CN114046968A - Two-step fault positioning method for process equipment based on acoustic signals - Google Patents
Two-step fault positioning method for process equipment based on acoustic signals Download PDFInfo
- Publication number
- CN114046968A CN114046968A CN202111167338.2A CN202111167338A CN114046968A CN 114046968 A CN114046968 A CN 114046968A CN 202111167338 A CN202111167338 A CN 202111167338A CN 114046968 A CN114046968 A CN 114046968A
- Authority
- CN
- China
- Prior art keywords
- signal
- arrival
- sound source
- time difference
- equipment
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 239000011159 matrix material Substances 0.000 claims abstract description 36
- 239000013598 vector Substances 0.000 claims description 42
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000005284 excitation Effects 0.000 claims description 8
- 238000003491 array Methods 0.000 claims description 7
- 230000003595 spectral effect Effects 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 5
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 2
- 230000003068 static effect Effects 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000012544 monitoring process Methods 0.000 description 15
- 238000003745 diagnosis Methods 0.000 description 13
- 238000011161 development Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 229910052618 mica group Inorganic materials 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
本发明提供了一种基于声学信号的过程装备两步故障定位方法。本发明结合MUSIC算法与时差矩阵算法，对封闭空间内过程装备的故障定位采用先整体区域多型动、静设备粗定位，后故障设备精确定位的两步方法，可以识别出多台运行设备中的故障设备，并且对该设备进行精确故障定位，减少了检测成本，降低了操作要求，可实现封闭空间内过程装备同时运行状态下的无接触故障定位。The present invention provides a two-step fault location method for process equipment based on acoustic signals. Combining the MUSIC algorithm and the time difference matrix algorithm, the present invention adopts a two-step method for the fault location of the process equipment in the closed space: firstly, the overall area multi-type dynamic and static equipment is roughly located, and then the faulty equipment is precisely located. It can accurately locate the fault of the equipment, reduce the detection cost, reduce the operation requirements, and can realize the non-contact fault location of the process equipment in the closed space when the process equipment is running at the same time.
Description
Technical Field
The invention provides a fault source positioning method of process equipment in a closed space in a simultaneous operation state, relates to a multi-type dynamic and static equipment non-contact monitoring method, and particularly relates to a two-step fault positioning method of process equipment based on acoustic signals.
Background
In recent years, with the rapid development of modern industry in China, the demand quantity of mechanical equipment is increased year by year, the mechanical equipment is an important material basis for national economic construction and production development, and has a great importance in daily industrial production, and the operation condition of the mechanical equipment directly influences the benefit of industrial enterprises. Moreover, most enterprises establish a flow production system, and once some parts of mechanical equipment have problems, the operation of the whole equipment is influenced, so that the whole production system is influenced, huge economic loss is caused, and even serious disaster events are caused. Therefore, condition monitoring and fault diagnosis techniques capable of ensuring the normal operation and the prolonged service life of equipment are increasingly showing necessity and urgency. By using the advanced state monitoring and fault diagnosis technology, the fault head of the equipment can be found in advance, the occurrence of serious accidents caused by untimely discovery is avoided, and the problems of 'overhaul and overhauling' in the conventional regular maintenance process of the equipment can be fundamentally solved. Therefore, in the daily management process of mechanical equipment, it is required to identify a minor failure phenomenon as early as possible through on-line monitoring and monitor the development trend of the failure, so as to adjust the operation parameters or scheduled daily maintenance before the failure causes major damage.
Because the complexity of the equipment is higher and higher, the loss of equipment shutdown is larger and larger, so the state monitoring problem of the equipment is generally regarded by the industry and academia, the health state monitoring and diagnosis of the mechanical equipment is rapidly developed abroad, and the equipment becomes a new idea of the current advanced equipment management. From the 90 s of the last century, China just starts to popularize the technology and obtains good effects. The health state monitoring and fault diagnosis technology of the mechanical equipment comprises two aspects: firstly, real-time monitoring is carried out; secondly, fault diagnosis. The health state monitoring of the equipment refers to monitoring some characteristic parameters of the equipment, comparing the acquired parameter values with normal theoretical values, and detecting whether the mechanical parts normally operate. The fault diagnosis is divided into two parts: one is to judge whether the equipment has a fault; and secondly, for the mechanical equipment with faults, the fault position is found, the influence reason is analyzed, the severity is judged, and an effective method is provided for processing. The literature for studying the equipment monitoring problem is in the tens of thousands, but only various models of single equipment are more, and the monitoring and fault diagnosis problem of the single equipment is more fully studied. However, due to the complexity of process equipment systems, research literature on monitoring and fault diagnosis location of process equipment systems is sparse. In the production application of the actual process industry, the traditional single-equipment monitoring and diagnosing method is difficult to identify and position in time. In addition, the fault diagnosis of the current mechanical equipment is mainly based on the measurement and analysis of vibration signals, can only be carried out on a fixed-point qualitative monitoring mode, cannot be carried out on a large scale, and the installation of a sensor is inconvenient or the vibration signals cannot be fed back under certain special working condition environments, so that a contact type fault diagnosis method represented by the measured vibration signals is limited, and the proposal of a non-contact type full-scene fault positioning method of process equipment is imperative.
The mechanical noise collected by the non-contact sensor contains rich equipment state information, which is the result of mechanical vibration transmitted to the outside through an elastic medium and is in causal relation with the mechanical vibration. Non-contact fault diagnosis methods typified by measuring acoustic signals may partially replace vibration signals as a means of fault diagnosis in some special environments. With the progress of research, acoustic technology has been applied to many fields and plays an important role therein. Microphone arrays are a common tool for picking up sound signals. The array signal contains many pieces of information, and one of the important items is the position information of the sound source. The MUSIC algorithm is an array signal processing algorithm which is widely applied, has a long development history, and has been researched in many fields such as radar, sonar, seismology, communication and the like. It can be applied in many different fields, such as signal detection, direction of arrival estimation (DOA), etc.
Therefore, aiming at some technical problems urgently solved by technical personnel in the field, according to the fault diagnosis thought, the acoustic signal-based process equipment two-step fault positioning method is provided by combining the MUSIC algorithm and the time difference matrix, the non-contact fault positioning of multiple movable and static devices in a closed space in a simultaneous operation state can be realized, the fault devices in the multiple movable and static devices can be identified, the precise fault positioning can be carried out on the devices, and the project provides a brand-new visual angle for the development of the field of sound source positioning.
Disclosure of Invention
The invention provides a two-step fault positioning method of process equipment based on acoustic signals, which adopts a two-step method of firstly roughly positioning multi-type movable and static equipment in an integral area and accurately positioning rear fault equipment for fault positioning of the process equipment in a closed space, can identify fault equipment in a plurality of running equipment and accurately position faults of the equipment, reduces detection cost, reduces operation requirements and can realize non-contact fault positioning of the process equipment in the closed space in a simultaneous running state.
The method is characterized in that: the specific operation steps are as follows:
the method comprises the following steps: abstracting the whole closed area and all equipment into point-line combinations, numbering each node, selecting positions of four or more nodes to arrange a microphone sensor array, wherein the number of the sensor arrays is enough to monitor all the equipment in the closed space. Using a modal force hammer, applying active excitation to each device, selecting one of the sensors in each sensor array, and extracting the acoustic signal arrival time using a Two-Step Chikuchi information criterion (Two-Step-AIC, TS-AIC).
Step two: and calculating the time difference between every two sensors according to the extracted arrival time, and constructing an integral time difference matrix database. Taking the closed space with N devices and arranging M sensors as an example, after active excitation, signal acquisition and time difference extraction, the method can generate
Group time difference, which may constitute one
The time difference vector of (a). Finally, after all N devices are excited, one device can be obtained
Time difference matrix database of dimensions, note
Step three: in the whole area positioning test of the closed space, applying a simulated fault sound source signal to obtain the arrival time of the sound source signal to each sensor to obtain a corresponding time difference vector, and comparing the time difference vector with a time difference matrix database
N vectors inAnd comparing, calculating the Pearson correlation coefficient, screening out the most similar vector in the database, and taking the corresponding node number as the target node. Wherein the Pearson correlation coefficient is:
wherein X is the arrival time difference of the simulated fault sound source signal, and YiWhere i is 1, 2, …, and N is a vector in the time difference matrix library, μ is the mathematical expectation of the vector, and σ is the standard deviation of the vector.
Step four: after the fault equipment is determined, the MUSIC algorithm is used, the microphone array closest to the fault equipment is used for accurate positioning, and the fault component of the fault equipment is determined. The MUSIC algorithm comprises the following specific steps:
after receiving the excitation signal, a covariance matrix is obtained from the signal vectors received by the L microphone sensors, i.e.
Wherein L is the number of sensors in the microphone array, S (n) is the time domain waveform of the signal, n is the number of sampling points of the signal, SHAnd (n) is the conjugate transpose of S (n), and H is the conjugate transpose sign.
And performing eigenvalue decomposition on the obtained covariance matrix to obtain R ═ U Σ UHWhere U is the eigenvector, Σ is the matrix of eigenvalues, and H is the conjugate transposed symbol. Sorting the eigenvalues according to magnitude, regarding the eigenvector corresponding to the largest eigenvalue as a signal space, and regarding the eigenvectors corresponding to the remaining eigenvalues as a noise space, i.e.
Wherein U isSAs a signal space feature vector, sigmaSIs a matrix of eigenvalues of the signal space, UNAs a noise spatial feature vector, sigmaNH is the conjugate transposed symbol, which is the eigenvalue matrix of the noise space.
The sensor array used by the invention is a uniform circular array with 10 array elements and fixed diameter, and 10 same omnidirectional arrays are uniformly distributed on a plane x-y, and the radius of the array is RmicAs shown in fig. 5. The direction of arrival of the incident plane wave is represented by a spherical coordinate system, and the origin O of the coordinate system is at the center of the array, namely the center of a circle. The elevation angle theta of the sound source is an included angle between a connecting line from the array origin to the signal source and the z axis, and the azimuth angle phi is an included angle between the projection of the connecting line from the array origin to the sound source on the plane x-y and the x axis.
Constructing spatial spectral functions
Where a (θ, φ) is a direction vector, is an array response with the sound source direction of arrival (θ, φ), and can be expressed as
In the formula, gammamRepresents the included angle between the connecting line of the mth sensor and the origin and the positive half shaft of the x axis, gammam2 pi m/10, m 0,1, L,9, j is expressed as an imaginary number, and λ is the wavelength of the sound source.
Since the direction matrices are related only to the angle of arrival (θ, φ) of the signals for differently shaped arrays, the spatial spectral function is a function of the angle of arrival. By changing the angle of arrival (theta, phi), after scanning two angles, seeking the peak value of the spectrum function, the estimation value of the angle of arrival direction can be obtained, and the accurate positioning can be realized.
Drawings
FIG. 1 is a flow chart of the method of the present invention
FIG. 2 is a model of process equipment operation in an enclosed space
FIG. 3 is a flow chart of the positioning of the whole area of the enclosed space
FIG. 4 is a flow chart of MUSIC fine positioning
FIG. 5 is a 10-element uniform circular array used
FIG. 6 is a diagram illustrating the results of the MUSIC algorithm
Detailed Description
In the following description, the present invention will be described in detail according to exemplary embodiments.
(1) Referring to the closed space process equipment operational model (12 devices) shown in fig. 2, the whole area and all devices are abstracted into a dotted line combined structure, and nodes therein are numbered. Microphone arrays (10 microphone sensors per array) are arranged at the top four corners of the enclosed space, one sensor of each microphone array is selected for area localization, active excitation is applied to each device, acoustic signals are collected, and arrival time is extracted by using a TS-AIC method. The TS-AIC method is an optimization of the conventional AIC method, and the arrival time of the conventional AIC extracted signal is specifically as follows:
the AIC function returns a minimum value, which occurs at the very beginning of the signal, and the signal start position is easily determined by image analysis, and is expressed as:
AIC(tw)＝tw·log(var(Sw(tw,1)))+(Tw-tw-1)·log(var(Sw(1+tw,Tw)))
wherein, the time sequence S is divided into two parts, w is used as a division point, w belongs to [1, n ]]N is the length of the acquired signal, var is the variance function, TwFor the last sample of the time series, twFor any sample in the time series, Sw(tw1) represents the calculation range of the variance function as a starting point to a current value tw，Sw(1+tw,Tw) Represents the calculation range of the variance function as 1+ twTo TwAll sample values of (a).
The TS-AIC method is improved aiming at the traditional AIC method, and the specific steps are as follows:
collecting sound source signals;
computing signal envelope by using Hilbert transform, and finding out global maximum t by computing characteristic function CF (n) ═ S (n) (wherein S is time domain expression of collected sound source signal)MAXSetting the AIC calculation range at the start point t0And tMAX+AMWherein t isAMThis value is experimentally measured as a period of the fastest wave velocity mode in the received signal for the set time delay. Thus, the time interval of the original time window is shortened;
thirdly, performing first AIC calculation, searching a global minimum value point in a redefined range, and determining first arrival time estimation by the point;
fourthly, the second AIC calculation is carried out, and the global minimum value point AIC determined in the first step1stThe area of (2) is subjected to AIC calculation. Sphere of global minimum point AIC1stFront tAMA/2 time delay until tBMA time delay of/2, where tBM＝tAM. The global minimum of the AIC function, i.e. the exact sound source signal arrival time, is recalculated.
(2) The arrival time difference of each sensor pair is calculated according to the arrival time obtained in the first step, and 12 devices can obtain one
Dimensional time difference matrix database
(3) In the sound source area localization test, a simulated sound source is applied. In the closed space, a certain component of certain equipment is randomly selected to apply excitation as a fault sound source, and after signals are collected, the arrival time is extracted by using a TS-AIC method, so that a group of time difference vectors X of the simulated fault signals can be obtained. Database of X and time difference matrix
The vectors in (1) are compared to calculate the Pearson correlation coefficient, i.e.
Wherein X is the arrival time difference of the simulated fault sound source signal, and YiWhere i is a vector in the moveout matrix library, 2, …, 12, μ is the mathematical expectation of the vector, and σ is the vectorStandard deviation of the amounts.
And obtaining the most similar vector in the database according to the Pearson correlation coefficient, wherein the number of the corresponding node is the target node. This makes it possible to find a faulty device from among the 12 operating devices.
(4) And after the fault equipment is determined, the MUSIC algorithm is adopted for accurate positioning. Obtaining a covariance matrix from the signal vectors received by the 10 microphone sensors, i.e.
Wherein S (n) is the time domain waveform of the signal, n is the number of sampling points of the signal, SHAnd (n) is the conjugate transpose of S (n), and H is the conjugate transpose sign.
And performing eigenvalue decomposition on the obtained covariance matrix to obtain R ═ U Σ UH. Where U is the eigenvector, Σ is the matrix of eigenvalues, and H is the conjugate transpose symbol. Sorting the eigenvalues according to magnitude, regarding the eigenvector corresponding to the largest eigenvalue as a signal space, and regarding the eigenvectors corresponding to the remaining eigenvalues as a noise space, i.e.
Wherein U isSAs a signal space feature vector, sigmaSIs a matrix of eigenvalues of the signal space, UNAs a noise spatial feature vector, sigmaNH is the conjugate transposed symbol, which is the eigenvalue matrix of the noise space.
Constructing a spatial spectral function PMUSIC(θ,φ)＝1/[aH(θ,φ)UNUN Ha(θ,φ)]The sensor array is in the shape of a uniform circular array with 10 array elements of fixed diameter, and the peak value of a spectrum function is searched after two angles are scanned by changing the arrival angle (theta, phi), so that the estimation value of the arrival direction angle can be obtained, and accurate positioning is realized.
The invention randomly selects 6 devices from 12 dynamic and static devices to carry out experiments, and each device is divided into regions according to the structure. The fault identification accuracy of the multi-type equipment in the whole area is 100%, which shows that the invention can realize the non-contact accurate positioning of the fault equipment in the closed space under the condition that the multi-type equipment simultaneously operates; meanwhile, after the faulty equipment is accurately identified, the accurate identification accuracy rate is 92% through the MUSIC algorithm, as shown in FIG. 6, the accurate identification accuracy rate is a MUSIC algorithm positioning result diagram of one of the equipment, and the method can accurately identify the specific fault position of the faulty equipment. The invention provides a new idea for accurately positioning the non-contact fault in the running state of the multi-type equipment.
Claims (2)
1. A two-step fault location method of process equipment based on acoustic signals is characterized by comprising the following specific operation steps:
the method comprises the following steps: abstracting the whole closed area and all equipment into point-line combinations, numbering each node, selecting four or more node positions to arrange a microphone sensor array, wherein the number of the sensor arrays is enough to monitor all equipment in the closed space; using a modal force hammer to apply active excitation on each device, selecting one sensor in each sensor array, and extracting the arrival time of an acoustic signal;
step two: calculating the time difference between every two sensors according to the extracted arrival time, and constructing an integral time difference matrix database; the closed space with N devices is provided with M sensors, and the M sensors are actively excited each time, acquire signals and extract time differences to generate
Group time difference, constitute one
The time difference vector of (a); finally, after all N devices are excited, one device is obtained
Time difference matrix database of dimensions, note
Step three: in the whole area positioning test of the closed space, applying a simulated fault sound source signal to obtain the arrival time of the sound source signal to each sensor to obtain a corresponding time difference vector, and comparing the time difference vector with a time difference matrix database
Comparing the N vectors, calculating a Pearson correlation coefficient, screening out the most similar vectors in the database, wherein the corresponding node number is the target node; wherein the Pearson correlation coefficient is:
wherein X is the arrival time difference of the simulated fault sound source signal, and YiThe vectors in the time difference matrix library are shown, mu is the mathematical expectation of the vectors, and sigma is the standard deviation of the vectors;
step four: after the fault equipment is determined, the MUSIC algorithm is used, the microphone array closest to the fault equipment is used for accurate positioning, and a fault part of the fault equipment is determined; the MUSIC algorithm comprises the following specific steps:
after receiving the excitation signal, a covariance matrix is obtained from the signal vectors received by the L microphone sensors, i.e.
Wherein L is the number of sensors in the microphone array, S (n) is the time domain waveform of the signal, n is the number of sampling points of the signal, SH(n) is the conjugate transpose of S (n), H is the conjugate transpose symbol;
and performing eigenvalue decomposition on the obtained covariance matrix to obtain R ═ U Σ UHWhere U is the eigenvector, Sigma is the matrix of eigenvalues, and H is the conjugate transposed symbol(ii) a Sorting the eigenvalues according to magnitude, regarding the eigenvector corresponding to the largest eigenvalue as a signal space, and regarding the eigenvectors corresponding to the remaining eigenvalues as a noise space, i.e.
Wherein U isSAs a signal space feature vector, sigmaSIs a matrix of eigenvalues of the signal space, UNAs a noise spatial feature vector, sigmaNIs a characteristic value matrix of a noise space, and H is a conjugate transpose symbol;
expressing the direction of arrival of the incident plane wave by using a spherical coordinate system, wherein the origin O of the coordinate system is at the center of the array, namely the circle center; the elevation angle theta of the sound source is an included angle between a connecting line from the array origin to the information source and the z axis, and the azimuth angle phi is an included angle between the projection of the connecting line from the array origin to the sound source on the plane x-y and the x axis;
constructing a spatial spectral function PMUSIC(θ,φ)＝1/[aH(θ,φ)UNUN Ha(θ,φ)](ii) a Where a (θ, φ) is the direction vector, and is the array response of the sound source with direction of arrival (θ, φ), expressed as
In the formula, gammamRepresents the included angle between the connecting line of the mth sensor and the origin and the positive half shaft of the x axis, gammam2 pi m/10, m 0,1, L,9, j is an imaginary expression, and λ is the wavelength of the sound source;
since the direction matrices are related only to the arrival angles (theta, phi) of the signals for different shapes of arrays, the spatial spectrum function is a function of the arrival angles; by changing the arrival angle (theta, phi), after scanning two angles, seeking the peak value of the spectrum function, and obtaining the estimated value of the arrival direction angle.
2. The method of claim 1, wherein:
the specific steps for extracting the arrival time of the acoustic signal are as follows:
collecting sound source signals;
calculating signal envelope by using Hilbert transform, and calculating a characteristic function CF (n) ═ S (n) | wherein S is time domain expression of the acquired sound source signal; find global maximum tMAXSetting the AIC calculation range at the start point t0And tMAX+tAMWherein t isAMFor a set time delay, the value is experimentally measured as a period of the fastest wave velocity mode in the received signal;
thirdly, performing first AIC calculation, searching a global minimum value point in a redefined range, and determining first arrival time estimation by the point;
fourthly, the second AIC calculation is carried out, and the global minimum value point AIC determined in the first step1stPerforming AIC calculations; sphere of global minimum point AIC1stFront tAMA/2 time delay until tBMA time delay of/2, where tBM＝tAM(ii) a Recalculating the global minimum value of the AIC function, namely the accurate sound source signal arrival time;
(2) calculating the arrival time difference of each sensor pair according to the arrival time obtained in the first step to obtain a time difference matrix database;
(3) in the sound source area positioning test, applying a simulated sound source; in the closed space, randomly selecting a certain component of certain equipment to apply excitation as a fault sound source, collecting signals and then extracting arrival time to obtain a group of time difference vectors X of simulated fault signals; comparing the X with the vector in the time difference matrix database, and calculating the Pearson correlation coefficient, i.e.
Wherein X is the arrival time difference of the simulated fault sound source signal, and YiThe vectors in the time difference matrix library are shown, mu is the mathematical expectation of the vectors, and sigma is the standard deviation of the vectors;
obtaining the most similar vector in the database according to the Pearson correlation coefficient, wherein the number of the corresponding node is the target node;
(4) after the fault equipment is determined, the MUSIC algorithm is adopted for accurate positioning; obtaining a covariance matrix from the signal vectors received by the 10 microphone sensors, i.e.
Wherein, L is the number of sensors in the microphone array, wherein S (n) is the time domain waveform of the signal, n is the number of sampling points of the signal, SH(n) is the conjugate transpose of S (n), H is the conjugate transpose symbol;
and performing eigenvalue decomposition on the obtained covariance matrix to obtain R ═ U Σ UH(ii) a Wherein U is an eigenvector, sigma is a matrix formed by eigenvalues, and H is a conjugate transpose symbol; sorting the eigenvalues according to magnitude, regarding the eigenvector corresponding to the largest eigenvalue as a signal space, and regarding the eigenvectors corresponding to the remaining eigenvalues as a noise space, i.e.
Wherein U isSAs a signal space feature vector, sigmaSIs a matrix of eigenvalues of the signal space, UNAs a noise spatial feature vector, sigmaNIs a characteristic value matrix of a noise space, and H is a conjugate transpose symbol;
constructing a spatial spectral function PMUSIC(θ,φ)＝1/[aH(θ,φ)UNUN Ha(θ,φ)]And a (theta, phi) is a direction vector, the array response of the sound source with the direction of arrival (theta, phi) is obtained, and the peak value of the spectrum function is searched after two angles are scanned by changing the angle of arrival (theta, phi) to obtain the estimated value of the direction of arrival angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111167338.2A CN114046968B (en) | 2021-10-04 | 2021-10-04 | Two-step fault positioning method for process equipment based on acoustic signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111167338.2A CN114046968B (en) | 2021-10-04 | 2021-10-04 | Two-step fault positioning method for process equipment based on acoustic signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114046968A true CN114046968A (en) | 2022-02-15 |
| CN114046968B CN114046968B (en) | 2024-09-17 |
Family
ID=80204873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111167338.2A Active CN114046968B (en) | 2021-10-04 | 2021-10-04 | Two-step fault positioning method for process equipment based on acoustic signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114046968B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114783462A (en) * | 2022-05-11 | 2022-07-22 | 安徽理工大学 | A fault source location analysis method for mine hoist based on CS-MUSIC |
| CN118999776A (en) * | 2024-10-24 | 2024-11-22 | 国网山西省电力公司超高压变电分公司 | GIS fault diagnosis method and system based on acoustic monitoring |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03265325A (en) * | 1990-03-15 | 1991-11-26 | Nec Corp | Fault deciding device |
| US20160037277A1 (en) * | 2014-07-30 | 2016-02-04 | Panasonic Intellectual Property Management Co., Ltd. | Failure detection system and failure detection method |
| CN105893256A (en) * | 2016-03-30 | 2016-08-24 | 西北工业大学 | Software failure positioning method based on machine learning algorithm |
| CN106969928A (en) * | 2017-03-28 | 2017-07-21 | 辽宁机电职业技术学院 | A kind of mechanical fault detection diagnostic method |
| CN108613820A (en) * | 2018-05-22 | 2018-10-02 | 国网陕西省电力公司榆林供电公司 | A kind of online allophone monitoring algorithm for GIS bulk mechanicals defect diagonsis and positioning |
| US20180299527A1 (en) * | 2015-12-22 | 2018-10-18 | Huawei Technologies Duesseldorf Gmbh | Localization algorithm for sound sources with known statistics |
| CN108896962A (en) * | 2018-06-14 | 2018-11-27 | 河北工业大学 | Iteration localization method based on sound position fingerprint |
| CN109211548A (en) * | 2018-08-31 | 2019-01-15 | 沃德传动（天津）股份有限公司 | A kind of mechanical failure diagnostic method |
| CN110310667A (en) * | 2019-07-12 | 2019-10-08 | 莫毓昌 | A kind of equipment fault detection method based on acoustic feature |
| CN110488223A (en) * | 2019-07-05 | 2019-11-22 | 东北电力大学 | A sound source localization method |
| US20200011921A1 (en) * | 2017-03-29 | 2020-01-09 | Baur Gmbh | Method for accurately locating a cable defect of a cable laid in the ground |
| CN112461934A (en) * | 2020-11-20 | 2021-03-09 | 北京化工大学 | Aero-engine blade crack source positioning method based on acoustic emission |
| CN113231651A (en) * | 2015-06-11 | 2021-08-10 | 瑞尼斯豪公司 | Additive manufacturing apparatus and method |
-
2021
- 2021-10-04 CN CN202111167338.2A patent/CN114046968B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03265325A (en) * | 1990-03-15 | 1991-11-26 | Nec Corp | Fault deciding device |
| US20160037277A1 (en) * | 2014-07-30 | 2016-02-04 | Panasonic Intellectual Property Management Co., Ltd. | Failure detection system and failure detection method |
| CN113231651A (en) * | 2015-06-11 | 2021-08-10 | 瑞尼斯豪公司 | Additive manufacturing apparatus and method |
| US20180299527A1 (en) * | 2015-12-22 | 2018-10-18 | Huawei Technologies Duesseldorf Gmbh | Localization algorithm for sound sources with known statistics |
| CN105893256A (en) * | 2016-03-30 | 2016-08-24 | 西北工业大学 | Software failure positioning method based on machine learning algorithm |
| CN106969928A (en) * | 2017-03-28 | 2017-07-21 | 辽宁机电职业技术学院 | A kind of mechanical fault detection diagnostic method |
| US20200011921A1 (en) * | 2017-03-29 | 2020-01-09 | Baur Gmbh | Method for accurately locating a cable defect of a cable laid in the ground |
| CN108613820A (en) * | 2018-05-22 | 2018-10-02 | 国网陕西省电力公司榆林供电公司 | A kind of online allophone monitoring algorithm for GIS bulk mechanicals defect diagonsis and positioning |
| CN108896962A (en) * | 2018-06-14 | 2018-11-27 | 河北工业大学 | Iteration localization method based on sound position fingerprint |
| CN109211548A (en) * | 2018-08-31 | 2019-01-15 | 沃德传动（天津）股份有限公司 | A kind of mechanical failure diagnostic method |
| CN110488223A (en) * | 2019-07-05 | 2019-11-22 | 东北电力大学 | A sound source localization method |
| CN110310667A (en) * | 2019-07-12 | 2019-10-08 | 莫毓昌 | A kind of equipment fault detection method based on acoustic feature |
| CN112461934A (en) * | 2020-11-20 | 2021-03-09 | 北京化工大学 | Aero-engine blade crack source positioning method based on acoustic emission |
Non-Patent Citations (6)
| Title |
|---|
| LIU, MZ (LIU, MINGZHOU) [2] ; YANG, JX (YANG, JIANGXIN): "A new method for arrival time determination of impact signal based on HHT and AIC", MECHANICAL SYSTEMS AND SIGNAL PROCESSING, vol. 86, 1 March 2017 (2017-03-01), pages 177 - 187, XP029806711, DOI: 10.1016/j.ymssp.2016.10.003 * |
| 刘增华;彭秋玲;何存富;吴斌;: "复合材料板声发射源定位的时间差映射方法", 声学学报, no. 03, 14 May 2020 (2020-05-14), pages 99 - 107 * |
| 姜志鹏;唐加能;梁瑞宇;: "实时声源定位算法研究与实现", 电子器件, no. 06, 20 December 2015 (2015-12-20), pages 192 - 197 * |
| 尹莘新: "基于时差法的声发射源定位方法研究", 中国博士学位论文全文数据库 (基础科学辑), no. 8, 15 August 2020 (2020-08-15), pages 005 - 7 * |
| 李铁军，王宁: "基于传感器阵列的机械故障声源定位系统", 机械设计与制造, no. 4, 30 April 2019 (2019-04-30), pages 166 - 170 * |
| 贾光，杨国安，沈江，张淼: "海水对海洋平台声发射传播特性影响研究", 海洋工程, vol. 31, no. 3, 31 May 2013 (2013-05-31), pages 84 - 88 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114783462A (en) * | 2022-05-11 | 2022-07-22 | 安徽理工大学 | A fault source location analysis method for mine hoist based on CS-MUSIC |
| CN118999776A (en) * | 2024-10-24 | 2024-11-22 | 国网山西省电力公司超高压变电分公司 | GIS fault diagnosis method and system based on acoustic monitoring |
| CN118999776B (en) * | 2024-10-24 | 2025-01-10 | 国网山西省电力公司超高压变电分公司 | GIS fault diagnosis method and system based on acoustic monitoring |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114046968B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114046968B (en) | Two-step fault positioning method for process equipment based on acoustic signals | |
| CN109975762B (en) | An underwater sound source localization method | |
| CN109947047B (en) | A method for diagnosing unbalance faults of electric spindle | |
| CN112461934B (en) | Aero-engine blade crack source positioning method based on acoustic emission | |
| CN110837076A (en) | Tensor decomposition-based vector hydrophone array orientation estimation method | |
| CN118604733B (en) | Lightning positioning method, device and storage medium | |
| CN109597021B (en) | Direction-of-arrival estimation method and device | |
| KR20210129942A (en) | Acoustic inspection device and inspection method thereof | |
| Cao et al. | From pseudo to real: Generalized subspace method for power spectrum reconstruction | |
| CN110702042B (en) | A thickness estimation method under the condition of signal aliasing in ultrasonic pulse echo thickness measurement | |
| Bagherahmadi et al. | Structural damage detection using a damage probability index based on frequency response function and strain energy concept | |
| CN112051063A (en) | Method and system for locating sound source of equipment fault | |
| CN110108474B (en) | Online monitoring and evaluating method and system for operation stability of rotary machine | |
| CN115474108A (en) | Event monitoring system and method based on edge calculation | |
| CN116256026B (en) | Health monitoring system of multidimensional splice welding structure under dynamic service working condition | |
| US20240249117A1 (en) | State classification method, state classification device, and state classification program | |
| CN117116293A (en) | Machine equipment fault diagnosis system in complex sound field environment | |
| CN113050036B (en) | GIS beam forming positioning method based on multi-resonance point microphone array | |
| CN113050037B (en) | Transformer substation equipment abnormal sound source positioning method and system | |
| Li et al. | A new iterative near-field coherent subspace method for rub-impact fault localization using AE technique | |
| CN112729742B (en) | A Shock Localization Method Based on Probabilistic Imaging | |
| CN116299176A (en) | Target space feature extraction and fusion positioning method based on Hough transformation | |
| CN114964581A (en) | Stress detection method based on ultrasonic phased array focusing principle | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| CN112363106A (en) | Signal subspace direction of arrival detection method and system based on quantum particle swarm |
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 |