Cable partial discharge signal multi-terminal synchronous detection method, device and medium

Abstract

The embodiment of the invention discloses a cable partial discharge signal multi-end synchronous detection method and device and a medium, belongs to the technical field of cable partial discharge detection, and solves the problem that the misjudgment rate is high due to the fact that real partial discharge pulses and interference signals are difficult to effectively separate through traditional fixed threshold value judgment. The method comprises the following steps: respectively acquiring original acquisition signals through signal acquisition devices at two ends of a to-be-detected cable, and preprocessing the original acquisition signals; performing sliding time window division on the preprocessed original acquisition signal, and determining short-time energy and spectral entropy features corresponding to each sliding time window; a joint detection index is constructed based on the short-time energy and spectral entropy features, and partial discharge signal identification is carried out based on an adaptive statistical threshold and the joint detection index; under the condition that partial discharge signals exist in the original acquisition signals, partial discharge signal segmentation is carried out on the preprocessed original acquisition signals; and performing partial discharge signal positioning based on partial discharge signals corresponding to the two ends of the to-be-detected cable through a frequency domain reflection method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable partial discharge detection, and particularly relates to a cable partial discharge signal multi-end synchronous detection method, device and medium. BACKGROUND

[0002] Partial discharge (PD) is a common discharge phenomenon in the early stage of insulation deterioration of high-voltage cables and their auxiliary equipment, and is an important precursor signal of insulation failure. It is usually caused by the existence of air gaps, impurities or structural defects in the insulation material, and generates a short-time high-frequency pulse signal under the action of high voltage. Such signals have a short duration and low amplitude, and are often covered by power frequency interference, electromagnetic noise and external interference signals, which brings challenges to effective detection.

[0003] In the daily operation of the power system, partial discharge detection is a core means to ensure the safe and stable operation of the power transmission and distribution line. The currently widely used detection methods include the ultra-high frequency method, the pulse current method and the time domain reflection method, etc. These methods can evaluate the cable insulation state by detecting the partial discharge signal. However, with the rapid expansion of urban power grids, the number and length of cable lines are increasing, and the traditional single-point detection and offline analysis mode cannot meet the real-time, accuracy and large-scale operation and maintenance requirements. Secondly, the partial discharge pulse signal is often hidden in a complex noise background, has a weak amplitude and unstable characteristics, and the traditional fixed threshold discrimination or simple filtering algorithm cannot effectively separate the real partial discharge pulse from the interference signal, resulting in a high misjudgment rate and low positioning reliability. SUMMARY

[0004] The embodiments of the present application provide a cable partial discharge signal multi-end synchronous detection method, device and medium, which are used to solve the technical problem that the traditional fixed threshold discrimination or simple filtering algorithm cannot effectively separate the real partial discharge pulse from the interference signal, resulting in a high misjudgment rate and low positioning reliability.

[0005] The embodiments of the present application adopt the following technical solutions: The embodiments of the present application provide a cable partial discharge signal multi-end synchronous detection method. The method comprises the following steps: acquiring original collection signals through signal collection devices at two ends of a to-be-detected cable, and pre-processing the original collection signals; dividing the pre-processed original collection signals into sliding time windows, and determining short-time energy and spectral entropy features corresponding to each sliding time window; constructing a joint detection index based on the short-time energy and spectral entropy features, and performing partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the median and absolute deviation of the adaptive statistical threshold and the joint detection index are related; in the case that there is a partial discharge signal in the original collection signal, performing partial discharge signal segmentation on the pre-processed original collection signal; and performing partial discharge signal positioning based on the frequency domain reflection method and the partial discharge signals respectively corresponding to the two ends of the to-be-detected cable.

[0006] In an implementation form of the application, the original acquisition signals are acquired by signal acquisition devices arranged at both ends of the cable to be tested, specifically including: transmitting the original acquisition signal acquired by the first signal acquisition device arranged at the first end of the cable to be tested to the host computer; and transmitting the original acquisition signal acquired by the second signal acquisition device arranged at the second end of the cable to be tested to the slave computer, so as to perform partial discharge signal segmentation on the original acquisition signal in the slave computer and send the segmented partial discharge signal to the host computer; wherein the host computer and the slave computer are time-synchronized by clock signals.

[0007] In an implementation form of the application, the original acquisition signal is preprocessed, specifically including: filtering the power frequency signal and the low frequency signal in the original acquisition signal by a high-pass filter; performing multi-layer wavelet packet decomposition on the filtered signal based on a preset mother wavelet to obtain each frequency band coefficient, and obtaining an adaptive threshold value according to the noise level of each frequency band; performing soft threshold compression on the frequency noise coefficient less than the adaptive threshold value, and retaining the partial discharge pulse characteristic coefficient higher than the adaptive threshold value, and obtaining the denoised signal through inverse wavelet packet reconstruction.

[0008] In an implementation form of the application, the preprocessed original acquisition signal is divided into sliding time windows, and the short-time energy and the spectral entropy feature corresponding to each sliding time window are determined, specifically including: dividing the preprocessed original acquisition signal into a plurality of sliding time windows based on a preset sliding step; determining the short-time energy corresponding to each time window through the sum of squares of signal amplitudes in each sliding time window; performing short-time Fourier transform on the signal corresponding to each sliding time window to obtain the time-frequency matrix corresponding to each sliding time window; determining the normalized power spectrum distribution corresponding to each time-frequency matrix in a preset analysis frequency band, so as to obtain the spectral entropy feature corresponding to each sliding time window based on the normalized power spectrum distribution.

[0009] In an implementation form of the application, a joint detection index is constructed based on the short-time energy and the spectral entropy feature, specifically including: performing first ratio processing on the short-time energy corresponding to each sliding time window with the maximum short-time energy; and performing second ratio processing on the spectral entropy feature corresponding to each sliding time window with the maximum spectral entropy feature; superimposing the first ratio processing result and the second ratio processing result by a weight coefficient to obtain the joint detection index.

[0010] In an implementation form of the application, before the partial discharge signal recognition based on the adaptive statistical threshold value and the joint detection index, the method further includes: determining the median and the absolute deviation of the joint detection index corresponding to different detection ends; superimposing the median and the absolute deviation by a preset adjustment coefficient to obtain the adaptive statistical threshold value corresponding to both ends of the cable to be tested.

[0011] In an implementation manner of the present application, the preprocessed original acquisition signals are subjected to partial discharge signal segmentation, specifically including: determining the earliest occurrence time of the partial discharge signals corresponding to the two ends of the cable to be tested in the preprocessed original acquisition signals; determining the effective time period corresponding to each original acquisition signal based on the full length of the cable, wave speed and the earliest occurrence time of the partial discharge signals; and segmenting the partial discharge signals based on the effective time period.

[0012] In an implementation manner of the present application, the partial discharge signals are positioned based on the frequency domain reflection method and the partial discharge signals corresponding to the two ends of the cable to be tested, specifically including: taking the union set of the effective time periods corresponding to each partial discharge signal as a unified time axis, filling the missing sampling points so as to align the partial discharge signals in the time domain; obtaining the frequency domain expression corresponding to the aligned partial discharge signals through Hilbert transform and discrete Fourier transform, and constructing a cable transfer function based on the frequency domain expression; performing spectrum analysis on the cable transfer function, constructing a partial discharge positioning function, and positioning the partial discharge signals based on the partial discharge positioning function.

[0013] The embodiment of the present application provides a cable partial discharge signal multi-end synchronous detection device, including: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: acquire original acquisition signals through signal acquisition devices at the two ends of a cable to be tested, and pre-process the original acquisition signals; divide the preprocessed original acquisition signals into sliding time windows, and determine the short-time energy and spectral entropy features corresponding to each sliding time window; construct a joint detection index based on the short-time energy and spectral entropy features, and perform partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the median and absolute deviation of the adaptive statistical threshold and the joint detection index are related; in the case that there is a partial discharge signal in the original acquisition signals, segment the preprocessed original acquisition signals; and position the partial discharge signals based on the frequency domain reflection method and the partial discharge signals corresponding to the two ends of the cable to be tested.

[0014] The nonvolatile computer storage medium provided by the embodiment of the application stores computer executable instructions, and the computer executable instructions are configured to: acquire original acquisition signals through signal acquisition devices at two ends of a to-be-tested cable, and pre-process the original acquisition signals; divide the pre-processed original acquisition signals into sliding time windows, and determine short-time energy and spectral entropy features corresponding to each sliding time window; construct a joint detection index based on the short-time energy and spectral entropy features, and perform partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the adaptive statistical threshold is related to a median and an absolute deviation corresponding to the joint detection index; in the case that there is a partial discharge signal in the original acquisition signal, the pre-processed original acquisition signal is segmented into partial discharge signals; and partial discharge signal positioning is performed based on the frequency domain reflection method and the partial discharge signals respectively corresponding to the two ends of the to-be-tested cable.

[0015] The above at least one technical solution adopted by the embodiment of the application can achieve the following beneficial effects: the embodiment of the application divides the pre-processed signal into sliding time windows, extracts short-time energy and spectral entropy features, can segment continuous signals, adapts to partial discharge short-time characteristics, highlights pulse amplitude by short-time energy, distinguishes partial discharge from noise from the frequency domain by spectral entropy, adaptively sets a threshold by combining the median absolute deviation method, realizes automatic recognition of partial discharge pulses, and improves detection accuracy and adaptability. Secondly, the embodiment of the application segments signals when there is a partial discharge signal, can intercept effective segments, eliminates redundant data, effectively reduces the amount of data transmitted from a slave to a master, while accurately retaining partial discharge characteristics, and realizes massive and efficient transmission of partial discharge signals. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings: Figure 1 A flow chart of a cable partial discharge signal multi-end synchronous detection method provided by the embodiment of the present application; Figure 2 An experimental circuit diagram provided by the embodiment of the present application; Figure 3 A partial diagram of a first-end original acquisition signal provided by the embodiment of the present application; Figure 4 A partial diagram of a last-end original acquisition signal provided by the embodiment of the present application; Figure 5 A first-end signal energy distribution diagram provided by the embodiment of the present application; Figure 6An end signal energy distribution diagram provided by an embodiment of the present application; Figure 7 A first end partial discharge signal diagram after pulse segmentation provided by an embodiment of the present application; Figure 8 An end partial discharge signal diagram after pulse segmentation provided by an embodiment of the present application; Figure 9 A positioning result diagram provided by an embodiment of the present application; Figure 10 A structural schematic diagram of a cable partial discharge signal multi-end synchronous detection device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] The present application provides a cable partial discharge signal multi-end synchronous detection method, device and medium.

[0018] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0019] The technical solutions proposed by the embodiments of the present application will be described in detail below with reference to the drawings.

[0020] Figure 1 A cable partial discharge signal multi-end synchronous detection method flowchart provided by an embodiment of the present application is shown in Figure 1 The cable partial discharge signal multi-end synchronous detection method includes the following steps: S101, through the signal acquisition devices at both ends of the cable to be tested, respectively acquire original acquisition signals, and pre-process the original acquisition signals.

[0021] In an embodiment of the present application, the original acquisition signal collected by the first signal acquisition device arranged at the first end of the cable to be tested is transmitted to the host computer. And the original acquisition signal collected by the second signal acquisition device arranged at the end of the cable to be tested is transmitted to the slave computer, so as to perform partial discharge signal segmentation on the original acquisition signal in the slave computer, and send the segmented partial discharge signal to the host computer, wherein the host computer and the slave computer are time-synchronized through a clock signal.

[0022] Specifically, ultra-high frequency current transformers (UHFCTs) are respectively installed at both ends of the cable to be tested, for real-time acquisition of partial discharge signals generated during operation of the cable. The partial discharge signals collected at the first end of the cable are transmitted to the host computer, denoted as The partial discharge signal collected by the cable end is transmitted to the slave machine, denoted as To ensure that the data at both ends have a unified time reference, the master and the slave machine are time-synchronized through a unified clock signal.

[0023] In an embodiment of the present application, the power frequency signal and the low frequency signal in the original collected signal are filtered through a high-pass filter. The filtered signal is decomposed into multiple layers of wavelet packet based on a preset mother wavelet, to obtain each frequency band coefficient, and an adaptive threshold is obtained according to the noise level of each frequency band. The noise coefficient less than the adaptive threshold is suppressed by soft threshold, and the partial discharge pulse characteristic coefficient higher than the adaptive threshold is reserved, and the denoised signal is obtained through inverse wavelet packet reconstruction.

[0024] Specifically, the original collected signal and often contains a large amount of strong background noise and power frequency interference, so it needs to be preprocessed before processing the data, and the preprocessing process is as follows: (1) A high-pass filter is used to preliminarily process the original collected signal, and the cutoff frequency is set to 1 MHz to filter out the power frequency and low frequency interference components, so that the high frequency characteristics of the partial discharge signal are retained.

[0025] (2) Select a mother wavelet suitable for the characteristics of the partial discharge pulse (such as db4 wavelet), and decompose the filtered signal into multiple layers of wavelet packet to obtain each frequency band coefficient; calculate the adaptive threshold : ; wherein is the noise standard deviation of the i th sub-band, N is the number of sampling points.

[0026] The noise coefficient less than the threshold is suppressed by soft threshold, and the partial discharge pulse characteristic coefficient higher than the threshold is reserved, and the denoised signal is obtained through inverse wavelet packet reconstruction. After preprocessing, the partial discharge signal with high signal-to-noise ratio and clear characteristics is obtained and : ; ; wherein, is the partial discharge signal corresponding to the primary end, is the partial discharge signal corresponding to the end, is the wavelet denoising.

[0027] S102, the original collected signal after preprocessing is divided into sliding time windows, and the short-time energy and spectral entropy features corresponding to each sliding time window are determined.

[0028] In an embodiment of the present application, the pre-processed original acquisition signal is divided into a plurality of sliding time windows based on a preset sliding step, and the short-time energy corresponding to each time window is determined by the sum of the square of the signal amplitude in each sliding time window. The short-time Fourier transform is performed on the signal corresponding to each sliding time window to obtain the time-frequency matrix corresponding to each sliding time window. In a preset analysis frequency band, the normalized power spectrum distribution corresponding to each time-frequency matrix is determined, and the spectral entropy feature corresponding to each sliding time window is obtained based on the normalized power spectrum distribution.

[0029] Specifically, after denoising filtering of the original detection signal, the short-time energy and spectral entropy feature are extracted by using the sliding time window, the joint detection index is constructed, and the partial discharge pulse is automatically identified by the adaptive statistical threshold. The specific process is as follows: (1) The input signal is divided into sliding time windows with a length of M , and the short-time energy in each window is calculated: ; ; wherein, represents the energy of the window starting at n corresponding to the beginning, represents the energy of the window starting at n corresponding to the end, and the sliding step can be set to M / 2 or smaller to ensure the time resolution of the detection.

[0030] (2) In order to more accurately depict the signal structure feature, the short-time Fourier transform is performed on the signal in each time window to obtain the time-frequency matrix: ; ; wherein, is a Hamming window.

[0031] Secondly, the normalized power spectrum distribution is calculated: ; ; wherein, and represent the starting frequency and ending frequency of the analysis frequency band.

[0032] Then, the spectral entropy of the time window is calculated: ; ; The background signal energy is dispersed, and the spectral entropy higher; partial discharge pulse energy is concentrated in a few frequencies, spectral entropy significantly reduced.

[0033] S103, based on the short-time energy and spectral entropy features, a joint detection index is constructed, and based on the adaptive statistical threshold and the joint detection index, partial discharge signal recognition is performed.

[0034] In an embodiment of the present application, the short-time energy corresponding to each sliding time window is respectively subjected to first ratio processing with the maximum short-time energy. And the spectral entropy features corresponding to each sliding time window are respectively subjected to second ratio processing with the maximum spectral entropy features. Through a weight coefficient, the first ratio processing result and the second ratio processing result are superimposed to obtain a joint detection index.

[0035] Specifically, the embodiment of the present application defines a joint detection index by comprehensively considering the amplitude and structure information: ; ; wherein, α 1, β 1, α 2, β 2 are adjustable weight coefficients, and satisfy α 1+ β 1=1, α 2+ β 2=1.

[0036] In an embodiment of the present application, the median and absolute deviation of the joint detection index corresponding to different detection ends are determined, and the median and absolute deviation are superimposed based on a preset adjustment coefficient to obtain the adaptive statistical threshold corresponding to the two ends of the cable to be tested.

[0037] Specifically, based on the median of the joint detection index of the first end , the median of the joint detection index of the end , the absolute deviation of the first end , and the absolute deviation of the end , the adaptive statistical threshold is constructed: ; ; wherein T 1 is the adaptive statistical threshold corresponding to the first end, T 2 is the adaptive statistical threshold corresponding to the end, is the adjustment coefficient corresponding to the first end, is the adjustment coefficient corresponding to the end, and is usually 3-5.

[0038] When and When the local discharge pulse exists in the signal, the determination result is that the local discharge pulse exists in the signal.

[0039] For the selection of the adjustment coefficient, the embodiments of the present application can adopt the following steps: first, the local discharge signal joint detection indicators of the cable head and tail ends are collected in real time, the wavelet packet energy proportion of the signal in the 100-500 MHz local discharge characteristic frequency band is calculated, the pulse density is obtained by counting the number of pulses per unit time, and the signal kurtosis is analyzed. When the characteristic frequency band energy proportion is high and the pulse density is normal, the adjustment coefficient is reduced to improve the detection sensitivity; if the pulse density exceeds twice the historical average or the kurtosis deviates from the normal distribution, the adjustment coefficient is increased to suppress noise interference. When it is monitored that the cable load current fluctuation exceeds 15% or the ambient temperature change exceeds 5°C, the smooth transition mechanism is triggered, and the current coefficient is adjusted to the optimal coefficient in the historical preset period, such as 30 minutes. At the same time, the fuzzy logic adjustment coefficient, the coefficient dynamically calculated according to the signal-to-noise ratio, and the initial fixed coefficient are used in parallel, and the final coefficient is determined according to the preset weight. If a certain coefficient causes the false rejection rate to exceed 10% or the missed detection rate to exceed 5% for 5 consecutive detection periods, the weight of the coefficient is reduced by 20%, so that the adjustment coefficient can adapt to the actual local discharge detection scene.

[0040] S104, in the case that the local discharge signal exists in the original acquisition signal, the local discharge signal in the preprocessed original acquisition signal is segmented.

[0041] In an embodiment of the present application, in the preprocessed original acquisition signal, the earliest occurrence time of the local discharge signal corresponding to each end of the cable to be detected is determined. Based on the cable full length, the wave speed and the earliest occurrence time of the local discharge signal, the effective time period corresponding to each original acquisition signal is determined, and based on the effective time period, the local discharge signal of each original acquisition signal is segmented.

[0042] Specifically, after determining that the local discharge pulse exists in the signal, the cable end local discharge signal detected by the slave needs to be transmitted to the host. However, the local discharge signal data is large, so the local discharge signal needs to be segmented to intercept the effective local discharge segment. The host detects the local discharge pulse in the signal The minimum time corresponding to the local discharge pulse is recorded as The minimum time corresponding to the local discharge pulse is recorded as The minimum time corresponding to the local discharge pulse is recorded as The time period intercepted by the master and the slave is respectively: ; ; Among them, is the intercepted time period corresponding to the head end, is the intercepted time corresponding to the tail end, is the cable full length, is the wave speed.

[0043] The time period intercepted by the slave and the corresponding partial discharge segment is sent to the host through the 4G network. The pulse segmentation technology can effectively reduce the amount of data transmitted from the slave to the host, realizing the mass and efficient transmission of partial discharge signals.

[0044] S105, based on the frequency domain reflection method and the partial discharge signals corresponding to the two ends of the cable to be measured, the partial discharge signal positioning is performed.

[0045] In an embodiment of the present application, the union of the effective time periods corresponding to each partial discharge signal is taken as a unified time axis, and the missing sampling points are filled to align the partial discharge signals in the time domain. Through Hilbert transform and discrete Fourier transform, the frequency domain expression corresponding to the aligned partial discharge signals is obtained, and the cable transfer function is constructed based on the frequency domain expression. The cable transfer function is analyzed in the frequency domain, and the partial discharge positioning function is constructed to position the partial discharge signals based on the partial discharge positioning function.

[0046] Specifically, after the host receives the end effective partial discharge segment from the slave, the host intercepts the head effective partial discharge segment , and the union of the two time periods , is taken as a unified time axis, and the missing sampling points are filled with zero to ensure the strict alignment of the signals in the time domain. After Hilbert transform and discrete Fourier transform, the frequency domain expressions of the partial discharge signals measured at the ends of the cable are obtained and . The cable transfer function is calculated as: ; wherein, is the cable transfer function; and are the upper and lower limits of the cable transfer function.

[0047] The transfer function is analyzed in the frequency domain to obtain the partial discharge positioning function: ; wherein, is the partial discharge positioning function; x is the distance from the head of the cable; is the phase constant of the cable; C is the Chebyshev window function; is the frequency interval.

[0048] The maximum value of the modulus of the cable partial discharge positioning function is taken as the x positioning position.

[0049] Figure 2 An experimental circuit diagram provided for the embodiment of the present application uses a signal generator to generate partial discharge signals, which are connected to the middle of a 70m cable and a 40m cable through a three-way joint to simulate the situation of partial discharge occurring at a distance of 70m from the head end of a 110m long cable. The UHF CT is used to collect the partial discharge signals at the head and tail ends of the cable. The signal joint detection index is extracted by using a sliding time window. The effective partial discharge segment is intercepted according to the adaptive statistical threshold. The signal length before interception is 0.02s. After pulse segmentation, the slave only needs to transmit the effective partial discharge segment of 500ns, which greatly improves the transmission efficiency of the partial discharge signal. The partial discharge occurs at a distance of 70.29m from the head end of the cable, which meets the expectation and proves the reliability of the method.

[0050] wherein, Figure 3 A partial graph of the original collected signal at the head end provided for the embodiment of the present application is shown in Figure 3 The signal amplitude in the local period fluctuates between 0.05 and 0.20, and sharp pulse peaks appear in some areas. However, the overall signal is disturbed by background noise, the waveform edge is blurred, and the signal details are difficult to identify clearly due to the noise. Figure 4 A partial graph of the original collected signal at the tail end provided for the embodiment of the present application is shown in Figure 4 The signal morphology has similarity with the head end, and also has short-time pulse characteristics. The local amplitude fluctuation range is large, and the energy distribution is relatively dispersed. Figure 5 A signal energy distribution graph at the head end provided for the embodiment of the present application is shown in Figure 5 The graph presents the energy distribution of the partial discharge signal at the head end over time. As can be seen from the curve, the energy distribution has obvious period characteristics. Energy peaks appear at some time points, indicating that partial discharge pulses are generated in these periods. The energy distribution curve corresponds to the pulse characteristics in the original signal, and the peak position reflects the occurrence time of the partial discharge pulse. Figure 6 A signal energy distribution graph at the tail end provided for the embodiment of the present application is shown in Figure 6 The graph shows the energy variation of the partial discharge signal at the tail end over time. The amplitude and number of peaks in the energy distribution curve reflect the energy characteristics and occurrence frequency of the partial discharge pulses in the tail end signal. Figure 7 A partial discharge signal graph at the head end after pulse segmentation provided for the embodiment of the present application is shown in Figure 7 After segmentation, the signal only retains the effective segment containing the partial discharge pulse. The partial discharge pulse characteristics are clearly visible in the waveform, the peak is obvious, and the noise is effectively eliminated. The signal becomes smooth and regular. Compared with the original signal, the data amount of the segmented signal is significantly reduced, while the key characteristics of the partial discharge pulse are completely retained. Figure 8 A partial discharge signal graph at the tail end after pulse segmentation provided for the embodiment of the present application is shown in Figure 8As shown, the time range of the segmented signal is also reduced, and the pulse characteristics in the waveform are highlighted. Compared with the original end signal, the segmented signal removes a large amount of redundant noise and invalid data, the energy is concentrated in the pulse area, and the data amount is effectively compressed. Figure 9 A positioning result graph provided for an embodiment of the present application is shown in the figure, which shows the positioning result curve of the partial discharge source. The horizontal axis represents the length from the cable head, and the vertical axis represents the modulus of the partial discharge positioning function. Figure 9 As shown, the figure shows the positioning result curve of the partial discharge source.

[0051] Figure 10 A structural schematic diagram of a cable partial discharge signal multi-end synchronous detection device provided for an embodiment of the present application is shown in the figure. Figure 10 As shown, the cable partial discharge signal multi-end synchronous detection device comprises at least one processor and a memory in communication connection with the at least one processor. The memory stores instructions executable by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to: acquire original acquisition signals through signal acquisition devices at both ends of the cable to be tested, and pre-process the original acquisition signals; divide the pre-processed original acquisition signals into sliding time windows to determine the short-time energy and spectral entropy features corresponding to each sliding time window; construct a joint detection index based on the short-time energy and spectral entropy features, and perform partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the median and absolute deviation of the adaptive statistical threshold and the joint detection index are related; in the case that there is a partial discharge signal in the original acquisition signal, segment the pre-processed original acquisition signal into partial discharge signals; and perform partial discharge signal positioning based on the frequency domain reflection method and the partial discharge signals respectively corresponding to both ends of the cable to be tested.

[0052] A non-volatile computer storage medium provided by an embodiment of the present application stores computer executable instructions. The computer executable instructions are configured to: acquire original acquisition signals through signal acquisition devices at both ends of the cable to be tested, and pre-process the original acquisition signals; divide the pre-processed original acquisition signals into sliding time windows to determine the short-time energy and spectral entropy features corresponding to each sliding time window; construct a joint detection index based on the short-time energy and spectral entropy features, and perform partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the median and absolute deviation of the adaptive statistical threshold and the joint detection index are related; in the case that there is a partial discharge signal in the original acquisition signal, segment the pre-processed original acquisition signal into partial discharge signals; and perform partial discharge signal positioning based on the frequency domain reflection method and the partial discharge signals respectively corresponding to both ends of the cable to be tested.

[0053] The various embodiments in the application are described in a progressive manner, and the same or similar parts among the various embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, the device, apparatus, and non-transitory computer storage medium embodiments are described simply because they are basically similar to the method embodiments, and the relevant parts can be referred to the part of the method embodiments.

[0054] The above only describes the embodiments of the application and is not intended to limit the application. The embodiments of the application can be variously changed and modified by those skilled in the art. The modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for multi-terminal synchronous detection of partial discharge signals in a cable, characterized in that, The method comprises: obtaining original acquisition signals through signal acquisition devices arranged at two ends of the cable to be tested, and preprocessing the original acquisition signals; dividing the preprocessed original acquisition signals into sliding time windows, and determining the short-time energy and spectral entropy features corresponding to each sliding time window; constructing a joint detection index based on the short-time energy and the spectral entropy features, and performing partial discharge signal recognition based on an adaptive statistical threshold and the joint detection index; wherein the adaptive statistical threshold is related to the median and absolute deviation of the joint detection index; in the case that the original acquisition signals contain the partial discharge signal, segmenting the partial discharge signal from the preprocessed original acquisition signals; locating the partial discharge signal based on the frequency domain reflection method and the partial discharge signals corresponding to the two ends of the cable to be tested.

2. The method of claim 1, wherein the method comprises: The method comprises: transmitting the original acquisition signal collected by the first signal acquisition device arranged at the first end of the cable to be tested to the host computer; and transmitting the original acquisition signal collected by the second signal acquisition device arranged at the second end of the cable to be tested to the slave computer, so that the slave computer segments the partial discharge signal from the original acquisition signal and sends the segmented partial discharge signal to the host computer; wherein the host computer and the slave computer are time-synchronized through a clock signal.

3. The method of claim 1, wherein the method comprises: The method comprises: filtering the power frequency signal and low frequency signal in the original acquisition signal through a high-pass filter; performing multi-layer wavelet packet decomposition on the filtered signal based on a preset mother wavelet to obtain frequency band coefficients, and obtaining an adaptive threshold based on the noise level of each frequency band; performing soft threshold compression on the frequency noise coefficients less than the adaptive threshold, and retaining the partial discharge pulse characteristic coefficients higher than the adaptive threshold, and reconstructing the denoised signal through inverse wavelet packet.

4. The method of claim 1, wherein the method is characterized by, The method comprises: dividing the preprocessed original acquisition signal into a plurality of sliding time windows based on a preset sliding step; determining the short-time energy corresponding to each time window through the sum of squares of the signal amplitudes in each sliding time window; performing short-time Fourier transform on the signal corresponding to each sliding time window to obtain a time-frequency matrix corresponding to each sliding time window; determining the normalized power spectrum distribution corresponding to each time-frequency matrix in a preset analysis frequency band, and obtaining the spectral entropy feature corresponding to each sliding time window based on the normalized power spectrum distribution.

5. The method of claim 1, wherein the method comprises: The method comprises: performing first ratio processing on the short-time energy corresponding to each sliding time window with respect to the maximum short-time energy; and performing second ratio processing on the spectral entropy feature corresponding to each sliding time window with respect to the maximum spectral entropy feature; The first ratio processing result and the second ratio processing result are superimposed by a weight coefficient to obtain the joint detection index.

6. The method of claim 1, wherein, Before the partial discharge signal identification based on the adaptive statistical threshold and the joint detection index, the method further comprises: The median and the absolute deviation of the joint detection index corresponding to different detection ends are determined. The median and the absolute deviation are superimposed based on a preset adjustment coefficient to obtain the adaptive statistical threshold corresponding to two ends of the cable to be detected.

7. The method of claim 1, wherein the method is a multi-terminal synchronous detection method for detecting partial discharge signals of a cable. The method for performing partial discharge signal segmentation on the preprocessed original acquisition signal comprises: In the preprocessed original acquisition signal, the earliest occurrence time of the partial discharge signal corresponding to two ends of the cable to be detected is determined. Based on the full length of the cable, the wave speed and the earliest occurrence time of the partial discharge signal, the effective time period corresponding to each original acquisition signal is determined. Based on the effective time period, the partial discharge signal segmentation is performed on each original acquisition signal.

8. The method of claim 7, wherein the plurality of synchronous detection is performed at a plurality of different frequencies. The method for performing partial discharge signal positioning based on the frequency domain reflection method and the partial discharge signal corresponding to two ends of the cable to be detected comprises: The union set of the effective time period corresponding to each partial discharge signal is taken as a unified time axis, and the missing sampling points are filled to align the partial discharge signals in the time domain. The frequency domain expression corresponding to the aligned partial discharge signal is obtained by Hilbert transform and discrete Fourier transform, and the cable transfer function is constructed based on the frequency domain expression. The cable transfer function is subjected to frequency spectrum analysis to construct a partial discharge positioning function, and the partial discharge signal positioning is performed based on the partial discharge positioning function.

9. A multi-terminal synchronous detection device for partial discharge signals of a cable, characterized in that The device comprises a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute the method of any one of claims 1-8.

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions comprising instructions for: receiving a request to access a file; determining whether the file is stored in a cache; and in response to determining that the file is stored in the cache, providing access to the file from the cache. The computer executable instructions can execute the method of any one of claims 1-8. The computer executable instructions can execute the method of any one of claims 1-8.

Images