phase measurement amplitudes during electromagnetic transients; f e represents the corrected frequency during electromagnetic transient period; e1, e2 are the amplification factors of frequency domain information entropy; H(x) represents the frequency domain information entropy of electrical quantity x; Represents R XE to the power of i; Represents RX to the power of .

[0034] Preferably, in step S3, the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy are calculated according to the frequency spectrum, and the specific operation is as follows:

[0035] The fundamental signal model of voltage and current in the power system is:

[0036]

[0037] Where x(t) represents the voltage and current signal sampling value; X m (t) and They represent the amplitude and phase angle at time t respectively; w(t) represents noise;

[0038] The spectrum analysis of voltage and current signals is performed, and the results are expressed as follows:

[0039]

[0040] Wherein, X(k) represents the amplitude of the kth spectrum line in the spectrum; j represents the imaginary part; x(c) represents the discrete voltage and current sampling values; c represents the cth sampling value in the sampling value sequence;

[0041] The power spectrum of voltage and current signals is expressed as:

[0042] P(k)=X 2 (k), k=0,1,…N-1

[0043] Where P(k) represents the amplitude of the kth spectral line in the power spectrum;

[0044] The amplitude of the kth spectral line in the normalized power spectrum of the electrical quantity represents the probability of occurrence of the corresponding frequency in the signal:

[0045]

[0046] Among them, p p (k) represents the probability of the frequency corresponding to the kth spectral line in the power spectrum appearing in a certain observation in the current time window;

[0047] The frequency domain information entropy of electrical quantity is:

[0048]

[0049] The mutation amount of the information entropy of the electrical quantity in the frequency domain is:

[0050]

[0051] Among them, dH(t k ) indicates t k The mutation amount of information entropy in the frequency domain at a certain moment; fr Indicates the calculation frequency.

[0052] Preferably, in step S4, two bandpass filters are used to filter and calculate the signal sampling value sequence to obtain two phasors, and the specific operation is:

[0053]

[0054] Among them, X1(t k ) and X2(t k ) represent the t calculated by two band-pass filters. k The phasor at the moment; n represents the nth sampling point in the data window; h1(n) and h2(n) represent the nth coefficients of the two bandpass filters respectively.

[0055] Preferably, in step S6, the corrected frequency is calculated according to the corrected frequency calculation model of step S2 and the phasor calculated in step S4, and the specific operation is:

[0056]

[0057] Preferably, in step S7, the uncorrected frequency is calculated according to the frequency calculation model of step S2 and the phasor calculated in step S4, and the specific operation is:

[0058]

[0059] Preferably, the control-type PMU has the following requirements: low measurement delay, fast response speed, narrow frequency band for accurate phasor measurement, and low requirements for suppression of harmonic signals and out-of-band interference signals.

[0060] Preferably, the preset threshold selection method is: in order to accurately determine the beginning and end of the electromagnetic transient process, the frequency domain information entropy and mutation amount of the non-electromagnetic transient signal and the electromagnetic transient signal need to be calculated. Taking the 9 static and dynamic test signals in the IEC / IEEE60255-118-1 standard as an example, the maximum frequency domain information entropy and mutation amount of the non-electromagnetic transient signal are calculated; taking the 4 types of fault simulation data and multiple groups of recording data as an example, the electromagnetic transient signal calculates the frequency domain information entropy and mutation amount at the beginning and end of the electromagnetic transient in the fault. By analyzing and comparing the above frequency domain information entropy and mutation amount, a reasonable threshold is set to distinguish electromagnetic transient from non-electromagnetic transient signals.

[0061] The present invention also provides a device for implementing the above-mentioned electromagnetic transient frequency correction method based on frequency domain information entropy, including a frequency domain information entropy and mutation amount calculation module, a phasor calculation module, a system state judgment module and a frequency calculation and correction module;

[0062] According to the requirements of control PMU, design two bandpass filters with amplitude-frequency response symmetrical about 50Hz in the measurement frequency band; calculate the ratio R of the amplitude-frequency response of the two bandpass filters in the measurement frequency band H , for frequency and R H Perform linear fitting and construct frequency about R H The function is for offline computing work;

[0063] The frequency calculation and correction module is used to estimate and correct the frequency according to the frequency calculation model, the corrected frequency calculation model and the phasor calculation result;

[0064] The frequency domain information entropy and mutation amount calculation module is used to obtain a signal sampling value sequence with a time stamp through the signal sampling module, perform fast Fourier transform on the signal sampling value sequence to obtain a spectrum, and calculate the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy according to the spectrum;

[0065] The phasor calculation module is used to filter and calculate the signal sampling value sequence using two bandpass filters to obtain two phasors;

[0066] The system state judgment module is used to judge whether the frequency domain information entropy and the mutation amount of the frequency domain information entropy exceed a preset threshold value according to the obtained frequency domain information entropy and its mutation amount.

[0067] Therefore, the present invention adopts the above-mentioned electromagnetic transient frequency correction method and device based on frequency domain information entropy, and the technical effects are as follows:

[0068] (1) Within the measurement frequency band, the ratio of the amplitude-frequency response of the two bandpass filters R H The function is approximately symmetric about (50Hz, 1);

[0069] (2) Accurately identify electromagnetic transient processes based on frequency domain information entropy and its mutation amount;

[0070] (3) Reduce the maximum frequency calculation offset and maximum frequency change rate calculation offset of the traditional PMU algorithm during electromagnetic transients to less than one-tenth of the original;

[0071] (4) The frequency correction effect can be flexibly adjusted by adjusting the parameters of the frequency correction calculation equation according to the on-site operating conditions and subsequent application requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flow chart of an electromagnetic transient frequency correction method based on frequency domain information entropy of the present invention;

[0073] Figure 2 A schematic diagram of the amplitude-frequency response of two bandpass filters provided in an embodiment of the present invention;

[0074] Figure 3 The amplitude-frequency response ratio of two bandpass filters provided in the embodiment of the present invention;

[0075] Figure 4 are the maximum FO and RFO calculated by different methods during different types of faults; where, Figure 4 (a) shows the maximum FO calculated by different methods during different types of faults; Figure 4 (b) shows the maximum RFO calculated by different methods during different types of faults. DETAILED DESCRIPTION

[0076] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0077] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0078] Embodiment 1

[0079] like Figure 1 FIG. 1 is a flow chart of an electromagnetic transient frequency correction method based on frequency domain information entropy according to the present invention, which specifically includes the following steps:

[0080] S1. According to the requirements of control PMU, design two bandpass filters with amplitude-frequency response symmetrical about 50Hz in the measurement frequency band (width is 48-52Hz). The specific operation is: the length of the bandpass filter is less than or equal to two cycles.

[0081] Design a bandpass filter h1(m), m=1,2,…2N+1, with a passband center at 48Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points;

[0082] Wherein, m represents the mth coefficient in the bandpass filter coefficient, and N represents the number of sampling points in one cycle;

[0083] Design another bandpass filter h2(m), m=1,2,…2N+1, with a passband center at 52Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points.

[0084] The amplitude-frequency response of the two bandpass filters is shown as Figure 2 shown. Figure 2 In the figure, h1 is the band-pass filter h1(m); h2 is the band-pass filter h2(m);

[0085] S2. Calculate the ratio R of the amplitude-frequency response of the two bandpass filters within the measurement frequency band. H , for frequency and R H Perform linear fitting and construct frequency about R H Function, establish frequency calculation model and correct frequency calculation model;

[0086] like Figure 3 As shown in the figure, the ratio of the amplitude-frequency response of the two bandpass filters within the measurement frequency band is R H The expression is:

[0087]

[0088] Wherein, f0 represents frequency, ranging from 48 Hz to 52 Hz; H1(f0) represents the amplitude-frequency response of the bandpass filter h1(m) at f0; H2(f0) represents the amplitude-frequency response of the bandpass filter h2(m) at f0.

[0089] Frequency vs. R H Perform linear fitting and construct frequency about R H The function is:

[0090]

[0091] Among them, p i is the polynomial fitting coefficient; q is the fitting order; For R H to the power of .

[0092] Establish a frequency calculation model and modify the frequency calculation model. The specific operations are:

[0093] because:

[0094]

[0095] Among them, R X Represents the ratio of the amplitudes of two phase quantities; X1(t k ) and X2(t k ) represent the t calculated by bandpass filters h1(m) and h2(m) respectively. k The phasor at the moment; X m represents the magnitude of the real phasor;

[0096] Therefore, the frequency calculation model is:

[0097]

[0098] Further, the correction frequency calculation model is established as:

[0099]

[0100] Among them, R XE Represents the ratio of the phase measurement amplitudes during electromagnetic transients; f e represents the corrected frequency during electromagnetic transient period; e1, e2 are the amplification factors of frequency domain information entropy; H(x) represents the frequency domain information entropy of electrical quantity x; Represents R XE to the power of i; Represents R X to the power of .

[0101] S3, obtaining a signal sampling value sequence with a time stamp through a signal sampling module, performing a fast Fourier transform on the signal sampling value sequence to obtain a spectrum, and calculating the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy according to the spectrum;

[0102] In step S3, the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy are calculated according to the frequency spectrum. The specific operation is as follows:

[0103] The fundamental signal model of voltage and current in the power system is:

[0104]

[0105] Where x(t) represents the voltage and current signal sampling value; X m (t) and They represent the amplitude and phase angle at time t respectively; w(t) represents noise;

[0106] The spectrum analysis of voltage and current signals is performed, and the results are expressed as follows:

[0107]

[0108] Wherein, X(k) represents the amplitude of the kth spectrum line in the spectrum; j represents the imaginary part; x(c) represents the discrete voltage and current sampling values; c represents the cth sampling value in the sampling value sequence;

[0109] The power spectrum of voltage and current signals is expressed as:

[0110] P(k)=X 2 (k), k=0,1,…N-1

[0111] Where P(k) represents the amplitude of the kth spectral line in the power spectrum;

[0112] The amplitude of the kth spectral line in the normalized power spectrum of the electrical quantity represents the probability of occurrence of the corresponding frequency in the signal:

[0113]

[0114] Among them, p p(k) represents the probability of the frequency corresponding to the kth spectral line in the power spectrum appearing in a certain observation in the current time window;

[0115] The frequency domain information entropy of electrical quantity is:

[0116]

[0117] The mutation amount of the information entropy of the electrical quantity in the frequency domain is:

[0118]

[0119] Among them, dH(t k ) indicates t k The mutation amount of information entropy in the frequency domain at a certain moment; f r Indicates the calculation frequency.

[0120] S4. Use two bandpass filters to filter and calculate the signal sampling value sequence to obtain two phasors, specifically:

[0121]

[0122] Among them, X1(t k ) and X2(t k ) represent the t calculated by two band-pass filters. k The phasor at the moment; n represents the nth sampling point in the data window; h1(n) and h2(n) represent the nth coefficients of the two bandpass filters respectively.

[0123] S5. According to the frequency domain information entropy and its mutation amount obtained in step S3, it is determined whether the frequency domain information entropy and the mutation amount of the frequency domain information entropy exceed a preset threshold;

[0124] S6. If the mutation amount of the frequency domain information entropy exceeds the preset threshold but the frequency domain information entropy does not fall below the preset threshold, it is determined that the current state is in a transient process, and the corrected frequency is calculated according to the corrected frequency calculation model of step S2 and the two phasors calculated in step S4:

[0125]

[0126] S7. If the mutation amount of the frequency domain information entropy does not exceed the preset threshold or the mutation amount exceeds the preset threshold but the frequency domain information entropy drops below the preset threshold, it is determined that no transient process has occurred or the transient process has ended, and the uncorrected frequency is calculated according to the frequency calculation model of step S2 and the two phasors calculated in step S4:

[0127]

[0128] The present invention is described below by means of specific examples.

[0129] To verify the effectiveness of the proposed method, a model with a wind turbine at one end and an infinite power grid at the other end was built in PSCAD. Four faults, namely, A-phase grounding fault (AG), BC-phase fault (BC), BC-grounding fault (BCG), and ABC-phase fault (ABC), were set on the line. The fault data were exported and played back to the developed device and the comparison method by the OMICRON 256plus signal source.

[0130] The comparison methods include three frequency estimation methods: phase difference method (M1), differential geometry method (M2), and dual complex bandpass filter method (M3). Among them, the phase difference method is a widely used traditional frequency calculation method, the differential geometry method is a novel frequency measurement method proposed in recent years, and the dual complex bandpass filter method is the uncorrected frequency estimation method in the present invention.

[0131] The data window length of all methods is set to 2 cycles, the sampling rate is 10kHz, and the upload frequency is 100Hz. The calculation frequency of the proposed method, M1 and M3 is 100Hz. To ensure the measurement accuracy, the calculation frequency of M2 is set to 10000Hz.

[0132] The maximum frequency offset (FO) and ROCOF offset (RFO) calculated by various methods during the fault process are as follows: Figure 4 As shown in Table 1, FO is the absolute value of the difference between the calculated frequency and the rated frequency, and RFO is the absolute value of the difference between the calculated ROCOF and 0. In the results, M2 and M3 have the largest FO and RFO, and both the maximum FO and RFO occur under the BCG fault type, where the maximum FO exceeds 25Hz and the maximum RFO exceeds 1500Hz / s. The maximum FO and RFO of M1 are between the proposed method and M2 and M3, and its maximum FO and RFO exceed 9Hz and 350Hz / s respectively. The proposed method has the smallest FO and RFO, with a maximum FO less than 0.75Hz and a maximum RFO less than 35Hz / s, and both maximum values ​​occur under the ABC fault.

[0133] Table 1 Comparison results

[0134]

[0135] Embodiment 2

[0136] The present invention also provides an electromagnetic transient frequency correction device based on frequency domain information entropy, including a frequency domain information entropy and mutation amount calculation module, a phasor calculation module, a system state judgment module and a frequency calculation and correction module;

[0137] The frequency calculation and correction module is used to estimate and correct the frequency according to the frequency calculation model, the corrected frequency calculation model and the phasor calculation result;

[0138] The frequency domain information entropy and mutation amount calculation module is used to obtain a signal sampling value sequence with a time stamp through the signal sampling module, perform fast Fourier transform on the signal sampling value sequence to obtain a spectrum, and calculate the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy according to the spectrum;

[0139] The phasor calculation module is used to filter and calculate the signal sampling value sequence using two bandpass filters to obtain two phasors;

[0140] The system state judgment module is used to judge whether the frequency domain information entropy and the mutation amount of the frequency domain information entropy exceed a preset threshold value according to the obtained frequency domain information entropy and its mutation amount.

[0141] It is worth noting that the contents not described in detail in the embodiments of the present invention belong to the prior art known to professional and technical personnel in the field.

[0142] Therefore, the present invention adopts the above-mentioned electromagnetic transient frequency correction method and device based on frequency domain information entropy to realize the correction of electromagnetic transient frequency measurement data of the power system.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for correcting electromagnetic transient frequency based on frequency domain information entropy, characterized in that: The following steps are involved: S1. According to the requirements of control PMU, design two bandpass filters with amplitude-frequency response symmetrical about 50Hz in the measurement frequency band; S2. Calculate the ratio of the amplitude-frequency response of the two bandpass filters within the measurement frequency band. , for frequency and Perform linear fitting and construct frequency about Function, establish frequency calculation model and correct frequency calculation model; S3, obtaining a signal sampling value sequence with a time stamp through a signal sampling module, performing a fast Fourier transform on the signal sampling value sequence to obtain a spectrum, and calculating the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy according to the spectrum; S4, using two bandpass filters to filter and calculate the signal sampling value sequence to obtain two phasors; S5, judging whether the frequency domain information entropy and the mutation amount of the frequency domain information entropy obtained in step S3 exceed a preset threshold value; S6. If the mutation amount of the frequency domain information entropy exceeds the first preset threshold but the frequency domain information entropy does not fall below the second preset threshold, it is determined that the current state is in a transient process, and the corrected frequency is calculated according to the corrected frequency calculation model of step S2 and the two phasors calculated in step S4; S7, if the mutation amount of the frequency domain information entropy does not exceed the preset threshold or the mutation amount exceeds the preset threshold but the frequency domain information entropy drops below the preset threshold, it is determined that no transient process has occurred or the transient process has ended, and the uncorrected frequency is calculated according to the frequency calculation model of step S2 and the two phasors calculated in step S4; In step S2, the frequency and Perform linear fitting and construct frequency about The function is: in, are the polynomial fitting coefficients; is the fitting order; for of Power; In step S2, a frequency calculation model and a modified frequency calculation model are established, and the specific operations are as follows: because: in, It represents the ratio of the magnitudes of two phasors; and They represent bandpass filters and Calculated The phasor of the moment; represents the magnitude of the real phasor; Represents a bandpass filter exist The amplitude-frequency response at Represents a bandpass filter exist The amplitude-frequency response at Therefore, the frequency calculation model is: Further, the correction frequency calculation model is established as: in, Represents the ratio of the magnitudes of phasor measurements during electromagnetic transients; Indicates the corrected frequency during electromagnetic transients; , is the amplification factor of frequency domain information entropy; Indicates electrical quantity Frequency domain information entropy; express of Power; express of Power; In step S3, the frequency domain information entropy of the electrical quantity and the mutation amount of the frequency domain information entropy are calculated according to the frequency spectrum. The specific operation is as follows: The fundamental signal model of voltage and current in the power system is: in, Indicates the sampling value of voltage and current signals; and Respectively Moment amplitude and phase angle; Indicates noise; The spectrum analysis of voltage and current signals is performed, and the results are expressed as follows: in, Indicates the frequency spectrum The amplitude of the root line; represents the imaginary part; Represents discrete voltage and current sampling values; Indicates the first Sample values; The power spectrum of voltage and current signals is expressed as: in, Indicates the power spectrum The amplitude of the root line; The normalized power spectrum of the electrical quantity The amplitude of the root spectrum line represents the probability of occurrence of the corresponding frequency in the signal: in, Indicates the power spectrum The probability of the frequency corresponding to the root spectral line appearing in a certain observation in the current time window; The frequency domain information entropy of electrical quantity is: The mutation amount of the information entropy of the electrical quantity frequency domain is: in, express The mutation amount of information entropy in the time-frequency domain; Indicates the calculation frequency.

2. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 1 is characterized in that: In step S1, the length of the bandpass filter is less than or equal to two cycles.

3. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 2 is characterized in that: In step S1, the width of the measurement frequency band is 48-52 Hz.

4. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 3 is characterized in that: In step S1, according to the requirements of the control PMU, two bandpass filters with amplitude-frequency responses symmetrical about 50 Hz in the measurement frequency band are designed. The specific operations are as follows: Design a bandpass filter with a passband center at 48Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points. ; in, The coefficients of the bandpass filter are coefficients, Indicates the number of sampling points in one cycle; Design another bandpass filter with a passband center at 52Hz, a passband width greater than 4Hz, a passband ripple less than 0.05dB, a gain less than -100dB at -50Hz and each harmonic frequency, and an order less than 2 cycle sampling points. .

5. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 4 is characterized in that: In step S2, the ratio of the amplitude-frequency response of the two bandpass filters within the measurement frequency band is The expression is: in, Indicates frequency, ranging from 48Hz to 52Hz; Represents a bandpass filter exist The amplitude-frequency response at Represents a bandpass filter exist The amplitude-frequency response at .

6. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 5 is characterized in that: In step S4, two bandpass filters are used to filter and calculate the signal sampling value sequence to obtain two phasors. The specific operation is: in, and Respectively represent the two bandpass filters calculated The phasor of the moment; Indicates the first sampling points; and Represents the first coefficients.

7. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 6 is characterized in that: In step S6, the corrected frequency is calculated according to the corrected frequency calculation model of step S2 and the phasor calculated in step S4. The specific operation is: 。 8. The electromagnetic transient frequency correction method based on frequency domain information entropy according to claim 7 is characterized in that: In step S7, the uncorrected frequency is calculated according to the frequency calculation model in step S2 and the phasor calculated in step S4. The specific operation is: 。 9. A device for implementing the electromagnetic transient frequency correction method based on frequency domain information entropy as described in any one of claims 1 to 8.

Images