New energy access pilot protection method based on generalized s transform and pearson coefficient

Abstract

The new energy access longitudinal protection method based on generalized S transform and Pearson coefficient belongs to the technical field of power grid protection control, solves the problem that the traditional power frequency relay protection method is no longer applicable due to the limited current amplitude and controlled phase angle of the sending line during short circuit caused by the high proportion of new energy access to the power grid, the frequency domain fault characteristic quantity is obtained by using the generalized S transform on the two sides of the current, and the Pearson coefficient is used to calculate the correlation degree of the transient energy on both sides to distinguish internal and external faults, the technical scheme of the present application can meet the rapidity requirement of protection, has strong resistance to transition resistance, is not affected by the new characteristics of short circuit current caused by the access of new energy power electronic devices, and can be well applied to the new energy sending line.

Description

Technical Field

[0001] This invention belongs to the field of power grid protection and control technology, and relates to a longitudinal protection method for new energy access based on generalized S-transform and Pearson coefficient. Background Technology

[0002] The construction of new power systems has become an important strategy for modern power development, and the high proportion of distributed renewable energy connected to the grid has become a trend. However, with the advancement of new power system construction, the capacity of distributed renewable energy has increased significantly, and more and more power electronic devices are being connected to the power system. This leads to limitations on the current amplitude and phase angle of the transmission lines during short circuits, and the generation of new non-power frequency currents. Traditional power frequency-based protection methods are no longer applicable, and problems such as protection maloperation and failure to operate are prone to occur, posing challenges to the safety and stability of the entire power system.

[0003] Traditional transmission line protection mainly includes frequency domain protection and time domain protection. Frequency domain protection uses fault characteristics in different frequency bands to form protection criteria, but this method uses only one type of frequency information, and its sensitivity decreases in high-noise and high-harmonic scenarios. Time domain protection uses measurable time-varying fault characteristics to form protection criteria, but this method relies excessively on the values ​​of voltage and current on both sides of the line. If the grid capacity is low and the current value is small, the protection may fail to operate.

[0004] In the prior art, the Chinese invention patent document "A Longitudinal Protection Method Based on Spearman's Rank Correlation Coefficient" (application publication date: March 8, 2019, publication number: CN109449899A) obtains the fault characteristics of transient currents of different types of power sources based on the transient short-circuit current expressions of different types of power sources; based on the fault characteristics, it uses the Spearman's rank correlation coefficient to measure the waveform difference of transient currents on both sides of the transmission line when there is a fault within the zone; based on the waveform difference of transient currents on both sides when there is a fault within the zone, it distinguishes between faults inside and outside the zone, and realizes longitudinal protection of the transmission line of the new energy power station; however, in the face of a power grid model with a large number of power equipment connected, the Spearman coefficient cannot be used for preprocessing, the calculation parameters are numerous, the calculation speed is slow, and the judgment rate of the correlation between variables is low, which cannot achieve the effect of rapid fault identification.

[0005] Therefore, it is urgent to develop a new protection method to adapt to power grids with a high proportion of renewable energy access and to protect renewable energy transmission lines. Summary of the Invention

[0006] The purpose of this invention is to design a longitudinal protection method for new energy access based on generalized S-transform and Pearson coefficient, so as to solve the problem that the current amplitude of the transmission line is limited and the phase angle is controlled during a short circuit caused by the grid with a high proportion of new energy access, which makes the traditional power frequency relay protection method no longer applicable.

[0007] The present invention solves the above-mentioned technical problems through the following technical solutions:

[0008] The longitudinal protection method for new energy access based on generalized S-transform and Pearson coefficient includes the following steps: S1. Read the current signals of the protection devices on both sides of the line, set the sampling rate, and use the protection criterion formula to determine whether the protection start-up conditions are met. S2. Perform a generalized S-transform on the current signals on both sides of the line to extract the transient energy of the current signals on both sides. S3. Calculate the transient energy Pearson coefficients on both sides to determine whether the fault is inside or outside the zone. If it is an internal fault, send a trip signal to the switches on both sides to disconnect the faulty line. If it is an external fault, send a protection blocking signal to the switches on both sides to prevent the switches from malfunctioning.

[0009] Furthermore, the protection criterion formula mentioned in step S1 is:

[0010] In the formula, The measured change in current; This refers to the rated current.

[0011] Furthermore, the method for extracting the transient energy of the currents on both sides of the line by performing a generalized S-transform in step S2 is as follows: S21. Input the current signals of the protection devices on both sides of the line read in step S1 to define the S-transform; S22. Add a positive-zero adjustment factor g to the time window function of the S-transform to construct the generalized S-transform; S23. Discretize the generalized S-transform to obtain the discrete generalized S-transform; S24. Extract the transient energy of the discrete generalized S-transform of the current signals on both sides in a specific frequency band.

[0012] Furthermore, the calculation formula for the S-transform described in step S21 is as follows: Given the original current signal x(t), the S-transform is defined as follows:

[0013] In the formula, The S-transform function, Let x(t) be the time window function of the S-transform, x(t) be the current signal read from the protection devices on both sides of the line, and t be the time parameter. is the control parameter of the Gaussian time window function, j is a complex number flag, and f is the frequency.

[0014] Furthermore, the formula for the generalized S-transform described in step S22 is as follows:

[0015] In the formula, The time window function of the generalized S-transform. It is the time window function of the generalized S-transform.

[0016] Furthermore, the discretization process of the discrete generalized S-transform described in step S23 is as follows: Discretizing x(t) into a vector x[iT], the discrete Fourier transform of x[iT] is as follows:

[0017] In the formula, N is the number of sampling points, n is an integer between 0 and N-1, and T is the sampling time interval.

[0018] Let the above formula be , Then the generalized S-transform of the discrete vector x[iT] is as follows:

[0019] In the formula, k is an integer between 0 and N-1; Furthermore, the formula for extracting the transient energy of the discrete generalized S-transform of the current signals on both sides in a specific frequency band, as described in step S24, is as follows:

[0020] In the formula, It is a complex time-frequency matrix, where the row vectors represent the time-domain characteristics at a certain frequency, and the column vectors represent the amplitude-frequency characteristics at a certain time. for The absolute value of matrix elements.

[0021] Furthermore, the method for calculating the transient energy Pearson coefficients on both sides and determining whether the fault is inside or outside the region, as described in step S3, is as follows: The correlation coefficient is defined as follows:

[0022] In the formula, D(X) and D(Y) are the variances of the two variables X and Y, respectively; E is the mean. , Let X and Y be the standard deviations of the two variables X and Y, respectively; and let Cov(X,Y) be the covariance of the two variables X and Y. The correlation coefficient between two variables X and Y is the quotient of their covariance and standard deviation. It ranges from -1 to 1. The larger the absolute value, the more correlated the two variables are. When the value is 0, the two variables are not correlated. If the correlation coefficient is greater than 0, the two variables are positively correlated. If the correlation coefficient is less than 0, the two variables are negatively correlated.

[0023] Pearson's coefficient is:

[0024] In the formula, Pearson coefficient; , Let X and Y be the values ​​of the i-th sequence elements, respectively; , These are the means of the sequence variables X and Y, respectively; When an intra-zone fault occurs, the bilateral transient energies are approximately positively correlated, with a theoretical Pearson coefficient of 1. When an extra-zone fault occurs, the bilateral transient energies are approximately negatively correlated, with a theoretical Pearson coefficient of -1. A threshold is set to account for error. , The formula is set as follows:

[0025] In the formula, K mag K mar The proportions are the reliability coefficient and margin coefficient of the amplitude; The specific criteria are as follows: the correlation energy of the generalized S-transform transient energy on both sides of the line is greater than the threshold. When the fault is identified as being within the designated area, the generalized S-transform transient energy correlation energy on both sides of the line is less than the threshold. When the fault is located outside the designated area, it is determined to be caused by the following formula:

[0026] Among them, E m E n These represent the generalized S-transform transient energies on both sides of the line.

[0027] The advantages of this invention are: This invention derives fault characteristic quantities in the frequency domain by applying a generalized S-transform to the currents on both sides, and uses Pearson coefficients to calculate the correlation between transient energy on both sides, thereby distinguishing between faults within and outside the fault zone. Through generalized S-transform processing, time-frequency characteristics are extracted, enabling the extraction of high-frequency fault characteristics. This provides a pre-processing effect in power grid models with a large number of connected power devices, improving the accuracy of relay protection. Using Pearson coefficients for correlation coefficient calculation achieves fewer calculation parameters, faster calculation speed, and a high accuracy in judging the correlation between variables. Applied to new energy power grid protection, this allows for rapid fault identification, enabling quick protection action and better performance.

[0028] The technical solution of this invention can meet the requirements of rapid protection and has a strong ability to withstand transition resistance. It is not affected by the new characteristics of short-circuit current caused by the access of new energy power electronic devices and can be well applied to new energy transmission lines. Attached Figure Description

[0029] Figure 1 This is a flowchart of a new energy access longitudinal protection method based on generalized S-transform and Pearson coefficients according to an embodiment of the present invention; Figure 2 This is a power grid model for new energy access built in the PSCAD / EMTDC simulation software according to an embodiment of the present invention; Figure 3 The current and Pearson coefficient of the power grid model for new energy access in this embodiment of the invention are shown under normal conditions. Figure 4 The current and Pearson coefficient of the power grid model for new energy access in this embodiment of the invention during an external fault; Figure 5 The current and Pearson coefficient of the power grid model for new energy access in this embodiment of the invention during regional faults. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments: Example 1 like Figure 1As shown in the figure, the longitudinal protection method for new energy grid access based on S-transform and Pearson coefficient proposed in this embodiment mainly includes two parts: S-transform and Pearson coefficient calculation. The specific steps are as follows: 1. Read the current signals of the protection devices on both sides of the line, set the sampling rate, and use the protection criterion formula to determine whether the protection start conditions are met. If the protection criterion conditions are not met, then the protection is blocked to prevent false protection operation. The protection criterion formula is as follows: (1) In the formula, The measured change in current; This refers to the rated current.

[0032] The method for setting the sampling rate is as follows: The sampling rate needs to be selected appropriately. If it is too high, the acquisition time will be too long, affecting the protection speed. If it is too low, the power grid fault will not be accurately identified, resulting in the protection failing to operate. The sampling rate range is usually between 3kHz and 8kHz. Preferably, the sampling rate is taken as the middle value of 5kHz.

[0033] 2. Perform S-transformation on the currents on both sides of the line to extract the transient energy of the currents on both sides. The specific steps are as follows: The S-transform is a Fourier transform performed by adding a Gaussian time window function to the original signal. This Gaussian time window function varies with frequency, ultimately forming the S-matrix. This matrix has time-frequency characteristics and can represent the changes in amplitude and phase of signals at different frequencies.

[0034] Given the original signal x(t), the S-transform is defined as follows: (2) In the formula, t is the time parameter. Here, j is the control parameter for the Gaussian time window function, j is a complex number flag, and f is the frequency. This is the time window function for the S-transform.

[0035] As shown in formula (2), the width of the time window function is negatively correlated with the frequency. When the frequency decreases, the width of the time window function increases, and when the frequency increases, the width of the time window function decreases. Using the S-transform can make the time window function wider in the low-frequency range and narrower in the high-frequency range.

[0036] The generalized S-transform adds a positive-zero adjustment factor g to the Gaussian time window function of the original S-transform, which can further optimize the time-frequency resolution of the S-transform. The formula for the time window function of the generalized S-transform is as follows: (3) In the formula, is the time window function of the generalized S transform.

[0037] The time window function can further optimize the time-frequency resolution of the generalized S transform by adjusting the value of g. Then the formula of the generalized S transform is as follows: (4) If g = 0, then the generalized S transform is equivalent to the S transform; if g > 1, then the generalized S transform has a higher frequency resolution and can form a clear high-frequency signal; if 0 < g < 1, then the generalized S transform has a higher time resolution and can detect mutation points.

[0038] The current signal collected in this embodiment belongs to a discrete signal, so it is necessary to study the discrete generalized S transform. If the discrete vector of the input signal x(t) is x[iT], then the discrete Fourier transform of x[iT] is shown as follows: (5) In the formula, N is the number of sampling points; n ranges from 0 to an integer between N - 1; T is the sampling time interval.

[0039] Let , , then the generalized S transform of the discrete vector x[iT] is shown as follows: (6) In the formula, k is an integer from 0 to N - 1; The sum of the transient energies of the generalized S transform of the signal in a specific frequency band is: (7) In the formula, is the complex time-frequency matrix, the row vector is the time-domain feature at a certain frequency, and the column vector is the amplitude-frequency feature at a certain moment; is the absolute value of the matrix element.

[0040] 3. Calculate the Pearson coefficient of the transient energies on both sides to judge the in-zone and out-of-zone faults; if it is an in-zone fault, then send a tripping signal to the switches on both sides to disconnect the faulty line; if it is an out-of-zone fault, then send a protection blocking signal to the switches on both sides to prevent the switches from malfunctioning. The specific steps are as follows: The Pearson coefficient is a method to measure the linear correlation between two variables and evaluates the relationship degree of the variables through the covariance matrix of the data. The definition of the correlation coefficient is shown as follows: (8) In the formula, D(X) and D(Y) are the variances of the two variables X and Y respectively; E is the mean value; , Let X and Y be the standard deviations of the two variables X and Y, respectively; and let Cov(X,Y) be the covariance of the two variables X and Y. The correlation coefficient between two variables X and Y is the quotient of their covariance and standard deviation. It ranges from -1 to 1. The larger the absolute value, the more correlated the two variables are. When the value is 0, the two variables are not correlated. If the correlation coefficient is greater than 0, the two variables are positively correlated. If the correlation coefficient is less than 0, the two variables are negatively correlated.

[0041] Pearson's coefficient is: (9) In the formula, Pearson coefficient; , Let X and Y be the values ​​of the i-th sequence elements, respectively; , These are the means of the sequence variables X and Y, respectively.

[0042] When an intra-zone fault occurs, the bilateral transient energies are approximately positively correlated, with a theoretical Pearson coefficient of 1. When an extra-zone fault occurs, the bilateral transient energies are approximately negatively correlated, with a theoretical Pearson coefficient of -1. A threshold is set to account for error. , The formula is set as follows: (10) In the formula, K mag K mar The proportions are the reliability coefficient and margin coefficient of the amplitude; The specific criteria are as follows: the correlation energy of the generalized S-transform transient energy on both sides of the line is greater than the threshold. When the fault is identified as being within the designated area, the generalized S-transform transient energy correlation energy on both sides of the line is less than the threshold. When the fault is located outside the designated area, it is determined to be caused by the following formula: (11) Among them, E m E n These represent the generalized S-transform transient energies on both sides of the line.

[0043] To account for the possibility of the maximum error, some margin needs to be reserved. In this embodiment, a reliability coefficient K is set. mag The value is 0.95, and the margin coefficient K mar The value is 0.95. The value was set to 0.95 × 0.95 = 0.9025, and the final protection threshold was obtained through calculation. It is 0.9025.

[0044] 4. Verification To verify the effectiveness of the present invention, the following example illustrates the advancement of the proposed longitudinal protection method for new energy grid access based on generalized S-transform and Pearson coefficient.

[0045] like Figure 2 As shown, a grid model for renewable energy integration was built in PSCAD / EMTDC simulation software. There are 100 photovoltaic generators with a capacity of 1MW each, and 20 wind turbine generators with a capacity of 2MW each. The simulation step size was set to 1 second, and the sampling rate to 5kHz. The transmission line length is 300km, and the parameters are as follows: r1=0.0758Ω / km, l1=0.83922mH / km, c1=0.014μF / km, r0=0.15421Ω / km, l0=2.6439mH / km, c0=0.009296μF / km; During normal operation, the simulation results of the current on both sides are as follows: Figure 3 As shown in the figure, the dashed line represents the current acquisition result on the left side of the line, and the solid line represents the current acquisition result on the right side of the line. The two currents in the figure have a certain phase difference, which is caused by the impedance of the transmission line itself. Furthermore, the simulation results show that the transient energy Pearson coefficient is 0, indicating that no fault has occurred and the protection does not operate.

[0046] When an external fault occurs, the fault occurs at a time of 0.2 seconds. The simulation results of the currents on both sides are as follows: Figure 4 As shown in the figure, when an external fault occurs, the current on both sides suddenly increases and then stabilizes. At this time, the Pearson coefficient drops rapidly, eventually reaching -0.993, indicating that an external fault has occurred. The protection on both sides of the line is blocked to prevent false tripping.

[0047] When a fault occurs within the fault zone, the fault occurs at a time of 0.2 seconds. The simulation results of the currents on both sides are as follows: Figure 5 As shown in the figure, before the fault occurs, the current waveforms on both sides are similar. When a fault occurs within the zone, the current waveforms on both sides become opposite, and the Pearson coefficient at the fault point rises rapidly to 0.998, indicating that a fault has occurred within the zone and the conditions for a fault within the zone have been met. The switches on both sides of the line receive the trip signal, and the protection system operates quickly.

[0048] Single-phase ground faults often pass through a transition resistance, therefore, it is necessary to verify the protection against this transition resistance. The transition resistances are set to 10Ω, 20Ω, 30Ω, 40Ω, 50Ω, 60Ω, 70Ω, 80Ω, 90Ω, and 100Ω. Table 1 shows that when an internal fault occurs through the transition resistance, the Pearson coefficient is greater than 0.9025, meeting the protection conditions, and the protection device operates reliably. When an external fault occurs, the Pearson coefficient is less than -0.9025, and the double-sided switches are locked to prevent maloperation.

[0049] Table 1 Protection Verification under Transition Resistance

[0050] To verify that the protection's action time meets the requirements, the fault protection action time in the recording area is only 16ms, which meets the protection's speed requirement.

[0051] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for longitudinal protection of new energy access based on generalized S-transform and Pearson coefficient, characterized in that, Includes the following steps: S1. Read the current signals of the protection devices on both sides of the line, set the sampling rate, and use the protection criterion formula to determine whether the protection start-up conditions are met. S2. Perform a generalized S-transform on the current signals on both sides of the line to extract the transient energy of the current signals on both sides. S3. Calculate the transient energy Pearson coefficients on both sides to determine whether the fault is inside or outside the zone. If it is an internal fault, send a trip signal to the switches on both sides to disconnect the faulty line. If it is an external fault, send a protection blocking signal to the switches on both sides to prevent the switches from malfunctioning. The method for calculating the transient energy Pearson coefficients on both sides and determining whether a fault is inside or outside the zone is as follows: The correlation coefficient is defined as follows: In the formula, D(X) and D(Y) are the variances of the two variables X and Y, respectively; E is the mean. , Let X and Y be the standard deviations of the two variables X and Y, respectively; and let Cov(X,Y) be the covariance of the two variables X and Y. The correlation coefficient between two variables X and Y is the quotient of their covariance and standard deviation. It ranges from -1 to 1. The larger the absolute value, the more correlated the two variables are. When the value is 0, the two variables are not correlated. If the correlation coefficient is greater than 0, the two variables are positively correlated. If the correlation coefficient is less than 0, the two variables are negatively correlated. Pearson's coefficient is: In the formula, Pearson coefficient; , Let X and Y be the values ​​of the i-th sequence elements, respectively; , These are the means of the sequence variables X and Y, respectively; When an intra-zone fault occurs, the bilateral transient energies are approximately positively correlated, with a theoretical Pearson coefficient of 1. When an extra-zone fault occurs, the bilateral transient energies are approximately negatively correlated, with a theoretical Pearson coefficient of -1. A threshold is set to account for errors. , The formula is set as follows: In the formula, K mag K mar These are the reliability coefficient and margin coefficient of the amplitude, respectively; The specific criteria are as follows: the correlation energy of the generalized S-transform transient energy on both sides of the line is greater than the threshold. When the fault is identified as being within the designated area, the generalized S-transform transient energy correlation energy on both sides of the line is less than the threshold. When the fault is located outside the designated area, it is determined to be caused by the following formula: Among them, E m E n These represent the generalized S-transform transient energies on both sides of the line.

2. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 1, characterized in that, The protection criterion formula mentioned in step S1 is: In the formula, The measured change in current; This refers to the rated current.

3. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 2, characterized in that, The method for extracting the transient energy of the currents on both sides of the line by performing a generalized S-transform in step S2 is as follows: S21. Input the current signals of the protection devices on both sides of the line read in step S1 to define the S-transform; S22. Add a positive-zero adjustment factor g to the time window function of the S-transform to construct the generalized S-transform; S23. Discretize the generalized S-transform to obtain the discrete generalized S-transform; S24. Extract the transient energy of the discrete generalized S-transform of the current signals on both sides in a specific frequency band.

4. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 3, characterized in that, The formula for calculating the S-transform mentioned in step S21 is as follows: Given the original current signal x(t), the S-transform is defined as follows: In the formula, The S-transform function, Let x(t) be the time window function of the S-transform, x(t) be the current signal read from the protection devices on both sides of the line, and t be the time parameter. is the control parameter of the Gaussian time window function, j is a complex number flag, and f is the frequency.

5. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 4, characterized in that, The formula for the generalized S-transform described in step S22 is as follows: In the formula, For the generalized S-transform function, It is the time window function of the generalized S-transform.

6. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 5, characterized in that, The discretization process of the discrete generalized S-transform described in step S23 is as follows: Discretizing x(t) into a vector x[iT], the discrete Fourier transform of x[iT] is as follows: In the formula, N is the number of sampling points, n is an integer between 0 and N-1, and T is the sampling time interval; Let the above formula be , Then the generalized S-transform of the discrete vector x[iT] is as follows: In the formula, k is an integer between 0 and N-1.

7. The renewable energy access longitudinal protection method based on generalized S-transform and Pearson coefficient as described in claim 6, characterized in that, The formula for extracting the transient energy of the discrete generalized S-transform of the current signals on both sides in a specific frequency band, as described in step S24, is as follows: In the formula, It is a complex time-frequency matrix, where the row vectors represent the time-domain characteristics at a certain frequency, and the column vectors represent the amplitude-frequency characteristics at a certain time. for The absolute value of matrix elements.

Images