CN108361560A - A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet - Google Patents
A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet Download PDFInfo
- Publication number
- CN108361560A CN108361560A CN201810242500.4A CN201810242500A CN108361560A CN 108361560 A CN108361560 A CN 108361560A CN 201810242500 A CN201810242500 A CN 201810242500A CN 108361560 A CN108361560 A CN 108361560A
- Authority
- CN
- China
- Prior art keywords
- signal
- wavelet packet
- neural network
- pipe
- natural gas
- 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
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
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
- G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
The present invention discloses a kind of pipe safety recognition methods of the natural gas line safety monitoring assembly based on wavelet packet, belongs to Monitoring Pinpelines technical field.The process includes：Utilize the voice signal of multifunctional data acquisition card acquisition Signal-regulated kinase output；The feature vector of detection signal is calculated using WAVELET PACKET DECOMPOSITION；Several collecting samples are selected respectively to hydrate cohesion and pipe leakage event；BP neural network is trained and is tested；Anomalous event type can be positioned using desired BP neural network has been reached.Compared with the prior art, the present invention can automatically extract fault signature and according to measurement parameter automatic decision fault type, overcome judging result of the prior art caused by artificial judgment subjectivity and one-sidedness problem with a low credibility, solving prior art deterministic process, there are the defects of time delay, the efficiency for improving malfunction monitoring, reduces implementation cost.
Description
Technical field
The present invention relates to it is a kind of based on wavelet packet be used for natural gas line safety monitoring assembly pipe safety recognition methods,
Belong to Monitoring Pinpelines technical field.
Background technology
With the fast development of Gas Industry, to meet needs of the national economy to natural gas supply, natural gas line
Construction, operation, maintenance and ensure as energy construction field the most important thing.Natural gas need to pass through from producing to selling to be adopted
The processes such as collection, purification, transport, distribution, most of completed in closed pipeline, and the natural gas supply in the whole world 95%
Pass through pipeline.General in-service natural gas line has the characteristics that Large Diameter Pipeline, long range, high pressure, huge throughput rate, in its transport
Hydrate cohesion easily occurs in the process and blocks the accidents such as pipeline and pipe leakage, will cause once accident occurs huge
Life and property loss and environmental pollution.
Therefore, the abnormal conditions in natural gas line how are found in time and are accurately judged, and then reduce it to just
The influence that often produces simultaneously avoids accident, is people in the industry's focus.Currently, both at home and abroad for being hydrated in natural gas line
Object Study on Monitoring Technology is in the starting stage, some external scholars have reported some related works, but it is domestic there is not yet it is similar at
Fruit is reported.
North Dakota, United States university A.R.Hasan et al. has carried out based on transient pressure analysis method come in the natural gas well
Partial Blocking positioned.This method can theoretically estimate the volume of tamper and position, but not propose to calculate stifled
The specific method of the position of plug thing, can not also be monitored gas pipeline leakage, in addition the article pointed out in this method and block
The thickness and length of object are affected to tamper positioning result.
The research group of Univ Manchester UK Barry professors Lennox leader after emitting sound wave to pipeline by examining
Sound echo is surveyed to determine along pipeline with the presence or absence of leakage, is tested by short distance experimental channel.The detection of the technology is former
Reason is that the acoustic impedance of pipe leakage is different from smooth pipeline, thus Acoustic Wave Propagation is at this when incident point to will produce back wave, from
And can receiving end capture, system can be positioned according to back wave time of return.Above-mentioned technology can not detect simultaneously
Pipeline internal leakage and hydrate agglomerate situation, and can not be identified.
Invention content
It is an object of the invention to propose it is a kind of can effectively detect the abnormal conditions inside pipeline and positioned, letter
The pipe safety recognition methods of single, reliable natural gas line safety monitoring assembly.
The present invention is to be realized by the following technical programs：One kind being used for natural gas line safety monitoring based on wavelet packet
The pipe safety recognition methods of device.Optionally, above-mentioned natural gas safety monitoring assembly includes：Online prison based on acoustic excitation
Survey device.
Optionally, the on-Line Monitor Device based on acoustic excitation includes：Loud speaker, power amplifier, multi-functional data are adopted
Truck and microphone.Wherein microphone is for recording reflection wave signal, to determine hydration according to reflection wave propagation time
The position of the tampers such as object or leakage point.
Optionally, the multifunctional data acquisition card uses
The pipe safety recognition methods of natural gas line safety monitoring assembly is used for based on wavelet packet with above-mentioned apparatus, including
Following procedure：
Establish the BP neural network pattern base for endangering pipe safety event type；
The BP neural network for completing training is endangered into pipe safety event for monitoring in real time.
Optionally, it establishes and endangers the BP neural network pattern base of pipe safety event type and include：
Vibration signal characteristics vector extracts；
Using the feature vector of event as the input of BP neural network, training and test b P neural networks are to establish event schema
Library.
Optionally, vibration signal characteristics vector extraction includes the following steps：
(1) sample frequency of vibration signal is set as 2f, and j layers of WAVELET PACKET DECOMPOSITION are carried out to signal, then form 2jA equal widebands
Band, each frequency band section bandwidth are f/2j, after WAVELET PACKET DECOMPOSITION, obtain j layers of wavelet packet coefficientK=0,1 ... 2j- 1,
M identifies for wavelet packet spatial position in formula, if k-th of node wavelet packet coefficient length is n, m=0,1 ... n-1 in j layers,
(2) energy of signal x (t) in the time domain is：
The wavelet conversion coefficient of above formula and x (t)Your energy integral is cut down by pa house：
Equation simultaneous gets up, and obtains：
(3) j layers of WAVELET PACKET DECOMPOSITION are carried out to signal to be processed；Then selection n is most sensitive to the signal energy
Frequency range finally finds out the energy of this n frequency band then into following calculating process respectively：
(4) above process, that is, normalized, then normalized energy is exactly the feature vector of the vibration signal, it is as follows：
T=[T1', T2' ..., Tn′] (6)
(5) due to wavelet packet coefficientThe dimension for square possessing energy, so carrying out the feature parameter vectors extraction.
Each frequency band that the number of winning the confidence is obtained by WAVELET PACKET DECOMPOSITION, the energy of each frequency band are denoted as：
Wherein NiFor the data length of i-th of sub-band, then the energy root mean square of the signal is：
Each frequency band energy value is normalized simultaneously, obtaining feature vector T is：
Optionally, using the feature vector of event as the input of BP neural network, training and test b P neural networks are to establish
Event schema library includes：
(1) network weight through network and the setting of threshold value initial value：
In order to avoid initial value it is excessive caused by network saturation, while taking into account the convergence rate of network and the complexity of sample
Property, the weights and threshold term of BP neural network of the present invention are set to equally distributed smaller random number, take in advance
It is -0.5~0.5.
(2) the network topology structure selection of BP neural network：
BP neural network uses three-decker, i.e. input layer, single hidden layer and output layer in the present invention, input layer number by
Element number in signal characteristic vector determines that output layer number of nodes is consistent with event type number, i.e., system judges event class
The total i of type, and system judge event j (1≤j≤i) occur when, BP neural network output 1 × i forms row vector
[a0a1...ai] in only aj=1, other elements are 0, if each element is all in event generating system without exception output row vector
It is 0, single hidden layer internal segment points should be few as possible in the case where ensureing system approximation accuracy, to improve network convergence rate.
(3) trained BP neural network is tested：
Each anomalous event chooses 10~20 test samples, will to having trained the BP neural network finished to test
The feature vector of collected anomalous event signal inputs trained BP neural network, by BP neural network by calculating
The output gone out is compared with actual anomalous event type, and system erroneous judgement event number and test sample sum are divided by
To system erroneous judgement rate, if the False Rate of test result is less than or equal to the False Rate of design requirement, illustrate the BP nerve nets established
Network model meets design requirement, can be used for safety monitoring along actual pipeline；If False Rate is more than design requirement, mould is adjusted
Shape parameter repeats the above steps, and re -training test b P neural network models are set until BP neural network False Rate meets system
Meter requires.
Optionally, the BP neural network for completing training is endangered pipe safety event for real time monitoring includes：
After BP neural network is test indicate that meet design requirement, monitoring system can acquire the reflection of monitoring system in real time
Signal, the feature vector of extraction reflection signal, input BP neural network are realized and are sent out along online recognition gas hydrates pipeline
Raw anomalous event type, once judging that anomalous event present in BP neural network pattern base occurs along pipeline, system is to different
Ordinary affair part is positioned.
After early warning system is abnormal event, signal propagation time is reflected by calculating, system can be realized to natural gas tube
Incident point in road is positioned, shown in calculating process such as formula (10)
Wherein, y is the position of hydrate in pipeline, and c is the velocity of sound inside pipeline, and t is the time for detecting transmitting signal,
The time can be calculated according to the waveform of time domain acquisition.
The present invention utilizes the function of classifying automatically of BP neural network, can effectively identify the harm being happened at along pipeline
Pipe safety event type, once it is determined that after event type, so that it may run through various means and be subject to safety precaution.With existing skill
Art is compared, and the present invention utilizes BP neural network, can automatically extract fault signature and according to measurement parameter automatic decision failure classes
Type overcomes judging result of the prior art caused by artificial judgment subjectivity and one-sidedness problem with a low credibility, solves
There are the defect of time delay for prior art deterministic process, improves the efficiency of malfunction monitoring, reduces implementation cost.
Description of the drawings
Fig. 1 is a kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
Flow chart；
Fig. 2 is a kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
On-Line Monitor Device structure chart based on acoustic excitation；
Fig. 3 is a kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
Extract vibration signal characteristics vector calculation flow chart；
Fig. 4 is a kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
BP neural network trains flow chart；
Fig. 5-1 is a kind of pipe safety identification side being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
There are reflect signal graph when hydrate cohesion in the collected pipeline of method；
Fig. 5-2 is a kind of pipe safety identification side being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
Signal graph is reflected when there is leakage in the collected pipeline of method；
Fig. 6-1 is a kind of pipe safety identification side being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
Hydrate agglomerates feature vector chart in the typical pipeline of method；
Fig. 6-2 is a kind of pipe safety identification side being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
The typical pipeline internal leakage feature vector chart of method；
Fig. 7 is a kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet of the present invention
Hydrate agglomerates the comparison with the feature vector of the signal of leakage.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Not
It is detached from the case of the principle of the present invention, variation, modification, replacement and deformation is made to the embodiment of the present invention and belong to protection of the present invention
Range.Specific steps of the present invention are as shown in Figure 1：
S1, foundation endanger the BP neural network pattern base of pipe safety event type.
Specifically, the BP neural network pattern base that foundation endangers pipe safety event type includes：
Vibration signal characteristics vector extracts；
Using the feature vector of event as the input of BP neural network, training and test b P neural networks are to establish event schema
Library.
Specifically, the extraction of vibration signal characteristics vector includes：
System electrical signal collection；
Detect the feature extraction of signal.
Specifically, system electrical signal collection equipment therefor is as shown in Figure 1：
In fig. 2, sound wave is sent out by the sound source (5) for being positioned over pipeline head end inlet, and the pumping signal of the sound source is by counting
Software systems in calculation machine (1) generate and export to multifunctional data acquisition card (2) digital-to-analogue conversion part after through sound source drive mould
Block (4) exports to drive sound source to send out required acoustic signals.Sound wave in pipeline will from head end along pipe transmmision, once encounter
Hydrate (8) or leakage point (7), part energy will be returned in the form of back wave to head end, and it is attached to be fixed on entrance
Close microphone (6) is for recording reflection wave signal, to determine hydrate or leakage point according to reflection wave propagation time
The position of equal tampers.Microphone output turns entering signal conditioning module (3), the rear modulus into multifunctional data acquisition card
Part is changed, computer is finally entered and is analyzed.
Specifically, the feature extraction of detection signal includes：
The abnormal signal that above-mentioned pipe safety intelligent monitor system obtains is calculated by WAVELET PACKET DECOMPOSITION as shown in Figure 3
Energy on sensitive frequency section, as the signal characteristic vector that the feature vector of vibration signal, the step obtain, according to
Obtained feature vector is classified.
Each abnormal conditions selects several representational 20 collecting samples, is extracted and is examined by signal characteristic abstraction flow
The feature vector of signal is surveyed, this feature vector is 1 × 8 type vector of normalized energy composition.In order to make sample signal have generation
Table reflects letter caused by choosing reflection signal caused by the hydrate of different cross-sectional and the leak by different location respectively
Number as carry out feature vector extraction vibration signal sample.
Specifically, using the feature vector of event as the input of BP neural network, training and test b P neural networks are to establish
Event schema library includes：
BP neural network is established and training；
Verification gained BP neural network.
Specifically, BP neural network is established and training includes：
The present invention uses the BP neural network model of 8 × 10 × 2 structures, i.e., input number of nodes is 8, implicit number of nodes is
10, output node number is 2, is trained and tests to neural network model.
The feature vector of vibration signal is caused to be inputted as BP neural network above two anomalous event, corresponding exception thing
Part type is trained BP neural network as output, and specific training process is as shown in Fig. 4 flows.
Specifically, verification gained BP neural network includes：
The present invention carries out simulating hydrate cohesion and leak test inside experimental channel.Using one section long in experiment
21.6m, the closed steel pipe that internal diameter of the pipeline is 100mm carry out simulated experiment.Pipe safety intelligent monitor system is installed in experiment
On experimental channel.In order to simulate pipe leakage experiment, the present embodiment in inside by placing artificial ice for simulating in pipeline
Hydrate agglomerates situation, and experimental channel used is in A (4.56m), B (11.02m), four positions C (14.43m) and D (18.65m)
It sets and machined leak for simulating pipe leakage situation.
The frequency for the pumping signal that the present invention takes is 800Hz, and Fig. 5-1 is that signal is reflected caused by typical hydrate,
Fig. 5-2 is to reflect signal caused by typical leakage.Then by the carry out wavelet packet processing of the signal to acquisition, according to anti-
The frequency distribution feature of signal is penetrated, ranges of the 0~3.125KHz as 8 frequency bands is chosen.In the extraction work to feature vector
Later, different experiments data reflection signal will obtain different feature vectors.Fig. 6-1 and Fig. 6-2 is respectively typical hydration
There is apparent difference in the character vectogram of object and leakage reflection signal, the two.
Cause the difference for emitting the feature vector of signal to further probe into hydrate and leakage, respectively to being cut by difference
Signal caused by the hydrate cohesion of area (D=30mm, 40mm, 50mm, 60mm, 70mm) carries out the extraction of feature vector.Together
When will caused by the leak by different location reflect signal carry out feature vector extraction.20 groups are all chosen for each signal
Data carry out analyzing processing, and after carrying out characteristic vector pickup to these signals, the average value of the energy of each frequency band distribution is such as
Shown in the following table 1.
1 hydrate of table agglomerates the feature vector with reflection signal caused by leakage
Fig. 7 is the comparison for the Energy distribution that hydrate causes the feature vector for reflecting signal with leakage, as we know from the figure water
It is more apparent to close the feature vector difference of both object and leakage.By Energy extraction as a result, can obtain to draw a conclusion：
(1) most of energy of the feature vector of reflection signal caused by hydrate all concentrates on first frequency band, and
Remaining energy is mainly distributed on second band and the 4th frequency band.
(2) the most energy for the feature vector for reflecting signal are caused all to be distributed in first frequency band by leakage, and the
The energy occupancy volume of one frequency band is all 90% or more.
(3) there is apparent difference in the feature vector of hydrate and reflection signal caused by leakage, illustrate that this method is available
In identification.
S2, the BP neural network for completing training is endangered into pipe safety event for monitoring in real time.
Specifically, detecting system is enabled to be monitored in real time along pipeline, and detection signal characteristic input is trained
BP neural networks, when the neural network judges to occur along pipeline the anomalous event of threat tube safety, system will swash
Locating module living.System locating module reflects signal propagation time by measurement and is positioned to incident point.
The above, therefore one of preferred embodiment only of the present invention is not departing from the principle of the present invention, right
The embodiment of the present invention is made variation, modification, replacement and deformation and be shall fall within the protection scope of the present invention.
Claims (6)
1. a kind of pipe safety recognition methods based on wavelet packet and natural gas line safety monitoring assembly, it is characterised in that：One
Pipe safety recognition methods of the kind based on wavelet packet and natural gas line safety monitoring assembly includes the following steps：
S1, foundation endanger the BP neural network pattern base of pipe safety event type；
S2, the BP neural network for completing training is endangered into pipe safety event for monitoring in real time.
2. according to claim 1, a kind of based on the knowledge of the pipe safety of wavelet packet and natural gas line safety monitoring assembly
Other method, which is characterized in that using the feature vector of event as the input of BP neural network, training and test b P neural networks are to build
The vertical BP neural network pattern base for endangering pipe safety event type includes：It includes that pipeline hydrate is solidifying to endanger pipe safety event
Poly- and two aspects of pipe leakage, cause the hydrate by different cross-sectional (D=30mm, 40mm, 50mm, 60mm and 70mm)
Reflection signal and leak by different location caused by reflection signal choose 15~20 respectively and be used as sample, according to above-mentioned
Feature extraction flow extracts its feature vector, using the signal characteristic vector of extraction as the input of BP neural network, corresponding event
Type is trained BP neural network as output, until BP neural network False Rate meets requirement of system design.
3. according to claim 1 a kind of based on the identification of the pipe safety of wavelet packet and natural gas line safety monitoring assembly
Method, which is characterized in that establish vibration signal characteristics vector in the BP neural network pattern base for endangering pipe safety event type
Extraction includes：
1) pumping signal that the software in computer generates is exported through sound source drive module, and pipeline head end entrance is positioned over to drive
The sound source at place sends out acoustic signals, when the sound wave along pipeline Propagation encounters the hydrate or pipe leakage of tube wall attachment
It will produce reflection signal；
2) when entrance is placed around microphone acquisition incident acoustic wave and encounters the hydrate or pipe leakage of tube wall attachment
Reflection signal is generated, the hydrate or leak position are determined according to the time of its propagation by acquiring the reflection signal
Position；
3) then the reflection signal entering signal conditioning module of microphone output carries out modulus by multifunctional data acquisition card and turns
It changes, finally enters computer and analyzed；
4) its feature vector is extracted by WAVELET PACKET DECOMPOSITION for collected reflection signal.
4. according to claim 3 a kind of based on the identification of the pipe safety of wavelet packet and natural gas line safety monitoring assembly
Method, it is characterised in that：The pumping signal of sound source is generated by the software systems in computer and is exported to multifunctional data acquiring
The analog-to-digital conversion part of card sends out required pulsed sound signal through sound source drive module output driving sound source.
5. according to claim 3 a kind of based on the identification of the pipe safety of wavelet packet and natural gas line safety monitoring assembly
Method, which is characterized in that the characteristic vector pickup of vibration signal includes：
(1) sample frequency of vibration signal is set as 2f, and j layers of WAVELET PACKET DECOMPOSITION are carried out to signal, then form 2jThe broadbands such as a, often
A frequency band section bandwidth is f/2j, after WAVELET PACKET DECOMPOSITION, obtain j layers of wavelet packet coefficientK=0,1 ... 2j- 1, m in formula
It is identified for wavelet packet spatial position, if k-th of node wavelet packet coefficient length is n, m=0,1 ... n-1 in j layers,
(2) energy of signal x (t) in the time domain is
The wavelet conversion coefficient of above formula and x (t)Your energy integral is cut down by pa house
Equation simultaneous gets up, and obtains
(3) j layers of WAVELET PACKET DECOMPOSITION are carried out to signal to be processed；Then n frequency bands most sensitive to the signal energy of selection
Range finally finds out the energy of this n frequency band then into following calculating process respectively：
(4) above process, that is, normalized, then normalized energy is exactly the feature vector of the vibration signal, it is as follows：
(5) due to wavelet packet coefficientThe dimension for square possessing energy, so carrying out the feature parameter vectors extraction；The number of winning the confidence
The each frequency band obtained by WAVELET PACKET DECOMPOSITION, the energy of each frequency band are denoted as
Wherein NiFor the data length of i-th of sub-band, then the energy root mean square of the signal is
Each frequency band energy value is normalized simultaneously, obtaining feature vector T is
。
6. according to claim 1, a kind of based on the knowledge of the pipe safety of wavelet packet and natural gas line safety monitoring assembly
Other method, which is characterized in that the BP neural network for completing training is endangered pipe safety event for real time monitoring includes：Work as BP
Neural network shows after meeting design requirement that monitoring system can acquire the reflection signal of monitoring system in real time after tested, and extraction is anti-
The feature vector of signal is penetrated, BP neural network is inputted, realizes and is abnormal event along online recognition gas hydrates pipeline
Type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242500.4A CN108361560A (en) | 2018-03-21 | 2018-03-21 | A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242500.4A CN108361560A (en) | 2018-03-21 | 2018-03-21 | A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108361560A true CN108361560A (en) | 2018-08-03 |
Family
ID=63001244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810242500.4A Pending CN108361560A (en) | 2018-03-21 | 2018-03-21 | A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108361560A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109058771A (en) * | 2018-10-09 | 2018-12-21 | 东北大学 | The pipeline method for detecting abnormality of Markov feature is generated and is spaced based on sample |
| CN110285332A (en) * | 2019-06-20 | 2019-09-27 | 天津科技大学 | A kind of natural gas line safety monitoring key technology based on acoustic pulses compression |
| CN110309558A (en) * | 2019-06-13 | 2019-10-08 | 中国石油大学(北京) | Pipe safety coefficient generation method and device based on radial basis function neural network |
| CN110398320A (en) * | 2019-07-09 | 2019-11-01 | 北京讯腾智慧科技股份有限公司 | A kind of gas leakage detection localization method and system being easy to Continuous optimization |
| WO2020136477A1 (en) * | 2018-12-27 | 2020-07-02 | Atlas Copco Airpower, Naamloze Vennootschap | Method for detecting obstructions in a gas network under pressure or under vacuum and gas network |
| BE1026836B1 (en) * | 2018-12-27 | 2021-01-06 | Atlas Copco Airpower Nv | A method for detecting obstructions in a gas network under pressure or under vacuum and gas network |
| CN113468804A (en) * | 2021-06-10 | 2021-10-01 | 电子科技大学 | Underground pipeline identification method based on matrix bundle and deep neural network |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101183899A (en) * | 2007-12-19 | 2008-05-21 | 天津大学 | BP network based pipeline security identifying method for optical fiber pipeline leakage monitoring device |
| CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
| JP2010025715A (en) * | 2008-07-18 | 2010-02-04 | Nippon Oil Corp | Pipe crack diagnosing apparatus and pipe crack diagnosing method |
| CN101832471A (en) * | 2010-04-19 | 2010-09-15 | 哈尔滨工程大学 | Signal identification and classification method |
| US20130066568A1 (en) * | 2010-04-15 | 2013-03-14 | Julio Roberto Alonso | Integrated system with acoustic technology, mass imbalance and neural network for detecting, locating and quantifying leaks in ducts |
| CN104061445A (en) * | 2014-07-09 | 2014-09-24 | 中国石油大学(华东) | Pipeline leakage detection method based on neural network |
| CN104373821A (en) * | 2014-11-21 | 2015-02-25 | 天津科技大学 | Natural gas pipeline safety monitoring device based on acoustical science active spurring |
| CN106525349A (en) * | 2016-11-18 | 2017-03-22 | 山西中天信科技股份有限公司 | Combustible gas leakage detection method and system |
-
2018
- 2018-03-21 CN CN201810242500.4A patent/CN108361560A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101183899A (en) * | 2007-12-19 | 2008-05-21 | 天津大学 | BP network based pipeline security identifying method for optical fiber pipeline leakage monitoring device |
| JP2010025715A (en) * | 2008-07-18 | 2010-02-04 | Nippon Oil Corp | Pipe crack diagnosing apparatus and pipe crack diagnosing method |
| CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
| US20130066568A1 (en) * | 2010-04-15 | 2013-03-14 | Julio Roberto Alonso | Integrated system with acoustic technology, mass imbalance and neural network for detecting, locating and quantifying leaks in ducts |
| CN101832471A (en) * | 2010-04-19 | 2010-09-15 | 哈尔滨工程大学 | Signal identification and classification method |
| CN104061445A (en) * | 2014-07-09 | 2014-09-24 | 中国石油大学(华东) | Pipeline leakage detection method based on neural network |
| CN104373821A (en) * | 2014-11-21 | 2015-02-25 | 天津科技大学 | Natural gas pipeline safety monitoring device based on acoustical science active spurring |
| CN106525349A (en) * | 2016-11-18 | 2017-03-22 | 山西中天信科技股份有限公司 | Combustible gas leakage detection method and system |
Non-Patent Citations (1)
| Title |
|---|
| 曲志刚等: "基于支持向量机的油气管道安全监测信号识别方法", 《天津大学学报》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109058771A (en) * | 2018-10-09 | 2018-12-21 | 东北大学 | The pipeline method for detecting abnormality of Markov feature is generated and is spaced based on sample |
| CN114072612A (en) * | 2018-12-27 | 2022-02-18 | 阿特拉斯·科普柯空气动力股份有限公司 | Method for detecting blockage in gas pipe network under pressure or vacuum and gas pipe network |
| WO2020136477A1 (en) * | 2018-12-27 | 2020-07-02 | Atlas Copco Airpower, Naamloze Vennootschap | Method for detecting obstructions in a gas network under pressure or under vacuum and gas network |
| BE1026836B1 (en) * | 2018-12-27 | 2021-01-06 | Atlas Copco Airpower Nv | A method for detecting obstructions in a gas network under pressure or under vacuum and gas network |
| CN114072612B (en) * | 2018-12-27 | 2023-11-10 | 阿特拉斯·科普柯空气动力股份有限公司 | Method for detecting blockage in gas pipe network under pressure or vacuum and gas pipe network |
| US12055165B2 (en) | 2018-12-27 | 2024-08-06 | Atlas Copco Airpower, Naamloze Vennootschap | Method for detecting obstructions in a gas network under pressure or under vacuum and gas network |
| CN110309558A (en) * | 2019-06-13 | 2019-10-08 | 中国石油大学(北京) | Pipe safety coefficient generation method and device based on radial basis function neural network |
| CN110309558B (en) * | 2019-06-13 | 2021-05-25 | 中国石油大学(北京) | Pipeline safety coefficient generation method and device based on radial basis function neural network |
| CN110285332A (en) * | 2019-06-20 | 2019-09-27 | 天津科技大学 | A kind of natural gas line safety monitoring key technology based on acoustic pulses compression |
| CN110398320A (en) * | 2019-07-09 | 2019-11-01 | 北京讯腾智慧科技股份有限公司 | A kind of gas leakage detection localization method and system being easy to Continuous optimization |
| CN110398320B (en) * | 2019-07-09 | 2020-11-20 | 北京讯腾智慧科技股份有限公司 | Gas leakage detection positioning method and system easy for continuous optimization |
| CN113468804A (en) * | 2021-06-10 | 2021-10-01 | 电子科技大学 | Underground pipeline identification method based on matrix bundle and deep neural network |
| CN113468804B (en) * | 2021-06-10 | 2023-09-19 | 电子科技大学 | Underground pipeline identification method based on matrix bundles and deep neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108361560A (en) | A kind of pipe safety recognition methods being used for natural gas line safety monitoring assembly based on wavelet packet | |
| CN101539241B (en) | Hierarchical multi-source data fusion method for pipeline linkage monitoring network | |
| CN103033567B (en) | Pipeline defect signal identification method based on guided wave | |
| CN104421620B (en) | Leakage signal analysis method | |
| CN107748314A (en) | Transformer Faults Analysis system based on sound wave shock detection | |
| CN113124929A (en) | Transformer substation multi-parameter signal acquisition comprehensive analysis system and method | |
| CN101718634B (en) | Equipment state comprehensive dynamic alarming method based on multivariate probability model | |
| CN111982273B (en) | Noise monitoring method for rail transit station platform | |
| CN109086888A (en) | Automobile engine failure judgment method and its device based on voice recognition | |
| CN104373821A (en) | Natural gas pipeline safety monitoring device based on acoustical science active spurring | |
| CN101592288B (en) | Method for identifying pipeline leakage | |
| CN109838696A (en) | Pipeline fault diagnostic method based on convolutional neural networks | |
| CN105679313A (en) | Audio recognition alarm system and method | |
| CN107355686B (en) | A kind of detection method of drain line blockage failure | |
| CN106198749B (en) | A kind of data fusion method of the multiple sensor based on Metal Crack monitoring | |
| CN105160359B (en) | A kind of labyrinth damage collaboration recognition methods based on supersonic guide-wave | |
| CN107844067B (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN109886433A (en) | The method of intelligent recognition city gas pipeline defect | |
| CN113566948A (en) | Fault audio recognition and diagnosis method for robot coal pulverizer | |
| CN111007365A (en) | Ultrasonic partial discharge identification method and system based on neural network | |
| CN105118511A (en) | Thunder identification method | |
| CN112985574B (en) | High-precision classification identification method for optical fiber distributed acoustic sensing signals based on model fusion | |
| CN111520615A (en) | Pipe network leakage identification and positioning method based on line spectrum pair and cubic interpolation search | |
| CN114383054A (en) | Pipe gallery gas pipeline leakage experiment system and method | |
| CN110365639A (en) | A kind of malicious traffic stream detection method and system based on depth residual error network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180803 |
|
| WD01 | Invention patent application deemed withdrawn after publication |