An overhead line partial discharge monitoring indicator automatic detection system and detection method

Abstract

The application discloses an overhead line partial discharge monitoring indicator automatic detection system and a detection method, and particularly relates to the technical field of discharge detection. The system collects ultrasonic original signals of the surface of an insulator and the structure area of an overhead line, amplifies and filters the signals to generate time domain pretreatment signals, and obtains multi-frequency band decomposition signals by using a frequency domain feature extraction method. The system extracts multiple reflection features and scattering features based on the multi-frequency band decomposition signals to form path related signal features. The system performs waveform distortion compensation and energy attenuation compensation on the ultrasonic signals according to the path related signal features to obtain corrected partial discharge feature signals. The system performs integral operation on the corrected partial discharge feature signals in a preset time window to generate a partial discharge intensity integral value sequence, compares the sequence with a preset threshold value, and automatically determines whether to trigger an alarm control action of the overhead line partial discharge monitoring indicator, so that automatic identification and alarm of the partial discharge state of the overhead line are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of discharge detection, and more particularly, to an overhead line partial discharge monitoring indicator automatic detection system and method. BACKGROUND

[0002] In the operation process of the overhead line of the power system, the partial discharge is an important factor affecting the insulation performance of the insulator and the line structure. The monitoring of the partial discharge of the overhead line is usually judged by ultrasonic detection to determine the insulation condition. In the actual operation environment, due to the complexity of the insulator surface and the line structure, the ultrasonic signal is easily affected by the multiple reflections, scattering and other complex effects in the propagation process, resulting in differences between the finally detected ultrasonic signal and the actual partial discharge source signal.

[0003] In the prior art, the ultrasonic signal is affected by the multiple path reflections and scattering in the propagation process in the complex structure environment of the overhead line, causing the waveform distortion and energy attenuation of the ultrasonic signal, resulting in the difficulty in accurately identifying the real partial discharge characteristic signal, and further causing the misjudgment or omission of the partial discharge monitoring, and reducing the reliability and accuracy of the partial discharge monitoring. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide an overhead line partial discharge monitoring indicator automatic detection system and method to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] An overhead line partial discharge monitoring indicator automatic detection method, comprising the following steps:

[0007] S1: collecting the ultrasonic original signal of the insulator surface and the line structure area of the overhead line, and performing amplification and filtering processing on the ultrasonic original signal to generate a time domain pretreatment signal;

[0008] S2: based on the time domain pretreatment signal, using a frequency domain feature extraction method to decompose the frequency components in the ultrasonic signal, and outputting a multi-band decomposition signal;

[0009] S3: based on the multi-band decomposition signal, analyzing the propagation path of the ultrasonic signal on the insulator surface and the line structure, identifying the multiple reflection and scattering characteristics, and outputting the path related signal characteristics;

[0010] S4: based on the path related signal characteristics, compensating the waveform distortion and energy attenuation of the ultrasonic signal, and outputting the corrected partial discharge characteristic signal;

[0011] S5: Based on the corrected partial discharge characteristic signal, the integral value of the corrected partial discharge characteristic signal in the preset time window is calculated, and a partial discharge intensity integral value sequence is generated;

[0012] S6: According to the partial discharge intensity integral value sequence and the preset threshold value, it is judged whether the alarm control action of the partial discharge monitoring indicator is triggered, and the alarm state of the monitoring indicator is output.

[0013] In a preferred embodiment, S2, specifically:

[0014] Performing fast Fourier transform on the time domain pretreatment signal converts the time domain pretreatment signal from time domain to frequency domain to obtain the frequency spectrum of the ultrasonic signal;

[0015] According to the preset frequency band division rule, the frequency spectrum of the ultrasonic signal is divided into multiple frequency bands, and each frequency band after division corresponds to a frequency sub-band signal;

[0016] All frequency sub-band signals are summarized to form a multi-band decomposition signal.

[0017] In a preferred embodiment, S3, specifically:

[0018] Using each frequency sub-band signal in the multi-band decomposition signal, the time delay sequence of the ultrasonic signal in the propagation process on the surface of the insulator and the line structure is calculated respectively, and the multiple reflection characteristics are detected according to the time delay sequence;

[0019] Using each frequency sub-band signal in the multi-band decomposition signal, the energy distribution change of the ultrasonic signal in the propagation process on the surface of the insulator and the line structure is calculated respectively, and the scattering characteristics are detected according to the energy distribution change;

[0020] Integrate the detected multiple reflection characteristics and scattering characteristics to generate path-related signal characteristics.

[0021] In a preferred embodiment, S4, specifically:

[0022] Using the time delay sequence in the path-related signal characteristics, the time offset of the ultrasonic signal in each frequency sub-band is calculated;

[0023] According to the time offset, the ultrasonic signal is phase corrected and time realigned to correct the waveform distortion caused by multiple reflections and scattering in the propagation path;

[0024] Using the energy distribution change in the path-related signal characteristics, the attenuation coefficient of the ultrasonic signal in each frequency sub-band is calculated;

[0025] According to the attenuation coefficient, the ultrasonic signal is amplitude scaled and energy normalized to recover the energy loss caused by the propagation path;

[0026] The ultrasonic signal, after waveform distortion compensation and energy attenuation compensation, is integrated to output the corrected partial discharge characteristic signal.

[0027] In a preferred embodiment, S5 specifically refers to:

[0028] Based on the corrected partial discharge characteristic signal, a preset time window is set for calculating the integral value of partial discharge intensity.

[0029] The corrected partial discharge characteristic signal is segmented according to the length of a preset time window to obtain multiple time window signal segments;

[0030] Perform integration on each time window signal segment to calculate the integral value of the absolute value of the signal amplitude within each time window signal segment;

[0031] Record the integral values ​​of the signal segments in all time windows to generate a sequence of partial discharge intensity integral values.

[0032] In a preferred embodiment, S6 specifically refers to:

[0033] Each partial discharge intensity integral value in the generated partial discharge intensity integral value sequence is compared with a preset threshold.

[0034] When at least three consecutive partial discharge intensity integral values ​​in the partial discharge intensity integral value sequence exceed the preset threshold, the alarm triggering condition is determined to be met.

[0035] When the alarm triggering conditions are met, an alarm control signal is generated and sent to the partial discharge monitoring indicator, setting the current state of the partial discharge monitoring indicator to the alarm state.

[0036] When the alarm triggering conditions are not met, maintain the current state of the partial discharge monitoring indicator.

[0037] On the other hand, the present invention provides an automatic detection system for overhead line partial discharge monitoring indicators, comprising:

[0038] Acquisition and filtering module: Acquires raw ultrasonic signals from the surface of overhead line insulators and the line structure area, and performs amplification and filtering on the raw ultrasonic signals to generate time-domain preprocessed signals;

[0039] Frequency domain decomposition module: Based on the time domain preprocessed signal, it uses frequency domain feature extraction methods to decompose the frequency components in the ultrasonic signal and outputs a multi-band decomposed signal;

[0040] Path Feature Module: Based on multi-band decomposition signals, analyze the propagation path of ultrasonic signals on insulator surfaces and line structures, identify multiple reflections and scattering characteristics, and output path-related signal features;

[0041] The distortion compensation module compensates the waveform distortion and energy attenuation of the ultrasonic signal based on the path-related signal characteristics, and outputs the corrected partial discharge characteristic signal.

[0042] The integral sequence module calculates the integral value of the corrected partial discharge characteristic signal within a preset time window based on the corrected partial discharge characteristic signal, and generates a partial discharge intensity integral value sequence.

[0043] The alarm judgment module judges whether to trigger the alarm control action of the partial discharge monitoring indicator according to the partial discharge intensity integral value sequence and the preset threshold, and outputs the alarm state of the monitoring indicator.

[0044] The technical effects and advantages of the overhead line partial discharge monitoring indicator automatic detection system and method of the present application are as follows:

[0045] The signal-to-noise ratio of the partial discharge related signal is improved by amplifying and filtering the ultrasonic raw signal on the surface of the insulator and the structure area of the line. The frequency components in the ultrasonic signal are decomposed using a frequency domain feature extraction method to obtain a multi-band decomposition signal, which separates different frequency components and distinguishes between structural noise and partial discharge characteristics. The propagation path of the ultrasonic signal on the surface of the insulator and the structure is analyzed based on the multi-band decomposition signal to identify multiple reflection and scattering characteristics, which can describe the source of signal distortion from the propagation path level. The waveform distortion and energy attenuation of the ultrasonic signal are compensated based on the path-related signal characteristics to obtain a corrected partial discharge characteristic signal that is closer to the actual discharge process, thereby providing a basis for quantitative characterization. The corrected partial discharge characteristic signal is integrated within a preset time window to form a partial discharge intensity integral value sequence, which can distinguish between accidental interference and sustained discharge. By comparing the partial discharge intensity integral value sequence with the preset threshold and driving the partial discharge monitoring indicator to automatically alarm, the accuracy of the overhead line partial discharge monitoring and the reliability of the on-site indication are improved. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 The present application is an overhead line partial discharge monitoring indicator automatic detection method schematic diagram.

[0047] Figure 2 The present application is an overhead line partial discharge monitoring indicator automatic detection system structure schematic diagram. DETAILED DESCRIPTION

[0048] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application. Embodiment 1

[0049] Figure 1 An automatic detection method of an overhead line partial discharge monitoring indicator is given, which comprises the following steps:

[0050] S1: collecting ultrasonic original signals of an insulator surface and a line structure area of an overhead line, and performing amplification and filtering processing on the ultrasonic original signals to generate a time domain pretreatment signal;

[0051] S2: based on the time domain pretreatment signal, decomposing frequency components in the ultrasonic signal by using a frequency domain feature extraction method to output a multi-band decomposition signal;

[0052] S3: based on the multi-band decomposition signal, analyzing a propagation path of the ultrasonic signal on the insulator surface and the line structure, identifying multiple reflection and scattering characteristics, and outputting path-related signal characteristics;

[0053] S4: based on the path-related signal characteristics, compensating waveform distortion and energy attenuation of the ultrasonic signal, and outputting a corrected partial discharge characteristic signal;

[0054] S5: based on the corrected partial discharge characteristic signal, calculating an integral value of the corrected partial discharge characteristic signal in a preset time window to generate a partial discharge intensity integral value sequence;

[0055] S6: according to the partial discharge intensity integral value sequence and a preset threshold, judging whether to trigger an alarm control action of the partial discharge monitoring indicator, and outputting an alarm state of the monitoring indicator.

[0056] S1: collecting ultrasonic original signals of an insulator surface and a line structure area of an overhead line, and performing amplification and filtering processing on the ultrasonic original signals to generate a time domain pretreatment signal, comprising:

[0057] collecting ultrasonic original signals of an insulator surface and a line structure area of an overhead line;

[0058] The piezoelectric ultrasonic sensor is installed at the key position of the surface of the insulator and the structure of the overhead line, such as the metal accessory of the insulator string or the vicinity of the conductor connection point, to capture the ultrasonic signal generated by the partial discharge event. The sensitivity range of the piezoelectric ultrasonic sensor is 50 dB to 80 dB, and the frequency response range is 20 kHz to 100 kHz, which is set based on the characteristics of typical partial discharge ultrasonic signals, such as the ultrasonic signal energy generated by partial discharge mainly concentrated in the 40 kHz to 80 kHz frequency band. The sensor output is an analog electrical signal. The analog electrical signal is converted into a digital signal using an analog-to-digital converter, the sampling rate of which is set to 1 MHz, the sampling resolution is set to 12 bits, and the sampling time window is set to 10 milliseconds. The setting of the sampling rate is based on the Nyquist sampling theorem, which ensures that the sampling rate is at least twice the highest frequency component of the signal; for example, the highest frequency component of the ultrasonic signal is 100 kHz, so the sampling rate is set to 1 MHz to meet the requirements. The setting method of the sampling time window includes selecting a long enough time to capture the complete signal waveform based on the typical duration of the partial discharge event; for example, the partial discharge pulse width is usually 1 to 100 microseconds, so a 10 millisecond window can cover multiple period signals. The collected ultrasonic raw signal is stored in the form of a digital sequence, and the sequence length is determined by the sampling rate and the time window, for example, a sampling rate of 1 MHz and a time window of 10 milliseconds correspond to 10000 sampling points.

[0059] Amplification and filtering processing is performed on the ultrasonic raw signal, and the signal after amplification and filtering processing is output as a time domain preprocessed signal;

[0060] Amplification processing is performed by a programmable gain amplifier circuit, and the setting method of the amplification factor includes first calculating the average amplitude of the ultrasonic raw signal, and adjusting the amplification factor according to the input voltage range of the data acquisition system so that the signal amplitude occupies 70% to 90% of the input range. For example, the input range of the data acquisition system is ±1.5 V, and the average amplitude of the ultrasonic raw signal is 10 mV, then the amplification factor can be set to 500 times, so that the average amplitude of the amplified signal reaches 1.5 V. An operational amplifier is used to build an inverting or non-inverting amplification structure, and a resistor network with an accuracy of 1% is used to provide stable gain.

[0061] The filtering process is performed by a digital band-pass filter, the filter type is Butterworth filter, and the order is set to 4. The cutoff frequency setting method of the band-pass filter includes selecting a passband frequency to retain useful signals and suppress interference based on the frequency range and noise characteristics of the ultrasonic signal; for example, partial discharge ultrasonic signals are mainly distributed in 40 kHz to 80 kHz, so the lower limit frequency of the passband is set to 30 kHz, and the upper limit frequency of the passband is set to 90 kHz. The filter coefficient is calculated using a bilinear transformation method to convert the analog filter transfer function to a digital filter transfer function, the passband ripple is set to 1 dB, and the stopband attenuation is set to 40 dB. The filtering process is performed in the digital domain, the input is the amplified digital signal, and the output is the filtered signal. The signal after amplification and filtering is called a time-domain preprocessed signal, which is retained in the time-domain representation, the amplitude range is normalized to the full-scale range of the data acquisition system, and the signal length is consistent with the original signal, which is 10000 sampling points.

[0062] S2: Based on the time-domain preprocessed signal, the frequency components in the ultrasonic signal are decomposed using a frequency-domain feature extraction method, and a multi-band decomposed signal is output, including:

[0063] Performing fast Fourier transform on the time-domain preprocessed signal converts the time-domain preprocessed signal from time domain to frequency domain to obtain the frequency spectrum of the ultrasonic signal;

[0064] The time-domain preprocessed signal data sequence is zero-padded to the nearest power of 2 length, for example, if the original data length is 10000 points, 6384 zero-amplitude sampling points are added at the end of the data to expand the total data length to 16384 sampling points, and 16384 is the 14th power of 2; the zero-padded data sequence is arranged in reverse binary order to form a reverse bit sequence data sequence; the reverse bit sequence data sequence is used as the starting data, and the butterfly operation is performed in stages, the interval and number of butterfly operations in each stage vary gradually according to the data length, and finally a complex sequence in frequency domain is output, which is the frequency spectrum of the ultrasonic signal. The frequency spectrum is represented in complex form, and the length is consistent with the length of the zero-padded data sequence, for example, the length of the zero-padded data sequence is 16384 sampling points, so the length of the frequency spectrum sequence is also 16384 frequency spectrum data points. Each data point in the frequency spectrum sequence represents the amplitude and phase information of the corresponding frequency.

[0065] According to a predetermined frequency band division rule, the frequency spectrum of the ultrasonic signal is divided into a plurality of frequency bands, and each frequency band after division corresponds to a frequency sub-band signal;

[0066] The setting method of the frequency band division rule is: firstly, the starting frequency and the ending frequency of the division are determined, and the frequency response range of the piezoelectric ultrasonic sensor and the passband frequency of the digital band-pass filter are determined, for example, the frequency response range of the piezoelectric ultrasonic sensor is 20 kHz to 100 kHz, and the passband frequency range of the digital band-pass filter is 30 kHz to 90 kHz, so the starting frequency of the divided frequency band is set to 30 kHz, and the ending frequency is set to 90 kHz; the number of frequency bands is determined according to the frequency range of the division and the frequency band resolution requirement, and the determination method of the number of frequency bands is: the number of frequency bands is determined according to the principle that each frequency band signal can clearly reflect different frequency characteristics, for example, the width of each frequency band is set to 10 kHz, and in the frequency range of 30 kHz to 90 kHz, a total of 6 frequency bands are divided, which are 30 kHz to 40 kHz, 40 kHz to 50 kHz, 50 kHz to 60 kHz, 60 kHz to 70 kHz, 70 kHz to 80 kHz and 80 kHz to 90 kHz; finally, the frequency spectrum data sequence is divided into a plurality of corresponding frequency bands according to the determined frequency band boundary, each frequency band contains all frequency data points in a specific frequency range, and the determination method of the number of frequency data points is: the frequency spectrum data points and the frequency interval are determined based on the relationship between the frequency resolution and the data length, for example, when the sampling rate is 1 MHz and the data sequence length after zero padding is 16384, the frequency spectrum resolution is the sampling rate divided by the data length, that is, 61.035 Hz, therefore, the number of frequency data points contained in each 10 kHz frequency band is about 164.

[0067] For each set of frequency spectrum data points corresponding to each frequency band, the set of frequency spectrum data points is placed at the corresponding frequency position, and the amplitude values of the frequency spectrum data points of other frequency bands are set to zero, thereby forming a frequency domain sequence containing only the frequency spectrum information of the frequency band; the inverse fast Fourier transform is performed on the formed frequency domain sequence to obtain the corresponding sub-band signal of the frequency band in the time domain, and the sub-band signal length is consistent with the data sequence length after zero padding, for example, the length of each sub-band signal is 16384 sampling points; finally, the end invalid data points introduced due to the initial zero padding in the inverse transform result are removed, and only the first 10000 sampling points are retained, that is, the effective representation of the frequency sub-band signal in the time domain is obtained. The above operations are sequentially performed on each frequency band to obtain the time domain frequency sub-band signal corresponding to each frequency band.

[0068] All frequency sub-band signals are summarized to form a multi-frequency band decomposition signal;

[0069] The multi-band decomposed signal is in the form of arranging each frequency sub-band signal in order of frequency interval to form a signal set with ordered frequency, and each sub-band signal is marked with a corresponding frequency band boundary, such as including a sub-band signal of a frequency band of 30 kHz to 40 kHz, a sub-band signal of a frequency band of 40 kHz to 50 kHz, a sub-band signal of a frequency band of 50 kHz to 60 kHz, a sub-band signal of a frequency band of 60 kHz to 70 kHz, a sub-band signal of a frequency band of 70 kHz to 80 kHz, and a sub-band signal of a frequency band of 80 kHz to 90 kHz.

[0070] S3: Based on the multi-band decomposed signal, analyze the propagation path of the ultrasonic signal on the surface of the insulator and the line structure, identify the multiple reflection and scattering characteristics, and output the path-related signal characteristics, including:

[0071] Using each frequency sub-band signal in the multi-band decomposed signal, the time delay sequence of the ultrasonic signal in the propagation process on the surface of the insulator and the line structure is calculated respectively, and the multiple reflection characteristics are detected according to the time delay sequence;

[0072] For each frequency sub-band signal, a corresponding reference signal is selected. The reference signal is determined by extracting from a pre-acquired standard state ultrasonic signal. The acquisition conditions of the standard state ultrasonic signal are as follows: when there is no partial discharge on the surface of the insulator and the line structure region of the overhead line, the piezoelectric ultrasonic sensor installed at the same position is used to collect the signal under the stable operating condition at a sampling rate of 1 MHz and a sampling time window of 10 milliseconds. Taking the standard state ultrasonic signal as a reference, cross-correlation operation is performed on each frequency sub-band signal and the reference signal. The specific implementation method of the cross-correlation operation is as follows: the frequency sub-band signal and the reference signal are converted to the frequency domain by fast Fourier transform, the conjugate multiplication of the two signal spectrum sequences is calculated, and then the inverse fast Fourier transform is performed to convert back to the time domain to obtain the cross-correlation function sequence. The position of the peak value in the cross-correlation function sequence determines the time delay of each frequency sub-band signal relative to the reference signal, and the time delay sequence of each frequency sub-band signal is obtained. For example, if the maximum peak value of the cross-correlation function sequence is located at the 200th sampling point, the time delay of the corresponding frequency sub-band signal is calculated to be 200 microseconds under the condition of a sampling rate of 1 MHz. The time delay values calculated for each frequency sub-band signal are combined to form a complete time delay sequence.

[0073] The multiple reflection feature is defined as a plurality of time delay differences exhibited by multiple echoes generated by the ultrasonic signal during propagation on the surface of the insulator and the line structure due to the interfaces of different impedance materials. The detection method of the multiple reflection feature is: first, statistical analysis is performed on the time delay values of each frequency sub-band signal in the time delay sequence, the statistical method includes calculating the probability distribution of the delay values, and according to the case that a plurality of peak values are present in the probability distribution, it is judged that multiple reflections have occurred. For example, when there are two or more peak value points with amplitudes higher than the average probability in the time delay probability distribution curve of a specific frequency sub-band signal, it indicates that the frequency sub-band signal has a multiple reflection feature. By setting a time delay difference threshold, the difference between the time delay values corresponding to each peak value is compared with the time delay difference threshold, and when the difference is greater than or equal to the time delay difference threshold, it is determined that there is multiple reflection. The determination method of the time delay difference threshold is: according to historical experimental data, the typical reflection delay difference values of different material interfaces in the ultrasonic signal propagation are measured, and the average value of the experimentally measured delay difference values is taken as the time delay difference threshold.

[0074] Each frequency sub-band signal in the multi-band decomposition signal is used to calculate the energy distribution change of the ultrasonic signal during propagation on the surface of the insulator and the line structure, and the scattering feature is detected according to the energy distribution change.

[0075] The energy distribution change is defined as the change trend of the ultrasonic signal intensity gradually attenuating along the propagation path due to factors such as surface roughness of the insulator material, differences in structural surface morphology, or local defects in the material. The calculation method of the energy distribution change is: first, the signal energy of each frequency sub-band signal is calculated respectively, the calculation method of the signal energy is: the amplitude of the frequency sub-band signal at each sampling point is squared and summed, and then divided by the number of sampling points; the distribution curve of the signal energy along the propagation path is plotted with the propagation path length as the abscissa and the signal energy as the ordinate, the path length is determined by the time delay of the ultrasonic wave on the propagation path combined with the ultrasonic signal propagation speed, for example, the propagation speed of the ultrasonic signal in the insulator material can be obtained by experiment; the corresponding propagation path length is calculated by the experimentally measured propagation time and speed, and the energy distribution curve is plotted.

[0076] The scattering feature is defined as a feature in the energy distribution curve that shows a significant decrease in signal energy or a point of energy jump. The scattering feature detection method is to perform a first-order differential operation on the drawn energy distribution curve, calculate the difference in signal energy at adjacent propagation path lengths and divide by the path length difference, that is, obtain the energy change rate by difference calculation; compare the energy change rate curve with a threshold value, and the threshold value determination method is to determine the maximum value of the energy change rate measured under non-scattering conditions in the preliminary experiment, for example, the maximum value of the energy change rate measured under non-scattering conditions is 0.01 joule per meter, then the threshold value is set to twice the maximum value of the energy change rate, that is, 0.02 joule per meter; when the energy change rate exceeds the set threshold value, it is determined that there is a scattering feature.

[0077] Integrate the detected multiple reflection features and scattering features to generate path-related signal features;

[0078] The time delay difference information corresponding to the multiple reflection features determined by each frequency sub-band signal, and the propagation path position and energy change rate information corresponding to the scattering features determined in the energy distribution curve, jointly constitute the path-related signal features. The detected multiple reflection features and scattering features are marked respectively, and the frequency sub-band signal frequency band range, the propagation path position where the feature occurs, and the numerical value of the feature corresponding to each feature are marked, for example, for the frequency sub-band signal in the frequency band of 30 kHz to 40 kHz, the time delay difference of the multiple reflection feature is recorded as 50 microseconds, the scattering feature position is located at the propagation path length of 0.3 meters, and the energy change rate is 0.025 joule per meter; after sequentially marking each frequency band frequency sub-band signal, the path-related signal features are formed.

[0079] S4: Based on the path-related signal features, compensate for the waveform distortion and energy attenuation of the ultrasonic signal, and output the corrected partial discharge feature signal, including:

[0080] Using the time delay sequence in the path-related signal features, calculate the time offset of the ultrasonic signal in each frequency sub-band;

[0081] For each frequency sub-band signal, the median value of the time delay is determined as the time offset of the corresponding frequency sub-band signal by statistical processing of the time delay sequence of the corresponding frequency sub-band signal. The time delay sequence of each frequency sub-band signal is sorted, and the delay value at the middle position is selected from the sorted time delay sequence as the time offset; for example, when the time delay sequence of the frequency sub-band signal contains 101 delay values, the 51st delay value in the sorted sequence is taken as the time offset. The time delay sequences of all frequency sub-band signals are calculated in sequence to obtain the time offsets corresponding to all frequency sub-band signals.

[0082] The ultrasonic signals are phase-corrected and time-realigned according to the time offset, to correct waveform distortion caused by multiple reflections and scattering in the propagation path;

[0083] For each frequency sub-band signal, the corresponding time offset is converted into a phase offset for compensation. The calculation method of the phase offset is as follows: the center frequency of each frequency sub-band signal and the corresponding time offset are used to calculate the phase offset according to the relationship between phase and frequency and time, i.e. the phase offset is equal to the frequency multiplied by the time offset and then multiplied by 360 degrees. For example, the center frequency of the frequency band from 40 kHz to 50 kHz is 45 kHz, and the time offset is 50 microseconds, so the phase offset is calculated as 45,000 Hz multiplied by 0.00005 seconds and then multiplied by 360 degrees, resulting in a phase offset of 810 degrees. After determining the phase offset, the frequency sub-band signal is phase-rotated by the corresponding angle in the frequency domain, i.e. the frequency spectrum data points of the frequency sub-band signal are rotated in the opposite direction of the phase offset of 810 degrees, so as to compensate for the phase distortion caused by multiple reflections and scattering.

[0084] The phase-corrected frequency sub-band signals are converted back to the time domain by inverse fast Fourier transform to form respective corrected time-domain signal sequences. Each corrected time-domain signal sequence is time-shifted along the time axis according to the corresponding time offset, i.e. the data of the sampling points in each sequence is moved in the negative direction of the time axis by the number of sampling points corresponding to the time offset. The determination method of the number of sampling points is as follows: the time offset corresponding to each frequency sub-band signal is multiplied by the sampling rate of the analog-to-digital converter; for example, when the sampling rate is 1 MHz, for a frequency sub-band signal with a time offset of 50 microseconds, the corrected time-domain signal sequence is time-shifted in the negative direction by 50 sampling points, so as to realize the synchronization and alignment of each frequency sub-band signal, thereby eliminating the waveform distortion caused by multiple reflections and scattering in the propagation path.

[0085] The energy distribution change in the path-dependent signal characteristics is used to calculate the attenuation coefficient of the ultrasonic signal in each frequency sub-band;

[0086] For each frequency sub-band signal, an exponential decay model of signal energy changing with propagation path length is obtained by energy distribution curve and least square fitting; the form of the decay model is that signal energy equals initial energy multiplied by negative exponential function with base e, wherein the exponential term of the negative exponential is attenuation coefficient multiplied by propagation path length. With all measured points of the energy distribution curve as input data, the best parameters of the model are obtained by least square fitting, wherein the best parameters include the attenuation coefficient. The specific implementation steps of the least square fitting are: constructing a data matrix of the energy distribution curve, calculating the sum of squares of residuals between the data matrix and the model function, and obtaining the attenuation coefficient by solving the optimization problem of minimizing the sum of squares of residuals; for example, for the frequency sub-band signal of 50 kHz to 60 kHz band, the attenuation coefficient is 0.03 per meter after fitting calculation, which indicates that the proportion of energy reduction per meter of the frequency band ultrasonic signal conforms to the exponential term coefficient of the exponential function, which is 0.03.

[0087] The ultrasonic signal is scaled in amplitude and normalized in energy according to the attenuation coefficient, so as to restore the energy loss caused by the propagation path;

[0088] First, the theoretical initial energy value of each frequency sub-band signal at the propagation path length of zero is determined, and the initial energy value is obtained by substituting the propagation path length into the decay model and setting the path length to zero; the initial energy of each frequency sub-band signal actually measured at the path length of zero is calculated, and the amplitude of each frequency sub-band signal is adjusted according to the ratio of the theoretical initial energy value to the actual initial energy value; for example, if the theoretical initial energy of the frequency band of 50 kHz to 60 kHz is 2 joules, and the actual measured initial energy is 1 joule, the amplitude of the frequency sub-band signal is amplified by 2 times as a whole to achieve the effect of energy compensation. The amplitude-scaled frequency sub-band signals are divided by the maximum amplitude of the frequency sub-band signals respectively, so that the amplitude range of each sub-band signal is uniformly normalized to 0 to 1.

[0089] The ultrasonic signals subjected to waveform distortion compensation and energy decay compensation processing are integrated, and the corrected partial discharge characteristic signal is output;

[0090] First, the amplitude-scaled and energy-normalized frequency sub-band signal sequences are arranged in the original frequency order to form a frequency-ordered compensated sub-band signal sequence set; the compensated sub-band signal sequence set is synthesized by linear superposition, and the values of each frequency sub-band signal at the same sampling point position are directly summed to obtain the synthesized complete time domain waveform sequence, which is the corrected partial discharge characteristic signal.

[0091] S5: based on the corrected partial discharge characteristic signal, calculating the integral value of the corrected partial discharge characteristic signal in a preset time window to generate a partial discharge intensity integral value sequence, including:

[0092] Based on the corrected partial discharge characteristic signal, a preset time window for calculating the partial discharge intensity integral value is set;

[0093] The length setting method of the preset time window is: the typical duration of the actual overhead line partial discharge event and the sampling parameters of the ultrasonic signal are comprehensively determined. Through a large number of early experimental data statistical analysis, the single complete discharge pulse duration range of the typical partial discharge signal is determined, for example, the typical discharge pulse duration range obtained by statistics is 50 microseconds to 200 microseconds; then according to the sampling rate of the analog-to-digital converter, it is ensured that the length of the preset time window can at least cover multiple continuous partial discharge pulses, so as to improve the accuracy and reliability of the integral calculation. For example, when the sampling rate of the analog-to-digital converter is 1MHz, each sampling point corresponds to 1 microsecond, then the length of the preset time window can be determined as 1 millisecond, which is enough to cover more than 5 typical partial discharge pulses.

[0094] The corrected partial discharge characteristic signal is segmented according to the length of the preset time window to obtain a plurality of time window signal segments;

[0095] First, the number of segments is calculated according to the total number of sampling points of the partial discharge characteristic signal and the length of the preset time window, and the number of segments is equal to the integer part of the quotient obtained by dividing the total length of the partial discharge characteristic signal by the length of a single time window; for example, the total length of the corrected partial discharge characteristic signal is 10000 sampling points, and the length of a single time window is 1 millisecond corresponding to 1000 sampling points, then the number of segments is determined as 10. The corrected partial discharge characteristic signal is divided into equal length, each segment contains an equal number of sampling points, and there is no overlap between each adjacent segment; for example, the first 1000 sampling points are divided into the first time window signal segment, the 1001th to 2000th sampling points are divided into the second time window signal segment, and so on, finally a plurality of equal length time window signal segments are obtained.

[0096] The integral operation is performed on each time window signal segment to calculate the integral value of the absolute value of the signal amplitude in each time window signal segment;

[0097] Taking absolute value of signal amplitude of all sampling points in each time window signal segment, an absolute value signal amplitude sequence is formed; for example, if the signal amplitudes of sampling points in a time window signal segment are -0.5 volts, 0.4 volts and -0.2 volts, the absolute value sequence obtained after taking absolute value is 0.5 volts, 0.4 volts and 0.2 volts. Numerical integral calculation is performed on the absolute value amplitude sequence, and the integral method adopts trapezoidal integral method. The sum of absolute value amplitudes of adjacent two sampling points in the time window signal segment is divided by 2, and then multiplied by the sampling interval time (i.e. sampling period), so as to calculate the integral subinterval area formed by each pair of adjacent sampling points; then all subinterval areas are added, and the integral value of the entire time window signal segment can be obtained. Integral operation is sequentially performed on each time window signal segment, and the integral value corresponding to each time window is obtained.

[0098] The integral values of all time window signal segments are recorded to generate a partial discharge intensity integral value sequence.

[0099] A data record table indexed by time window order is established, the integral value corresponding to each time window signal segment is recorded to the data record table, and the corresponding window index and time interval information are marked. For example, the first row in the generated partial discharge intensity integral value sequence record table corresponds to the first time window, the integral value is 0.75 microvolt seconds, the second row corresponds to the second time window, the integral value is 0.85 microvolt seconds, and the last time window is sequentially added to form a partial discharge intensity integral value sequence with complete time information.

[0100] In order to ensure the effectiveness and accuracy of the integral value, normalization processing is performed on the partial discharge intensity integral value sequence, all integral values in the partial discharge intensity integral value sequence are divided by the maximum value of the integral values in the sequence, so that the integral values in the sequence are uniformly between 0 and 1, and a normalized integral value sequence is obtained.

[0101] S6: According to the partial discharge intensity integral value sequence and the preset threshold value, it is judged whether the alarm control action of the partial discharge monitoring indicator is triggered, and the alarm state of the monitoring indicator is output, including:

[0102] Each partial discharge intensity integral value in the generated partial discharge intensity integral value sequence is compared with the preset threshold value;

[0103] The setting method of the preset threshold is: according to the statistical analysis of the integral values collected from a large number of insulators and line structure regions of overhead lines under normal operation state and partial discharge state, the maximum value of the integral value under normal state is determined, and a reasonable safety margin is set based on the maximum value to form an integral threshold that can effectively distinguish between normal state and abnormal partial discharge state. For example, through experiments, the maximum integral value of the partial discharge intensity integral value sequence under normal state is 0.3 microvolts-second, and a certain percentage of safety margin is added, for example, 50%, then the preset threshold can be set to 0.45 microvolts-second, to effectively reduce the possibility of false alarm.

[0104] When at least three partial discharge intensity integral values in the partial discharge intensity integral value sequence continuously exceed the preset threshold, it is determined that the alarm triggering condition is met;

[0105] The partial discharge intensity integral value sequence is sequentially scanned to identify the data segment whose integral values continuously exceed the preset threshold; for example, in the partial discharge intensity integral value sequence, the first to third integral values are 0.50 microvolts-second, 0.52 microvolts-second and 0.49 microvolts-second, respectively, at this time, the three integral values all exceed the preset threshold of 0.45 microvolts-second, which meets the condition of continuous three integral values exceeding the limit, that is, it is determined that the alarm triggering condition is met. If there are no three or more integral values continuously exceeding the preset threshold, it is determined that the alarm triggering condition is not met.

[0106] When the alarm triggering condition is met, an alarm control signal is generated and sent to the partial discharge monitoring indicator, and the current state of the partial discharge monitoring indicator is set to the alarm state;

[0107] The digital representation of the alarm control signal is defined, for example, the alarm state is represented by a standard binary control signal, 1 represents the alarm state, and 0 represents the normal state; according to the alarm triggering condition determination result, the current state of the partial discharge monitoring indicator is changed from the original normal state to the alarm state, and the alarm control signal is set to 1. After the alarm control signal is generated, it is sent to the overhead line partial discharge monitoring indicator through the communication interface.

[0108] After the partial discharge monitoring indicator receives the alarm control signal, the current state is set to the alarm state. The alarm state includes but is not limited to: the red warning light of the indicator is on, the sound buzzer is started or the combination of the two, to ensure that the alarm information can be perceived by the on-site staff. For example, the red warning light and the sound buzzer can be used in combination, when the alarm state is triggered, the red warning light is turned on in a flickering manner, and the buzzer emits a warning sound at a fixed frequency, thereby improving the on-site alarm effect and ensuring that the staff respond quickly and take appropriate measures.

[0109] When the alarm triggering condition is not met, the current state of the partial discharge monitoring indicator is maintained;

[0110] At this time, the alarm control signal remains as a digital signal 0, the current indication state of the partial discharge monitoring indicator maintains the original setting, and the alarm indicator light and the buzzer are both in the off or silent state. By maintaining the current normal state, unnecessary false alarm interference is effectively avoided, and normal monitoring work of the on-site staff is ensured not to be disturbed. Embodiment 2

[0111] The difference between the embodiment 2 and the embodiment 1 of the present application is that the embodiment 2 is to introduce an automatic detection system of an overhead line partial discharge monitoring indicator.

[0112] Figure 2 The structural schematic diagram of the automatic detection system of the overhead line partial discharge monitoring indicator is given, and the automatic detection system of the overhead line partial discharge monitoring indicator comprises:

[0113] The acquisition and filtering module: acquires the ultrasonic original signal of the surface of the insulator and the structure area of the line, and performs amplification and filtering processing on the ultrasonic original signal to generate a time domain pretreatment signal;

[0114] The frequency domain decomposition module: based on the time domain pretreatment signal, the frequency components in the ultrasonic signal are decomposed by using a frequency domain feature extraction method, and a multi-frequency band decomposition signal is output;

[0115] The path feature module: based on the multi-frequency band decomposition signal, the propagation path of the ultrasonic signal on the surface of the insulator and the structure of the line is analyzed, the multiple reflection and scattering features are identified, and a path related signal feature is output;

[0116] The distortion compensation module: based on the path related signal feature, the waveform distortion and energy attenuation of the ultrasonic signal are compensated, and a corrected partial discharge feature signal is output;

[0117] The integral sequence module: based on the corrected partial discharge feature signal, the integral value of the corrected partial discharge feature signal in a preset time window is calculated, and a partial discharge intensity integral value sequence is generated;

[0118] The alarm judgment module: according to the partial discharge intensity integral value sequence and a preset threshold, whether the alarm control action of the partial discharge monitoring indicator is triggered is judged, and the alarm state of the monitoring indicator is output.

[0119] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center through a wired (for example, infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0120] Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0121] Those of ordinary skill in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device, and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0122] In several embodiments provided in the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed ones can be indirect coupling or communication connection through some interfaces, devices, or modules, which can be electrical, mechanical, or other forms.

[0123] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0124] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0125] If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0126] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0127] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. An automatic detection method of an overhead line partial discharge monitoring indicator, characterized in that, The method comprises the following steps: S1: Collecting the ultrasonic raw signal of the insulator surface and the line structure area of the overhead line, and performing amplification and filtering processing on the ultrasonic raw signal to generate a time domain pretreatment signal; S2: Based on the time domain pretreatment signal, the frequency components in the ultrasonic signal are decomposed by using a frequency domain feature extraction method, and a multi-frequency band decomposition signal is output; S3: Based on the multi-frequency band decomposition signal, the propagation path of the ultrasonic signal on the insulator surface and the line structure is analyzed, the multiple reflection and scattering characteristics are identified, and a path-related signal feature is output; S4: Based on the path-related signal feature, the waveform distortion and energy attenuation of the ultrasonic signal are compensated, and a corrected partial discharge characteristic signal is output, specifically: Using the time delay sequence in the path-related signal feature, the time offset of the ultrasonic signal in each frequency sub-band is calculated; According to the time offset, the phase correction and time realignment of the ultrasonic signal are performed, and the waveform distortion caused by multiple reflections and scattering in the propagation path is corrected; Using the energy distribution change in the path-related signal feature, the attenuation coefficient of the ultrasonic signal in each frequency sub-band is calculated; According to the attenuation coefficient, the amplitude scaling and energy normalization of the ultrasonic signal are performed, and the energy loss caused by the propagation path is recovered; Integrate the ultrasonic signal after waveform distortion compensation and energy attenuation compensation processing, and output the corrected partial discharge characteristic signal; S5: Based on the corrected partial discharge characteristic signal, the integral value of the corrected partial discharge characteristic signal in the preset time window is calculated, and a partial discharge intensity integral value sequence is generated, specifically: Based on the corrected partial discharge characteristic signal, a preset time window for calculating the partial discharge intensity integral value is set; The corrected partial discharge characteristic signal is segmented according to the length of the preset time window to obtain a plurality of time window signal segments; Integrate each time window signal segment to calculate the integral value of the absolute value of the signal amplitude in each time window signal segment; Record the integral value of all time window signal segments to generate a partial discharge intensity integral value sequence; S6: According to the partial discharge intensity integral value sequence and the preset threshold, it is judged whether the alarm control action of the partial discharge monitoring indicator is triggered, and the alarm state of the monitoring indicator is output.

2. The overhead line partial discharge monitoring indicator automatic detection method according to claim 1, characterized in that, S2, specifically: Performing fast Fourier transform on the time domain pretreatment signal, converting the time domain pretreatment signal from time domain to frequency domain, and obtaining the frequency spectrum of the ultrasonic signal; According to the preset frequency band division rule, the frequency spectrum of the ultrasonic signal is divided into a plurality of frequency bands, and each frequency band after division corresponds to a frequency sub-band signal; All frequency sub-band signals are integrated to form a multi-frequency band decomposition signal.

3. The overhead line fault monitoring indicator automatic detection method according to claim 2, wherein, S3, specifically: Using each frequency sub-band signal in the multi-frequency band decomposition signal, the time delay sequence of the ultrasonic signal in the propagation process on the insulator surface and the line structure is calculated, and the multiple reflection characteristics are detected according to the time delay sequence; Using each frequency sub-band signal in the multi-frequency band decomposition signal, the energy distribution change of the ultrasonic signal in the propagation process on the insulator surface and the line structure is calculated, and the scattering characteristics are detected according to the energy distribution change; Integrate the detected multiple reflection characteristics and scattering characteristics to generate a path-related signal feature.

4. The overhead line partial discharge monitoring indicator automatic detection method according to claim 3, characterized in that, S6, specifically: Compare each partial discharge intensity integral value in the generated sequence of partial discharge intensity integral values with a preset threshold value; When at least three partial discharge intensity integral values in the sequence of partial discharge intensity integral values continuously exceed the preset threshold value, it is determined that the alarm triggering condition is met; When the alarm triggering condition is met, an alarm control signal is generated and sent to the partial discharge monitoring indicator, and the current state of the partial discharge monitoring indicator is set to an alarm state; When the alarm triggering condition is not met, the current state of the partial discharge monitoring indicator is maintained.

5. An automatic detection system for an overhead line partial discharge monitoring indicator for implementing the automatic detection method of an overhead line partial discharge monitoring indicator according to any one of claims 1 to 4, characterized in that, Comprise: Acquisition and filtering module: acquire ultrasonic raw signals of the surface of the insulator and the structure area of the line, and perform amplification and filtering processing on the ultrasonic raw signals to generate time domain preprocessed signals; Frequency domain decomposition module: based on the time domain preprocessed signals, decompose the frequency components in the ultrasonic signals using a frequency domain feature extraction method to output multi-frequency band decomposition signals; Path feature module: based on the multi-frequency band decomposition signals, analyze the propagation path of the ultrasonic signals on the surface of the insulator and the structure of the line, identify multiple reflection and scattering characteristics, and output path-related signal features; Distortion compensation module: based on the path-related signal features, compensate for the waveform distortion and energy attenuation of the ultrasonic signals to output corrected partial discharge feature signals; Integral sequence module: based on the corrected partial discharge feature signals, calculate the integral value of the corrected partial discharge feature signals within a preset time window to generate a sequence of partial discharge intensity integral values; Alarm judgment module: according to the sequence of partial discharge intensity integral values and the preset threshold value, judge whether to trigger the alarm control action of the partial discharge monitoring indicator, and output the alarm state of the monitoring indicator.

Images