Partial discharge anti-interference method and system based on time-frequency analysis

Abstract

The application discloses a partial discharge anti-interference method and system based on time-frequency analysis, and the method comprises the following steps: collecting a partial discharge signal of a discharge period and converting the partial discharge signal into a corresponding pulse sequence; calculating the equivalent time length and the equivalent frequency width of each pulse in the pulse sequence respectively, and calculating the quadrant in which each pulse is located; deleting the pulses in the target quadrant, and deleting the pulses in the remaining quadrants which are equivalent to the pulses in the target quadrant in terms of the equivalent time length and the equivalent frequency width; and performing waveform restoration according to the remaining pulses to obtain a partial discharge signal after anti-interference. The application can effectively eliminate various interference signals such as 12-pulse wave interference caused by a converter transformer, a corona and a space electromagnetic interference signal, and achieves the purpose of improving the accuracy of analyzing partial discharge information.

Description

Technical Field

[0001] This invention relates to power monitoring technology, and more particularly to a partial discharge anti-interference method and system based on time-frequency analysis. Background Technology

[0002] During the on-site partial discharge test of high-voltage power equipment, abnormal partial discharge signals exist. Specifically, the partial discharge signals are mixed with numerous interference signals, which mainly include the following aspects:

[0003] 1. Electromagnetic interference at the partial discharge detection site;

[0004] 2. Noise present in the test power supply;

[0005] 3. Interference signals caused by various wiring defects;

[0006] Among them, spatial electromagnetic interference signals, 12-pulse interference caused by converter transformers, and corona interference can all enter the signal circuit of partial discharge detection equipment and be mixed with partial discharge signals, which will have a great impact on partial discharge analysis and localization. The main methods of anti-interference for conventional partial discharge are antenna anti-interference, threshold method anti-interference, and frequency domain anti-interference. None of these methods can effectively eliminate interference such as 12-pulse interference caused by converter transformers.

[0007] Patent document CN 113325277 A discloses a partial discharge processing method, in which the anti-interference method is wavelet transform denoising. The method involves extracting time-frequency clustering features for clustering and then restoring the partial discharge pattern (PRPD) for discharge type identification. However, verification has shown that wavelet transform denoising cannot effectively remove interference such as 12-pulse interference caused by commutation transformers.

[0008] Therefore, there is an urgent need for an effective method to suppress partial discharge interference and remove interference such as 12-pulse interference caused by the operation of large converter transformers in the field. Summary of the Invention

[0009] The technical problem to be solved by this invention is: In view of the technical problems existing in the prior art, this invention provides a partial discharge anti-interference method and system based on time-frequency analysis, which effectively eliminates a variety of interference signals such as 12-pulse interference, corona, and spatial electromagnetic interference signals caused by commutation transformers, thereby improving the accuracy of partial discharge information analysis.

[0010] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0011] A partial discharge anti-interference method based on time-frequency analysis includes the following steps:

[0012] Acquire the partial discharge signal of one discharge cycle and convert it into a corresponding pulse sequence;

[0013] Calculate the equivalent duration and equivalent bandwidth of each pulse in the pulse sequence, and calculate the quadrant in which each pulse is located;

[0014] Delete pulses in the target quadrant and delete pulses in the other quadrants that have equivalent duration and bandwidth to the pulses in the target quadrant.

[0015] The waveform is restored based on the remaining pulses to obtain the partial discharge signal after interference suppression.

[0016] Furthermore, the expression for the equivalent duration is as follows:

[0017]

[0018] Where t represents the time corresponding to each data point in the pulse sequence, t0 represents the time scale of the first pulse in the pulse sequence, and x(t) represents the time-domain pulse sequence of the pulse.

[0019] Furthermore, the expression for the equivalent bandwidth is as follows:

[0020]

[0021] Where X(ω) represents the frequency domain pulse sequence after the Fourier transform of the time domain pulse sequence, and ω represents the frequency independent variable corresponding to the frequency domain pulse sequence.

[0022] Furthermore, before acquiring the partial discharge signal for one discharge cycle, the method further includes: acquiring a synchronization control signal, and synchronizing the acquired signal of the partial discharge detector and the partial discharge signal according to the synchronization control signal.

[0023] Furthermore, before synchronizing the phase of the partial discharge detector's acquisition signal and the partial discharge signal according to the synchronization control signal, it is determined whether the synchronization control signal is a sine wave starting from zero. If it is not zero, software phase shifting is performed.

[0024] Furthermore, when calculating the quadrant in which each pulse lies, the following steps are taken:

[0025] The sampling period T is calculated based on the frequency H of the synchronization control signal. S The sampling period T S Divide by 4 to get the cycle time T for each quadrant. q ;

[0026] Calculate the time scale t of each pulse in the pulse sequence relative to the sampling period T. S An integer multiple of the error, and the error value relative to the cycle time T occupied by each quadrant.q The quadrant Q of each pulse is calculated based on the multiple of the given multiple.

[0027] Furthermore, the Q expression for each pulse in its quadrant is as follows:

[0028]

[0029] in, The floor function is used for rounding down, where t represents the time index of the current pulse in the pulse sequence. S T represents the sampling period. q This indicates the cycle time occupied by each quadrant.

[0030] Furthermore, the target quadrant is the second quadrant and the fourth quadrant, and the remaining quadrants are the first quadrant and the third quadrant.

[0031] Furthermore, when deleting pulses in the target quadrant and deleting pulses in other quadrants that have equivalent duration and bandwidth to those in the target quadrant, this includes:

[0032] Construct a first matrix, in which each element of each row is, in order, the time index, quadrant, equivalent duration, and equivalent bandwidth of the corresponding pulse in the pulse sequence;

[0033] The equivalent duration and equivalent bandwidth of the rows with values ​​of 2 or 4 in the quadrant of the first matrix are calculated separately, and the maximum equivalent duration T is obtained. max Minimum value T min And the maximum value B of the equivalent bandwidth max Minimum value B min ;

[0034] Iterate through each row of the first matrix. If the equivalent duration in the current row is at the maximum value T... max With minimum value T min Between, and the equivalent bandwidth in the current row is at the maximum value B. max With minimum value B min Delete the current line between them.

[0035] This invention also proposes a partial discharge anti-interference system, comprising a synchronous power supply and a partial discharge detector, wherein:

[0036] The synchronous power supply is used to provide synchronous control signals for the signal acquisition of the partial discharge detector.

[0037] The partial discharge detector is programmed or configured to perform any of the time-frequency analysis-based partial discharge anti-interference methods described above.

[0038] Compared with the prior art, the advantages of the present invention are as follows:

[0039] This invention requires no additional hardware devices, uses a synchronization signal to ensure that the acquired signal and the discharge signal are in phase, and fuses the time and frequency domain characteristics of the pulse as the pulse characteristics, which can more clearly characterize the pulse characteristics. It has the advantages of good anti-interference effect and ease of use.

[0040] This invention calculates the equivalent frequency and equivalent time of each pulse in the acquired pulse sequence, and calculates the quadrant in which each pulse is located. Since partial discharge of large power transformers generally occurs during the voltage rise process, that is, in the first and third quadrants, and partial discharge signals generally do not occur in the second and fourth quadrants, based on this theory, pulse sequences with the same characteristics as pulses in the second and fourth quadrants are deleted in all full cycles to achieve the purpose of anti-interference. This can effectively eliminate various interference signals such as 12-pulse interference caused by converter transformers, and meet certain real-time requirements. Attached Figure Description

[0041] Figure 1 This is a flowchart of a method according to an embodiment of the present invention.

[0042] Figure 2 This is a diagram illustrating the anti-interference effect of partial discharge signals in an embodiment of the present invention. Detailed Implementation

[0043] The present invention will be further described below with reference to the accompanying drawings and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0044] To eliminate interference such as 12-pulse interference caused by the commutator, this embodiment designs a partial discharge anti-interference system, including a synchronous power supply and a partial discharge detector, wherein:

[0045] The synchronization power supply is a test power supply with a frequency of H, which is used to provide a synchronization control signal for the signal acquisition of the partial discharge detector. The use of a synchronization signal can ensure that the acquired signal and the discharge signal are in phase.

[0046] The partial discharge detector has a built-in high-speed data acquisition card that acquires signals at a sampling rate of no less than 80 MSPS and converts them into pulse sequences with precise time stamps.

[0047] In this embodiment, the microprocessor of the partial discharge detector is programmed or configured to execute the partial discharge anti-interference method based on time-frequency analysis proposed in this embodiment, such as... Figure 1 As shown, the method includes the following steps:

[0048] 1. Connect to a synchronous power supply and acquire the synchronization control signal generated by the synchronous power supply. Synchronize the phase of the acquired signal from the partial discharge detector and the partial discharge signal according to the synchronization control signal. Before phase synchronization, determine whether the synchronization control signal is a sine wave starting from zero. If it is not zero, perform software phase shifting.

[0049] 2. Perform high-speed data acquisition (not less than 80 MSPS) to obtain the partial discharge signal of one discharge cycle, and convert it into a corresponding pulse sequence N with time stamp t;

[0050] In this embodiment, the pulse sequence N with time stamp t is n sets of pulses N(1:n) collected by the partial discharge detector for a partial discharge signal of one discharge cycle through automatic triggering or threshold triggering. Pulse 1 is represented by N(1). In the data of N(1), the first data is the trigger time t of the pulse, and then the data sequence of the pulse is in sequence (the specific data volume is determined according to the sampling rate and pulse duration); the data sequences of pulse 2 to pulse n, that is, N(2) to N(n), are the same as N(1).

[0051] 3. Based on the time-frequency characteristics of the acquired pulse sequence, calculate the equivalent duration and equivalent bandwidth of each pulse;

[0052] In this embodiment, the formula for calculating the equivalent duration is as follows:

[0053]

[0054] Where t represents the time corresponding to each data point in the pulse data sequence, t0 represents the time scale of the first pulse, and x(t) represents the time-domain pulse sequence of the partial discharge detector's high-speed data acquisition, i.e., the time-domain pulse sequence of the pulse.

[0055] In this embodiment, the formula for calculating the equivalent bandwidth is as follows:

[0056]

[0057] Where X(ω) represents the frequency domain pulse sequence after Fourier transform of the acquired pulse sequence, and ω represents the frequency independent variable corresponding to the frequency domain pulse sequence.

[0058] In equation (1), the time stamp t0 of the first pulse, i.e., the starting position of the first pulse acquisition time, is calculated by the following formula:

[0059]

[0060] Where x(t) represents the time-domain pulse sequence of the first pulse collected, and t represents the time corresponding to each data point in the data sequence of the first pulse.

[0061] In this embodiment, the time stamp of the i-th pulse in the pulse sequence is t. i The equivalent duration is obtained from equation (1), denoted as T. i The corresponding equivalent bandwidth, denoted as B, is obtained from equation (2). i The range of i is 0 to N-1.

[0062] 4. Calculate the quadrant in which each pulse is located;

[0063] First, the sampling period T is calculated based on the frequency H of the synchronization control signal. S ,

[0064]

[0065] Sampling period T S The unit is milliseconds;

[0066] Then, the sampling period T S Divide by 4 to get the cycle time T for each quadrant. q , The unit is milliseconds;

[0067] Finally, the time scale t of each pulse in the pulse sequence relative to the sampling period T is calculated. S An integer multiple of the error, and the error value relative to the cycle time T occupied by each quadrant. q The quadrant Q of each pulse is calculated based on the multiple of the given multiple. The expression for the quadrant Q of each pulse is as follows:

[0068]

[0069] in, The floor function is used for rounding down, where t represents the time index of the current pulse in the pulse sequence. S T represents the sampling period. q This indicates the cycle time occupied by each quadrant.

[0070] 5. Delete pulses in the target quadrant and delete pulses in the remaining quadrants that have equivalent duration and bandwidth to those in the target quadrant. In this embodiment, the target quadrants are the second and fourth quadrants, and the remaining quadrants are the first and third quadrants. This is because partial discharge generally occurs during the periodic voltage rise process, i.e., quadrants 1 and 3. Pulses in quadrants 2 and 4 are generally noise signals. Since noise pulses have similar equivalent duration and bandwidth, and the equivalent duration and bandwidth of pulses in quadrants 1 and 3 are comparable to those in quadrants 2 and 4, they are considered noise.

[0071] First, construct a first matrix P1. The elements of each row in the first matrix P1 are, in order, the time index, quadrant, equivalent duration, and equivalent bandwidth of the corresponding pulse in the pulse sequence, as shown below:

[0072]

[0073] Then, the equivalent duration and equivalent bandwidth of the rows with values ​​of 2 or 4 in the quadrant where the pulse is located in the first matrix P1 are calculated respectively, and the maximum value T of the equivalent duration is obtained. max Minimum value T min And the maximum value of the equivalent bandwidth B max Minimum value B min ;

[0074] Finally, iterate through each row of the first matrix. If the equivalent duration in the current row is at the maximum value T... max With minimum value T min Between, and the equivalent bandwidth in the current row is at the maximum value B. max With minimum value B min Between these rows, delete the current row to form a new matrix P2, as shown below:

[0075]

[0076] Where M≤N, after the quadrant transformation is completed, pulse sequences with the same characteristics as pulses in the second and fourth quadrants are deleted in all full cycles, thereby removing interference such as 12-pulse interference caused by commutation.

[0077] 6. The waveform is restored based on the remaining pulses to obtain the partial discharge signal after interference suppression.

[0078] Specifically, the pulses with corresponding time scales in matrix P2 are obtained. Using the time scales of these pulses, the corresponding positions in the periodic waveform data source are determined to redraw a new waveform, thus forming a new waveform after interference suppression.

[0079] 7. If a stop command is received from the user, the process ends and exits; otherwise, it returns to step 2 to begin anti-interference processing of the partial discharge signal for the next discharge cycle.

[0080] like Figure 2 As shown in the figure, the waveform in the upper half represents the original partial discharge signal of one discharge cycle, which shows obvious interference. The lower half represents the partial discharge signal output by the partial discharge anti-interference method based on time-frequency analysis in this embodiment. Compared with the original partial discharge signal, it eliminates a lot of interference. It can be seen that the partial discharge anti-interference method based on time-frequency analysis in this embodiment can achieve a better anti-interference effect.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention should fall within the protection scope of the present invention.

Claims

1. A partial discharge anti-interference method based on pulse time-frequency analysis, characterized in that, Includes the following steps: Acquire the partial discharge signal of one discharge cycle and convert it into a corresponding pulse sequence; Calculate the equivalent duration and equivalent bandwidth of each pulse in the pulse sequence, and calculate the quadrant in which each pulse is located; Delete pulses in the target quadrant and delete pulses in the other quadrants that have equivalent duration and bandwidth to the pulses in the target quadrant. The waveform is restored based on the remaining pulses to obtain the partial discharge signal after interference suppression; The target quadrants are the second and fourth quadrants, and the remaining quadrants are the first and third quadrants. Deleting pulses from the target quadrants and deleting pulses from the remaining quadrants that have equivalent duration and bandwidth to those from the target quadrants includes: Construct a first matrix, in which each element of each row is, in order, the time index, quadrant, equivalent duration, and equivalent bandwidth of the corresponding pulse in the pulse sequence; The equivalent duration and equivalent bandwidth of the rows with values ​​of 2 or 4 in the quadrant of the first matrix are calculated separately, and the maximum value of the equivalent duration is obtained. T max Minimum value T min and the maximum value of the equivalent bandwidth B max Minimum value B min ; Iterate through each row of the first matrix. If the equivalent duration in the current row is at the maximum value... T max and minimum value T min Between, and the equivalent bandwidth in the current row is at the maximum value. B max and minimum value B min Delete the current line between them.

2. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 1, characterized in that, The expression for the equivalent duration is as follows: in, t This represents the time corresponding to each data point in the pulse data sequence. t 0 indicates the time stamp of the first pulse in the pulse sequence. x ( t ) represents the time-domain pulse sequence of a pulse.

3. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 1, characterized in that, The expression for the equivalent bandwidth is as follows: in, This represents the frequency domain pulse sequence obtained after the Fourier transform of a time-domain pulse sequence. This represents the frequency independent variable corresponding to the frequency domain pulse sequence.

4. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 1, characterized in that, Before acquiring the partial discharge signal for one discharge cycle, the method further includes: acquiring a synchronization control signal, and synchronizing the acquired signal of the partial discharge detector and the partial discharge signal according to the synchronization control signal.

5. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 4, characterized in that, Before synchronizing the phase of the partial discharge detector's acquisition signal and the partial discharge signal according to the synchronization control signal, it is determined whether the synchronization control signal is a sine wave starting from zero. If it is not zero, software phase shifting is performed.

6. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 4, characterized in that, When calculating the quadrant in which each pulse lies, the following is included: Based on the frequency of the synchronization control signal H The sampling period was calculated. T S The sampling period T S Divide by 4 to get the cycle time for each quadrant. T q ; Calculate the time scale of each pulse in the pulse sequence. t Relative to sampling period T S Integer multiple of error ， And the time period occupied by this error value for each quadrant. T q The quadrant in which each pulse is located is calculated based on the multiple of the given multiple. Q .

7. The partial discharge anti-interference method based on pulse time-frequency analysis according to claim 6, characterized in that, The quadrant where each pulse is located Q The expression is as follows: in, This is the floor operator. t This indicates the time stamp of the current pulse in the pulse sequence. T S Indicates the sampling period. T q This indicates the cycle time occupied by each quadrant.

8. A partial discharge anti-interference system, characterized in that, Includes a synchronous power supply and a partial discharge detector, among which: The synchronous power supply is used to provide synchronous control signals for the signal acquisition of the partial discharge detector. The partial discharge detector is programmed or configured to perform the partial discharge anti-interference method based on pulse time-frequency analysis as described in any one of claims 1 to 7.

Images