CN113205828A - SF based on deep learning6Gas leakage detection method - Google Patents
SF based on deep learning6Gas leakage detection method Download PDFInfo
- Publication number
- CN113205828A CN113205828A CN202110469596.XA CN202110469596A CN113205828A CN 113205828 A CN113205828 A CN 113205828A CN 202110469596 A CN202110469596 A CN 202110469596A CN 113205828 A CN113205828 A CN 113205828A
- Authority
- CN
- China
- Prior art keywords
- time
- background noise
- neural network
- gas leakage
- network model
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000013135 deep learning Methods 0.000 claims abstract description 20
- 230000009467 reduction Effects 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 25
- 238000013527 convolutional neural network Methods 0.000 claims description 22
- 238000003062 neural network model Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 12
- 238000001228 spectrum Methods 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 7
- 241000282414 Homo sapiens Species 0.000 abstract description 3
- 238000013459 approach Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 239000007789 gas Substances 0.000 description 37
- 229910018503 SF6 Inorganic materials 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 5
- 230000005236 sound signal Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 230000005684 electric field Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- SFZCNBIFKDRMGX-UHFFFAOYSA-N sulfur hexafluoride Chemical compound FS(F)(F)(F)(F)F SFZCNBIFKDRMGX-UHFFFAOYSA-N 0.000 description 2
- 229960000909 sulfur hexafluoride Drugs 0.000 description 2
- AZFKQCNGMSSWDS-UHFFFAOYSA-N MCPA-thioethyl Chemical compound CCSC(=O)COC1=CC=C(Cl)C=C1C AZFKQCNGMSSWDS-UHFFFAOYSA-N 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
- G01M3/04—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point
- G01M3/24—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point using infrasonic, sonic, or ultrasonic vibrations
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
The invention relates to SF based on deep learning6Gas leak detection method using deep learning method using SF6The reduction degree of the air leakage noise reaches 90 percent, and the SF is improved6Accuracy of gas leak detection; by using a deep learning approach, SF6The efficiency and the precision of detection reduce the labor cost and prevent SF6The gas causes harm to human bodies; by pre-processing noise reduction with SF6Gas leakage detection is combined, and a set of SF specific to high-voltage power equipment in a transformer substation scene is constructed6The gas leakage detection scheme can obtain a relatively remarkable effect under the working condition of low signal-to-noise ratio.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to SF (sulfur hexafluoride) based on deep learning6Gas leak detectorAnd (4) a measuring method.
Background
In the operation process of the existing high-voltage power equipment, due to factors such as installation mode, material quality, aging caused by long-time operation and the like, SF is generated at the flange connection part of the high-voltage power equipment and in the equipment sealing structure6The problem of gas leakage influences the safe and reliable operation of equipment, leads to the insulating properties of equipment to descend to produce the potential safety hazard.
SF today6The gas leakage detection method comprises a gas density detection technology, a negative corona discharge technology, a laser imaging technology and the like, and SF (sulfur hexafluoride) is detected6Gas leak detection has a great impact. The gas density detection technology is mainly used for measuring the air pressure and temperature in a gas chamber of high-voltage power equipment to realize SF6Although this technique is simple in structure, the measurement accuracy is not high in monitoring the gas concentration. The negative corona discharge technique is based on the principle of corona discharge, which refers to the phenomenon of local self-sustaining discharge in a gas or liquid medium on the surface of a charged body in a non-uniform electric field region with high electric field intensity, while SF6The gas has certain inhibiting effect on negative corona discharge to reduce corona current, and the current change value is converted into concentration indication value via the amplifier circuit to detect leaked SF6Gas, the negative corona sensor is vulnerable, short lived and costly, although the sensitivity of the technique is high. The basic principle of the laser imaging technology is to emit laser to the equipment to be detected, image a part of the reflected or backscattered laser on the detection equipment, and finally generate a video image.
Disclosure of Invention
The invention provides an SF based on deep learning to solve the problems mentioned above6The gas leakage detection method comprises the steps of collecting noise signals in a transformer substation through an acoustic array, converting the noise signals into a time-frequency diagram through preprocessing such as short-time Fourier transform and windowing slicing, extracting redundant features of frequency spectrums of the signals through a Convolutional Neural Network (CNN), denoising the signals to recover pure audio signals, and processing the signals into a VGG neural networkZero-one identification is carried out to determine whether SF exists in the transformer substation6And the air leakage condition is quickly and accurately detected.
The technical scheme adopted by the invention for solving the technical problems is as follows: SF based on deep learning is constructed6A gas leak detection method, comprising:
obtaining SF in high-voltage power equipment in transformer substation6Carrying out noise reduction pretreatment on background noise under the condition of gas leakage and background noise under the condition of no gas leakage, and dividing a pretreated background noise signal into a training set and a test set;
building a VGG neural network model, inputting a background noise signal serving as a training set into the VGG neural network model for training, and adjusting parameters and functions of the VGG neural network model according to an output result until the output result is consistent with the type of the background noise; after training is finished, verifying the accuracy of detection and classification of the VGG neural network model through a test set;
inputting the background noise signals acquired in real time in the high-voltage power equipment in the transformer substation into the trained VGG neural network model, and outputting the result as SF6And judging and detecting the gas leakage condition.
Wherein the noise reduction preprocessing comprises the following steps:
carrying out short-time Fourier transform (STFT) on the background noise signal, converting the background noise signal into a time-frequency graph, and making the time-frequency graph into a data set;
building a convolutional neural network model (CNN), and training and adjusting the model by using a time-frequency graph data set;
and inputting the background noise signal acquired in real time into the trained convolutional neural network model, and outputting a pure signal to finish noise reduction preprocessing.
In the step of converting the acquired background noise signal into a time-frequency graph through short-time Fourier transform (STFT), the short-time Fourier transform (STFT) is carried out on the background noise signal, a window function is selected in a time domain, power spectrums at different moments are calculated to form the time-frequency graph, and the time-frequency graph is manufactured into a data set and divided into a training set and a testing set.
The convolutional neural network model (CNN) structure comprises two convolutional layers, two pooling layers and three full-connection layers, and an image with the resolution of 32 x 32 is used as input.
The VGG neural network model structure comprises 13 convolutional layers and 3 fully-connected layers, and takes an image with the resolution of 224 x 224 as an input.
Wherein, the short-time Fourier transform expression is as follows:
where x (t) is the input signal, h (τ -t) is the analysis window function, and STFT (t, f) is the spectrum at a given time t.
Unlike the prior art, the deep learning-based SF of the present invention6Gas leak detection method using deep learning method using SF6The reduction degree of the air leakage noise reaches 90 percent, and the SF is improved6Accuracy of gas leak detection; by using a deep learning approach, SF6The efficiency and the precision of detection reduce the labor cost and prevent SF6The gas causes harm to human bodies; by pre-processing noise reduction with SF6Gas leakage detection is combined, and a set of SF specific to high-voltage power equipment in a transformer substation scene is constructed6The gas leakage detection scheme can obtain a relatively remarkable effect under the working condition of low signal-to-noise ratio.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is an SF diagram based on deep learning provided by the present invention6Schematic flow chart of the gas leakage detection method.
FIG. 2 is an SF diagram based on deep learning provided by the present invention6CNN model parameter structure schematic diagram in gas leakage detection method.
FIG. 3 is an SF diagram based on deep learning provided by the present invention6And the VGG16 model parameter structure schematic diagram in the gas leakage detection method.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described are only for illustrating the present invention and are not to be construed as limiting the present invention. 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 invention.
Referring to FIG. 1, the present invention provides an SF data based on deep learning6A gas leak detection method, comprising:
obtaining SF in high-voltage power equipment in transformer substation6Carrying out noise reduction pretreatment on background noise under the condition of gas leakage and background noise under the condition of no gas leakage, and dividing a pretreated background noise signal into a training set and a test set;
building a VGG neural network model, inputting a background noise signal serving as a training set into the VGG neural network model for training, and adjusting parameters and functions of the VGG neural network model according to an output result until the output result is consistent with the type of the background noise; after training is finished, verifying the accuracy of detection and classification of the VGG neural network model through a test set;
inputting the background noise signals acquired in real time in the high-voltage power equipment in the transformer substation into the trained VGG neural network model, and outputting the result as SF6And judging and detecting the gas leakage condition.
Wherein the noise reduction preprocessing comprises the following steps:
carrying out short-time Fourier transform (STFT) on the background noise signal, converting the background noise signal into a time-frequency graph, and making the time-frequency graph into a data set;
building a convolutional neural network model (CNN), and training and adjusting the model by using a time-frequency graph data set;
and inputting the background noise signal acquired in real time into the trained convolutional neural network model, and outputting a pure signal to finish noise reduction preprocessing.
In the step of converting the acquired background noise signal into a time-frequency graph through short-time Fourier transform (STFT), the short-time Fourier transform (STFT) is carried out on the background noise signal, a window function is selected in a time domain, power spectrums at different moments are calculated to form the time-frequency graph, and the time-frequency graph is manufactured into a data set and divided into a training set and a testing set.
The convolutional neural network model (CNN) structure comprises two convolutional layers, two pooling layers and three full-connection layers, and an image with the resolution of 32 x 32 is used as input.
The VGG neural network model structure comprises 13 convolutional layers and 3 fully-connected layers, and takes an image with the resolution of 224 x 224 as an input.
Wherein, the short-time Fourier transform expression is as follows:
where x (t) is the input signal, h (τ -t) is the analysis window function, and STFT (t, f) is the spectrum at a given time t.
The selected application example of the invention is SF of high-voltage power equipment in a transformer substation6Gas leakage detection, major transformer in transformer substation and equipment such as high voltage and the like generate large noise during operation, so SF is subjected to detection6The detection of the gas leakage signal requires a noise reduction preprocessing. Due to the fact that SF is in practical condition6The gas is harmful gas and can cause harm to the living environment and personal safety of human beings, so that nitrogen or air is selected to replace SF when actually collecting noise signals6Gas experiment is carried out to simulate SF under actual conditions6Leakage of gas. And (3) carrying out short-time Fourier transform and windowing cutting on the acquired noise signals to obtain a time-frequency diagram, making the time-frequency diagram into a data set, and carrying out the following steps on the data set according to the ratio of 7: and 3, dividing the signal into a training set and a testing set, extracting redundant features by using a convolutional neural network model, reducing the acquired signal into a pure air leakage noise signal, performing zero-one recognition on the signal subjected to noise reduction pretreatment by using a VGG model, and judging whether the air leakage condition exists.
In particular, SF is simulated in the transformer substation under different distances and different discharge pressures6In the gas leakage experiment, a microphone array is used for collecting leakage sound signals, and simultaneously background noise signals under the condition of no gas leakage are collected, the sampling rate is set to be 44100Hz, and the sampling time is 30 s.
And carrying out short-time Fourier transform on the two types of audio signals, wherein the frame length of the short-time Fourier transform is 0.025 seconds, the length of the sliding block is one half of the frame length, and a time-frequency diagram is obtained after the short-time Fourier transform, so that the signal characteristics are conveniently analyzed. The short-time fourier transform expression of signal x (t) is as follows:
where x (t) is the input signal, h (τ -t) is the analysis window function, and STFT (t, f) is the spectrum at a given time t.
From the experimental sampling results, SF6The emitted audio signal is difficult to find a single characteristic spectrum, but a regional dense distribution characteristic exists in a high frequency band, so a regional window is adopted to calculate the spectral distribution characteristic to judge whether SF exists or not6A parameter of the leakage.
Taking a distribution characteristic time-frequency graph as a data set, and taking the distribution characteristic time-frequency graph as a data set according to the following steps of 7: and 3, dividing the model into a training set and a testing set in proportion, so as to facilitate subsequent model training and processing.
Generally, a convolutional neural network model comprises a convolutional layer, a pooling layer and a full-link layer, and a DnCNN is taken as a typical CNN model and comprises a convolutional layer, a Batch Normalization (BN) and a rectifying linear unit (Relu) activation layer, wherein the pooling layer is removed, and a time-frequency picture with a resolution of 32 × 32 is taken as an input to output a denoised spectrum. The model removes a full connection layer, reduces training parameters and enables a network to be easier to train, test and predict. The model architecture is shown in fig. 2. The mathematical expression for the convolution operation is as follows:
S(i，j)＝(I*J)(i，j)＝∑∑I(i+m，j+m)K(m，n)
where I is the input picture and K is the convolution kernel.
Training a model by using a corresponding training set and testing the performance of the model by using a test set, defining and minimizing a loss function, and adjusting and optimizing the built convolutional neural network model, wherein the loss function expression is as follows:
wherein y istIs the input signal, f (x (n)) is the output signal, and x (n) is the signal with noise.
And taking the output pure signal spectrum as a data set of leakage signal identification for training and testing the VGG model.
And performing feature extraction on the denoised frequency spectrum by using a VGG model, and detecting the leakage state. The VGG model comprises 13 convolution layers and 3 full-connection layers, a picture with the resolution of 224 x 224 is input, after two convolutions of 64 convolution kernels, one pooling operation is adopted, after two convolutions of 128 convolution kernels, the pooling operation is adopted again, after three convolution kernels of 512 are repeated twice, pooling is carried out again, and finally, a classification result is output through three full-connection. The model architecture is shown in fig. 3.
The output of the model is the gas leakage condition, the trained VGG model is verified by using a test set, and the parameters of the VGG model are optimized to obtain the optimal accuracy.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (6)
1. SF based on deep learning6A gas leak detection method, comprising:
obtaining high-voltage power equipment in transformer substationMiddle SF6Carrying out noise reduction pretreatment on background noise under the condition of gas leakage and background noise under the condition of no gas leakage, and dividing a pretreated background noise signal into a training set and a test set;
building a VGG neural network model, inputting a background noise signal serving as a training set into the VGG neural network model for training, and adjusting parameters and functions of the VGG neural network model according to an output result until the output result is consistent with the type of the background noise; after training is finished, verifying the accuracy of detection and classification of the VGG neural network model through a test set;
inputting the background noise signals acquired in real time in the high-voltage power equipment in the transformer substation into the trained VGG neural network model, and outputting the result as SF6And judging and detecting the gas leakage condition.
2. Deep learning based SF according to claim 16The gas leakage detection method is characterized in that the noise reduction pretreatment step comprises the following steps:
carrying out short-time Fourier transform (STFT) on the background noise signal, converting the background noise signal into a time-frequency graph, and making the time-frequency graph into a data set;
building a convolutional neural network model (CNN), and training and adjusting the model by using a time-frequency graph data set;
and inputting the background noise signal acquired in real time into the trained convolutional neural network model, and outputting a pure signal to finish noise reduction preprocessing.
3. Deep learning based SF according to claim 26The gas leakage detection method is characterized in that in the step of converting the acquired background noise signal into a time-frequency graph through short-time Fourier transform (STFT), the short-time Fourier transform (STFT) is carried out on the background noise signal, a window function is selected in the time domain, power spectrums at different moments are calculated to form the time-frequency graph, and the time-frequency graph is manufactured into a data set and divided into a training set and a testing set.
4. Deep learning based SF according to claim 36The gas leakage detection method is characterized in that a Convolutional Neural Network (CNN) structure comprises two convolutional layers, two pooling layers and three full-connection layers, and an image with the resolution of 32 x 32 is used as input.
5. Deep learning based SF according to claim 16The gas leakage detection method is characterized in that the VGG neural network model structure comprises 13 convolutional layers and 3 full-connected layers, and an image with the resolution of 224 x 224 is used as an input.
6. Deep learning based SF according to claim 16The gas leakage detection method is characterized in that the short-time Fourier transform expression is as follows:
where x (t) is the input signal, h (τ -t) is the analysis window function, and STFT (t, f) is the spectrum at a given time t.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110469596.XA CN113205828A (en) | 2021-04-28 | 2021-04-28 | SF based on deep learning6Gas leakage detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110469596.XA CN113205828A (en) | 2021-04-28 | 2021-04-28 | SF based on deep learning6Gas leakage detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113205828A true CN113205828A (en) | 2021-08-03 |
Family
ID=77027118
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110469596.XA Pending CN113205828A (en) | 2021-04-28 | 2021-04-28 | SF based on deep learning6Gas leakage detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113205828A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114018498A (en) * | 2021-09-22 | 2022-02-08 | 国网河北省电力有限公司营销服务中心 | Air valve device air leakage state evaluation method and device, terminal and storage medium |
| CN117452865A (en) * | 2023-12-22 | 2024-01-26 | 中测智联(深圳)科技有限公司 | Intelligent monitoring system for environmental parameters of power distribution room |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110185939A (en) * | 2019-05-16 | 2019-08-30 | 西北工业大学 | Gas pipe leakage recognition methods based on convolutional neural networks |
| CN111223493A (en) * | 2020-01-08 | 2020-06-02 | 北京声加科技有限公司 | Voice signal noise reduction processing method, microphone and electronic equipment |
| CN111750283A (en) * | 2020-06-26 | 2020-10-09 | 西北工业大学 | Deep learning-based gas pipeline leakage identification method in strong background noise environment |
-
2021
- 2021-04-28 CN CN202110469596.XA patent/CN113205828A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110185939A (en) * | 2019-05-16 | 2019-08-30 | 西北工业大学 | Gas pipe leakage recognition methods based on convolutional neural networks |
| CN111223493A (en) * | 2020-01-08 | 2020-06-02 | 北京声加科技有限公司 | Voice signal noise reduction processing method, microphone and electronic equipment |
| CN111750283A (en) * | 2020-06-26 | 2020-10-09 | 西北工业大学 | Deep learning-based gas pipeline leakage identification method in strong background noise environment |
Non-Patent Citations (1)
| Title |
|---|
| 张瑞程等: "基于一维卷积神经网络的燃气管道泄漏声发射信号识别", 《中国安全生产科学技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114018498A (en) * | 2021-09-22 | 2022-02-08 | 国网河北省电力有限公司营销服务中心 | Air valve device air leakage state evaluation method and device, terminal and storage medium |
| CN117452865A (en) * | 2023-12-22 | 2024-01-26 | 中测智联(深圳)科技有限公司 | Intelligent monitoring system for environmental parameters of power distribution room |
| CN117452865B (en) * | 2023-12-22 | 2024-03-26 | 中测智联(深圳)科技有限公司 | Intelligent monitoring system for environmental parameters of power distribution room |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109856517B (en) | Method for distinguishing partial discharge detection data of extra-high voltage equipment | |
| CN109034046B (en) | Method for automatically identifying foreign matters in electric energy meter based on acoustic detection | |
| CN113205828A (en) | SF based on deep learning6Gas leakage detection method | |
| CN110987434A (en) | Rolling bearing early fault diagnosis method based on denoising technology | |
| CN108693448B (en) | Partial discharge mode recognition system applied to power equipment | |
| Hou et al. | Research on audio-visual detection method for conveyor belt longitudinal tear | |
| CN113763986B (en) | Abnormal sound detection method for air conditioner indoor unit based on sound classification model | |
| CN112329914A (en) | Fault diagnosis method and device for buried transformer substation and electronic equipment | |
| CN113345443A (en) | Marine mammal vocalization detection and identification method based on mel-frequency cepstrum coefficient | |
| CN116778956A (en) | Transformer acoustic feature extraction and fault identification method | |
| Hwang et al. | Application of cepstrum and neural network to bearing fault detection | |
| Han et al. | A novel rolling bearing fault diagnosis method based on generalized nonlinear spectral sparsity | |
| CN113642417B (en) | Denoising method for partial discharge signal of insulated overhead conductor based on improved wavelet algorithm | |
| Krause et al. | Detection of impulse-like airborne sound for damage identification in rotor blades of wind turbines | |
| CN105928666B (en) | Leakage acoustic characteristic extracting method based on Hilbert-Huang transform and blind source separating | |
| Prodeus et al. | Objective and subjective assessment of the quality and intelligibility of noised speech | |
| CN110706721A (en) | Electric precipitation spark discharge identification method based on BP neural network | |
| CN111755025A (en) | State detection method, device and equipment based on audio features | |
| Zhang et al. | Flaw classification in ultrasonic guided waves signal using Wavelet Transform and PNN classifier | |
| Howard | Speech fundamental period estimation using a neural network | |
| Yang et al. | Research on Voiceprint recognition method of buried drainage pipe based on MFCC and GMM-HMM | |
| Joseph et al. | Indian accent detection using dynamic time warping | |
| CN113933658B (en) | Dry-type transformer discharge detection method and system based on audible sound analysis | |
| Jiang et al. | ICA and PIND based Remainder Particle Detection Technique for Space-borne Equipment | |
| Gourishetti et al. | Partial discharge monitoring using deep neural networks with acoustic emission |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210803 |