CN107355686B - A kind of detection method of drain line blockage failure - Google Patents
A kind of detection method of drain line blockage failure Download PDFInfo
- Publication number
- CN107355686B CN107355686B CN201710431959.4A CN201710431959A CN107355686B CN 107355686 B CN107355686 B CN 107355686B CN 201710431959 A CN201710431959 A CN 201710431959A CN 107355686 B CN107355686 B CN 107355686B
- Authority
- CN
- China
- Prior art keywords
- pipeline
- signal
- vector set
- feature vector
- kinds
- 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.)
- Active
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
- F17D5/06—Preventing, monitoring, or locating loss using electric or acoustic means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a kind of detection methods of drain line blockage failure.Method are as follows: the detection device of installation drain line blockage failure；Known pipeline fault-free section/pipeline fault section is selected to be detected, computer obtains the signal under two kinds of operating conditions from receiving end；Obtain the sound induction signal under two kinds of operating conditions of drainage pipeline；Sound induction signal carries out SVD decomposition；Energy-Entropy index and approximate entropy index are extracted, the extraction result of two kinds of indexs is incorporated as initial characteristics vector set；Obtained Energy-Entropy feature vector set, approximate entropy feature vector set, initial characteristics vector set are obtained into class discrete matrix between discrete matrix and class using apart from Separability Criterion method；Calculate weight, the weight of approximate entropy index of Energy-Entropy index；Each feature vector set is weighted to obtain new feature vector set；By feature vector set training random forest grader, fault identification model is obtained.The present invention improves the accuracy and reliability of testing result.
Description
Technical field
The present invention relates to a kind of detection methods of drain line blockage failure, belong to pipeline fault detection field.
Background technique
The horizontal positive continuous improvement of Chinese Urbanization, with urban economy development, acceleration is presented every year and increases for drainage pipeline length
Long trend, but initial design existing defects or using the time it is long due to, the failure rate of drainage pipeline also ten
The height divided.Wherein, the blocking of pipeline belongs to the hidden danger for threatening urban safety draining, if cannot find in time and purging line event
Barrier, will affect the normal operation of city water supply and sewage.Therefore the drain line blockage detection method of efficiently and accurately is for pipeline fault
Detection is very necessary.
Summary of the invention
In order to solve the problems, such as drain line blockage fault detection, the present invention provides a kind of drain line blockage failures
Detection method.
The technical scheme is that a kind of detection method of drain line blockage failure, the method specific steps are such as
Under:
S1, the detection device for installing drain line blockage failure；
Pipeline fault-free section known to S2, selection is detected, and computer controls the signal X (t) of sound card generation through power
Amplifier, transmitting terminal, loudspeaker are received by hydrophone, and computer obtains the signal Y of normal pipeline from receiving end1(t)；
S3, the signal X generated in the artificial placement tamper of above-mentioned known pipeline fault-free section, computer control sound card
(t) it is received through power amplifier, transmitting terminal, loudspeaker by hydrophone, computer obtains the signal of blocking pipeline from receiving end
Y2(t)；
Sound induction signal G (t) under S4, acquisition two kinds of operating conditions of drainage pipeline, calculation method areWherein water
Listening the collected pipeline acoustical signal of device is Y (t), and the sinusoidal signal of sound card transmitting is X (t)；Signal calculation method according to response, point
The response signal of normal pipeline is not obtainedBlock the response signal of pipeline
S5, SVD decomposition is carried out to the sound induction signal under normal pipeline, blocking two kinds of operating conditions of pipeline respectively, after decomposition
Signal component under normal pipeline, blocking two kinds of operating conditions of pipeline；
S6, Energy-Entropy index and approximate entropy index, the Energy-Entropy index that will be obtained are extracted to signal component under two kinds of operating conditions
The extraction result of two kinds of indexs is closed using obtained approximate entropy index as feature vector set B as feature vector set A
And as initial characteristics vector set C1；
S7, by Energy-Entropy feature vector set A that step S6 is obtained using apart from Separability Criterion method obtain in class from
Dissipate matrix SW1The discrete matrix S between classB1；
S8, by approximate entropy feature vector set B that step S6 is obtained using apart from Separability Criterion method obtain in class from
Dissipate matrix SW2The discrete matrix S between classB2；
S9, the initial characteristics vector set C for obtaining step S61It is discrete in class using being obtained apart from Separability Criterion method
Matrix SW3The discrete matrix S between classB3；
S10, the weight for calculating Energy-Entropy indexThe weight of approximate entropy index is
Each feature vector set is weighted to obtain new feature vector set C2, C2=A × w1+B×w2；Tr () expression takes matrix
Mark；
S11, using feature vector set C2Training random forest grader, obtains fault identification model；
S12, other sections of use step S4~step S10 methods acquisition feature vector set for pipeline, using step
Trained random forest grader carries out the identification of failure to other sections of pipeline in rapid S11.
SVD decomposition is carried out by building Hankel matrix.
The detection device of the drain line blockage failure include two telescopic rods, sound card, power amplifier, loudspeaker,
Sieve, hydrophone, filter and computer；
Two telescopic rods stretch to shaft bottom from two different Sewage well covers, and loudspeaker is fixed under telescopic rod I
The ground surface end at end, telescopic rod I connects computer by power amplifier, sound card；It places sieve and sound is facilitated to believe in II bottom end of telescopic rod
Number aggregation, hydrophone are fixed on the lower end of telescopic rod II and are located above sieve, and the ground surface end of telescopic rod II passes through filter company
Connect computer；
The signal that computer control sound card generates amplifies through power amplifier, and amplified signal is via in telescopic rod I
It sets conducting wire and is transferred to the transmitting that loudspeaker carries out sound；The telescopic rod II of voice signal inner conductors in hydrophone receiving pipeline
It reaches filter to be filtered, filtered signal is input in computer and carries out data processing.
The loudspeaker, hydrophone are in place above shaft bottom one on the other.
The beneficial effects of the present invention are:
Detection method is extracted two kinds of features of signal, compensates for single feature failure and extracts insufficient situation, can be more complete
The characteristic of the reflection pipeline in face, reaches and more reliably extracts result than traditional single failure extracting mode.This method is by special
Weighted Fusion and classifier recognition training are levied, can be with the detect line clogging failure of efficiently and accurately, therefore there is Practical valence
Value.
In conclusion introducing acoustic conductance wave detecting method, cooperation the present invention is based on the actual conditions of drainage pipeline fault detection
Detection device, the detection acoustical signal for extracting drainage pipeline carry out fault identification, optimize testing process.Detection proposed by the present invention
Method has carried out failure multi-feature extraction, and has carried out the Weighted Fusion of feature, improve the accuracy of testing result with can
By property.
Detailed description of the invention
Fig. 1 is a kind of detection device of drain line blockage failure；
Fig. 2 is the method for the present invention flow chart；
Fig. 3 is the normal pipeline time domain waveform of 0.1s；
Fig. 4 is the blocking pipeline time domain waveform of 0.1s；
Fig. 5 is blocking pipeline SVD decomposition result；
Fig. 6 is normal pipeline SVD decomposition result.
Specific embodiment
Embodiment 1: a kind of detection method of drain line blockage failure is tested using Bradford, Britain pipeline
The number of chambers is according to progress case verification.Duct length is 15.4 meters, and pipe material is clay pipeline, diameter 150mm.Choose 0.1s's
Experimental data carry out illustration (due to signal in 0.1s can round-trip pipeline it is multiple back and forth, carry letter in pipeline abundant
Breath, has met testing requirements).
Detailed process is as follows for Computer signal processing:
Extract signal: experiment flow is as shown in Fig. 2.Before detection, in order to obtain normal pipeline, blocking pipeline, two kinds
Training data under operating condition selects a segment pipe fault-free section to be detected.Computer obtains the training signal of normal pipeline, such as
Shown in Fig. 3.Then barrier is artificially placed in the normal pipeline of this section, is detected.To obtain the instruction of blocking pipeline
Practice signal, as shown in Figure 4.Normal pipeline training signal is 52 groups, 62 groups of jam signal.
It carries out Signal Pretreatment: the signal under two kinds of operating conditions being input in MATLAB and is analyzed, two kinds of signals are made
SVD decomposition is carried out with 6 rank Hankel matrixes.The SVD decomposition result of one group of normal pipeline is as shown in fig. 6, one group of blocking pipeline
SVD decomposes as shown in Figure 5.
Feature extraction: and then the characteristic extraction procedure of signal is carried out, feature is carried out to the component signal under two kinds of operating conditions and is mentioned
It takes, extracts its Energy-Entropy, approximate entropy two indices, obtained Energy-Entropy index is as feature vector set A, the approximation that will be obtained
Entropy index will extract result joint and be used as feature vector set C as feature vector set B1, partial results (its as shown in table 1
In the whole of one group of signal extract results).
The feature extraction result of 1 one groups of normal pipelines of table and blocking pipeline
Characteristic weighing fusion process: and then the characteristic weighing fusion of signal is carried out, calculate the identification power of each characteristic index
Weight.By A, B, C1Feature vector set is used respectively apart from Separability Criterion, is obtained three groups and is spread between within-class scatter matrix and class
Matrix.Then the identification weight of Energy-Entropy index and approximate entropy approximate entropy in total feature vector set is calculated, by calculating
The weight calculation result of Energy-Entropy is 0.41, and approximate entropy index weights are calculated as 0.59.In order to obtain the feature of Weighted Fusion to
Duration set；Two feature vector set weighted arrays are obtained into new feature vector set C2, combined method C2=A × 0.41+
B×0.59。
Training random forest grader is identified: by feature vector set C250% be used as training sample, 50% make
For test sample training random forest grader, fault identification model is obtained；(the feature vector set after dimensionality reduction is inputted,
By feature vector set C257 groups of samples as training sample, 57 groups of samples are as test sample training random forest classification
Device obtains fault identification model.) as shown in table 2 (will be in embodiment with the resulting discrimination power of feature vector set of embodiment
The feature vector set that known pipeline obtains is identified as test group part using fault identification model, by what is obtained
Classification is compared with the good known class of model mark, is equally correct.).
Feature vector set is inputted the discrimination table after random forest by table 2
| Correctly | Mistake | Recognition correct rate | |
| Normal pipeline | 26 | 2 | 92.31% |
| Big blocking pipeline | 31 | 1 | 96.77% |
Detect remaining unknown situation duct section: last experimenter detects the duct section of remaining unknown situation, and repeats
(present invention is equivalent to the detection of unknown segment the part of test group to above-mentioned detection process, to unknown duct section, acquisition letter
Number, and double faults detection process obtains feature vector set, is input in classifier, and classifier will recognise that a class
Not, this classification is exactly testing result.).
Experimental provision is as shown in Fig. 1, and the detection device of drain line blockage failure can be with are as follows:
The detection device of the drain line blockage failure include two telescopic rods, sound card, power amplifier, loudspeaker,
Sieve, hydrophone, filter and computer；
Two telescopic rods stretch to shaft bottom from two different Sewage well covers, and loudspeaker is fixed under telescopic rod I
The ground surface end at end, telescopic rod I connects computer by power amplifier, sound card；It places sieve and sound is facilitated to believe in II bottom end of telescopic rod
Number aggregation, hydrophone are fixed on the lower end of telescopic rod II and are located above sieve, and the ground surface end of telescopic rod II passes through filter company
Connect computer；
The signal that computer control sound card generates amplifies through power amplifier, and amplified signal is via in telescopic rod I
It sets conducting wire and is transferred to the transmitting that loudspeaker carries out sound；The telescopic rod II of voice signal inner conductors in hydrophone receiving pipeline
It reaches filter to be filtered, filtered signal is input in computer and carries out data processing.
The loudspeaker, hydrophone can be to place apart from bottom hole location one on the other, can also be apart from the same.
When it is implemented, needing to open two well cover for sewer on ground, two telescopic rods are protruded into well.Two flexible
Bar gos deep into pipeline and bottoms out simultaneously.Sieve is placed in the bottom end of telescopic rod II, and facilitating acoustical signal aggregation, (sieve can guarantee that sound generates
Certain rebound, and then facilitate hydrophone collecting signal).The loudspeaker model can select Wei Shatong company (Germany)
K50WP model；Hydrophone can be the SQ31 model hydrophone produced by sensor technology Co., Ltd (Canada)；Telescopic rod
Kernel is metallic conductor, and shell is insulator, can transmit signal.When detection, by the computer control sound that WinMLS software is housed
Card generates 20 seconds sine sweep acoustical signal, and the frequency range of signal is 100-6000 hertz.The acoustical signal that sound card generates
It need to be amplified by power amplifier, loudspeaker is transferred to via the inner conductors of telescopic rod I by transmitting terminal and carries out sound
Transmitting.The power amplifier of selection is the power amplifier of 2708 models of Konstanze Bachmann-Wayne Kramer company (Germany) production, function
Rate amplifier drive the speaker sounding.After speaker operation, hydrophone is reception state.Hydrophone connects via telescopic rod I
Receiving end connects filter, and Kemo VBF 10M filter is selected to be filtered, and it is 100-4000 hertz that signal frequency range, which is controlled,
It is hereby input to afterwards in computer and carries out subsequent processing.
Above in conjunction with attached drawing, the embodiment of the present invention is explained in detail, but the present invention is not limited to above-mentioned
Embodiment within the knowledge of a person skilled in the art can also be before not departing from present inventive concept
Put that various changes can be made.
Claims (4)
1. a kind of detection method of drain line blockage failure, it is characterised in that: specific step is as follows for the method:
S1, the detection device for installing drain line blockage failure；
Pipeline fault-free section known to S2, selection is detected, and computer controls the signal X (t) of sound card generation through power amplification
Device, loudspeaker are received by hydrophone, and computer obtains the signal Y of normal pipeline from receiving end1(t)；
S3, the signal X (t) generated in the artificial placement tamper of above-mentioned known pipeline fault-free section, computer control sound card
It is received through power amplifier, loudspeaker by hydrophone, computer obtains the signal Y of blocking pipeline from receiving end2(t)；
Sound induction signal G (t) under S4, acquisition two kinds of operating conditions of drainage pipeline, calculation method areWherein hydrophone
Collected pipeline acoustical signal is Y (t), and the sinusoidal signal of sound card transmitting is X (t)；Signal calculation method according to response, respectively
To the response signal of normal pipelineBlock the response signal of pipeline
S5, SVD decomposition is carried out to the sound induction signal under normal pipeline, blocking two kinds of operating conditions of pipeline respectively, is obtained just after decomposition
Signal component under Chang Guandao, blocking two kinds of operating conditions of pipeline；
S6, Energy-Entropy index and approximate entropy index are extracted to signal component under two kinds of operating conditions, using obtained Energy-Entropy index as
The extraction result of two kinds of indexs is merged and is made using obtained approximate entropy index as feature vector set B by feature vector set A
For initial characteristics vector set C1；
S7, the Energy-Entropy feature vector set A that step S6 is obtained is obtained into discrete square in class using apart from Separability Criterion method
Battle array SW1The discrete matrix S between classB1；
S8, the approximate entropy feature vector set B that step S6 is obtained is obtained into discrete square in class using apart from Separability Criterion method
Battle array SW2The discrete matrix S between classB2；
S9, the initial characteristics vector set C for obtaining step S61Discrete matrix in class is obtained using apart from Separability Criterion method
SW3The discrete matrix S between classB3；
S10, the weight for calculating Energy-Entropy indexThe weight of approximate entropy index is
Each feature vector set is weighted to obtain new feature vector set C2, C2=A × w1+B×w2；Tr () expression takes matrix
Mark；
S11, using feature vector set C2Training random forest grader, obtains fault identification model；
S12, other sections of use step S4~step S10 methods acquisition feature vector set for pipeline, using step S11
In trained random forest grader to pipeline other sections carry out failure identification.
2. the detection method of drain line blockage failure according to claim 1, it is characterised in that: by constructing Hankel
Matrix carries out SVD decomposition.
3. the detection method of drain line blockage failure according to claim 1, it is characterised in that: the drainage pipeline is stifled
The detection device for filling in failure includes two telescopic rods, sound card, power amplifier, loudspeaker, sieve, hydrophone, filter and meter
Calculation machine；
Two telescopic rods stretch to shaft bottom from two different Sewage well covers, and loudspeaker is fixed on the lower end of telescopic rod I, stretches
The ground surface end of contracting bar I connects computer by power amplifier, sound card；It places sieve and facilitates acoustical signal poly- in II bottom end of telescopic rod
Collection, hydrophone are fixed on the lower end of telescopic rod II and are located above sieve, and the ground surface end of telescopic rod II is connected by filter to be counted
Calculation machine；
The signal that computer control sound card generates amplifies through power amplifier, and amplified signal is led via built in telescopic rod I
Line is transferred to the transmitting that loudspeaker carries out sound；The telescopic rod II of voice signal inner conductors in hydrophone receiving pipeline reaches
Filter is filtered, and filtered signal is input in computer and carries out data processing.
4. the detection method of drain line blockage failure according to claim 3, it is characterised in that: the loudspeaker, water
Listening device is in place above shaft bottom one on the other.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710431959.4A CN107355686B (en) | 2017-06-09 | 2017-06-09 | A kind of detection method of drain line blockage failure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710431959.4A CN107355686B (en) | 2017-06-09 | 2017-06-09 | A kind of detection method of drain line blockage failure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107355686A CN107355686A (en) | 2017-11-17 |
| CN107355686B true CN107355686B (en) | 2019-04-09 |
Family
ID=60272718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710431959.4A Active CN107355686B (en) | 2017-06-09 | 2017-06-09 | A kind of detection method of drain line blockage failure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107355686B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108734197A (en) * | 2018-04-17 | 2018-11-02 | 东北大学 | A kind of Fault monitoring and diagnosis method of the dense washing process of hydrometallurgy |
| CN108958301B (en) * | 2018-06-27 | 2021-08-31 | 北京小米移动软件有限公司 | Method and device for controlling equipment drainage and storage medium |
| CN110469522A (en) * | 2019-08-13 | 2019-11-19 | 浪潮通用软件有限公司 | A kind of method for detecting abnormality and device of drainage system |
| CN112198232A (en) * | 2020-09-14 | 2021-01-08 | 昆明理工大学 | Drainage pipeline working condition detection and identification method |
| CN112902029B (en) * | 2021-01-19 | 2022-03-18 | 昆明理工大学 | U-shaped pipe running state voiceprint recognition method based on VMD and PNCC |
| CN118194142B (en) * | 2024-05-17 | 2024-08-23 | 中电科大数据研究院有限公司 | Post-earthquake repair engineering analysis method and system for intelligent pipe network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509604A (en) * | 2009-03-20 | 2009-08-19 | 武汉大学 | Method and device for detecting and assessing deposit in metal pipe |
| CN101551064A (en) * | 2009-05-22 | 2009-10-07 | 重庆大学 | Water supply pipe leakage detection locating signal processing method |
| CN202040552U (en) * | 2010-08-12 | 2011-11-16 | 北京知容寓远软件科技有限公司 | System for monitoring the leakage or obstruction of pipeline base on low frequency sound wave |
| CN103836347A (en) * | 2014-03-07 | 2014-06-04 | 中国石油大学(华东) | Leakage monitoring device and method for crude oil gathering pipelines |
| CN104569153A (en) * | 2014-12-31 | 2015-04-29 | 天津大学 | Ship pool experiment system based on PCCP failure early-warning and monitoring signals |
| EP2955493A1 (en) * | 2014-06-13 | 2015-12-16 | Reece Innovation Centre Limited | System for monitoring and/or surveying conduits |
| CN205716445U (en) * | 2016-04-21 | 2016-11-23 | 淼淼众安（武汉）科技有限公司 | The monitoring system of sensor at each node of grid |
-
2017
- 2017-06-09 CN CN201710431959.4A patent/CN107355686B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509604A (en) * | 2009-03-20 | 2009-08-19 | 武汉大学 | Method and device for detecting and assessing deposit in metal pipe |
| CN101551064A (en) * | 2009-05-22 | 2009-10-07 | 重庆大学 | Water supply pipe leakage detection locating signal processing method |
| CN202040552U (en) * | 2010-08-12 | 2011-11-16 | 北京知容寓远软件科技有限公司 | System for monitoring the leakage or obstruction of pipeline base on low frequency sound wave |
| CN103836347A (en) * | 2014-03-07 | 2014-06-04 | 中国石油大学(华东) | Leakage monitoring device and method for crude oil gathering pipelines |
| EP2955493A1 (en) * | 2014-06-13 | 2015-12-16 | Reece Innovation Centre Limited | System for monitoring and/or surveying conduits |
| CN104569153A (en) * | 2014-12-31 | 2015-04-29 | 天津大学 | Ship pool experiment system based on PCCP failure early-warning and monitoring signals |
| CN205716445U (en) * | 2016-04-21 | 2016-11-23 | 淼淼众安（武汉）科技有限公司 | The monitoring system of sensor at each node of grid |
Non-Patent Citations (1)
| Title |
|---|
| 基于水中听音给水管道泄漏检测技术的研究;张禄;《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》;20111215(第S1期);全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107355686A (en) | 2017-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107355686B (en) | A kind of detection method of drain line blockage failure | |
| CN107218518B (en) | A kind of detection method of the detection device for drain line blockage failure | |
| CN107355687B (en) | A kind of sewer pipe fault detection method | |
| CN109357167B (en) | Gas pipeline leakage point detection device and detection method | |
| CN102606891B (en) | Water leakage detector, water leakage detecting system and water leakage detecting method | |
| CN105956526B (en) | Low signal-to-noise ratio microseismic event identification method based on multi-scale permutation entropy | |
| CN110454687A (en) | A kind of pipeline multipoint leakage localization method based on improvement VMD | |
| CN104634879B (en) | A kind of Metal Material Fatigue load test and fatigue damage Nondestructive Detection method | |
| CN109838696A (en) | Pipeline fault diagnostic method based on convolutional neural networks | |
| CN106447040B (en) | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion | |
| CN107014668A (en) | A kind of fatigue crack integrated monitoring based on piezoelectricity and smart coat sensor | |
| CN110398647B (en) | Transformer state monitoring method | |
| CN107314251B (en) | A kind of detection device and detection method of sewer pipe leakage failure | |
| CN101592288B (en) | Method for identifying pipeline leakage | |
| CN109827081A (en) | A kind of buried drain pipe road plugging fault and branch pipe tee connection part diagnostic method based on acoustics active detecting | |
| CN101846594A (en) | Fault detection device based on beam forming acoustic-image mode recognition and detection method thereof | |
| CN108361560A (en) | A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet | |
| CN103234121A (en) | Acoustic signal based device and method for detecting gas pipeline leakages | |
| CN107963239A (en) | A kind of booster failure detection device and detection method based on audio | |
| CN109116196B (en) | Intelligent power cable fault discharge sound identification method | |
| CN111007365A (en) | Ultrasonic partial discharge identification method and system based on neural network | |
| CN109239553A (en) | A kind of clustering method based on local density of partial discharge pulse | |
| CN108181059A (en) | Multiphase flow pipeline leakage acoustic signals recognition methods based on small echo signal | |
| CN115479219B (en) | Intelligent pipeline state monitoring method, monitoring device and intelligent pipeline system | |
| CN113029327A (en) | Tunnel fan embedded foundation damage identification method based on metric attention convolutional neural network |
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 |