CN115078538A - High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE - Google Patents
High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE Download PDFInfo
- Publication number
- CN115078538A CN115078538A CN202210723765.2A CN202210723765A CN115078538A CN 115078538 A CN115078538 A CN 115078538A CN 202210723765 A CN202210723765 A CN 202210723765A CN 115078538 A CN115078538 A CN 115078538A
- Authority
- CN
- China
- Prior art keywords
- voltage cable
- corrosion
- guided wave
- aluminum sheath
- ultrasonic guided
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 229910052782 aluminium Inorganic materials 0.000 title claims abstract description 114
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 title claims abstract description 114
- 230000007797 corrosion Effects 0.000 title claims abstract description 95
- 238000005260 corrosion Methods 0.000 title claims abstract description 95
- 238000001514 detection method Methods 0.000 title claims abstract description 74
- 230000007547 defect Effects 0.000 claims abstract description 79
- 230000009466 transformation Effects 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 230000004927 fusion Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 13
- 230000006378 damage Effects 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 5
- 238000013135 deep learning Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 239000000284 extract Substances 0.000 abstract 1
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009659 non-destructive testing Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 230000005347 demagnetization Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000003302 ferromagnetic material Substances 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000005415 magnetization Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4409—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison
- G01N29/4418—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison with a model, e.g. best-fit, regression analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/26—Scanned objects
- G01N2291/262—Linear objects
- G01N2291/2626—Wires, bars, rods
Landscapes
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Signal Processing (AREA)
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, a system and a medium based on SWT-AE, wherein the method comprises the following steps: obtaining corrosion defect samples such as uniform corrosion, pitting corrosion, filiform corrosion and the like, and obtaining an original signal of the ultrasonic guided wave of the aluminum sheath of the high-voltage cable; converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, segmenting the guided wave signals by using a sliding window and performing discrete wavelet transformation; extracting statistical characteristics from the signal representation of each signal segment; and (4) performing defect identification regression by adopting a self-coding network, and finally identifying the corrosion defect of the high-voltage cable aluminum sheath. The method combines ultrasonic guided wave signal processing with a deep learning algorithm based on a sliding window wavelet transform-self-coding network, accurately extracts characteristic representation of the whole propagation process of guided wave signals under the condition of less damage to information quantity, and performs characteristic identification on the corrosion defect state of the high-voltage cable aluminum sheath by using local time-frequency characteristics of the guided wave signals.
Description
Technical Field
The invention belongs to the technical field of nondestructive testing, and particularly relates to a high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE.
Background
In recent years, the main insulation of the cable is damaged by the frequent and serious defects of the cable accessory caused by the corrosion defect of the joint of the tail pipe of the cable accessory and the metal aluminum sheath of the cable. At present, effective means for preventing and detecting the defects are unavailable, and the defects can be found only when the defects happen to be serious, so that how to accurately detect and identify the corrosion defects at the aluminum sheath of the high-voltage cable is urgent.
In the face of the harm caused by the corrosion of the aluminum sheath of the high-voltage cable, a new detection means for the aluminum sheath of the high-voltage cable in use is urgently needed. The nondestructive testing of the high-voltage cable aluminum sheath can protect the high-voltage cable aluminum sheath in use to the maximum extent, namely, the process of applying a certain testing technology and an analysis method to test the state characteristics of the high-voltage cable aluminum sheath and evaluating the state characteristics according to a certain criterion under the condition of not damaging the use state of the high-voltage cable aluminum sheath. Currently, the magnetic detection method, the eddy current detection method, the infrared detection method and the ultrasonic detection method are mainly applied to the nondestructive detection of the high-voltage cable. The magnetic detection method needs to be carried out in a ferromagnetic material, complex procedures of pre-magnetization, detection and demagnetization need to be carried out during detection, and meanwhile, the detection precision is not high, so that the magnetic detection is not suitable for detecting corrosion damage of high-voltage cable accessories; the coil does not need to be in direct contact with a measured object during eddy current detection, high-speed detection can be carried out, automation is easy to realize, but the coil is not suitable for parts with complex shapes, only the surface and near-surface defects of the conductive material can be detected, the detection depth is low, and the detection result is also easy to be interfered by the material and other factors; the infrared detection technology needs to use an excitation source, the excitation source and the thermal infrared imager are not provided with proper carriers and are difficult to use in the cable, the penetration force is poor, the infrared detection technology is mainly suitable for detecting shallow damage, the infrared radiation of a high-voltage cable accessory is easy to shield and is difficult to receive by a detector, and therefore the corrosion defect below the copper braided belt is difficult to detect. The ultrasonic guided wave method has the advantages of long propagation distance, large detection distance, high detection efficiency and the like, and can realize the positioning and damage evaluation of the corrosion defect of the aluminum sheath of the high-voltage cable. Meanwhile, researches show that the ultrasonic guided wave technology can have good sensitivity to early damage and micro defects. The ultrasonic guided waves are introduced into the corrosion damage detection of the high-voltage cable aluminum sheath, so that the corrosion defects can be effectively positioned and evaluated.
Signal processing techniques for feature processing and post-processing detection are required during the guided wave test. The deep learning can directly obtain the parameterized mapping relation between the original data and the detection result through the complex network structure and the extremely strong mapping capability. The deep learning algorithm is applied to guided wave detection, so that the detection precision and efficiency can be further improved.
Disclosure of Invention
The invention mainly aims to overcome the defects of the prior art and provide an ultrasonic guided wave detection method, a system and a medium for corrosion of a high-voltage cable aluminum sheath based on SWT-AE. And performing local time-frequency feature extraction on the obtained signal set by using sliding window wavelet transform, and completing unsupervised feature fusion by using a self-coding network model, thereby accurately obtaining the corrosion feature identification of the high-voltage cable aluminum sheath.
In order to achieve the purpose, the invention adopts the following technical scheme:
in one aspect of the invention, the invention provides a high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method based on SWT-AE, which comprises the following steps:
establishing an actual standard test block for the aluminum sheath of the high-voltage cable, collecting a corrosion defect sample of the aluminum sheath of the high-voltage cable, detecting the corrosion defect of the aluminum sheath of the high-voltage cable by using an ultrasonic guided wave detector, and acquiring an ultrasonic guided wave original signal of the aluminum sheath of the high-voltage cable;
converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, segmenting the guided wave signals by using a sliding window and performing discrete wavelet transformation;
extracting statistical characteristics from the signal representation of each signal segment;
performing defect identification regression by using a self-coding network, wherein the defect identification regression comprises unsupervised feature fusion and integral fine adjustment;
and identifying the corrosion defect of the high-voltage cable aluminum sheath.
As a preferable technical scheme, the high-voltage cable aluminum sheath corrosion defect sample comprises a sample with uniform corrosion, pitting corrosion and filiform corrosion, and is obtained by manually simulating or collecting an actual corrosion defect sample.
As a preferred technical solution, the segmenting of the guided wave signal by using the sliding window and performing the discrete wavelet transform specifically includes:
cutting the guided wave signal through a sliding window with the window length of k;
performing discrete wavelet transform, and setting its wavelet coefficient C ψ Extracting root mean square RMS, peak value P, kurtosis K, form factor W and absolute mean value X from the windowed high-voltage cable aluminum sheath ultrasonic guided wave signal a The statistical indicator is calculated as a signal characteristic by the following formula:
P＝max(C ψ )-min(C ψ )
preferably, when extracting statistical features from the signal representation of each signal segment, feature selection is performed according to the feature monotonicity of the training set.
As a preferred technical solution, the self-coding network includes an input layer, a plurality of encoder layers, a plurality of regression layers, and an output layer.
As an optimal technical scheme, selected features are input into a high-voltage cable aluminum sheath corrosion detection self-coding network model for feature fusion and defect identification, and a high-dimensional input signal X is obtained by an encoder in the following mode k ∈R n×1 Low dimensional representation of (c):
h k ＝f(W×X k +b 1 )
wherein h is k Is a k-dimensional feature representation and k<n, f are nonlinear activation functions, W is a weight matrix of the encoder, b 1 Is the bias of the hidden layer;
the reconstruction of the original input by a decoder in the high-voltage cable aluminum sheath corrosion detection self-coding network model is realized by the following steps:
wherein
Is an input reconstructed signal, W T Is the weight matrix of the decoder, h m Is an activation function of the output layer, b 2 Is the bias of the output layer.
As a preferred technical solution, the overall fine adjustment specifically includes:
and adjusting the weight of the neural network by using the labeled training sample set.
As a preferred technical scheme, the identifying the corrosion defect of the aluminum sheath of the high-voltage cable specifically comprises the following steps: synchronously carrying out detection and defect characteristic identification by large-scale and multiple groups of data; when in detection, the ultrasonic guided wave detection equipment is used for detecting multiple groups of high-voltage cable aluminum sheaths, characteristic extraction is carried out through sliding window wavelet transformation, characteristic signals are marked by codes, and characteristic identification is carried out on the multiple groups of currently obtained high-voltage cable aluminum sheath characteristic signals through a trained self-coding network model capable of automatically identifying corrosion defect types.
In another aspect of the invention, an ultrasonic guided wave detection system for high-voltage cable aluminum sheath corrosion based on SWT-AE is provided, and is applied to the ultrasonic guided wave detection method for high-voltage cable aluminum sheath corrosion based on SWT-AE, and the ultrasonic guided wave detection system comprises a signal acquisition module, a statistical characteristic extraction module, a model training module and a model identification module;
the signal acquisition module is used for establishing an actual high-voltage cable aluminum sheath standard test block, collecting a high-voltage cable aluminum sheath corrosion defect sample, detecting the high-voltage cable aluminum sheath corrosion defect by using an ultrasonic guided wave detector and acquiring an ultrasonic guided wave original signal of the high-voltage cable aluminum sheath;
the statistical characteristic extraction module is used for converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, the guided wave signals are segmented by utilizing a sliding window and discrete wavelet transformation is carried out; extracting statistical characteristics from the signal representation of each signal segment;
the model training module is used for performing defect recognition regression through a self-coding network, and comprises unsupervised feature fusion and integral fine tuning;
the model identification module is used for identifying the corrosion defect of the high-voltage cable aluminum sheath.
In another aspect of the invention, a storage medium is provided, which stores a program, and when the program is executed by a processor, the SWT-AE based ultrasonic guided wave detection method for detecting corrosion of an aluminum sheath of a high-voltage cable is realized.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention adopts ultrasonic guided waves to detect the high-voltage cable aluminum sheath, and the detection and the defect characteristic identification can be synchronously carried out on a large scale by a plurality of groups of data; when in detection, the ultrasonic guided wave detection equipment is used for detecting multiple groups of high-voltage cable aluminum sheaths, local time-frequency characteristic extraction is carried out through sliding window wavelet transformation, coding marks are made on characteristic signals, and efficient and accurate characteristic identification is carried out on the currently obtained multiple groups of high-voltage cable aluminum sheath corrosion defect characteristic signals through a trained self-coding network model capable of automatically identifying corrosion defect types.
Drawings
FIG. 1 is a flow chart of an ultrasonic guided wave detection method for corrosion of an aluminum sheath of a high-voltage cable based on SWT-AE in an embodiment of the invention;
FIG. 2 is a schematic diagram of a feature selection process of an embodiment of the invention;
FIG. 3 is a schematic structural diagram of a three-layer self-coding network according to an embodiment of the present invention;
FIG. 4 is a self-coding network model structure for corrosion defect detection of an aluminum sheath of a high-voltage cable according to an embodiment of the invention;
FIG. 5 is a schematic structural diagram of an SWT-AE based high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection system according to an embodiment of the invention;
fig. 6 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Examples
The invention aims to provide a Sliding window Wavelet Transform-self-encoding network SWT-AE (Sliding Wavelet Transform-automatic encoder) based high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method. The detection method comprises the steps of analyzing the defect type of the high-voltage cable aluminum sheath, obtaining manually simulated defect signals by an ultrasonic guided wave detection method and actually obtaining a plurality of groups of defect echo signal samples on site. And converting the ultrasonic guided wave original signals of the aluminum sheath of the high-voltage cable into a time-frequency domain for analysis by utilizing sliding window wavelet transformation, and extracting statistical characteristics from the signal representation of each signal segment. And performing defect identification regression by adopting an auto-coding network model to form a sample set and finish the matching of the corrosion characteristics of the sample set and the corrosion characteristics of the high-voltage cable aluminum sheath, thereby accurately obtaining the corrosion defect characteristic identification of the high-voltage cable aluminum sheath.
As shown in fig. 1, the present embodiment provides a flow of a high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method based on a sliding window wavelet transform-self-coding network, including the following steps:
s1, establishing an actual standard test block of the aluminum sheath of the high-voltage cable, collecting a corrosion defect sample of the aluminum sheath of the high-voltage cable, detecting the corrosion defect of the aluminum sheath of the high-voltage cable by using an ultrasonic guided wave detector, and acquiring an ultrasonic guided wave original signal of the aluminum sheath of the high-voltage cable.
The corrosion defect samples of the high-voltage cable aluminum sheath comprise uniform corrosion, pitting corrosion and filiform corrosion, and are obtained by manually simulating or collecting actual corrosion defect samples;
s2, converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, meanwhile, dividing the guided wave signals by using a sliding window, performing discrete wavelet transformation, and extracting statistical characteristics of signal representation of each signal segment, wherein the method specifically comprises the following steps:
the wavelet transform provides a time-frequency representation of a signal, and decomposes the signal into a series of finite attenuation signals, and the process mainly comprises the calculation of mother wavelets and the signal transformation.
For the mother wavelet function ψ (t) in the wavelet transform, it must satisfy the square integrable condition:
wherein
Is a fourier transform of ψ (t).
The mother wavelet function can obtain a mother wavelet psi via a stretching transformation ab (t) } acquisition mode:
where a is a scale parameter and b is a translation parameter.
The discrete wavelet transform is developed by a discrete scale parameter a and a translation parameter b, and a and b can be discretized by the following formula:
a＝a 0 j ,b＝ka 0 j b 0 ,j,k∈Z
under the current assumption, the discrete wavelet transform can be calculated by the following formula:
the sliding window wavelet transform cuts off ultrasonic guided wave signals of the aluminum sheath of the high-voltage cable by using the sliding window, and reduces the interference of other modes on useful detection waveforms. Cutting the guided wave signal by a sliding window with the window length of k, then performing discrete wavelet transform, and setting the wavelet coefficient C ψ Extracting root mean square RMS, peak value P, kurtosis K, form factor W and absolute mean value X from the windowed high-voltage cable aluminum sheath ultrasonic guided wave signal a And waiting for statistical indexes to serve as signal characteristics, wherein the statistical characteristics can be calculated through the following formula:
P＝max(C ψ )-min(C ψ )
after the statistical characteristics are extracted from the signal representation of each signal segment, the characteristics are selected according to the characteristic monotonicity of the training set in order to reduce calculation and improve regression results. The process of feature selection is shown in fig. 2. Taking N random samples of defect states F ∈ R N×K Let the negative correlation factor α, the positive correlation factors β all be 0, and the serial number h be 0. Judging whether the adjacent characteristic value trends are consistent with the defect state change, if so, determining positive correlation factors beta +1 and h + 1; if not, the correlation factors alpha +1, h +1 are reversed. Judging whether h is larger than N, if h is not larger than N, returning to judge whether the trend of the adjacent characteristic values is consistent with the defect state change; if h is greater than N, the monotonic score S of the feature m Is [ alpha, beta ]]Is measured. Iterating all the k characteristics to obtain their correlation factor scores, and calculating the average correlation factor score S of each characteristic mean Then selecting S m Greater than S mean The characteristics of (1).
When wavelet transformation is applied, a sliding window is used for segmenting signals, and local information and time-frequency information are obtained at the same time. Then, all the extracted features are pre-screened according to the correlation of the wavelet coefficients and the structural state. Only the selected features are input into the self-coding network for feature fusion and state recognition.
And S3, performing defect identification regression by adopting an autocorrelation network, wherein the defect identification regression comprises unsupervised feature fusion and integral fine adjustment.
The high voltage cable aluminum sheath detection self-coding network established by the application can reconstruct the signal as much as possible by setting the target output as input, but the purpose is to obtain a low-dimensional representation of the hidden layer. A three-layer automatic coding network as shown in fig. 3, the mapping from the input layer to the hidden layer is called coding, and the mapping from the hidden layer to the output layer is decoding.
The calculation of the high-voltage cable aluminum sheath detection self-coding network comprises three steps including forward calculation, regression calculation and integral fine adjustment, and aims to obtain low-dimensional representation. Giving a high-dimensional input signal X k ∈R n×1 The encoder may obtain the low-dimensional representation by:
h k ＝f(W×X k +b 1 )
wherein h is k Is a k-dimensional feature representation and k<n, f () are nonlinear activation functions, W is the weight matrix of the encoder, b 1 Is the bias of the hidden layer.
The decoder may perform the reconstruction of the original input by:
wherein
Is an input reconstructed signal, W T Is the weight matrix of the decoder, h m Is an activation function of the output layer, b 2 Is the bias of the output layer.
Specifically, the high-voltage cable aluminum sheath corrosion defect detection self-coding network model adopted in the embodiment has seven layers, as shown in fig. 4, including 1 input layer, 3 encoder layers, 2 regression layers, and 1 output layer. The extracted features are input into a self-encoding network with three encoder layers to achieve unsupervised feature fusion. The fused features are then input into a three-layer neural network to quantitatively detect corrosion defects. And finally, updating the weight matrixes of all the encoders and the regression layers at a smaller learning rate so as to perform integral fine adjustment (the weight of the neural network is adjusted by using a labeled training sample set), thereby realizing the identification of the corrosion defect of the aluminum sheath of the high-voltage cable. And S4, identifying the corrosion defect of the high-voltage cable aluminum sheath by using the trained self-coding network.
Synchronously carrying out detection and defect feature identification by large-scale and multi-group data; when in detection, the ultrasonic guided wave detection equipment is used for detecting multiple groups of high-voltage cable aluminum sheaths, characteristic extraction is carried out through sliding window wavelet transformation, coding marks are made on characteristic signals, and characteristic identification is carried out on the currently obtained multiple groups of high-voltage cable aluminum sheath characteristic signals through a trained self-coding network model capable of automatically identifying corrosion defect types.
In another embodiment of the present application, as shown in fig. 5, there is provided an SWT-AE based ultrasonic guided wave detection system for aluminum sheath corrosion of a high voltage cable, the system includes a signal acquisition module, a statistical feature extraction module, a model training module, and a model identification module;
the signal acquisition module is used for establishing an actual high-voltage cable aluminum sheath standard test block, collecting a high-voltage cable aluminum sheath corrosion defect sample, detecting the high-voltage cable aluminum sheath corrosion defect by using an ultrasonic guided wave detector and acquiring an ultrasonic guided wave original signal of the high-voltage cable aluminum sheath;
the statistical characteristic extraction module is used for converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, the guided wave signals are segmented by utilizing a sliding window and discrete wavelet transformation is carried out; extracting statistical characteristics from the signal representation of each signal segment;
the model training module is used for performing defect recognition regression through a self-coding network, and comprises unsupervised feature fusion and integral fine tuning;
the model identification module is used for identifying the corrosion defect of the high-voltage cable aluminum sheath.
It should be noted that the system provided in the above embodiment is only exemplified by the division of the above functional modules, and in practical applications, the above function distribution can be completed by different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the above described functions.
As shown in fig. 6, in another embodiment of the present application, there is further provided a storage medium storing a program, which when executed by a processor, implements an SWT-AE based ultrasonic guided wave detection method for detecting corrosion of an aluminum sheath of a high voltage cable, specifically:
establishing an actual standard test block for the aluminum sheath of the high-voltage cable, collecting a corrosion defect sample of the aluminum sheath of the high-voltage cable, detecting the corrosion defect of the aluminum sheath of the high-voltage cable by using an ultrasonic guided wave detector, and acquiring an ultrasonic guided wave original signal of the aluminum sheath of the high-voltage cable;
converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, segmenting the guided wave signals by using a sliding window and performing discrete wavelet transformation;
extracting statistical characteristics from the signal representation of each signal segment;
performing defect identification regression by using a self-coding network, wherein the defect identification regression comprises unsupervised feature fusion and integral fine adjustment;
and identifying the corrosion defect of the high-voltage cable aluminum sheath.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (10)
1. A high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method based on SWT-AE is characterized by comprising the following steps:
establishing an actual standard test block for the aluminum sheath of the high-voltage cable, collecting a corrosion defect sample of the aluminum sheath of the high-voltage cable, detecting the corrosion defect of the aluminum sheath of the high-voltage cable by using an ultrasonic guided wave detector, and acquiring an ultrasonic guided wave original signal of the aluminum sheath of the high-voltage cable;
converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, segmenting the guided wave signals by using a sliding window and performing discrete wavelet transformation;
extracting statistical characteristics from the signal representation of each signal segment;
performing defect identification regression by using a self-coding network, wherein the defect identification regression comprises unsupervised feature fusion and integral fine adjustment;
and identifying the corrosion defect of the high-voltage cable aluminum sheath.
2. The SWT-AE based ultrasonic guided wave detection method for the corrosion of the aluminum sheath of the high-voltage cable according to claim 1, wherein the corrosion defect samples of the aluminum sheath of the high-voltage cable comprise samples of uniform corrosion, pitting corrosion and filiform corrosion, and are obtained by manually simulating or collecting actual corrosion defect samples.
3. The SWT-AE-based ultrasonic guided wave detection method for high-voltage cable aluminum sheath corrosion according to claim 1, wherein the guided wave signals are segmented by using a sliding window and discrete wavelet transform is performed, specifically:
cutting the guided wave signal through a sliding window with the window length of k;
performing discrete wavelet transform, and setting its wavelet coefficient C ψ Extracting root mean square RMS, peak value P, kurtosis K, form factor W and absolute mean value X from the windowed high-voltage cable aluminum sheath ultrasonic guided wave signal a The statistical indicator is calculated as a signal characteristic by the following formula:
P＝max(C ψ )-min(C ψ )
4. the SWT-AE based ultrasonic guided wave detection method for the corrosion of the aluminum sheath of the high-voltage cable is characterized in that when the statistical characteristics are extracted from the signal representation of each signal section, the characteristics are selected according to the characteristic monotonicity of a training set.
5. The SWT-AE based ultrasonic guided wave detection method for high voltage cable aluminum sheath corrosion according to claim 1, wherein the self-encoding network comprises an input layer, a plurality of encoder layers, a plurality of regression layers, and an output layer.
6. The SWT-AE based ultrasonic guided wave detection method for high-voltage cable aluminum sheath corrosion according to claim 1, wherein selected features are input into a high-voltage cable aluminum sheath corrosion detection self-coding network model for feature fusion and defect identification, and a high-dimensional input signal X is obtained by an encoder in the following way k ∈R n×1 Low dimensional representation of (c):
h k ＝f(W×X k +b 1 )
wherein h is k Is a k-dimensional feature representation and k<n, f are nonlinear activation functions, W is a weight matrix of the encoder, b 1 Is the bias of the hidden layer;
the reconstruction of the original input by a decoder in the high-voltage cable aluminum sheath corrosion detection self-coding network model is realized by the following steps:
wherein
Is an input reconstructed signal, W T Is the weight matrix of the decoder, h m Is an activation function of the output layer, b 2 Is the bias of the output layer.
7. The SWT-AE-based ultrasonic guided wave detection method for high-voltage cable aluminum sheath corrosion according to claim 1, wherein the overall fine tuning specifically comprises:
and adjusting the weight of the neural network by using the labeled training sample set.
8. The SWT-AE-based ultrasonic guided wave detection method for corrosion of the aluminum sheath of the high-voltage cable according to claim 1, wherein the identification of the corrosion defect of the aluminum sheath of the high-voltage cable is specifically as follows: synchronously carrying out detection and defect feature identification by large-scale and multi-group data; when in detection, the ultrasonic guided wave detection equipment is used for detecting multiple groups of high-voltage cable aluminum sheaths, characteristic extraction is carried out through sliding window wavelet transformation, coding marks are made on characteristic signals, and characteristic identification is carried out on the currently obtained multiple groups of high-voltage cable aluminum sheath characteristic signals through a trained self-coding network model capable of automatically identifying corrosion defect types.
9. A high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection system based on SWT-AE is characterized in that the high-voltage cable aluminum sheath corrosion ultrasonic guided wave detection system based on SWT-AE is applied to any one of claims 1 to 8, and comprises a signal acquisition module, a statistical characteristic extraction module, a model training module and a model identification module;
the signal acquisition module is used for establishing an actual high-voltage cable aluminum sheath standard test block, collecting a high-voltage cable aluminum sheath corrosion defect sample, detecting the high-voltage cable aluminum sheath corrosion defect by using an ultrasonic guided wave detector and acquiring an ultrasonic guided wave original signal of the high-voltage cable aluminum sheath;
the statistical characteristic extraction module is used for converting the ultrasonic guided wave original signals of the high-voltage cable aluminum sheath into a time-frequency domain for analysis, and meanwhile, the guided wave signals are segmented by utilizing a sliding window and discrete wavelet transformation is carried out; extracting statistical characteristics from the signal representation of each signal segment;
the model training module is used for performing defect recognition regression through a self-coding network, and comprises unsupervised feature fusion and integral fine tuning;
the model identification module is used for identifying the corrosion defect of the high-voltage cable aluminum sheath.
10. A storage medium storing a program, characterized in that: when being executed by a processor, the program realizes the SWT-AE based ultrasonic guided wave detection method for the corrosion of the aluminum sheath of the high-voltage cable according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210723765.2A CN115078538B (en) | 2022-06-24 | 2022-06-24 | High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210723765.2A CN115078538B (en) | 2022-06-24 | 2022-06-24 | High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115078538A true CN115078538A (en) | 2022-09-20 |
| CN115078538B CN115078538B (en) | 2024-07-09 |
Family
ID=83255272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210723765.2A Active CN115078538B (en) | 2022-06-24 | 2022-06-24 | High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115078538B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118501272A (en) * | 2024-07-19 | 2024-08-16 | 国网山东省电力公司东营供电公司 | Method, system, medium and equipment for identifying salt spray corrosion of insulating partition plate |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105203632A (en) * | 2015-10-12 | 2015-12-30 | 江苏大学 | Pipeline defect type distinguishing method based on ultrasonic guided waves |
| CN110567558A (en) * | 2019-08-28 | 2019-12-13 | 华南理工大学 | Ultrasonic guided wave detection method based on deep convolution characteristics |
| WO2021128510A1 (en) * | 2019-12-27 | 2021-07-01 | 江苏科技大学 | Bearing defect identification method based on sdae and improved gwo-svm |
| CN113740434A (en) * | 2021-09-23 | 2021-12-03 | 南京尚华电力科技有限公司 | Method for detecting lead seal corrosion of high-voltage cable based on ultrasonic guided wave technology |
-
2022
- 2022-06-24 CN CN202210723765.2A patent/CN115078538B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105203632A (en) * | 2015-10-12 | 2015-12-30 | 江苏大学 | Pipeline defect type distinguishing method based on ultrasonic guided waves |
| CN110567558A (en) * | 2019-08-28 | 2019-12-13 | 华南理工大学 | Ultrasonic guided wave detection method based on deep convolution characteristics |
| WO2021128510A1 (en) * | 2019-12-27 | 2021-07-01 | 江苏科技大学 | Bearing defect identification method based on sdae and improved gwo-svm |
| CN113740434A (en) * | 2021-09-23 | 2021-12-03 | 南京尚华电力科技有限公司 | Method for detecting lead seal corrosion of high-voltage cable based on ultrasonic guided wave technology |
Non-Patent Citations (1)
| Title |
|---|
| 祁宏昌;张斌;黄嘉盛;刘远;吴倩;洪晓斌;: "基于STFT和DBN的高压电缆瓷套式终端液位智能检测", 中国测试, no. 04, 30 April 2019 (2019-04-30) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118501272A (en) * | 2024-07-19 | 2024-08-16 | 国网山东省电力公司东营供电公司 | Method, system, medium and equipment for identifying salt spray corrosion of insulating partition plate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115078538B (en) | 2024-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109649432B (en) | System and method for monitoring integrity of steel rail of cloud platform based on guided wave technology | |
| CN109212392B (en) | Method and system for identifying partial discharge defect fault of direct current cable and Caffe convolutional neural network | |
| CN113642417B (en) | Denoising method for partial discharge signal of insulated overhead conductor based on improved wavelet algorithm | |
| CN105223482A (en) | The wavelet decomposition two-value denoising method of partial-discharge ultrahigh-frequency signal waveform | |
| CN115078538B (en) | High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE | |
| Li et al. | An improved AIC onset-time picking method based on regression convolutional neural network | |
| CN112884134A (en) | Time domain-based convolutional neural network model for seismic facies identification and application | |
| CN113777650B (en) | Sparse time-frequency spectrum decomposition method, device and equipment based on mixed norm and wavelet transform and storage medium | |
| CN112364296A (en) | P wave arrival time automatic picking method based on deep learning | |
| CN114839269B (en) | Nondestructive testing method and device for internal defect stress of GIS solid insulator | |
| CN107727749B (en) | Ultrasonic quantitative detection method based on wavelet packet fusion feature extraction algorithm | |
| CN115656335A (en) | Method for quickly detecting and identifying internal defects of bearing ring | |
| CN115840119A (en) | Power cable line degradation diagnostic system and method using database samples | |
| CN113219054B (en) | Magnetic shoe internal defect detection device and detection method | |
| CN114487129A (en) | Flexible material damage identification method based on acoustic emission technology | |
| CN113536894A (en) | Defect detecting method, defect detecting apparatus, and computer-readable storage medium | |
| CN117371515A (en) | Partial discharge mode identification method based on multi-level gradual migration learning | |
| CN112305378A (en) | Photoelectric composite submarine cable fault distance measurement system and fault distance measurement method | |
| CN114047259B (en) | Method for detecting multi-scale steel rail damage defects based on time sequence | |
| CN114841200A (en) | Steel rail damage monitoring method based on wavelet transform and semi-supervised convolution self-encoder | |
| CN115406971B (en) | Ultrasonic first arrival picking method for rock mass test piece | |
| CN117851754A (en) | Multi-pulse test denoising method based on wavelet transformation | |
| Ma et al. | Fault diagnosis of motor bearing equipment based on vibration signals | |
| CN114997236A (en) | Cable accessory copper braided belt welding corrosion identification method based on deep learning | |
| CN111398755B (en) | Cable partial discharge waveform extraction method based on short-time FFT (fast Fourier transform) segmentation technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |