Two-section fault distance protection method for multi-terminal flexible direct current system

Abstract

The invention discloses a two-section fault distance protection method for a multi-terminal flexible DC system, and the method comprises the steps: starting a criterion, carrying out the protection starting judgment, and carrying out the fault direction recognition; when a positive direction fault occurs, taking a cyclic wave with the lowest frequency in the line mode voltage and the noise as a time window, and decoupling the voltage in the time window to obtain the line mode voltage; obtaining MUSIC spectrums of an original noisy signal and a noisy signal of the line mode voltage, and respectively extracting spectrum peak frequencies; the inherent frequency is obtained through clustering so as to calculate the fault distance; and designing an action criterion of I-section protection and an action criterion of II-section protection, and determining one of four situations that a fault occurs in the I section, the fault occurs in the II section, the fault occurs in the II section and a downstream direct-current circuit breaker trips and a local direct-current circuit breaker trips according to the action criterion of the I-section protection and the action criterion of the II-section protection and the fault distance. Compared with the prior art, selective action of single-terminal protection of the multi-terminal flexible direct current system can be realized.

Description

Technical Field

[0001] The invention belongs to the field of relay protection of power systems, and in particular relates to a two-stage distance protection method for a multi-terminal flexible direct current system based on natural frequency ranging. Background Art

[0002] With the large-scale access of new energy sources such as wind power and photovoltaic power generation to the power grid, the scale and capacity of grid connection have increased significantly. It is becoming increasingly important to ensure the safety of modern power systems and connect various new energy sources to the power grid, which has put forward more requirements for relay protection.

[0003] The modular multi-level multi-terminal flexible direct current (MMC-HVDC) system has shown unique advantages in coping with large-scale renewable energy grid connection, realizing long-distance large-capacity power transmission, and promoting asynchronous power grid interconnection. It is an important trend in the development of future power systems. However, due to its low system damping, when a DC fault occurs, the fault current will increase rapidly, which requires the protection to complete fault identification and removal in a very short time to ensure the reliable operation of the remaining network. Single-terminal traveling wave protection does not require communication and synchronization, and has a fast action speed, which can just meet the requirements of multi-terminal flexible direct current system for speed. In actual engineering, due to the presence of current limiting reactors at both ends of the line, the traveling wave will have obvious boundary effects due to the influence of the current limiting reactor. This feature is the core basis for constructing a single-terminal traveling wave protection algorithm. Therefore, most of the existing single-terminal protection methods in multi-terminal flexible direct current systems are designed based on the boundary characteristics of current limiting reactors. However, with the increase of system scale and topological complexity, there may be no boundaries at both ends of the line, and the existing protection will no longer be applicable to borderless multi-terminal flexible direct current systems. However, in the existing protection methods applied to borderless multi-terminal flexible DC systems, there are problems such as low accuracy of natural frequency measurement and fault location, and large errors caused by false spectral peaks. Summary of the invention

[0004] In view of the problems disclosed in the above background technology, the present invention proposes a two-stage distance protection method for a multi-terminal flexible DC system, which realizes the accurate calculation of fault distance by natural frequency measurement in a multi-terminal flexible DC system for the first time.

[0005] In order to achieve the above-mentioned invention object, the present invention is implemented by the following technical solutions:

[0006] The present invention proposes a two-stage fault distance protection method for a multi-terminal flexible DC system, comprising:

[0007] Start judgment criteria, make protection start judgment, and identify fault direction;

[0008] When a positive fault occurs, the line mode voltage and the lowest frequency cycle of the noise are taken as the time window, and the voltage within the time window is decoupled to obtain the line mode voltage.

[0009] Obtain the MUSIC spectra of the original noisy signal and the noisy signal of the line mode voltage, and extract the spectrum peak frequencies respectively;

[0010] Cluster the frequencies of the original noisy signal and the noisy signal, and use the clustered frequencies as the natural frequencies;

[0011] If the main natural frequency is greater than the threshold value, the fault distance is calculated using the main natural frequency, otherwise, the second natural frequency is selected from the natural frequencies according to the value and the slope of the df curve, and the fault distance is calculated using the second natural frequency;

[0012] The action criteria of section I protection and section II protection are designed. The action criteria of section I protection and section II protection are used together with the fault distance to determine one of the four situations: the fault occurs in section I, the fault occurs in section II, the fault occurs in section II and the downstream DC circuit breaker trips, and the local DC circuit breaker trips.

[0013] In some implementations, performing protection start determination further includes:

[0014] Measuring the positive / negative DC voltage U of a multi-terminal flexible DC system dcp , U dcn and positive / negative DC current I dcp , U dcn ,

[0015] If the DC voltage change of any pole during the measurement cycle exceeds the set threshold, the expression is as follows:

[0016] |du / dt|>|du / dt| set

[0017] Where, u represents the positive / negative DC voltage U dcp , U dcn ;

[0018] It is determined that the protection has been activated.

[0019] In some implementations, identifying the fault direction further includes:

[0020] The forward and reverse traveling wave amplitude ratio is used as a criterion to determine the fault direction:

[0021]

[0022] Among them, Δu 正 , Δu 反 They represent the forward and reverse traveling waves, respectively, and K set is the threshold value for the traveling wave direction criterion,

[0023] If the above formula is satisfied, it is determined that a positive direction fault occurs; otherwise, it is determined that a reverse direction fault occurs.

[0024] In some embodiments, clustering the frequencies of the original noisy signal and the noisy signal further includes: using a DBSCAN clustering algorithm with a minimum threshold set to the total number of groups of the original noisy signal and the noisy signal and a domain radius set to 10% of the minimum spectral peak radius.

[0025] In some implementations, the expression for the fault distance is as follows:

[0026]

[0027] Where θ1 and θ2 are the phase angles of the reflection coefficients Г1 and Г2 at the converter station at the start of section I protection and the fault point, v is the transmission speed of the traveling wave on the line at different frequencies, and f is n is the nth natural frequency component of the line port voltage at the starting converter station, and k is an arbitrary natural number;

[0028] In some embodiments, the operation criteria of the section I protection and the section II protection are used together with the fault distance to determine whether the fault occurs in section I, the fault occurs in section II, the fault occurs in section II and the downstream DCCB trips, and the local DCCB trips. The following further includes:

[0029] The action criterion for designing the protection of stage I is d≤k I l1,t I = 0s, the action criterion of stage II protection is d≤k II l1,t II =t delay ; If d≤k I l1, then the fault is determined to occur in section I; otherwise, continue. If d≤k is not satisfied II l1, then the fault is judged to occur in section II. If d≤k II l1, continue to judge in time window t I +t II Is the residual voltage traveling wave of the downstream DCCB tripping detected? If detected, it is determined that the fault occurs in Section II and the downstream DCCB is tripped. If not detected, it is determined that the local DCCB is tripped.

[0030] Where l1 represents the length of protected line 1, k I Represents the reliability coefficient of stage I protection, k II Represents the reliability coefficient of stage II protection, t I Indicates the delay time of the I-stage action criterion, t II Indicates the delay time of the II stage action criterion, t delayIt indicates the time required for the residual voltage traveling wave generated by the DCCB tripping at the downstream line protection to reach the protection fault protection 1 in the event of a fault on the downstream line.

[0031] Compared with the prior art, the advantages and positive technical effects achieved by the present invention are as follows:

[0032] 1) Obtain the MUSIC spectra of the original noisy signal and the noisy signal of the line-mode voltage, and extract the spectrum peak frequencies respectively; cluster the frequencies of the original noisy signal and the noisy signal, and use the clustered frequencies as the natural frequencies. This design is based on the idea of ​​active random noise injection, thereby effectively eliminating the measurement frequency deviation problem caused by the false spectrum peak phenomenon of the traditional frequency measurement method;

[0033] 2) The action criteria of section I protection and section II protection are designed to determine one of the four situations: the fault occurs in section I, the fault occurs in section II, the fault occurs in section II and the downstream DC circuit breaker trips, and the local DC circuit breaker trips. Therefore, a two-stage single-ended distance protection scheme for the multi-terminal DC system is designed, and the overvoltage information of the DC circuit breaker tripping is used to achieve reliable coordination of upstream and downstream protections, realizing the selective action of the single-ended quantity protection of the multi-terminal flexible DC system when the line boundary is missing. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The present invention is a flow chart of a two-stage distance protection method for a multi-terminal flexible DC system;

[0035] Figure 2 Figure 2 shows the topological structure and frequency domain equivalent model of a multi-terminal flexible DC system: (a) basic topological structure, (b) frequency domain equivalent model of a fault circuit;

[0036] Figure 3 The voltage and its MUSIC spectrum containing different noises when the bipolar fault distance is d=90km. DETAILED DESCRIPTION

[0037] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0038] like Figure 1 As shown, a two-stage distance protection method based on natural frequency ranging for a multi-terminal flexible DC system of the present invention specifically includes the following steps:

[0039] Step 1: First, measure the positive / negative DC voltage U of the multi-terminal flexible DC system. dcp , U dcn and positive / negative DC current I dcp , U dcn , if the DC voltage change of any pole exceeds the set threshold during the measurement period, the expression is as follows:

[0040] |du / dt|>|du / dt| set (1)

[0041] If the protection is judged to have been activated, the startup timing is started and the process goes to step 2. Otherwise, the protection startup judgment is repeated. u represents the positive / negative DC voltage U dcp , U dcn ;

[0042] Step 2: Use the forward and reverse wave amplitude ratio as a criterion to determine the fault direction;

[0043]

[0044] If equation (2) is satisfied, it is determined that a positive fault has occurred and step 3 is executed; otherwise, it is determined that a reverse fault has occurred and the process is exited; where Δu 正 , Δu 反 They represent the forward and reverse traveling waves, respectively, and K set The threshold value for the traveling wave direction criterion is usually set to 1 to 2;

[0045] The expressions of forward and reverse traveling waves are as follows:

[0046]

[0047] Where Δu represents the fault component of the measured voltage, Δi represents the fault component of the measured current, and Z c It represents the characteristic impedance of the line, and usually takes a fixed value in the high frequency band to ensure the accuracy of the traveling wave extraction calculation;

[0048] Step 3: Decouple the voltage in time window t1: Decouple the line mode voltage from the voltage in the lowest frequency cycle of the noise into the line mode voltage U l , the expression is as follows:

[0049]

[0050] Time window t1 is the lowest frequency cycle of line mode voltage and noise;

[0051] Line mode voltage U lA group of random Gaussian white noises are injected to obtain a group of noisy signals x1′, x2′, etc., and the frequencies of the original noisy signal x and the group of noisy signals are measured respectively. The MUSIC spectrum (Multiple Signal Classification) of the noisy signal x and the group of noisy signals is obtained by using the MUSIC algorithm (spatial spectrum estimation algorithm), and the spectrum peak frequency is extracted; in the MUSIC spectrum, the frequency corresponding to the relatively stable spectrum peak is regarded as the real signal frequency, and the other relatively fluctuating spectrum peaks are regarded as false spectrum peaks, so the real spectrum peaks and false spectrum peaks of the MUSIC spectrum are obtained, which are collectively referred to as spectrum peaks;

[0052] Extract the peaks of the MUSIC spectra of the original noisy signal x and the noisy signals x1′, x2′, etc. of the line mode voltage, use the DBSCAN clustering algorithm to perform clustering according to the peak frequency values, and obtain a clustering result consisting of multiple groups of clustering results of the original noisy signal and the noisy signal. The clustering result is the clustering frequency of the original noisy signal and the noisy signal, and the clustering result is represented by category + frequency value;

[0053] The minimum threshold in the DBSCAN algorithm is set to the total number of groups of the original noisy signal and the noisy signal, and the domain radius is set to 10% of the minimum spectrum peak radius;

[0054] The frequency average value in each category is calculated according to the total clustering result, and a set of signal real frequencies are obtained from the frequency average value in each category as the natural frequencies f1, f2, etc. The average value of the natural frequencies f1 and f2 obtained when the fault point is at the tail end of the I-stage protection is used as the threshold value f set , the calculation formula of the main natural frequency f1 when the line boundary is short-circuited and open-circuited is as follows:

[0055]

[0056] Among them, l m represents the fault distance, v represents the speed of the traveling wave;

[0057] The main natural frequency is a unique attribute of the line at a certain fault distance and a certain line fault type, indicating that a certain fault has occurred, including short circuit fault and open circuit fault.

[0058] When the measured main natural frequency (for example, f1) is greater than the threshold value f set When the main natural frequency is substituted into formula (5) for fault distance measurement, the expression of fault distance is as follows:

[0059]

[0060] In the above formula, θ1 and θ2 are the phase angles of the reflection coefficients Г1 and Г2 at the converter station at the beginning of section I protection and the fault point, v is the transmission speed of the traveling wave on the line at different frequencies, and f is n is the nth natural frequency component of the line port voltage at the starting converter station, and k is an arbitrary natural number;

[0061] When the measured main natural frequency f1 is less than the frequency threshold value f set When the second natural frequency (e.g. f2) with a larger value and a smaller slope of the df curve is substituted into formula (6) for fault distance measurement, the df curve refers to the fault distance-frequency curve;

[0062] Step 4: Design the action criterion of stage I protection to be d≤k I l1,t I = 0s, the action criterion of stage II protection is d≤k II l1,t II =t delay ; If d≤k I l1, then the fault is determined to occur in section I; otherwise, continue. If d≤k is not satisfied II l1, then the fault is judged to occur in section II. If d≤k II l1, continue to judge in time window T1+t II Is the residual voltage traveling wave that trips the downstream DCCB detected? If detected, it is determined that the fault occurs in Section II and the downstream DCCB trips. If not detected, it is determined that the local DCCB trips, where T1 is the wave propagation time on the protected line and the operation time of the protection algorithm;

[0063] Where l1 represents the length of protected line 1, k I It represents the reliability coefficient of stage I protection, usually 0.8~0.9, k II It represents the reliability coefficient of the protection of stage II, usually ranging from 1.2 to 1.5, t I Indicates the delay time of the I-stage action criterion, t II Indicates the delay time of the II stage action criterion, t delay In the case of a fault in the downstream line, it is the time required for the residual voltage traveling wave generated by the tripping of the DCCB at the downstream line protection to reach the protection fault protection 1 (P1), t II =t delay

[0064] DCCB is a direct current circuit breaker. The main reasons for DCCB tripping include voltage swell, voltage sag, short-term power outage and voltage flicker.

[0065] like Figure 2As shown in the figure, the I section protection is the area where the first section protection is activated, which is generally 0.8 to 0.9 of the length of the protected line, that is, the first section protection will be used when the first 0.8 to 0.9 of Line 1 is faulty; the II section protection means that the second section protection will be used when the fault is in the area between Line 1 and part of Line 2. Figure 2 The shaded area of ​​Line 1 and Line 2 in the figure. The second protection segment includes the first protection segment and is slower than the first protection segment.

[0066] As shown in the figure, it includes a startup distance determination module, a fault direction determination module, a frequency domain distance measurement module and a two-stage protection module, among which:

[0067] Start distance judgment module, used to start judgment criteria and make protection start judgment.

[0068] Fault direction identification module, to identify the fault direction;

[0069] The frequency domain distance measurement module is used to decouple the voltage within the time window of the line mode voltage and the lowest frequency cycle of the noise when a positive fault occurs, and obtain the line mode voltage;

[0070] Obtain the MUSIC spectra of the original noisy signal and the noisy signal of the line mode voltage, and extract the spectrum peak frequencies respectively;

[0071] Cluster the frequencies of the original noisy signal and the noisy signal, and use the clustered frequencies as the natural frequencies;

[0072] If the main natural frequency is greater than the threshold value, the fault distance is calculated using the main natural frequency, otherwise, the second natural frequency is selected from the natural frequencies according to the value and the slope of the df curve, and the fault distance is calculated using the second natural frequency;

[0073] The two-stage protection judgment module is used to design the action criteria of the stage I protection and the stage II protection. The action criteria of the stage I protection and the stage II protection are respectively used with the fault distance to judge one of the four situations: the fault occurs in the stage I, the fault occurs in the stage II, the fault occurs in the stage II and the downstream DC circuit breaker trips, and the local DC circuit breaker trips.

[0074] The above data and process can be used for simulation verification:

[0075] In PSCAD / EMTDC, Figure 2 In the three-terminal MMC-HVDC system shown, both line 1 and line 2 are DC cables, and a frequency-variable parameter model is used. The length is set to 80 km and the sampling rate is 20 kHz. Other key parameters are shown in Table 1. Reliability coefficient k I and k IIThey are set to 0.85 and 1.2 respectively. Therefore, the action criterion for section I is d≤68km, and the action criterion for section II is d≤96km. The delay time t1 of the protection calculation time window and the action criterion for section II is II Set to 3ms and 6ms respectively, f set Set to 1500Hz.

[0076] Take the case where a short circuit occurs between two poles of the protected line 2 at a distance of 10 km from the converter station S2 as an example. For protection P1, the fault distance d is 90 km and the fault occurs in section II. First, obtain the line mode voltage U at protection P1. l And inject five groups of noise line mode voltage U with different signal-to-noise ratios of 40dB, 50dB, 60dB and 70dB l1 ~U l5 ,like Figure 3 (a). Then, the MUSIC algorithm is used to calculate the original line mode voltage U l And five groups of noise-added line mode voltage U l1 ~U l5 Perform frequency measurement, six groups of MUSIC spectrum diagrams are shown as follows Figure 3 (b) as shown.

[0077] like Figure 3 As shown in (b), the six groups of MUSIC spectrum remain relatively stable at some peak frequencies, with a high distribution density. These frequencies correspond to real frequencies, and some relatively scattered peak frequencies correspond to false frequencies. The DBSCAN clustering algorithm can be clustered according to the density of the peak frequencies. Therefore, the DBSCAN clustering algorithm is used to obtain all the peak frequencies as clustering objects, the minimum threshold is set to the number of signal groups, that is, 6, and the domain radius is set to 10% of the minimum peak frequency. The clustering results are as follows: Figure 3 As shown in the bottom projection of (b), different symbols represent different cluster categories, and black empty circles represent noise points. The average value of the peak frequency in each cluster category is regarded as the final natural frequency value. It can be calculated that when d = 90km, the first two natural frequencies are 535Hz and 1600Hz respectively, and the main natural frequency is less than f set 1500Hz, so the second natural frequency 1600Hz is substituted into the distance measurement result of formula (2), which is 89.965km, and the distance measurement error is -0.039%. After judging the two-stage distance protection criteria, the fault protection area is identified as stage II, which is correct. II = Within 6ms, if protection P1 can detect the residual voltage traveling wave caused by the tripping of DC circuit breaker CB3 at the downstream line protection P3, protection P1 will return.

[0078] It should be noted that, although the technical solutions have been shown and described with reference to specific exemplary embodiments of the present invention, those skilled in the art should understand that the present invention is not limited to the above-mentioned implementation modes, and any replacements, improvements and supplements made without departing from the spirit of the present invention and without creative labor shall fall within the scope of protection of the present invention.

[0079] The above implementation steps are only used to help understand the specific methods and core ideas of the present invention, but do not limit the present invention in any way. Any changes and technical substitutions made by those skilled in the art in accordance with the protection scope of this application and the teachings and inspirations of the inventive concept without departing from the spirit of the present invention and without making creative work, all fall within the protection scope of the present invention.

[0080] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

Claims

1. A two-stage fault distance protection method for a multi-terminal flexible DC system, characterized in that: include: Start judgment criteria, make protection start judgment, and identify fault direction; When a positive fault occurs, the line mode voltage and the lowest frequency cycle of the noise are taken as the time window, and the voltage within the time window is decoupled to obtain the line mode voltage. Obtain the MUSIC spectra of the original noisy signal and the noisy signal of the line mode voltage, and extract the spectrum peak frequencies respectively; Cluster the frequencies of the original noisy signal and the noisy signal, and use the clustered frequencies as the natural frequencies; Calculate fault distance using natural frequency; The action criteria of section I protection and section II protection are designed. The action criteria of section I protection and section II protection are used together with the fault distance to determine one of the four situations: the fault occurs in section I, the fault occurs in section II, the fault occurs in section II and the downstream DC circuit breaker trips, and the local DC circuit breaker trips.

2. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1 is characterized in that: The protection start judgment further includes: Measuring the positive / negative DC voltage U of a multi-terminal flexible DC system dcp , U dcn and positive / negative DC current I dcp , U dcn , If the DC voltage change of any pole during the measurement cycle exceeds the set threshold, the expression is as follows: |you / dt|>|you / dt| set Where, u represents the positive / negative DC voltage U dcp , U dcn ; It is determined that the protection has been activated.

3. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1 is characterized in that: The fault direction identification further includes: The forward and reverse traveling wave amplitude ratio is used as a criterion to determine the fault direction: Among them, Δu 正 , Δu 反 They represent the forward and reverse traveling waves, respectively, and K set is the threshold value for the traveling wave direction criterion, If the above formula is satisfied, it is determined that a positive direction fault occurs; otherwise, it is determined that a reverse direction fault occurs.

4. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1, characterized in that: Clustering the original noisy signal and the noisy signal further includes: using a DBSCAN clustering algorithm with a minimum threshold set to the total number of groups of the original noisy signal and the noisy signal and a domain radius set to 10% of the minimum spectrum peak radius.

5. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1, characterized in that: The expression of fault distance is as follows: In the above formula, θ1 and θ2 are the phase angles of the reflection coefficients Г1 and Г2 at the converter station at the beginning of section I protection and the fault point, v is the transmission speed of the traveling wave on the line at different frequencies, and f is n is the nth natural frequency component of the line port voltage at the starting converter station, and k is an arbitrary natural number.

6. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1, characterized in that: Calculating fault distance using natural frequency includes: If the main natural frequency is greater than the threshold value, the fault distance is calculated using the main natural frequency. Otherwise, the second natural frequency is selected from the natural frequencies according to the value and the slope of the df curve, and the fault distance is calculated using the main natural frequency.

7. The two-stage fault distance protection method for a multi-terminal flexible DC system according to claim 1, characterized in that: The operation criteria of the section I protection and the section II protection are respectively used with the fault distance to determine that the fault occurs in section I, the fault occurs in section II, the fault occurs in section II and the downstream DCCB trips, and the local DCCB trips. One of the four situations further includes: The action criterion for designing the protection of stage I is d≤k I l1,t I = 0s, the action criterion of stage II protection is d≤k II l1,t II =t delay ; If d≤k I l1, then the fault is determined to occur in section I; otherwise, continue. If d≤k is not satisfied II l1, then the fault is judged to occur in section II. If d≤k II l1, continue to judge in time window T1+t II Is the residual voltage traveling wave that trips the downstream DCCB detected? If detected, it is determined that the fault occurs in Section II and the downstream DCCB trips. If not detected, it is determined that the local DCCB trips, where T1 is the wave propagation time on the protected line and the operation time of the protection algorithm; Where l1 represents the length of the protected line 1, k I Represents the reliability coefficient of stage I protection, k II Represents the reliability coefficient of stage II protection, t I Indicates the delay time of the I-stage action criterion, t II Indicates the delay time of the II stage action criterion, t delay It indicates the time required for the residual voltage traveling wave generated by the DCCB tripping at the downstream line protection to reach the protection fault protection 1 in the event of a fault on the downstream line.

Images