A small current grounding system single-phase grounding fault determination method

Abstract

This invention belongs to the field of relay protection technology for power systems, specifically relating to a method for determining single-phase grounding faults in low-current grounding systems. It includes the following steps: Step 1: Collect the zero-sequence voltage of the busbar and the zero-sequence current of each bay on the busbar, and calculate the instantaneous change in the zero-sequence voltage in real time; Step 2: When the zero-sequence voltage change trigger criterion is met, the system is considered to have a single-phase grounding fault, and the grounding fault determination program is initiated; Step 3: Extract the sampled data of the zero-sequence current and zero-sequence voltage of each bay at the start time, perform wavelet packet decomposition on the sampled data, and calculate the energy of each frequency band after decomposition for each bay; Step 4: Select the three characteristic frequency bands with the highest energy in each bay for wavelet packet weighted reconstruction; Step 5: Fault identification. The beneficial effects are: it can quickly and accurately obtain transient data at the time of the fault, perform weighted reconstruction of the waveforms of the three characteristic frequency bands with the highest energy in each bay, and use the phase comparison method to identify the grounding bay.

Description

Technical Field

[0001] This invention belongs to the field of relay protection technology for power systems, and specifically relates to a method for determining single-phase grounding faults in low-current grounding systems. Background Technology

[0002] In China's 3-66kV power supply and distribution networks and nuclear power plant power systems, most adopt the neutral point not directly grounded method. Because the current flowing through the grounding point is small when a grounding fault occurs, it is often called a low current grounding system (NUGS). It includes neutral point ungrounded system (NUS), neutral point grounded through arc suppression coil system (NES, also known as resonant grounding system) and neutral point grounded through resistance system (NRS).

[0003] Low-current grounding systems have the highest probability of single-phase grounding faults. However, when a single-phase grounding fault occurs, since it does not form a short-circuit loop, the grounding fault current is much smaller than the load current. Especially in systems where the neutral point is grounded via an arc suppression coil, the grounding current is very small, and the three-phase line voltages remain symmetrical, not affecting continuous power supply to the load. Therefore, immediate tripping is not necessary, and regulations stipulate that operation can continue for 1-2 hours. However, due to the grounding point, the voltage to ground of the non-faulty phases in the system rises to 1.732 times the original voltage, threatening the insulation of the power grid. This can easily induce another grounding point at a weak point in the grid, thus forming a phase-to-phase short circuit. As system capacity and total line length increase, the capacitive current becomes larger, and the overvoltage multiple caused by arcing grounding is very high.

[0004] To address these issues, domestic and international scientists have conducted research on wiring methods, overvoltage limiting measures, and grounding protection, aiming to minimize the probability of single-phase grounding faults. Furthermore, they hope to quickly identify the faulty line once a single-phase grounding fault occurs so that it can be eliminated promptly. However, when a single-phase grounding fault occurs in an arc-suppression coil grounding system, the phase and amplitude of the zero-sequence current in the faulty and non-faulty lines are quite similar, leading to a high misjudgment rate in line selection. Currently, most commonly used grounding fault determination methods utilize the zero-sequence steady-state electrical quantities after the fault. However, due to the small amplitude of the zero-sequence steady-state electrical quantities in single-phase grounding, accurate measurement is difficult, making practical application challenging. The injection method is limited by the voltage transformer capacity, resulting in a small injected signal, which is prone to errors in fault determination when a transition resistance grounding occurs. The transient method is currently widely used, but its accuracy is affected by unreasonable selection of characteristic frequency bands and difficulty in obtaining waveforms at the start-up moment. Summary of the Invention

[0005] The purpose of this invention is to provide a method for determining single-phase grounding faults in low-current grounding systems. It employs a rapid start-up algorithm based on instantaneous value floating thresholds and uses a wavelet packet algorithm to weight and reconstruct the waveforms of the three characteristic frequency bands with the highest energy in each interval. Finally, it uses the phase comparison method to identify the grounding interval, thereby improving the reliability and speed of device operation when a single-phase grounding fault occurs in a low-current grounding system and solving the problem of high misjudgment rate when a single-phase grounding fault occurs in a low-current grounding system.

[0006] The technical solution of the present invention is as follows: A method for determining single-phase grounding faults in a low-current grounding system, comprising the following steps:

[0007] Step 1: Collect the zero-sequence voltage of the bus and the zero-sequence current of each interval on the bus, and calculate the instantaneous change of the zero-sequence voltage in real time.

[0008] Step 2: When the zero-sequence voltage change trigger criterion is met, the system is considered to have a single-phase ground fault, and the ground fault determination procedure is initiated.

[0009] Step 3: Extract the sampling data of zero-sequence current and zero-sequence voltage at each interval during startup, perform wavelet packet decomposition on the sampling data, and calculate the energy of each frequency band after decomposition of each interval.

[0010] Step 4: Select the three characteristic frequency bands with the largest energy in each interval for wavelet packet weighted reconstruction;

[0011] Step 5: Fault diagnosis.

[0012] The method for calculating the instantaneous change in zero-sequence voltage in step 1 is as follows:

[0013] (1)

[0014] in, This is the current sampled value of the zero-sequence voltage. These are the sampled values ​​of the first 192 points of the zero-sequence voltage buffer.

[0015] Step 1 employs a fast mutation detection algorithm based on an instantaneous value floating threshold to calculate... As in equation (2):

[0016] (2)

[0017] in, This is the current calculated value of the zero-sequence voltage change. This is the calculated value at point (192+n) before the zero-sequence voltage change.

[0018] In step 2, when the time for satisfying equation (2) is greater than 4ms, the system is considered to have a single-phase ground fault, and the ground fault determination program is started.

[0019] Step 3 includes:

[0020]

[0021] in, Let n be the energy of the zero-sequence current at interval n in frequency band m. To determine the basic wavelet of zero-sequence current at interval n in frequency band m, the three frequency bands with the largest energy in each interval are selected as characteristic frequency bands. The weighting coefficients of the three characteristic frequency bands in each interval are determined according to the proportion of each characteristic frequency band to the sum of the three characteristic frequency bands.

[0022] Step 4 involves determining the weighting coefficients of the three characteristic frequency bands with the highest energy based on the proportion of the energy of each characteristic frequency band to the sum of the energy of the three characteristic frequency bands.

[0023] Step 5 includes comparing the direction of the zero-sequence current in each interval after wavelet packet weighted reconstruction. When the direction of one interval is opposite to the phase of the other intervals, the interval is determined to be a grounding interval. When the direction of the zero-sequence current in each interval is consistent, it is determined to be a bus fault.

[0024] The beneficial effects of this invention are as follows: It employs a rapid start-up algorithm based on a sudden change threshold, enabling quick and accurate acquisition of transient data at the moment of fault. This transient data is then decomposed using a wavelet packet algorithm. Afterward, the waveforms of the three characteristic frequency bands with the highest energy in each interval are weighted and reconstructed. Finally, the phase comparison method is used to identify the grounding interval. By accurately acquiring transient data at the moment of fault and performing weighted reconstruction calculations on the three characteristic frequency band waveforms, the reliability and speed of device operation during single-phase grounding faults in low-current grounding systems are improved, solving the problem of a high misjudgment rate in current low-current grounding systems when single-phase grounding faults occur. Attached Figure Description

[0025] Figure 1 This is a wiring diagram implemented in the grounding selection device;

[0026] Figure 2 The flowchart of a method for determining single-phase grounding faults in a low-current grounding system provided by the present invention is shown. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0028] like Figure 2 As shown, a method for determining single-phase grounding faults in a low-current grounding system includes the following steps:

[0029] Step 1: Collect the zero-sequence voltage of the bus and the zero-sequence current of each interval on the bus, and calculate the instantaneous change of the zero-sequence voltage in real time.

[0030] The instantaneous change in zero-sequence voltage is calculated as follows:

[0031] (1)

[0032] in, This is the current sampled value of the zero-sequence voltage. The first 192 sampled values ​​of the zero-sequence voltage buffer (i.e., one wavefront sampled value);

[0033] Step 2: When the zero-sequence voltage change trigger criterion is met, the system is considered to have a single-phase ground fault, and the ground fault determination procedure is initiated.

[0034] In step 1 A fast mutation detection algorithm based on an instantaneous value floating threshold is adopted, as shown in formula (2):

[0035] (2)

[0036] in, This is the current calculated value of the zero-sequence voltage change. The calculated value is the point (192+n) before the zero-sequence voltage change; when the time to satisfy formula (2) is greater than 4ms, the system is considered to have a single-phase ground fault, and the ground fault judgment program is started.

[0037] Step 3: Extract the sampled data of zero-sequence current and zero-sequence voltage at each interval during startup, perform wavelet packet decomposition on the sampled data, and calculate the energy of each frequency band after decomposition for each interval:

[0038]

[0039] in, Let n be the energy of the zero-sequence current at interval n in frequency band m. For the basic wavelet of zero-sequence current at interval n in frequency band m, the three frequency bands with the largest energy in each interval are selected as characteristic frequency bands. The weighting coefficients of the three characteristic frequency bands in each interval are determined according to the proportion of each characteristic frequency band to the sum of the three characteristic frequency bands.

[0040] Step 4: Select the three characteristic frequency bands with the largest energy in each interval for wavelet packet weighted reconstruction;

[0041] The weighting coefficients of the three characteristic frequency bands with the highest energy are determined according to the proportion of each characteristic frequency band's energy to the sum of the three characteristic frequency bands; and the three frequency bands with the highest energy at each interval are selected as characteristic frequency bands, with the weighting coefficients of the three characteristic frequency bands at each interval determined according to the proportion of each characteristic frequency band's energy to the sum of the three characteristic frequency bands; assuming , , Given three frequency bands with an interval of n, representing the three most maximally characteristic frequency bands, the weighting coefficients for the three frequency bands are:

[0042]

[0043]

[0044]

[0045] In the formula, k 1n k represents the weighting coefficients for frequency bands n-1 with intervals of n. mn k represents the weighting coefficients for frequency bands m at intervals of n. nn These are the weighting coefficients for frequency bands n at intervals of n.

[0046] Step 5: Compare the directions of the zero-sequence current in each interval after wavelet packet weighted reconstruction. When the direction of one interval is opposite to the phase of the other intervals, the interval is determined to be a grounding interval. When the directions of the zero-sequence current in each interval are consistent, it is determined to be a bus fault.

[0047] Example:

[0048] A method for determining single-phase grounding faults in a low-current system includes the following steps:

[0049] Step 1: Collect the zero-sequence voltage of the bus and the zero-sequence current of each interval on the bus at a sampling rate of 9.6K, and calculate the instantaneous change of the zero-sequence voltage in real time.

[0050] The method for calculating the instantaneous change in zero-sequence voltage is as follows:

[0051] (1)

[0052] in, This is the current sampled value of the zero-sequence voltage. The first 192 sampled values ​​(i.e., one cycle wavefront sampled value) are buffered for the zero-sequence voltage.

[0053] Step 2: When the zero-sequence voltage change start criterion is met, the system is considered to have a single-phase ground fault, and the ground fault judgment program is started. The algorithm for fast detection of sudden changes based on the instantaneous value floating threshold is adopted, as shown in formula (2):

[0054] (2)

[0055] in, This is the current calculated value of the zero-sequence voltage change. The calculated value is the point (192+n) before the zero-sequence voltage change; when the time to satisfy formula (2) is greater than 4ms, the system is considered to have a single-phase ground fault, and the ground fault judgment program is started.

[0056] Step 3: Take the zero-sequence current data for each interval of the first half-cycle and the second half-cycle, and perform wavelet packet decomposition of the data at 8 levels; then calculate the energy of the 32 frequency bands after decomposition for each interval:

[0057]

[0058] in, Let n be the energy of the zero-sequence current at interval n in frequency band m. The fundamental wavelet of zero-sequence current at interval n in frequency band m.

[0059] The three frequency bands with the largest energy in each interval are selected as characteristic frequency bands, and the weighting coefficients of the three characteristic frequency bands in each interval are determined according to the proportion of each characteristic frequency band to the sum of the three characteristic frequency bands.

[0060] Assumption , , Given three frequency bands with an interval of n, representing the three most maximally characteristic frequency bands, the weighting coefficients for the three frequency bands are:

[0061]

[0062]

[0063]

[0064] Step 4: Perform wavelet packet weighted reconstruction on the three characteristic frequency band data according to the calculated weighting coefficients; obtain the waveform after wavelet packet filtering.

[0065] Step 5: Compare the directions of the zero-sequence current in each interval after wavelet packet reconstruction. When the direction of one interval is opposite to that of the other intervals, the interval is determined to be a grounding interval. When the directions of the zero-sequence current in each interval are consistent, the fault is determined to be a bus fault.

[0066] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining single-phase grounding faults in a low-current grounding system, characterized in that, It includes the following steps: Step 1: Collect the zero-sequence voltage of the bus and the zero-sequence current of each interval on the bus, and calculate the instantaneous change of the zero-sequence voltage in real time. The method for calculating the instantaneous change in zero-sequence voltage in step 1 is as follows: （1）， in, This is the current sampled value of the zero-sequence voltage. These are the first 192 sampled values ​​of the zero-sequence voltage buffer; Step 1 employs a fast mutation detection algorithm based on an instantaneous value floating threshold to calculate... For example, in formula (2): （2）， in, This is the current calculated value of the zero-sequence voltage change. Before the zero-sequence voltage change ( 192 +n Calculated value of point ) Step 2: When the zero-sequence voltage change trigger criterion is met, the system is considered to have a single-phase ground fault, and the ground fault determination procedure is initiated. In step 2, when the time for satisfying formula (2) is greater than 4ms, the system is considered to have a single-phase ground fault, and the ground fault determination program is started. Step 3: Extract the sampling data of zero-sequence current and zero-sequence voltage at each interval during startup, perform wavelet packet decomposition on the sampling data, and calculate the energy of each frequency band after decomposition of each interval. Step 3 includes: ， in, Let n be the energy of the zero-sequence current at interval n in frequency band m. For the basic wavelet of zero-sequence current at interval n in frequency band m, the three frequency bands with the largest energy in each interval are selected as characteristic frequency bands. The weighting coefficients of the three characteristic frequency bands in each interval are determined according to the proportion of each characteristic frequency band to the sum of the three characteristic frequency bands. Step 4: Select the three characteristic frequency bands with the largest energy in each interval for wavelet packet weighted reconstruction; Step 4 includes determining the weighting coefficients of the three characteristic frequency bands with the highest energy according to the proportion of the energy of each characteristic frequency band to the sum of the energy of the three characteristic frequency bands; Step 5: Fault diagnosis; Step 5 includes comparing the direction of the zero-sequence current in each interval after wavelet packet weighted reconstruction. When the direction of one interval is opposite to the phase of the other intervals, the interval is determined to be a grounding interval. When the direction of the zero-sequence current in each interval is consistent, it is determined to be a bus fault.

Images