CN110657355A - Method for detecting leakage of thermal pipeline - Google Patents
Method for detecting leakage of thermal pipeline Download PDFInfo
- Publication number
- CN110657355A CN110657355A CN201910800692.0A CN201910800692A CN110657355A CN 110657355 A CN110657355 A CN 110657355A CN 201910800692 A CN201910800692 A CN 201910800692A CN 110657355 A CN110657355 A CN 110657355A
- Authority
- CN
- China
- Prior art keywords
- noise signal
- background noise
- leakage
- time
- noise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
Abstract
The application discloses a thermal pipeline leakage detection method, which solves the problems that in the prior art, the position of a thermal pipeline needs to be checked, and the precision is insufficient. And acquiring background noise of the pipeline, and constructing a background noise database according to the Manhattan distance. And extracting effective noise signals according to the maximum time-frequency oscillation ratio of the background noise database. And extracting a suspected leakage noise signal by a noise clustering method and a category delineation method. And judging whether the noise signal is a leakage noise signal or a background noise signal according to the Mahalanobis distance of the time-frequency domain combined feature similarity. The invention aims to provide a thermal pipeline leakage detection method, which utilizes a noise recorder to collect pipeline noise signals, detects and judges leakage noise of a pipeline in real time, finds pipeline leakage in time and gives an alarm, and provides a basis for scientifically maintaining pipeline operation.
Description
Technical Field
The application relates to the field of machine learning and pattern recognition, in particular to a thermal pipeline leakage detection method.
Background
With the acceleration of the urbanization process in China, the centralized heating is developed vigorously. However, the existing heat supply pipeline has low construction level and serious leakage, which causes huge energy waste and pipeline investment loss. Because the construction quality of a heat supply network is poor, the pipeline leakage is difficult to search, the detection of medium leakage faults in the pipeline basically depends on the methods of very low efficiency and lagging such as pressure rising, sectional valve closing and the like to search a leakage pipeline section, and then leakage points are searched through a correlator, a sound bar, an ultrasonic detector and the like. The leakage point is searched by using the instrument, the leakage point needs to be checked by manually arriving at the position of the thermal pipeline, the leakage of the thermal pipeline cannot be monitored for a long time, and the leakage is easy to cause.
Disclosure of Invention
The embodiment of the application provides a thermal pipeline leakage detection method, and solves the problems that in the prior art, the position of a thermal pipeline needs to be checked, and the precision is insufficient.
The application provides a thermal pipeline leakage detection method, which comprises the following steps:
calculating the Manhattan distance between the newly acquired background noise signal and all signals in the background noise database respectively to obtain a first maximum value, and adding the signal into the background noise database if the first maximum value is greater than a Manhattan distance threshold value so as to obtain a background noise database;
obtaining the maximum time-frequency oscillation starting ratio of the background noise database by using the ratio of the time domain oscillation peak value and the frequency domain energy density of all signals in the background noise database;
acquiring a noise signal of a pipeline in real time, calculating the ratio of a time domain oscillation peak value to a frequency domain energy density, and if the ratio of the time domain oscillation peak value to the frequency domain energy density is greater than the maximum time-frequency oscillation starting ratio determined by weighting, determining the noise signal to be an effective noise signal;
gathering all signals in the background noise database into m classes by using a mean shift algorithm, and calculating the minimum class interval of the background noise database by using the Euclidean distance between the centroid of each two classes of signals;
calculating Euclidean distance between the mass centers of the effective noise signal and the m types of signals in the background noise database to obtain a second maximum value, if the second maximum value is larger than the minimum type distance, judging the effective noise signal as a suspected leakage noise signal, and otherwise, judging the effective noise signal as background interference noise;
carrying out empirical mode decomposition on the suspected leakage noise signals, calculating time-frequency domain joint characteristic vectors of all signals in a background noise database, and forming a time-frequency domain joint characteristic sample set of the background noise database;
calculating the mahalanobis distance between two vectors in the TF through the mean vector and the covariance matrix of the time-frequency domain combined characteristic sample set, and calculating the mahalanobis distance of the maximum mahalanobis distance and the suspected leakage noise signal to the time-frequency domain combined characteristic sample set;
and if the mahalanobis distance from the suspected leakage noise signal to the time-frequency domain combined feature sample set is greater than the maximum mahalanobis distance subjected to weighted classification, the suspected leakage noise signal is a leakage noise signal, and otherwise, the suspected leakage noise signal is background interference noise.
Further, the method also comprises the following steps:
and adding the noise signal of the judged error category into a background noise database, and updating the background noise database and the judgment parameters.
Preferably, the acquisition interval of the background noise signal is 10 minutes, the acquisition time is 10 seconds, and the sampling frequency is 6000 points.
Preferably, the value of the weighting judgment coefficient is 0.6-0.75.
Preferably, the coefficient value of the weighted classification is 0.9-1.15.
Preferably, the acquisition time of the background noise signal is 15 days before the start of the heating day and 15 days after the start of the heating day.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects:
the invention aims to provide a thermal pipeline leakage detection method, which utilizes a noise recorder to collect pipeline noise signals, detects and judges leakage noise of a pipeline in real time, finds pipeline leakage in time and gives an alarm, and provides a basis for scientifically maintaining pipeline operation. The method starts from the background noise of the thermal pipeline which is convenient to collect, judges whether the actually collected noise signal is a leakage noise signal step by step from coarse to fine by establishing a background noise database, and the accuracy of a thermal pipeline leakage detection model is higher and higher along with the continuous update of the background noise database. The invention has strong applicability to the leakage detection of the thermal pipeline, can monitor the leakage state of the pipeline in real time and give an alarm in time when the leakage occurs, and is beneficial to improving the level of the operation and maintenance of the thermal pipeline.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a flow chart of an embodiment of a method for detecting a leakage in a thermal line;
FIG. 2 is a flow chart of another embodiment of a method for thermal line leak detection;
FIG. 3 is a flow chart of a method for detecting leakage in a thermal pipeline.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Example 1
FIG. 1 is a flow chart of an embodiment of a method for detecting a leakage in a thermal line.
The application provides a thermal pipeline leakage detection method, which comprises the following steps:
In step 101, a manhattan distance is calculated between the newly acquired background noise signal and all signals in the background noise database to obtain a first maximum value, and if the first maximum value is greater than a threshold of the manhattan distance, the signal is added to the background noise database to obtain the background noise database.
For example, let the Manhattan distance threshold of the background noise signal be TmdCalculating the Manhattan distance between the newly acquired background noise signal N and all the signals in the background noise database to obtain a first maximum value MmdIf M is presentmd＞TmdThe signal is added to the background noise database, otherwise N is discarded, thus obtaining the background noise database.
It should be noted that, if the background noise signal N is the background noise signal acquired first, the background noise signal N does not need to be added to the background noise database directly.
The T ismdAnd the artificially set threshold is used for comparing all the signals in the background noise database with a first maximum value obtained by calculating the Manhattan distance of the newly acquired background noise signal, and determining the acceptance or rejection of the newly acquired background noise signal.
E.g. Tmd＝0.4
The Manhattan distance, the Manhattan distance of signals X and Y is md (X, Y).
Usually, the heating mechanism performs a test operation on the thermal pipeline one month before the heating starts, so that the background noise signal can be collected at the time of the test operation, and therefore, the collection time of the background noise signal is preferably 15 days before the heating day starts and 15 days after the heating day starts.
Preferably, the acquisition interval of the background noise signal is 10 minutes, the acquisition time is 10 seconds, and the sampling frequency is 6000 points.
And 102, extracting effective noise signals according to the maximum time-frequency oscillation ratio of the background noise database.
In step 102, the time domain oscillation peak and the frequency domain energy density of the signal are calculated, and the maximum time-frequency oscillation ratio of the background noise database is obtained by using the ratio of the time domain oscillation peak and the frequency domain energy density of all the signals in the background noise database.
And acquiring a noise signal of the pipeline in real time, calculating the ratio of the time domain oscillation peak value to the frequency domain energy density, and if the ratio of the time domain oscillation peak value to the frequency domain energy density is greater than the maximum time-frequency oscillation starting ratio determined by weighting, determining that the noise signal is an effective noise signal.
For example, calculate the maximum time-frequency oscillation-starting ratio OM of the background noise databasetf，
Time-domain oscillation peak value op (X) of signal X:
in equation (2), max () is a maximum function and min () is a minimum function.
Frequency-domain energy density ed (X) of signal X:
ed(X)＝sum(fft(X)·∧2)/Fm (3)
in the formula (3), fft () is a fast Fourier transform function, sum () is a summation function, FmThe maximum frequency value of the X spectrum. Where Λ represents the same vector multiplied by multiple points.
Obtaining the maximum time-frequency oscillation starting ratio OM of the background noise database by the ratio of the time domain oscillation peak value and the frequency domain energy density of all the signals in the background noise databasetf：
In the formula (4), XkRepresenting the kth signal in the background noise database, and n is the number of signals in the database.
Collecting noise signal M of pipeline in real time, calculating op (M)/ed (M), if op (M)/ed (M) is greater than alpha × OMtfThen M is the valid noise signal, otherwise M is discarded. Said alpha is a coefficient of the weighted decision,preferably, the value of the weighting judgment coefficient is 0.6-0.75.
And 103, extracting a suspected leakage noise signal by a noise clustering method and a category delineation method.
In step 103, a noise signal of the pipeline is collected in real time, a ratio of the time domain oscillation peak value to the frequency domain energy density is calculated, and if the ratio of the time domain oscillation peak value to the frequency domain energy density is greater than the maximum time-frequency oscillation ratio determined by weighting, the noise signal is an effective noise signal.
And (3) utilizing a mean shift algorithm to gather all signals in the background noise database into m classes, and calculating the minimum class distance of the background noise database by using the Euclidean distance between the centroid of each two classes of signals.
And calculating the Euclidean distance between the mass centers of the effective noise signal and the m types of signals in the background noise database to obtain a second maximum value, if the second maximum value is larger than the minimum type distance, judging that the effective noise signal is a suspected leakage noise signal, and otherwise, judging that the effective noise signal is background interference noise.
For example, all signals in the background noise database are gathered into m classes by using a mean shift algorithm, and the Euclidean distance between the centroids of every two classes of signals is recorded as DkThere are a total of K ═ m (m-1)/2 class spacings, i.e., K ═ 1,2, …, K. Thus, the minimum class spacing D of the background noise databasemi：
Dmi＝min(Dk),k＝1,2,...,k (5)
Calculating Euclidean distance between the mass center of the M-type signals in the effective noise signal M and the background noise database to obtain a second maximum value DteIf D iste＞DmiIf yes, M is judged to be a suspected leakage noise signal, otherwise, the background interference noise is judged.
And step 104, judging whether the noise signal is a leakage noise signal or a background noise signal according to the Mahalanobis distance of the time-frequency domain combined feature similarity.
In step 104, empirical mode decomposition is performed on the suspected leakage noise signal, and time-frequency domain joint feature vectors of all signals in the background noise database are calculated to form a time-frequency domain joint feature sample set of the background noise database.
And calculating the mahalanobis distance between two vectors in the TF through the mean vector and the covariance matrix of the time-frequency domain combined characteristic sample set, and calculating the mahalanobis distance of the maximum mahalanobis distance and the suspected leakage noise signal to the time-frequency domain combined characteristic sample set.
And if the mahalanobis distance from the suspected leakage noise signal to the time-frequency domain combined feature sample set is greater than the maximum mahalanobis distance subjected to weighted classification, the suspected leakage noise signal is a leakage noise signal, and otherwise, the suspected leakage noise signal is background interference noise.
For example, the suspected leakage noise signal M is empirically modal decomposed,
in the formula (6), emd () represents an empirical mode decomposition function, imfkRepresenting the natural modal components, n in total, ref representing the residual component;
calculating the time-frequency domain joint characteristic vectors of all signals in the background noise database to form a time-frequency domain joint characteristic sample set TF of the background noise database
The time-frequency domain joint feature vector of M, the time-frequency domain joint feature vector tf (M), is:
tf(M)＝cat([mean(imfk),median(imfk),fem(imfk),ffm(imkf)],l) (7)
in the formula (7), k is 1,2, …, l, mean () represents a mean function, mean () represents a median function, fem () represents a spectrum energy maximum frequency value function, ffm () represents a spectrum maximum spectrum value function, and cat (, l) represents that l vectors are connected in series;
the time-frequency domain joint feature sample set TF ═ (TF)1,...,tfn) Wherein tf is1The time-frequency domain of the signal is combined with the feature vector, and so on. Setting the mean vector of TF as mu and the covariance matrix as S, and calculating the Mahalanobis distance M between two vectors in TFij：
In equation (8), () T denotes a vector transposition operation, i, j is 1,2, …, n, and M is the largestijRecording as MS;
calculating the mahalanobis distance MD of the signal M to TF:
if MD > β × MS, M is the leakage noise signal, otherwise, M is the background interference noise. And beta is a coefficient of weighted classification, and preferably, the value of the coefficient of weighted classification is 0.9-1.15.
Example 2
FIG. 2 is a flow chart of another embodiment of a method for thermal line leak detection.
The application provides a thermal pipeline leakage detection method, which comprises the following steps:
And 102, extracting effective noise signals according to the maximum time-frequency oscillation ratio of the background noise database.
And 103, extracting a suspected leakage noise signal by a noise clustering method and a category delineation method.
And step 104, judging whether the noise signal is a leakage noise signal or a background noise signal according to the Mahalanobis distance of the time-frequency domain combined feature similarity.
Further, the method also comprises the following steps:
and 105, adding the noise signal of the judged error category into a background noise database, and updating the background noise database and the judgment parameters.
In step 105, if the background interference noise is determined to be the leakage noise signal, the background interference noise is added to the background noise database, the database is updated, and the OM is updated at the same timetf、DmiAnd MS, etc.
And if the noise signal is judged to be a leakage noise signal, the leakage is considered to occur, whether the thermal pipeline is leaked or not is judged through field inspection of maintenance personnel, and if the thermal pipeline is not leaked, the noise signal is a judgment error.
FIG. 3 is a flow chart of a method for detecting leakage in a thermal pipeline.
And acquiring an actually measured signal, extracting an effective noise signal according to the maximum time-frequency oscillation ratio of the background noise database, and continuing to judge in the next step if the effective noise signal is obtained. And extracting suspected leakage noise signals through noise clustering and class delineation, and continuing to judge in the next step if the suspected leakage noise signals are judged. And judging the type of the noise signal according to the Mahalanobis distance of the time-frequency domain combined feature similarity, and if the noise signal is a leakage noise signal, giving a leakage alarm. Positioning and maintaining the leakage point by the maintenance personnel, judging whether the alarm is accurate, if not, adding the background interference noise into a background noise database, updating the database and updating an OM (open-close memory) at the same timetf、DmiAnd MS, etc.
It should be noted that the signal that is judged to be unnecessary for further judgment is discarded, and the noise signal is collected again for detection.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (6)
1. A method of detecting a leakage in a thermal line, comprising the steps of:
calculating the Manhattan distance between the newly acquired background noise signal and all signals in the background noise database respectively to obtain a first maximum value, and adding the signal into the background noise database if the first maximum value is greater than a Manhattan distance threshold value so as to obtain a background noise database;
obtaining the maximum time-frequency oscillation starting ratio of the background noise database by using the ratio of the time domain oscillation peak value and the frequency domain energy density of all signals in the background noise database;
acquiring a noise signal of a pipeline in real time, calculating the ratio of a time domain oscillation peak value to a frequency domain energy density, and if the ratio of the time domain oscillation peak value to the frequency domain energy density is greater than the maximum time-frequency oscillation starting ratio determined by weighting, determining the noise signal to be an effective noise signal;
gathering all signals in the background noise database into m classes by using a mean shift algorithm, and calculating the minimum class interval of the background noise database by using the Euclidean distance between the centroid of each two classes of signals;
calculating Euclidean distance between the mass centers of the effective noise signal and the m types of signals in the background noise database to obtain a second maximum value, if the second maximum value is larger than the minimum type distance, judging the effective noise signal as a suspected leakage noise signal, and otherwise, judging the effective noise signal as background interference noise;
carrying out empirical mode decomposition on the suspected leakage noise signals, calculating time-frequency domain joint characteristic vectors of all signals in a background noise database, and forming a time-frequency domain joint characteristic sample set of the background noise database;
calculating the mahalanobis distance between two vectors in the TF through the mean vector and the covariance matrix of the time-frequency domain combined characteristic sample set, and calculating the mahalanobis distance of the maximum mahalanobis distance and the suspected leakage noise signal to the time-frequency domain combined characteristic sample set;
and if the mahalanobis distance from the suspected leakage noise signal to the time-frequency domain combined feature sample set is greater than the maximum mahalanobis distance subjected to weighted classification, the suspected leakage noise signal is a leakage noise signal, and otherwise, the suspected leakage noise signal is background interference noise.
2. The method of claim 1, further comprising the steps of:
and adding the noise signal of the judged error category into a background noise database, and updating the background noise database and the judgment parameters.
3. The method for detecting leakage of thermal pipeline according to claim 1, wherein the background noise signal is collected at 10 min intervals, the collection time is 10 s, and the sampling frequency is 6000 points.
4. The thermal pipeline leakage detection method according to claim 1, wherein the coefficient value of the weighted decision is 0.6-0.75.
5. The thermal pipeline leakage detection method according to claim 1, wherein the coefficient value of the weighted classification is 0.9-1.15.
6. The method of claim 1, wherein the background noise signal is collected 15 days before the start of the heating day and 15 days after the start of the heating day.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910800692.0A CN110657355B (en) | 2019-08-28 | 2019-08-28 | Method for detecting leakage of thermal pipeline |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910800692.0A CN110657355B (en) | 2019-08-28 | 2019-08-28 | Method for detecting leakage of thermal pipeline |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110657355A true CN110657355A (en) | 2020-01-07 |
| CN110657355B CN110657355B (en) | 2021-06-01 |
Family
ID=69036476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910800692.0A Active CN110657355B (en) | 2019-08-28 | 2019-08-28 | Method for detecting leakage of thermal pipeline |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110657355B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113567063A (en) * | 2021-07-23 | 2021-10-29 | 宁波水表(集团)股份有限公司 | Construction and application method of water supply pipe network leakage noise database |
| CN115950590A (en) * | 2023-03-15 | 2023-04-11 | 凯晟动力技术（嘉兴）有限公司 | Gas engine leakage early warning system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5416724A (en) * | 1992-10-09 | 1995-05-16 | Rensselaer Polytechnic Institute | Detection of leaks in pipelines |
| CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
| CN103234121A (en) * | 2013-05-10 | 2013-08-07 | 中国石油大学(华东) | Acoustic signal based device and method for detecting gas pipeline leakages |
| CN108980630A (en) * | 2017-05-31 | 2018-12-11 | 西门子（中国）有限公司 | Pipeline leakage detection method and device |
| CN109284777A (en) * | 2018-08-28 | 2019-01-29 | 内蒙古大学 | Recognition methods is leaked based on signal time-frequency characteristics and the water supply line of support vector machines |
-
2019
- 2019-08-28 CN CN201910800692.0A patent/CN110657355B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5416724A (en) * | 1992-10-09 | 1995-05-16 | Rensselaer Polytechnic Institute | Detection of leaks in pipelines |
| CN101592288A (en) * | 2009-05-22 | 2009-12-02 | 重庆大学 | A kind of method for identifying pipeline leakage |
| CN103234121A (en) * | 2013-05-10 | 2013-08-07 | 中国石油大学(华东) | Acoustic signal based device and method for detecting gas pipeline leakages |
| CN108980630A (en) * | 2017-05-31 | 2018-12-11 | 西门子（中国）有限公司 | Pipeline leakage detection method and device |
| CN109284777A (en) * | 2018-08-28 | 2019-01-29 | 内蒙古大学 | Recognition methods is leaked based on signal time-frequency characteristics and the water supply line of support vector machines |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113567063A (en) * | 2021-07-23 | 2021-10-29 | 宁波水表(集团)股份有限公司 | Construction and application method of water supply pipe network leakage noise database |
| CN115950590A (en) * | 2023-03-15 | 2023-04-11 | 凯晟动力技术（嘉兴）有限公司 | Gas engine leakage early warning system |
| CN115950590B (en) * | 2023-03-15 | 2023-05-30 | 凯晟动力技术（嘉兴）有限公司 | Gas engine leakage early warning system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110657355B (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111853555B (en) | Water supply pipe network blind leakage identification method based on dynamic process | |
| CN111626153A (en) | Integrated learning-based partial discharge fault state identification method | |
| CN110454687A (en) | A kind of pipeline multipoint leakage localization method based on improvement VMD | |
| CN103033567B (en) | Pipeline defect signal identification method based on guided wave | |
| CN102623009B (en) | Abnormal emotion automatic detection and extraction method and system on basis of short-time analysis | |
| CN109375060B (en) | Method for calculating fault waveform similarity of power distribution network | |
| CN111520615B (en) | Pipe network leakage identification and positioning method based on line spectrum pair and cubic interpolation search | |
| CN110657355B (en) | Method for detecting leakage of thermal pipeline | |
| CN111027408A (en) | Load identification method based on support vector machine and V-I curve characteristics | |
| CN108444696A (en) | A kind of gearbox fault analysis method | |
| CN112599134A (en) | Transformer sound event detection method based on voiceprint recognition | |
| CN113556629B (en) | Intelligent ammeter error remote estimation method and device | |
| CN109524972A (en) | Low-frequency oscillation method for parameter estimation based on GSO and SVM algorithm | |
| CN108089097B (en) | Intelligent online distribution network ground fault positioning method | |
| CN111237646B (en) | Automatic identification and positioning method for leakage of water supply pipe network | |
| CN109657287B (en) | Hydrological model precision identification method based on comprehensive scoring method | |
| CN108615054B (en) | Method for constructing comprehensive index for measuring similarity between drainage pipe network nodes | |
| CN106251861A (en) | A kind of abnormal sound in public places detection method based on scene modeling | |
| CN114487129A (en) | Flexible material damage identification method based on acoustic emission technology | |
| CN115688015A (en) | Voltage sag type identification method and device based on non-invasive load monitoring | |
| CN108106717B (en) | A method of set state is identified based on voice signal | |
| CN109840386A (en) | Damnification recognition method based on factorial analysis | |
| CN115406522A (en) | Power plant equipment running state research and application based on voiceprint recognition | |
| CN112329535B (en) | CNN-based quick identification method for low-frequency oscillation modal characteristics of power system | |
| CN114969638A (en) | Bridge performance abnormity early warning method based on modal equivalent standardization |
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 |