CN-121656779-B - Acoustic-magnetic time difference calculation method based on signal quality self-adaption

CN121656779BCN 121656779 BCN121656779 BCN 121656779BCN-121656779-B

Abstract

The invention provides an acoustic-magnetic time difference calculation method based on signal quality self-adaption, and belongs to the technical field of cable fault detection. The method comprises the steps of obtaining acoustic signals and electromagnetic signals of partial discharge of a high-voltage cable, detecting the electromagnetic signals to determine trigger time, intercepting an acoustic signal section and a background noise signal section by taking the time as a reference, calculating a signal to noise ratio and a maximum peak value after preprocessing, comparing the signal to a preset threshold value, dynamically selecting a trigger point positioning scheme, determining an effective trigger point according to the scheme, calculating an original acousto-magnetic time difference, calculating the energy of the acoustic signals, and correcting the original time difference by combining reference energy to obtain a final time difference. According to the method, through the self-adaptive matching scheme, accuracy and reliability of time difference calculation are improved, different signal quality scenes are adapted, and application adaptability of complex field environments is enhanced.

Inventors

WANG FEI
WEI BI
ZHANG SHUO
ZHANG LIN
HE SEN
CHEN HONG
WU CHAO
YU JIANJUN
MENG CHAO
MA WEIGANG

Assignees

探博士电气科技(杭州)有限公司

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (7)

1. The signal quality self-adaption-based acousto-magnetic time difference calculation method is characterized by comprising the following steps of: acquiring an acoustic signal and an electromagnetic signal generated by partial discharge of a high-voltage cable, and performing trigger detection on the electromagnetic signal to determine the trigger moment of the electromagnetic signal; The signal preprocessing and quality evaluation step, namely intercepting a sound signal section and a background noise signal section with preset time length by taking the triggering time of the electromagnetic signal as a reference; Comparing the calculated signal-to-noise ratio and the maximum peak value with a preset threshold condition, and dynamically selecting a corresponding trigger point positioning scheme according to the comparison result; the preset threshold condition comprises a first signal-to-noise ratio threshold, a second signal-to-noise ratio threshold, a first peak-to-peak value threshold and a second peak-to-peak value threshold, wherein the trigger point positioning scheme comprises a direct peak positioning scheme, a similarity extremum matching positioning scheme and a segmentation correlation and accumulation average positioning scheme, and the corresponding trigger point positioning scheme is dynamically selected according to the comparison result, wherein the trigger point positioning scheme comprises the direct peak positioning scheme if the signal-to-noise ratio is larger than the first signal-to-noise ratio threshold and the maximum peak-to-peak value is larger than the first peak-to-peak value threshold, the direct peak positioning scheme is selected if the signal-to-noise ratio is not larger than the first signal-to-noise ratio threshold but is larger than the second signal-to-noise ratio threshold and the maximum peak-to-peak value is larger than the second peak-to-peak value threshold, and the similarity extremum matching positioning scheme is selected otherwise; The similar extremum matching positioning scheme comprises an extremum matching step, a gradient judging step, a trigger point judging step and a similarity judging step, wherein the extremum matching step is used for carrying out distribution matching on extremum points of an acquired sound signal and a reference sound signal, the distribution matching comprises the steps of matching the minima points of the two extremum points and reserving a minimum value point pair with position deviation within a preset range, positioning maximum value points between adjacent minima points based on the minimum value point pair, constructing respective extremum sequences, calculating position difference values of corresponding maximum value points and adjacent minimum value points in the two groups of sequences, reserving the maximum value points with position difference value deviation within the preset range as matched extremum point pairs, and calculating bilateral gradient symmetry scores of candidate peak points in the matched extremum point pairs; The segmentation correlation and accumulation average positioning scheme comprises a signal enhancement step, a signal enhancement step and a signal enhancement step, wherein the signal enhancement step is used for carrying out spectrum subtraction enhancement processing on the sound signal segment based on the background noise signal segment; the step of segment screening, namely carrying out segment cross-correlation analysis on the enhanced sound signals acquired for multiple times, calculating segment cross-correlation coefficients among the sound signals acquired for multiple times, removing abnormal signal segments with minimum correlation coefficients lower than the median and difference values exceeding a preset difference value, and obtaining screened signal segments; A trigger point positioning step of determining an effective trigger point of the sound signal according to the selected positioning scheme; And calculating and correcting the time difference, namely calculating the energy of the acoustic signal acquired at this time based on the acoustic signal segment, calculating the original acousto-magnetic time difference based on the effective trigger point and the trigger point of the electromagnetic signal, and carrying out energy self-adaptive correction on the original acousto-magnetic time difference by combining the energy of the acoustic signal and the preset reference energy to obtain the final acousto-magnetic time difference.
2. The method of claim 1, wherein the triggering detection includes detecting whether the magnitude of the electromagnetic signal exceeds a preset fixed threshold and determining a time when the fixed threshold is exceeded for the first time as the electromagnetic signal triggering time.
3. The method of claim 1, wherein the direct peak positioning scheme comprises: And in the sound signal section, judging a peak point with a first amplitude exceeding the preset multiple of the maximum peak value as the effective trigger point.
4. The method of claim 1, wherein the computing the bilateral gradient symmetry score comprises: And respectively taking a plurality of sampling points before and after the candidate peak point, calculating the symbol consistency proportion of the forward gradient and the backward gradient, and taking the proportion as the bilateral gradient symmetry score.
5. The acousto-magnetic time difference calculation method according to claim 1, wherein the spectrum subtraction enhancement processing includes: dividing a background noise signal section according to a preset frame length, screening out a stable noise frame, estimating a noise power spectrum based on the noise frame, and subtracting a fixed multiple of the noise power spectrum from the power spectrum of the sound signal section to realize signal enhancement.
6. The method for calculating the acousto-magnetic time difference according to claim 1, wherein the calculating the acoustic signal energy based on the acoustic signal segment specifically comprises: and calculating the square sum of the amplitudes of all sampling points in the sound signal segment, and taking the square sum as the sound signal energy of the current time.
7. The method of claim 1, wherein the reference energy is an acoustic signal energy corresponding to a time difference between the sound and the magnetic signal in the historical detection data.

Description

Acoustic-magnetic time difference calculation method based on signal quality self-adaption Technical Field The invention relates to the technical field of cable fault detection, in particular to an acousto-magnetic time difference calculation method based on signal quality self-adaption. Background High-voltage cables are a key part of power transmission networks of power systems, and the operation reliability of the high-voltage cables directly influences the safe power supply of the power network. Partial discharge is an important characterization of cable insulation degradation and is also a main cause of faults, so accurate detection and positioning of partial discharge are core requirements of safety guarantee of power equipment. The acoustic-magnetic time difference positioning technology becomes a main scheme of the partial discharge positioning of the high-voltage cable due to high positioning precision and wide application scene, and the core of the acoustic-magnetic time difference positioning technology is to capture the arrival time difference of acoustic signals and electromagnetic signals generated by partial discharge and reversely deduce the fault point position by utilizing the propagation speed difference of the acoustic signals and the electromagnetic signals. The existing acousto-magnetic positioning technology has obvious defects in practical application that on-site electromagnetic interference is prominent, background noise in complex industrial, underground or underwater scenes is complex and fluctuates greatly, so that the signal-to-noise ratio of an acoustic signal is low, trigger point judgment deviation is caused, positioning reliability is reduced, meanwhile, multi-focus hardware filtering optimization or post-data processing in the existing scheme is lack of self-adaptive capacity to dynamic change of signal quality, a hierarchical processing mechanism of different signal-to-noise ratio scenes is not established, diversified application environments cannot be flexibly adapted, and the problem of accurate positioning under complex noise is difficult to solve. Disclosure of Invention The invention provides an acoustic-magnetic time difference calculation method based on signal quality self-adaption, which is used for solving the problems of insufficient trigger point judgment deviation, insufficient positioning precision and reliability, lack of self-adaption capability on dynamic change of signal quality, hierarchical processing mechanism and the like caused by low signal-to-noise ratio of an acoustic signal in the prior art. In order to achieve the above purpose, the embodiment of the invention provides an acoustic-magnetic time difference calculation method based on signal quality adaptation, which comprises the following steps of a signal acquisition and trigger reference establishment step, a signal preprocessing and quality assessment step, a time difference calculation and correction step, a sound signal energy calculation and original magnetic energy correction step, wherein the sound signal segment and the background noise signal segment are cut off for a preset time length based on the trigger time of an electromagnetic signal, the signal noise ratio and the maximum peak value are preprocessed and calculated based on the preprocessed signals, the adaptive scheme selection step is used for comparing the calculated signal noise ratio and the maximum peak value with preset threshold conditions, a corresponding trigger point positioning scheme is dynamically selected according to the comparison result, the trigger point positioning step is used for determining the effective trigger point of the sound signal according to the selected positioning scheme, the time difference calculation and correction step is used for calculating the sound signal energy of the time difference based on the trigger point of the electromagnetic signal, and the sound signal energy of the time difference calculation and the original magnetic energy correction step is combined with the original acoustic-magnetic energy, and the original magnetic energy is finally obtained. Optionally, the triggering detection comprises detecting whether the amplitude of the electromagnetic signal exceeds a preset fixed threshold value, and determining the moment when the fixed threshold value is exceeded for the first time as the electromagnetic signal triggering moment. Optionally, the preset threshold condition comprises a first signal-to-noise ratio threshold, a second signal-to-noise ratio threshold, a first peak-to-peak value threshold and a second peak-to-peak value threshold, the trigger point positioning scheme comprises a direct peak positioning scheme, a similarity extremum matching positioning scheme and an accumulated average positioning scheme, the corresponding trigger point positioning scheme is dynamically selected according to the comparison result, the method comprises the steps that if the signal-