Method for discriminating lightning disturbance of flexible HVDC transmission line based on instantaneous dominant frequency characteristic

Abstract

The application discloses a lightning disturbance discrimination method for flexible HVDC transmission line based on instantaneous frequency characteristics, and specifically comprises the following steps: step 1, extracting voltage and current data in a time window from 1ms before protection starting to 1.0ms after protection starting, and obtaining line mode voltage forward and reverse traveling waves; step 2, performing discrete S transform on the line mode voltage reverse traveling wave obtained in step 1 to obtain a mode time-frequency matrix, and extracting an amplitude-frequency curve at a fault occurrence moment; step 3, obtaining a maximum value number Q of the instantaneous amplitude-frequency curve, if Q=1, it is lightning disturbance, and the discrimination ends; if Q>1, it is a fault, and step 4 is entered; step 4, defining a frequency of a point with the maximum amplitude in the amplitude-frequency curve as a main frequency F1, comparing the main frequency F1 with a threshold value F set , and judging a fault type according to a comparison result. The discrimination method provided by the application is simple in calculation, and can accurately judge lightning disturbance and lightning fault.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power systems, and relates to a lightning disturbance discrimination method for a flexible DC transmission line based on instantaneous main frequency characteristics. BACKGROUND

[0002] In terms of large-scale wind energy and solar energy centralized grid connection far away from a power grid, flexible DC transmission has a significant advantage and is a trend of future power grid technology development. However, the lightning strike probability of a long-distance overhead line flexible DC transmission system is obviously increased due to environmental and climate factors. When the lightning disturbance is slight, no permanent damage point is caused to the transmission line, which belongs to lightning interference, and the protection does not need to act, but still needs to be identified. If the lightning disturbance is strong enough to cause flashover of an insulator, the lightning fault needs to be removed by protection action. In actual operation, the lightning interference can introduce a large amount of high-frequency transient components in the DC line, which can easily cause misoperation of ultra-high-speed protection based on transient information, and seriously reduces the reliability of the protection, and even endangers the safe operation stability of the power system.

[0003] At present, the lightning recognition research of the flexible DC transmission line mainly focuses on single-ended protection. In the prior art, the lightning current wave head is recognized by using the time-domain waveform of the transient. However, the time-domain method often has poor adaptability and is easily disturbed by noise. Some documents use the rich characteristics of the high-frequency components of the lightning traveling wave for analysis. However, due to the large change of the high-frequency characteristics at the line boundary caused by the fold reflection of the traveling wave, if only the energy distribution method of the high-frequency and low-frequency bands is used, the disturbance distance has a great influence, the adaptability is low, and the calculation complexity is high, which is not conducive to accurate disturbance discrimination. SUMMARY

[0004] The purpose of the application is to provide a lightning disturbance discrimination method for a flexible DC transmission line based on instantaneous main frequency characteristics, which can accurately judge lightning interference and lightning fault by using the method.

[0005] The technical solution adopted by the application is a lightning disturbance discrimination method for a flexible DC transmission line based on instantaneous main frequency characteristics, which specifically includes the following steps:

[0006] Step 1, extract the voltage and current data in the time window of 1.0 ms before protection startup to 1.0 ms after protection startup, and calculate the line-mode voltage forward traveling wave and reverse traveling wave;

[0007] Step 2, perform discrete S transform on the line-mode voltage reverse traveling wave obtained in step 1 to obtain a mode-time-frequency matrix, and extract the amplitude-frequency curve at the fault occurrence time;

[0008] Step 3, calculate the number Q of maximum points of the instantaneous amplitude-frequency curve, if Q=1, it is judged as lightning interference, and the discrimination is ended; if Q>1, it is judged as a fault, and step 4 is entered;

[0009] Step 4, define the frequency of the point with the maximum amplitude in the amplitude-frequency curve as the main frequency F1, and compare the main frequency F1 with the threshold value F set Comparing, judging the fault type according to the comparison result.

[0010] The application also has the characteristics that:

[0011] The specific process of step 1 is:

[0012] Step 1.1, after the protection starting element is started, take 1.0 ms before starting and 1.0 ms after starting as the protection time window, read the positive voltage u p (k) of the protection installation in the time window, n (k), the positive current i p (k), the negative current i n (k) respectively, and the average value of the positive voltage u p0 , the average value of the negative voltage u n0 , the average value of the positive current i p0 , the average value of the negative current i n0 Subtract, get the positive voltage fault component Δu p (k), the negative voltage fault component Δu n (k), the positive current fault component Δi p (k), and the negative current fault component Δi n (k), the calculation process is shown in formula (1):

[0013]

[0014] Step 1.2, calculate the line mode voltage fault component Δu1(k) and the line mode current fault component Δi1(k) in the time window, the calculation process is shown in formula (2):

[0015]

[0016] Step 1.3, calculate the line mode voltage reverse wave u b (k) and the line mode voltage forward wave u f (k) in the time window, the calculation process is shown in formula (3):

[0017]

[0018] In the formula, Z c1 is the line mode wave impedance of the line.

[0019] The specific process of step 2 is:

[0020] Step 2.1, compare the line mode voltage reverse wave u b(k) performing a discrete S transform to obtain a complex time-frequency matrix S(k, n), the discrete S transform calculation formula being shown as formula (4) :

[0021]

[0022] In the formula, k is a time serial number, 0≤k≤N-1; T is a sampling time interval; N is a sampling point number; m is a discrete time point, and the value is 0≤m≤N-1; n is a frequency serial number, and the value is 1≤n≤fix(N / 2), fix indicating taking the maximum integer not exceeding N / 2; n / NT represents a frequency corresponding to the frequency serial number n; X(n) represents: a signal of a frequency corresponding to the frequency serial number n in the time window u b (k) performing a Fourier transform on the data of (k) to obtain a discrete signal;

[0023] Step 2.2, taking a modulus value of each element in the complex time-frequency matrix S(k, n) in step 2.1 to obtain a modulus time-frequency matrix D[k, n], as shown in formula (5) :

[0024]

[0025] In formula (5), D(k a ,n b ) represents an element in the a-th row and the b-th column of the modulus time-frequency matrix D[k, n];

[0026] Step 2.3, extracting an instantaneous amplitude-frequency curve g(n) at the moment k=fix(N / 2)+1 in formula (5), and defining the moment as k1, as shown in formula (6) :

[0027] g(n)=D(k1,n) (6).

[0028] The specific process of step 3 is as follows:

[0029] Step 3.1, extracting a maximum value in the instantaneous amplitude-frequency curve g(n), if a point n1 in the amplitude-frequency curve is greater than the amplitude value corresponding to the previous sampling point and the amplitude value corresponding to the subsequent sampling point, then n1 is a maximum value point in the amplitude-frequency curve, as shown in formula (7) :

[0030]

[0031] In the formula, n1∈[2, fix(N / 2)-1].

[0032] Step 3.2, counting the number Q of the maximum value points satisfying formula (7), and judging whether it is a lightning disturbance according to the number Q of the maximum values, if Q=1, it is judged as a lightning disturbance; if Q>1, it is judged as a fault, and step 4 is entered.

[0033] The specific process of step 4 is as follows: the main frequency F1 and the threshold value F setIn contrast, if formula (8) is satisfied, it is judged as a ground fault, otherwise it is judged as a lightning strike fault:

[0034] F1 < F set (8).

[0035] The beneficial effects of the present application are that the present application is a lightning disturbance identification method for flexible DC transmission lines, according to the modal time-frequency matrix of the reverse traveling wave of the line mode voltage after S transformation, the amplitude-frequency curve at the fault occurrence time is extracted, and the recognition criterion is formed according to the position of the main frequency. Compared with other methods, it can better tolerate the influence of transition resistance. The number of maximum values and the position of the main frequency are used to distinguish lightning disturbance. Secondly, this method only uses the curve at the fault occurrence time to extract fault features, reduces the interference of subsequent wave heads, and reduces the complexity of calculation. The method is simple in principle, high in sensitivity, strong in reliability, strong in anti-noise ability, almost not affected by transition resistance, and meets the requirements of engineering practice. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is the flow chart of the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application;

[0037] Figure 2 is the flexible DC transmission line topology structure used in the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application;

[0038] Figure 3 is the amplitude-frequency curve at the fault occurrence time extracted after S transformation by the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application in the simulation verification stage, which occurs at 100km;

[0039] Figure 4 is the amplitude-frequency curve at the fault occurrence time extracted after S transformation by the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application in the simulation verification stage, which occurs at 400km and the positive pole is grounded with a transition resistance of 500Ω;

[0040] Figure 5 is the amplitude-frequency curve at the fault occurrence time extracted after S transformation by the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application in the simulation verification stage, which occurs at 200km and the positive pole is grounded with a metallic short circuit under 20dB noise interference;

[0041] Figure 6 is the amplitude-frequency curve at the fault occurrence time extracted after S transformation by the lightning disturbance identification method for flexible DC transmission lines based on the instantaneous main frequency characteristics of the present application in the simulation verification stage, which occurs at 300km. DETAILED DESCRIPTION

[0042] The application will be described in detail below in combination with the drawings and specific embodiments.

[0043] The lightning disturbance discrimination method for flexible HVDC transmission lines based on the instantaneous dominant frequency characteristics comprises the following steps:

[0044] Step 1, after the protection starting element is started, 1.0 ms before starting and 1.0 ms after starting are taken as the protection time window, and the voltage and current data in the time window are read. The positive voltage fault component Δu(k) is calculated p (k), the negative voltage fault component Δu(k) is calculated n (k), the positive current fault component Δi(k) is calculated p (k), and the negative current fault component Δi(k) is calculated n (k). Then, the line mode voltage fault component Δu1(k) and the line mode current fault component Δi1(k) are calculated, and finally the line mode voltage reverse traveling wave u b (k) and the line mode voltage forward traveling wave u f (k) are calculated.

[0045] Step 1.1, after the protection starting element is started, 1.0 ms before starting and 1.0 ms after starting are taken as the protection time window, and the voltage and current data in the data window are read. The positive voltage u(k) at the protection installation is read p (k), the negative voltage u(k) is read n (k), the positive current i(k) is read p (k), the negative current i(k) is read n (k), and the positive voltage average value u(k-10) before the protection starting 10.0 ms, the negative voltage average value u(k-10), the positive current average value i(k-10), and the negative current average value i(k-10) are read p0 n0 p0 n0 respectively. The positive voltage fault component Δu(k) is calculated p (k), the negative voltage fault component Δu(k) is calculated n (k), the positive current fault component Δi(k) is calculated p (k), and the negative current fault component Δi(k) is calculated n (k). The calculation formula is shown in formula (1):

[0046]

[0047] Step 1.2, the line mode voltage fault component Δu1(k) and the line mode current fault component Δi1(k) in the time window are calculated, and the calculation formula is shown in formula (2):

[0048]

[0049] Step 1.3, the line mode voltage reverse traveling wave u b (k) in the time window is calculated.​b (k), the line mode voltage forward wave u f (k), the calculation formula is shown in formula (3):

[0050]

[0051] In the formula, Z c1 is the line mode wave impedance.

[0052] Step 2, the discrete S transform is performed on the line mode voltage reverse wave u b (k) in the time window in step 1 to obtain a corresponding mode time-frequency matrix, the corresponding amplitude at each frequency at the time of 1.0 ms in the mode time-frequency matrix is extracted to obtain an amplitude-frequency characteristic curve at the time of fault occurrence, which is recorded as an instantaneous amplitude-frequency curve.

[0053] Step 2.1, the discrete S transform is performed on u b (k) to obtain a complex time-frequency matrix, and the calculation formula of the discrete S transform is shown in formula (4):

[0054]

[0055] In the formula, k is a time sequence number, 0≤k≤N-1; T is a sampling time interval; N is a sampling point number; m is a discrete time point, and the value is: 0≤m≤N-1; n is a frequency sequence number, and the value is: 1≤n≤fix(N / 2), fix indicates taking the maximum integer not exceeding N / 2; n / NT represents the frequency corresponding to the frequency sequence number n; X(n) represents the discrete signal obtained by performing Fourier transform on the data of u b (k) in the time window.

[0056] Step 2.2, the S transform is performed on the collected discrete signal by using formula (4) to obtain a complex time-frequency matrix S(k, n) of (N / 2+1)×N, and the modulus value of each element of S is obtained to obtain a mode time-frequency matrix D[k, n] as shown in formula (5):

[0057]

[0058] In formula (5), D(k a ,n b ) represents the element in the a-th row and b-th column of the mode time-frequency matrix D[k, n].

[0059] Step 2.3, the instantaneous amplitude-frequency curve g(n) at the time of k=fix(N / 2)+1 in formula (5) is extracted, and the time is defined as k1 time, as shown in formula (6):

[0060] g(n)=D(k1,n) (6).

[0061] Step 3: Extract the maximum value in the instantaneous amplitude-frequency curve g(n). If a point n1 in the amplitude-frequency curve is greater than the amplitude of the previous sampling point and the amplitude of the next sampling point, then n1 is the maximum value point in the amplitude-frequency curve, as shown in equation (7), where n1∈[2, fix(N / 2)-1].

[0062]

[0063] The number of maximum points Q that satisfy equation (7) is counted. Based on the number of maximum points Q, it is determined whether it is lightning interference. If Q = 1, it is determined to be lightning interference. If Q > 1, it is determined to be a fault and proceed to step 4.

[0064] Step 4: Extract the frequency of the point with the largest amplitude in the amplitude-frequency curve, define the frequency of the point with the largest amplitude as the main frequency F1, and compare the obtained main frequency F1 with the threshold value F set In comparison, if equation (8) is satisfied, it is judged as a grounding fault; otherwise, it is judged as a lightning strike fault.

[0065] F1 < F set (8)

[0066] The threshold value F in step 4 set The tuning method is as follows:

[0067] Simulations were conducted at 5km, 50km, 250km, 450km, and 495km along the line, considering different types of disturbances and ground faults, with a transition resistance of 500 ohms. The corresponding setting values ​​were calculated based on the simulation results. The simulation results are shown in Table 1.

[0068] Table 1 shows that when a ground fault occurs on a transmission line, the dominant frequency on the amplitude-frequency curve of the fault occurrence time obtained from the line-mode voltage reverse traveling wave within the time window is consistently above 1kHz. However, when a lightning strike fault occurs on the transmission line, considering the worst-case scenario, the dominant frequency F1 = 4kHz at 5km, falls between the ground fault dominant frequency of 1kHz and the lightning strike dominant frequency of 4kHz. Therefore, F1 is taken as... set =3kHz as the threshold value.

[0069] Table 1 Simulation verification results under different fault distances

[0070]

[0071] As attached Figure 2 The figure shown is a simulation model diagram of a double-ended flexible DC transmission system. In the figure, L... dc Line boundary current limiting reactor L dc= 0.2H, MMC is a modular converter, R1 and R2 are line boundary protection measurement points, f1, f2 and f3 are disturbance occurrence positions, L is the distance of the fault distance protection measurement point, the system rated voltage is ±500kV, the rated transmission capacity is 3000MVA, the total length of the transmission line is 500km, the overhead line frequency-varying parameter model is used, and the line wave impedances are Z c0 = 320Ω, Z c1 = 260Ω. In the system simulation, the sampling frequency is 100kHz, the 1.2 / 50μs standard lightning current model is used to simulate lightning, the threshold value F set = 3kHz is defined according to the simulation verification result. At this time, the total length L = 500km of the line is identified according to the identification process as shown in the accompanying drawing Figure 1 .

[0072] Embodiment 1

[0073] As shown in the accompanying drawing Figure 3 , lightning interference occurs at 100km away from the protection installation, the protection element is started, the voltage and current data of 100 sampling points before starting and 100 sampling points after starting are read, and the line mode voltage reverse wave is calculated. The line mode voltage reverse wave in the data window is subjected to S transformation to obtain a mode time frequency matrix, and the amplitude-frequency curve at the fault occurrence time in the mode time frequency matrix is as shown in the accompanying drawing Figure 3 , the maximum value quantity Q = 1 is determined as lightning interference, and the protection returns.

[0074] Embodiment 2

[0075] As shown in the accompanying drawing Figure 4 , a positive pole grounding fault occurs at 400km away from the protection installation through a 500Ω transition resistance, the protection element is started, the voltage and current data of 100 sampling points before starting and 100 sampling points after starting are read, and the line mode voltage reverse wave is calculated. The line mode voltage reverse wave in the data window is subjected to S transformation to obtain a mode time frequency matrix, and the amplitude-frequency curve at the fault occurrence time in the mode time frequency matrix is as shown in the accompanying drawing Figure 4 , the maximum value quantity Q > 1 is determined as a fault, the main frequency F1 is calculated, F1 = 1kHz is less than the setting threshold value F set = 3kHz, and the protection is determined as a grounding fault and is acted.

[0076] Embodiment 3

[0077] As shown in the accompanying drawing Figure 5 , a positive pole metallic short circuit grounding fault occurs at 200km away from the protection installation under the condition of 20dB noise interference, the protection element is started, the voltage and current data of 100 sampling points before starting and 100 sampling points after starting are read, and the line mode voltage reverse wave is calculated. The line mode voltage reverse wave in the data window is subjected to S transformation to obtain a mode time frequency matrix, and the amplitude-frequency curve at the fault occurrence time in the mode time frequency matrix is as shown in the accompanying drawingFigure 5 , the maximum number Q > 1, it is judged that the fault, the calculation of the main frequency F1; F1 = 1kHz, less than the setting threshold F set = 3kHz, it is determined that the ground fault, protection action.

[0078] Example 4

[0079] As Figure 6 shown, the lightning fault occurs 350km away from the protection installation, the protection element starts, reads the 100 sampling points before starting, the voltage and current data after starting, calculates the line mode voltage reverse wave. The line mode voltage reverse wave in the data window is S-transformed to obtain the mode time-frequency matrix, and the amplitude-frequency curve at the fault occurrence time in the mode time-frequency matrix is as Figure 6 , the maximum number Q > 1, it is judged that the fault, the calculation of the main frequency F1; F1 = 14kHz, F1 is greater than the setting threshold F set = 3kHz, it is determined that the lightning fault, protection action.

[0080] In order to comprehensively verify the influence of fault distance, different lightning current model, different transition resistance on the judgment result, the in-zone ground fault, out-of-zone ground fault, transition resistance of 0Ω, 100Ω, 500Ω are set at 50km, 100km, 200km, 300km, 400km, 450km respectively, and the influence of 2.6 / 50μs and 5 / 100μs lightning current model is considered, and the out-of-zone fault and lightning interference identification method is verified according to the simulation results. The verification results are shown in Tables 2-4 as follows:

[0081] Table 2 Simulation verification results under different disturbance distances

[0082]

[0083]

[0084] From Table 2, it can be obtained that the maximum number of ground fault and lightning fault is at least two, and the main frequency of lightning fault is greater than 10kHz, and the main frequency of ground fault is 1kHz. The results show that the method has high margin, and is less affected by fault distance.

[0085] Table 3 Simulation verification results under different transition resistances

[0086]

[0087] From Table 3, it can be obtained that the method can still accurately identify under different transition resistances, and the position of the main frequency is not affected by the transition resistance.

[0088] Table 4 Simulation verification results under different lightning current models

[0089]

[0090]

[0091] From table 4, it can be obtained that different disturbances can still be accurately distinguished under different lightning current models, and simulation verification results show that the scheme has high adaptability.

Claims

1. A method for distinguishing lightning disturbance of flexible HVDC transmission line based on instantaneous dominant frequency characteristics, characterized in that: Specifically comprising the following steps: Step 1, extracting voltage and current data in the time window from 1.0 ms before the protection starting to 1.0 ms after the protection starting, and obtaining line-mode voltage forward and reverse traveling waves; The specific process of the step 1 is: Step 1.1, after the protection starting element is started, taking 1.0 ms before starting and 1.0 ms after starting as the protection time window, reading the positive voltage of the protection installation in the time window , negative voltage , positive current , negative current , respectively, and the positive voltage average , negative voltage average , positive current average , negative current average before the protection is started 10 ms, subtracting to obtain the positive voltage fault component , negative voltage fault component , positive current fault component and negative current fault component , the calculation process is shown in formula (1): （1） Step 1.2, calculating the line mode voltage fault component within the time window , line mode current fault component The calculation process is shown in equation (2): （2） Step 1.3, calculate the line mode voltage backward wave in the time window , line mode voltage forward wave , the calculation process is shown in equation (3): （3） wherein Z0is the line mode wave impedance; Step 2, performing discrete S transform on the line-mode voltage reverse traveling wave obtained in the step 1 to obtain a mode time-frequency matrix, and extracting an amplitude-frequency curve at the fault occurrence moment; The specific process of the step 2 is: Step 2.1, counter traveling wave of line-mode voltage Discrete S-transform is performed to obtain a complex time-frequency matrix The formula for the discrete S-transform is given by equation (4): （4） In the formula, k For time sequence number, 0≤ k ≤ N -1; T is the sampling time interval; N This represents the number of sampling points; m For discrete time points, the value is: 0 ≤ m ≤ N -1; n The frequency index has a value of 1 ≤ n ≤fix( N / 2), fix refers to no more than N The largest integer that is 2 / 2; n / NT indicates the frequency serial number. n The corresponding frequency; Indicates: within the time window The discrete signal obtained by performing a Fourier transform on the data; Step 2.

2. Taking the modulus of each element in the complex time-frequency matrix from Step 2.1, resulting in a modulus time-frequency matrix , as shown in equation (5). ​ （5） In formula (5), denotes the matrix of the time-frequency representation the element in the i-th row and j-th column of the matrix a row b column Step 2.3, extracting from equation (5) the instantaneous amplitude-frequency curve at the moment , defining the moment as k 1 the moment, as shown in equation (6): （6） Step 3, find the maximum number of instantaneous amplitude-frequency curve Q , if Q= 1, it is judged as lightning interference, and the discrimination ends; if Q> 1, it is judged as fault, and enters step 4; the specific process of the step 3 is: Step 3.1, extracting the instantaneous amplitude-frequency curve of the maximum value in the amplitude-frequency curve, if a point n 1 in the amplitude-frequency curve is greater than the amplitude value corresponding to the previous sampling point and the amplitude value corresponding to the next sampling point, then n 1 is the maximum value point in the amplitude-frequency curve, as shown in equation (7): （7） wherein, n 1 ∈ [2, fix( N / 2)-1] ; Step 3.2, count the number of maximum points satisfying formula (7) Q , according to the number of maximum points Q to determine whether it is lightning interference, if Q =1, it is determined to be lightning interference, if Q >1, it is determined to be a fault, go to step 4; Step 4, define the frequency of the maximum amplitude point in the amplitude-frequency curve as the main frequency F 1, compare the main frequency F 1 with the threshold value F set , and determine the fault type according to the comparison result; the specific process of step 4 is: compare the main frequency F 1 with the threshold value F set , and determine the fault type according to the comparison result; the specific process of step 4 is: compare the main frequency F 1 with the threshold value F set , and determine the fault type according to the comparison result; the specific process of step 4 is: compare the main frequency F 1 with the threshold value F set , and determine the fault type according to the comparison result; the specific process of step 4 is: compare the main frequency F 1 with the threshold value F set , and determine the fault type according to the comparison result; the specific process of step 4 is: （8） wherein F set is a threshold value.

Images