High-efficiency fault location method and device for double-H connection high-voltage filter capacitor bank

Abstract

The present invention discloses an efficient fault location method and device for a double-H connection high-voltage filter capacitor bank, belonging to the field of direct current transmission technology. The method comprises the following steps: collecting waveform data of two cycles before and after a fault in a double-H connection high-voltage filter capacitor bank; calculating the threshold for breakdown of a capacitor element, and finding the sampling point corresponding to the time when the ratio of the unbalanced current to the total current is greater than the threshold as the disturbance start time; performing first-layer wavelet decomposition on the branch current difference signals at different positions of the capacitor bank, and calculating the energy value based on the first-layer detail coefficients; obtaining the energy value threshold based on the mean and standard deviation of the first-layer coefficient energy value sequence; locating the fault to the 1 / 3 region based on the magnitude of the energy value and the energy value threshold; and locating the fault to the 1 / 6 region based on the sudden change direction of the pulse current difference at the disturbance start time, thereby obtaining the bridge arm where the fault is located. The present invention can quickly capture the characteristic signals generated by the fault, thereby quickly locating the faulty bridge arm.

Description

Technical Field

[0001] The present invention relates to the field of ultra-high voltage direct current (UHV) transmission of power systems, and in particular to a high-efficiency fault locating method and device for a double-H connection high-voltage filter capacitor bank. Background Art

[0002] In actual projects, the fault location and maintenance work of the double-H connection high-voltage filter capacitor group in the converter station mainly relies on manual operation. The technicians use a digital megohmmeter to test the capacitance value of each bridge arm capacitor group one by one, and determine the specific bridge arm location where the fault occurred by comparing the capacitance values obtained by the test, and then determine the faulty unit. This method requires technicians to use a digital megohmmeter to test the capacitance value of each bridge arm capacitor group one by one, which is a time-consuming process. Especially when the number of capacitor groups is large, the detection work becomes very cumbersome, resulting in low overall efficiency of fault location. In addition, due to the energy storage characteristics of capacitors, if the capacitors are not fully discharged before detection, residual charge may be released during the detection process, posing a threat to the personal safety of inspection personnel.

[0003] To improve maintenance efficiency and safety, existing methods propose adding a current transformer to the high-voltage side of a double H-bridge. This method determines the fault location by obtaining a quantitative relationship between the steady-state current before and after a fault. This method relies on the fact that the steady-state current changes before and after a fault, and uses this quantitative relationship to infer the fault location. However, since this method requires adding a current transformer to the high-voltage side of the double H-bridge, this not only involves the purchase cost of the current transformer itself, but also the costs of installation, commissioning, and subsequent maintenance, significantly increasing the operation and maintenance costs of the entire system. Furthermore, since high-voltage environments are inherently dangerous, adding equipment further increases the complexity and uncertainty of the system, thereby introducing certain safety risks. Furthermore, capacitor banks typically consist of multiple power capacitors, and the change in steady-state current before and after a fault is relatively small. In actual operation, factors such as system noise and measurement errors of the current transformers themselves can easily interfere with this small change, resulting in inaccurate fault location.

[0004] Power system operation experience shows that identifiable early disturbance characteristics often appear before a serious equipment failure occurs. When electrical equipment is operating, its voltage and current waveform signals often produce transient disturbances with sudden, periodic, and reversible characteristics. Compared with steady-state signals, which tend to stabilize after a fault, have weak characteristics, and contain limited information, transient disturbance signals have more prominent variation characteristics and richer fault information, and are of great research value in the field of fault diagnosis. Based on the current research status, this proposal specifically proposes an efficient fault location method and device for double-H connection high-voltage filter capacitor banks. Summary of the Invention

[0005] In order to solve the problem of rapid fault location of high-voltage filter capacitor banks with double H-bridge connections, this solution proposes an efficient fault location method and device for double H-bridge high-voltage filter capacitor banks. This method can quickly capture the characteristic signals generated by the fault, thereby quickly locating the position of the faulty bridge arm, greatly shortening the time required for fault location and significantly improving the efficiency of maintenance work.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution of the present invention is as follows:

[0007] An efficient fault location method for a double-H connection high-voltage filter capacitor bank includes the following steps:

[0008] Collect waveform data of two cycles before and after the double-H connection high-voltage filter capacitor bank fault. The waveform data includes the capacitor bank terminal voltage, total current, two unbalanced currents, and low-voltage terminal current.

[0009] The threshold for breaking down a capacitor element is calculated based on the capacitor bank parameters, and the sampling point corresponding to the time when the ratio of the unbalanced current to the total current is greater than the threshold is found as the disturbance start time;

[0010] Perform the first-layer wavelet decomposition on the branch current difference signals of the upper, middle and lower parallel bridge arms of the capacitor bank respectively to obtain the first-layer detail coefficients corresponding to each, and then obtain the corresponding energy values according to the first-layer detail coefficients;

[0011] The initial data of the three branch current difference signals are grouped, and the energy of the first-layer detail coefficient of each group of data is calculated to form their own energy value sequences. Then, the energy value thresholds of each energy value sequence are obtained based on the mean and standard deviation of the energy value sequence.

[0012] The fault is located in the 1 / 3 area according to the size of the energy value and the energy value threshold; then the fault is located in the 1 / 6 area according to the sudden change direction of the pulse current difference at the starting moment of the disturbance, and the bridge arm where the fault is located is obtained.

[0013] Preferably, the threshold value for breaking down a capacitor element is calculated based on the capacitor bank parameters, and the calculation formula is as follows:

[0014] ;

[0015] Where, is the breakdown threshold of a capacitor element, is the number of series sections within a single power capacitor unit, is the number of power capacitors connected in series in a single bridge arm.

[0016] Preferably, the first-layer wavelet decomposition is performed on the branch current difference signals of the upper, middle and lower parallel bridge arms of the capacitor bank respectively to obtain the first-layer detail coefficients corresponding to each of them. The calculation formula is as follows:

[0017] ;

[0018] Where, represents the first layer detail coefficient, n is the index of the detail coefficient, n =0,1,2,…, N / 2−1, N is the total number of sampling points, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n+1.

[0019] Preferably, the energy values corresponding to the first layer detail coefficients are obtained by the following calculation formula:

[0020] ;

[0021] Represents the energy value of the first layer detail coefficient, It represents the decomposition of the current difference signal of a parallel bridge arm branch. A detail factor.

[0022] Preferably, the energy value thresholds are obtained according to the mean value and standard deviation of the energy value sequence, and the calculation formula is as follows:

[0023] ;

[0024] Where, Indicates the energy value threshold, is the standard deviation of the energy value sequence, is the average value of the energy value sequence.

[0025] Preferably, the fault is located in the 1 / 3 area based on the size of the energy value and the energy value threshold, including: if the energy value of the first-layer detail coefficient of a parallel bridge arm is greater than its corresponding energy value threshold, and the energy values of the first-layer detail coefficients of the other two parallel bridge arms are less than or equal to the corresponding energy value threshold, then it is determined that the parallel bridge arm that is greater than the corresponding energy value threshold has a fault.

[0026] Preferably, the method of locating the fault to the 1 / 6 area according to the sudden change direction of the pulse current difference at the start of the disturbance includes: obtaining the average value of the current difference signal of the parallel bridge arm corresponding to the 1 / 3 area in the two cycles before the moment If The current difference at the time point is greater than the average value , then the bridge arm where the current suddenly increases is the fault bridge arm.

[0027] The present invention also proposes an efficient fault locating device for a double-H connection high-voltage filter capacitor bank, which is characterized by comprising:

[0028] The data acquisition module is configured to collect waveform data of two cycles before and after the double-H connection high-voltage filter capacitor bank fails, the waveform data including the capacitor bank terminal voltage, total current, two unbalanced currents and low-voltage terminal current;

[0029] The first data processing module is configured to calculate a threshold value for breakdown of a capacitor element according to the capacitor bank parameters, and find the sampling point corresponding to the time when the ratio of the unbalanced current to the total current is greater than the threshold value as the disturbance start time;

[0030] The second data processing module is configured to perform first-layer wavelet decomposition on the branch current difference signals of the upper, middle, and lower parallel bridge arms of the capacitor bank, respectively, to obtain the first-layer detail coefficients corresponding to each of the branch current difference signals, and then obtain the energy value of the first-layer detail coefficients;

[0031] The third data processing module is configured to group the initially collected data of the three branch current difference signals, calculate the first-layer detail coefficient energy of each group of data, and form respective energy value sequences; and then calculate respective energy value thresholds based on the mean and standard deviation of the energy value sequences;

[0032] The judgment output module is configured to locate the fault to the 1 / 3 area based on the size of the energy value and the energy value threshold; then locate the fault to the 1 / 6 area based on the sudden change direction of the pulse current difference at the starting moment of the disturbance, and obtain the bridge arm where the fault is located.

[0033] Preferably, the judgment output module locates the fault to the 1 / 3 area based on the size of the energy value and the energy value threshold, specifically including: if the energy value of the first-layer detail coefficient of a parallel bridge arm is greater than its corresponding energy value threshold, and the energy values of the first-layer detail coefficients of the other two parallel bridge arms are less than or equal to the corresponding energy value threshold, then it is determined that the parallel bridge arm that is greater than the corresponding energy value threshold has a fault.

[0034] Preferably, the judgment output module locates the fault to the 1 / 6 area according to the sudden change direction of the pulse current difference at the start of the disturbance, including: obtaining the current difference signal of the parallel bridge arm corresponding to the 1 / 3 area The average of the two waves before the moment ; If the current difference at the time point is greater than the average value , then the bridge arm where the current suddenly increases is the fault bridge arm.

[0035] In summary, the present invention has the following advantages:

[0036] 1. This solution is based on the capture of transient disturbance signals and uses the pulse current generated during a fault to locate the fault. The pulse signal is relatively obvious, and there is no need to test each bridge arm capacitor bank one by one. It can quickly capture the characteristic signal generated by the fault, thereby quickly locating the faulty bridge arm. This greatly shortens the time required for fault location and significantly improves the efficiency of maintenance work.

[0037] 2. This solution adds a current transformer at the low-voltage end to detect pulse current, avoiding direct contact detection with the capacitor, reducing the safety risks caused by the release of residual charge in the capacitor, and providing a safer working environment for maintenance personnel.

[0038] 3. This solution adds a current transformer at the low-voltage end. The procurement, installation and maintenance costs of the low-voltage equipment are relatively low, which can effectively reduce the overall operation and maintenance costs of the system.

[0039] 4. The implementation of this solution can be operated at the low-voltage end. The low-voltage environment is relatively safe and stable. Adding current transformers will not significantly increase the complexity and uncertainty of the system, reducing the safety hazards caused by the increase in equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Flow chart of the implementation of the present invention;

[0041] Figure 2 This is a schematic diagram of the wiring and internal structure of a double-H connection high-voltage filter capacitor bank. DETAILED DESCRIPTION

[0042] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and drawings. Those skilled in the art should understand that the following specific description is illustrative rather than restrictive and should not be used to limit the scope of protection of the present invention.

[0043] Example 1

[0044] The embodiment of the present invention provides an efficient fault location method for a double-H connection high-voltage filter capacitor bank. Assuming that each arm of the double-H bridge consists of N The capacitor units are connected in series, and the capacitance of the capacitor unit is C u , the capacitor unit has a total of m × n Only components, first by m Only components are connected in parallel to form a parallel module, and then n Parallel modules are connected in series, such as Figure 2 As shown in .

[0045] like Figure 1 As shown, this method specifically includes the following steps:

[0046] Step 1: Collect the waveform data of two cycles before and after the double H connection high voltage filter capacitor bank fault, including: capacitor bank terminal voltage u (t), total current i (t), unbalanced current of upper H bridge i u0 (t), unbalanced current of lower H bridge i d0 (t) and low voltage terminal current i 5(t).

[0047] Combined with attachment Figure 2 , in actual engineering, the total voltage can be directly obtained , total current , unbalanced current of upper H bridge and the unbalanced current of the lower H-bridge Waveform data of low voltage end current. The current transformer added to the branch where C5 is located is used to measure the low-voltage end current (the current transformer added to the branch where C6 is located can also be used to measure the low-voltage end current). ).

[0048] Step 2: According to the parameters of the capacitor bank and Calculate the breakdown threshold η of a capacitor element and find i u0 (t) / i (t)>η or i d0 (t) / i The sampling point where (t)>η is taken as the starting point of the disturbance, and the corresponding time t 0 is the starting time of the disturbance.

[0049] For details, please refer to the attached Figure 2 , this step is specifically performed as follows:

[0050] According to Kirchhoff's current law, the branch currents flowing through the C6, C4, C3, C2 and C1 arms of the capacitor bank are: , , , , .

[0051] Calculate the breakdown threshold η of a capacitor element based on the parameters of the capacitor bank:

[0052] ;(1)

[0053] Where, The breakdown threshold of a capacitor element, is the number of series sections within a single power capacitor unit, is the number of power capacitors connected in series in a single bridge arm.

[0054] Finally, find or The sampling point is taken as the starting point of the disturbance. is time (in ms), then the corresponding time is the starting time of the disturbance.

[0055] Step 3: Use Haar wavelet to perform first-layer wavelet decomposition on the branch current difference signals of the upper, middle and lower parallel bridge arms of the capacitor bank, obtain the first-layer detail coefficients corresponding to each, and then calculate the corresponding energy values according to the first-layer detail coefficients.

[0056] The branch current difference signals of the upper, middle and lower parallel bridge arms of the double H-bridge are 、 and Discrete wavelet transform (DWT) is a discrete form of wavelet transform, which is suitable for the analysis of digital signals. In this embodiment, Haar wavelet is used to analyze the starting moment of the disturbance. 1ms before and after (i.e. ) current difference signal 、 and Decompose each scale (first layer) to obtain detail coefficients (also called high-frequency coefficients); 、 、 、 、 and are the branch currents flowing through the C1, C2, C3, C4, C5 and C6 bridge arms respectively.

[0057] by For example, let ,in k is the sampling time point, k =0,1,2,…, N −1; N The total number of sampling points within 2ms.

[0058] The current difference signal The corresponding detail coefficient It can be calculated by the following formula (2):

[0059] ; (2)

[0060] in, n is the index of the detail coefficient, n =0,1,2,…, N / 2−1. is the branch current difference between the parallel bridge arms C1 and C2 when the sampling point is 2n, is the branch current difference between the parallel bridge arms C1 and C2 corresponding to the sampling point 2n+1.

[0061] Current difference signal The energy of the first layer detail coefficient is calculated according to the following formula (3):

[0062] ; (3)

[0063] Similarly, the current difference signal can be calculated according to the above steps and The first layer detail coefficient energy and .

[0064] Step 4: Group the initial data of the current difference signals at different locations (i.e., the data of two cycles before and after the fault), calculate the energy of the first-layer detail coefficient of each group of data, and form their own energy value sequences. For each energy value sequence, calculate its mean and standard deviation to obtain the corresponding energy value threshold. Based on the threshold judgment, locate the fault to the 1 / 3 area.

[0065] The specific operation process is as follows:

[0066] First, the current difference signal 、 and The initial data is divided into 40 groups, and the total number of sampling points in each group is N , N is the total number of sampling points within 2ms.

[0067] Secondly, the energy of the first-layer detail coefficients of each set of data is calculated according to formula (2) and formula (3) to form the energy value sequence of the first-layer detail coefficients. For example, the energy value sequence of the first layer detail coefficient is .

[0068] Then, the energy value sequence is calculated according to formula (4): The mean :

[0069] ; (4)

[0070] The energy value sequence is calculated according to the following formula (5): Standard deviation :

[0071] ; (5)

[0072] Then calculate the energy value threshold according to the following formula (6): :

[0073] ; (6)

[0074] Similarly, according to the above steps, we can calculate and The corresponding energy value threshold and .

[0075] Based on the first-layer detail coefficient energy of each current difference signal and its respective energy value threshold, the following judgment is made:

[0076] like and It is determined that there is a fault in bridge arm C1 or C2;

[0077] like and It is determined that there is a fault in bridge arm C3 or C4;

[0078] like It is determined that there is a fault in bridge arm C5 or C6.

[0079] When a capacitor element breakdown occurs in a capacitor bank, the normal elements in parallel and other bridge arm capacitor banks will quickly release charge to the fault point, forming a high-frequency, large-amplitude, and rapidly decaying transient discharge current. This transient process will show a significant pulse feature in the waveform of the current difference between the upper, middle, and lower end bridge arm branches of the double H-bridge. and The first-layer detail coefficient energy will increase significantly. The relevant threshold is calculated based on the standard deviation of the first-layer detail coefficient energy of the branch current difference before the fault. If a fault occurs in the bridge arm, the first-layer detail coefficient energy of the corresponding branch current difference will exceed the threshold and be identified, and the fault is located in the 1 / 3 range (that is, it is determined whether the fault occurs in the upper, middle, or lower bridge arm of the double H-bridge).

[0080] Step 5: Locate the fault to the 1 / 6 area according to the sudden change direction of the pulse current difference at the start of the disturbance, and obtain the bridge arm where the fault is located.

[0081] Assuming that according to step 4, it is determined that the bridge arm C1 or C2 is faulty, the specific operations are as follows:

[0082] First, calculate the current difference signal according to the following formula (7): exist The average of the two waves before the moment :

[0083] ; (7)

[0084] in, N is the total number of sampling points within 2ms.

[0085] Then, based on the current difference signal exist The value at the moment and The average of the two waves before the moment , make the following judgment:

[0086] like , it is determined that there is a fault in bridge arm C1;

[0087] like , it is determined that there is a fault in bridge arm C2.

[0088] Similarly, according to the above steps, we can calculate and The corresponding average and , and make the following judgments:

[0089] If it is determined according to step 4 that there is a fault in bridge arm C3 or C4, and , it is determined that there is a fault in the bridge arm C3, otherwise it is determined that there is a fault in the bridge arm C4;

[0090] If it is determined according to step 4 that there is a fault in bridge arm C5 or C6, and , it is determined that there is a fault in the bridge arm C5, otherwise it is determined that there is a fault in the bridge arm C6.

[0091] According to Kirchhoff's current law (KCL), when a component breakdown occurs, the capacitor banks in other arms connected in parallel with the faulty bridge arm branch will discharge into the broken component. Therefore, the direction of the pulse mutation of the current difference waveform of the upper, middle, and lower bridge arm branches of the double H-bridge depends on the location of the faulty bridge arm. Based on this, the fault location can be determined based on the positive or negative value of the bridge arm branch current difference at the moment of the fault. To eliminate the influence of the initial current difference, the steady-state one-cycle average of the bridge arm branch current difference before the fault is used as the positioning threshold, which can locate the fault to within the 1 / 6 range.

[0092] Example 2

[0093] Based on the same inventive concept, this embodiment provides an efficient fault locating device for a double-H connection high-voltage filter capacitor bank, including a data acquisition module, a first data processing module, a second data processing module, a third data processing module, and a judgment output module.

[0094] The data acquisition module is configured to collect waveform data of two cycles before and after the double-H connection high-voltage filter capacitor bank fault, the waveform data including the capacitor bank terminal voltage, total current, two unbalanced currents and low-voltage terminal current;

[0095] The first data processing module is configured to calculate a threshold value for breaking down a capacitor element according to the capacitor bank parameters, and find a sampling point corresponding to a time when the ratio of the unbalanced current to the total current is greater than the threshold value as the disturbance start time;

[0096] The second data processing module is configured to perform first-layer wavelet decomposition on the branch current difference signals of the upper, middle, and lower parallel bridge arms of the capacitor bank, respectively, to obtain the first-layer detail coefficients corresponding to each, and then obtain the energy values corresponding to each according to the first-layer detail coefficients;

[0097] The third data processing module is configured to group the initially collected data of the three branch current difference signals, calculate the first-layer detail coefficient energy of each group of data, and form respective energy value sequences; and then calculate respective energy value thresholds based on the mean and standard deviation of the energy value sequences;

[0098] The judgment output module is configured to locate the fault to the 1 / 3 area based on the size of the energy value and the energy value threshold; then locate the fault to the 1 / 6 area based on the sudden change direction of the pulse current difference at the starting moment of the disturbance, and obtain the bridge arm where the fault is located.

[0099] Preferably, the first data processing module breaks down a threshold value of a capacitor element according to the capacitor bank parameters. , the calculation formula is as follows:

[0100] ;

[0101] In the formula , is the threshold value for breaking down a capacitor element, is the number of series sections within a single power capacitor unit, is the number of power capacitors connected in series in a single bridge arm.

[0102] Preferably, the second data processing module performs a first-layer wavelet decomposition on the branch current difference signals of the upper, middle, and lower parallel bridge arms of the capacitor bank to obtain the first-layer detail coefficients corresponding to each of them. The calculation formula is as follows:

[0103] ;

[0104] Where, represents the first layer detail coefficient, is the index of the detail coefficient, n =0,1,2,…, N / 2−1, N is the total number of sampling points, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n+1.

[0105] The second data processing module obtains the corresponding energy values according to the first layer detail coefficients. The calculation formula is as follows:

[0106] ;

[0107] Represents the energy value of the first layer detail coefficient, It represents the decomposition of the current difference signal of a parallel bridge arm branch. A detail factor.

[0108] Preferably, the third data processing module calculates the respective energy value thresholds according to the mean value and standard deviation of the energy value sequence, and the calculation formula is as follows:

[0109] ;

[0110] Where, Indicates the energy value threshold, is the standard deviation of the energy value sequence, is the average value of the energy value sequence.

[0111] Furthermore, the judgment output module locates the fault to the 1 / 3 area based on the judgment of the energy value and the energy value threshold, including: if the energy value of the first-layer detail coefficient of a parallel bridge arm is greater than its corresponding energy value threshold, and the energy values of the first-layer detail coefficients of the other two parallel bridge arms are less than or equal to the corresponding energy value threshold, then it is determined that the parallel bridge arm greater than the corresponding energy value threshold has a fault.

[0112] Furthermore, the judgment output module locates the fault to the 1 / 6 area according to the sudden change direction of the pulse current difference at the start of the disturbance, including: obtaining the current difference signal of the parallel bridge arm corresponding to the 1 / 3 area The average of the two waves before the moment ; If the current difference at the time point is greater than the average value , then the bridge arm where the current suddenly increases is the fault bridge arm.

[0113] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. An efficient fault location method for a double-H connection high-voltage filter capacitor bank, characterized in that: The steps include: Collect waveform data of two cycles before and after the double-H connection high-voltage filter capacitor bank fault. The waveform data includes the capacitor bank terminal voltage, total current, two unbalanced currents, and low-voltage terminal current. The threshold for breaking down a capacitor element is calculated based on the parameters of the capacitor bank, and the sampling point corresponding to the time when the ratio of the unbalanced current to the total current is greater than the threshold is found as the disturbance start time; Perform the first-layer wavelet decomposition on the branch current difference signals of the upper, middle and lower parallel bridge arms of the capacitor bank respectively to obtain the first-layer detail coefficients corresponding to each, and then obtain the corresponding energy values according to the first-layer detail coefficients; The initial data of the three branch current difference signals are grouped, and the energy of the first-layer detail coefficient of each group of data is calculated to form their own energy value sequences. Then, the energy value thresholds of each energy value sequence are obtained based on the mean and standard deviation of the energy value sequence. The fault is located in the 1 / 3 area based on the difference between the energy value and the energy value threshold, including: if the energy value of the first-layer detail coefficient of a parallel bridge arm is greater than its corresponding energy value threshold, and the energy values of the first-layer detail coefficients of the other two parallel bridge arms are less than or equal to the corresponding energy value threshold, then it is determined that the parallel bridge arm greater than the corresponding energy value threshold has a fault; Then, according to the sudden change direction of the pulse current difference at the start of the disturbance, the fault is located in the 1 / 6 area, and the bridge arm where the fault is located is obtained, including: obtaining the average value of the current difference signal of the parallel bridge arm corresponding to the 1 / 3 area for two cycles before the start of the disturbance ; If the current difference at the disturbance starting point is greater than the average value , then the bridge arm where the current suddenly increases is the fault bridge arm.

2. The efficient fault location method for a double-H connection high-voltage filter capacitor bank according to claim 1, characterized in that: The calculation formula for calculating the breakdown threshold of a capacitor element based on the capacitor bank parameters is as follows: ； Where, is the breakdown threshold of a capacitor element, is the number of series sections within a single power capacitor unit, is the number of power capacitors connected in series in a single bridge arm.

3. The efficient fault location method for a double-H connection high-voltage filter capacitor bank according to claim 1, characterized in that: Perform the first-layer wavelet decomposition on the branch current difference signals of the upper, middle, and lower parallel bridge arms of the capacitor bank to obtain the first-layer detail coefficients corresponding to each of them. The calculation formula is as follows: ； Where, represents the first layer detail coefficient, n is the index of the detail coefficient, n =0,1,2,…, N / 2−1, N is the total number of sampling points, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n, is the branch current difference corresponding to a parallel bridge arm when the sampling point is 2n+1.

4. The efficient fault location method for a double-H connection high-voltage filter capacitor bank according to claim 3, characterized in that: The energy values corresponding to the first layer detail coefficients are obtained by the following calculation formula: ； Represents the energy value of the first layer detail coefficient, It represents the decomposition of the current difference signal of a parallel bridge arm branch. A detail factor.

5. The efficient fault location method for a double-H connection high-voltage filter capacitor bank according to claim 4, characterized in that: The energy value thresholds are obtained based on the mean and standard deviation of the energy value sequence, and the calculation formula is as follows: ； Where, Indicates the energy value threshold, is the standard deviation of the energy value sequence, is the average value of the energy value sequence.

6. An efficient fault location device for a double-H connection high-voltage filter capacitor bank, characterized in that: include: The data acquisition module is configured to collect waveform data of two cycles before and after the double-H connection high-voltage filter capacitor bank fails, the waveform data including the capacitor bank terminal voltage, total current, two unbalanced currents and low-voltage terminal current; The first data processing module is configured to calculate a threshold value for breakdown of a capacitor element according to the capacitor bank parameters, and find the sampling point corresponding to the time when the ratio of the unbalanced current to the total current is greater than the threshold value as the disturbance start time; The second data processing module is configured to perform first-layer wavelet decomposition on the branch current difference signals of the upper, middle, and lower parallel bridge arms of the capacitor bank, respectively, to obtain the first-layer detail coefficients corresponding to each of the branch current difference signals, and then obtain the energy value of the first-layer detail coefficients; The third data processing module is configured to group the initially collected data of the three branch current difference signals, calculate the first-layer detail coefficient energy of each group of data, and form respective energy value sequences; and then calculate respective energy value thresholds based on the mean and standard deviation of the energy value sequences; The judgment output module is configured to determine the fault location in the 1 / 3 region based on the difference between the energy value and the energy value threshold; and then determine the fault location in the 1 / 6 region based on the sudden change direction of the pulse current difference at the start of the disturbance, thereby obtaining the bridge arm where the fault is located; The judgment output module locates the fault to the 1 / 3 area based on the size of the energy value and the energy value threshold. Specifically, if the energy value of the first-layer detail coefficient of a parallel bridge arm is greater than its corresponding energy value threshold, and the energy values of the first-layer detail coefficients of the other two parallel bridge arms are less than or equal to the corresponding energy value threshold, then it is determined that the parallel bridge arm greater than the corresponding energy value threshold has a fault; The judgment output module locates the fault to the 1 / 6 area according to the sudden change direction of the pulse current difference at the start of the disturbance, including: obtaining the average value of the current difference signal of the parallel bridge arm corresponding to the 1 / 3 area for two cycles before the start of the disturbance ; If the current difference at the disturbance starting point is greater than the average value , then the bridge arm where the current suddenly increases is the fault bridge arm.

Images