CN117789768A - Method, system, equipment and medium for monitoring abnormal sound of fan blade - Google Patents
Method, system, equipment and medium for monitoring abnormal sound of fan blade Download PDFInfo
- Publication number
- CN117789768A CN117789768A CN202410019518.3A CN202410019518A CN117789768A CN 117789768 A CN117789768 A CN 117789768A CN 202410019518 A CN202410019518 A CN 202410019518A CN 117789768 A CN117789768 A CN 117789768A
- Authority
- CN
- China
- Prior art keywords
- fan blade
- sound
- blade
- signal
- real
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012544 monitoring process Methods 0.000 title claims abstract description 41
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 36
- 230000005236 sound signal Effects 0.000 claims abstract description 73
- 238000012545 processing Methods 0.000 claims abstract description 28
- 230000009467 reduction Effects 0.000 claims abstract description 18
- 238000010586 diagram Methods 0.000 claims abstract description 16
- 238000001514 detection method Methods 0.000 claims abstract description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 claims description 2
- 238000012806 monitoring device Methods 0.000 claims 1
- 238000013145 classification model Methods 0.000 description 13
- 238000012549 training Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000036541 health Effects 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 2
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 1
- 235000017491 Bambusa tulda Nutrition 0.000 description 1
- 241001330002 Bambuseae Species 0.000 description 1
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 1
- 239000011425 bamboo Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- KJONHKAYOJNZEC-UHFFFAOYSA-N nitrazepam Chemical compound C12=CC([N+](=O)[O-])=CC=C2NC(=O)CN=C1C1=CC=CC=C1 KJONHKAYOJNZEC-UHFFFAOYSA-N 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a method, a system, equipment and a medium for monitoring abnormal sound of a fan blade, which are characterized in that noise reduction processing is carried out on the sound signal of the fan blade by collecting the sound signal data of the fan blade to obtain a noise-reduced sound signal Z (t) of the fan blade, the noise-reduced sound signal Z (t) of the fan blade is visually processed to obtain a time sequence diagram, and the state of the fan blade is monitored and analyzed; the system, the equipment and the medium are used for realizing a fan blade abnormal sound monitoring method; the invention can timely and rapidly detect the abnormal sound signals of the fan blade, has the characteristics of simple operation, high detection efficiency and low time cost and labor cost, and simultaneously does not damage the fan structure and does not influence the normal operation of the fan blade.
Description
Technical Field
The invention belongs to the technical field of fan blade abnormality detection, and particularly relates to a fan blade abnormality sound online monitoring method, system, equipment and medium.
Background
With the rapid development of the wind power industry, the problem of fan faults is increasingly remarkable, the blade is one of important components of the wind turbine, and the health state of the blade directly influences the safety and reliability of the operation of the wind turbine, so that the method has important significance in detecting the operation state of the fan blade.
Modern fan blade monitoring methods employ techniques including vibration analysis, infrared thermal imaging, computer vision, and the like. However, these methods have the disadvantages of damaging the blade structure, expensive equipment cost, being limited by environmental factors, etc., so that the relevant students monitor the health condition of the blade by using the sound signals of the blade.
The Chinese patent application with publication number of CN115641871A discloses a fan blade abnormality detection method based on voiceprint, which is characterized by collecting the sound signals of the blades and describing the Mel spectrogram of the blades, extracting the voiceprint characteristics of the signals through a convolutional neural network, establishing a blade voiceprint capturing model based on a long-short-time memory neural network, and calculating the Fre chet distance between each blade and the characteristics according to the characteristic distribution to judge whether the fan blade is abnormal or not.
The Chinese patent application with publication number of CN116163894A discloses a method, a system and a storage medium for detecting the state of a fan blade of a wind power plant, and provides a method for collecting sound data of the fan blade during operation and establishing a sample database, training by utilizing a multi-scale convolutional neural network and fusing multi-scale characteristics to obtain a classification model, and importing a collected sound loudness time-frequency diagram into the model to judge the state of the fan blade.
The Chinese patent application with publication number of CN116153333A discloses a wind turbine blade fault diagnosis method based on aerodynamic noise, which is characterized in that noise of a wind turbine blade acoustic signal is reduced and then converted into image data, the image data is input into a convolutional neural network model for pre-training, a convolutional neural network model is built, a training set is built, aerodynamic noise of the wind turbine blade to be detected is converted into an image number, and then the image number is input into the convolutional neural network model, and the fault state of the wind turbine blade is determined.
The basic thought of the method is to extract the characteristics of the blade sound signals, or to convert the signals into images and extract the characteristics, then to construct a training set and to construct a classification model by using a neural network, and then to input the signals to be detected into the classification model to realize fault monitoring of the blade, but the method has the following defects:
1. the sound signal characteristics of different blades are not the same, and the classification model constructed according to the existing training set is not suitable for blade monitoring of a full field or other stations.
2. The sound signals of different blade faults are different, and the extracted acoustic features are different, so that the accuracy of the classification model can be ensured only by collecting fan blade sound signals of various faults when the classification model is constructed, and the classification model is difficult to realize in practical application.
3. When the classification model is built, the larger the training set is, the more accurate the classification model is, and in practical application, enough fault characteristics are difficult to collect, and the classification model is not accurate due to the fact that the training set is not large.
4. The classification accuracy of the classification model is 80% -95%, and the false diagnosis phenomenon exists.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention aims to provide a fan blade abnormal sound monitoring method and system, which utilize a Local Mean Decomposition (LMD) method to carry out noise reduction treatment on sound signals, and can monitor fan blades in time through a time sequence diagram of the noise-reduced blade sound signals; the invention has the characteristics of high judging speed, simple operation, high detecting efficiency and low time cost and labor cost, and simultaneously, the structure of the fan is not damaged, and the normal operation of the fan blade is not influenced.
In order to achieve the above object, the present invention has the technical scheme that:
a fan blade abnormal sound monitoring method comprises the following steps:
step 1: setting sound collection frequency and duration parameters, and collecting fan blade sound signal data, wherein the sound signal data comprise a sound signal X (t) when a blade rotates and a field real-time wind noise signal Y (t);
step 2: decomposing the sound signal X (t) acquired in the step 1 during rotation of the blade based on a Local Mean Decomposition (LMD) method to obtain k amplitude modulation and frequency modulation components (PF);
step 3: based on the k amplitude modulation frequency modulation components PF and the on-site real-time wind noise signals Y (t) obtained by decomposition in the step 2, the Pearson coefficients of the two components are calculatedObtaining k real-time wind noise similarities;
step 4: comparing the k real-time wind noise similarities obtained in the step 3 with a set threshold, discarding the corresponding amplitude modulation and frequency modulation components PF as wind noise signals when the real-time wind noise similarities are higher than the set threshold, and reconstructing the corresponding amplitude modulation and frequency modulation components PF to obtain a blade sound signal Z (t) after noise reduction when the real-time wind noise similarities are lower than the set threshold;
step 5: and (3) carrying out visual processing on the blade sound signal Z (t) subjected to noise reduction in the step (4) to obtain a time sequence diagram, and carrying out monitoring analysis on the state of the fan blade.
The Pierson coefficients of the k amplitude modulation frequency modulation components PF and the field real-time wind noise signal Y (t) in the step 3The calculation formula of (2) is as follows:
wherein cov is covariance and σ is standard deviation.
And in the step 4, setting the threshold value to be 0.7.
The invention also provides a fan blade abnormal sound monitoring system, which comprises:
the sound collection module is used for: the device is used for collecting a sound signal X (t) and a field real-time wind noise signal Y (t) when the blades rotate;
and a signal processing module: the method comprises the steps of decomposing a sound signal X (t) when a blade rotates to obtain an amplitude modulation frequency modulation component PF; the real-time wind noise similarity of the amplitude modulation component PF and the on-site real-time wind noise signal Y (t) is calculated, and the real-time wind noise similarity is compared with a set threshold value to obtain a blade sound signal Z (t) after noise reduction;
and a data transmission module: the blade sound signal Z (t) after noise reduction in the signal processing module is transmitted to the abnormal sound detection module;
abnormal sound detection module: the method is used for processing the blade sound signal Z (t) after noise reduction to obtain a time sequence diagram, and judging whether the fan blade is abnormal or not according to the time sequence diagram.
The sound collection module adopts a VS1053 audio decoding chip.
The signal processing module adopts an STM32F103ZET6 chip.
The data transmission module adopts an L768H-E18V communication module.
The outside of the sound collection module is wrapped with a shell, and the shell is arranged on the fan tower.
The invention also provides fan blade abnormal sound monitoring equipment, which comprises:
a memory: the computer program is used for storing and realizing a fan blade abnormal sound monitoring method;
a processor: the fan blade abnormal sound monitoring method is used for executing the computer program.
The present invention also provides a computer-readable storage medium: the computer readable storage medium stores a computer program which when executed by a processor is capable of implementing a fan blade abnormal sound monitoring method.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention utilizes the local mean decomposition LMD method to decompose the sound signals when the blades rotate, the local mean decomposition LMD method can adaptively separate a nonlinear and non-stationary signal step by step according to the envelope characteristics of the signals and the descending order of the frequencies, and compared with the decomposition methods such as EMD, EEMD and the like, the decomposition method is very suitable for decomposing the sound signals of the fan blades, and can well distinguish the on-site real-time wind noise signals from the sound signals when the blades rotate.
2. According to the invention, the Pelson correlation coefficient of each frequency modulation and amplitude modulation component and the on-site real-time wind noise signal is calculated, the frequency modulation and amplitude modulation component of the on-site real-time wind noise signal is abandoned, and the rest frequency modulation and amplitude modulation components are reconstructed, so that a purer blade sound signal can be obtained, and the influence of the wind noise signal is eliminated.
3. According to the invention, the blade monitoring is directly realized by describing the noise-reduced fan blade sound signal timing diagram and according to the periodicity and the volatility of the noise-reduced fan blade sound signal, the steps of extracting features in other technical schemes and constructing a classification model by using machine learning or a neural network are omitted, the monitoring method is simpler and more convenient and rapid, meanwhile, the defect of wrong classification in the classification model is avoided, and the monitoring result is more accurate.
4. Compared with the method based on vibration, ultrasonic wave and acoustic emission signals, the method has the advantages of higher efficiency and lower cost, and can monitor the abnormal condition of the blade without stopping the machine.
Drawings
FIG. 1 is a flow chart of a fan blade abnormal sound monitoring method of the present invention.
Fig. 2 is a timing chart before processing of the sound signal of the faulty blade.
Fig. 3 is a timing chart after processing of the fault blade sound signal.
Fig. 4 is a timing chart before normal blade sound signal processing.
Fig. 5 is a timing chart after normal blade sound signal processing.
Detailed Description
The technical scheme of the invention is described in detail below with reference to the accompanying drawings and examples.
As shown in fig. 1, a fan blade abnormal sound monitoring method includes the following steps:
step 1: setting sound collection frequency and duration parameters, collecting fan blade sound signal data, wherein the sound collection frequency is set to be 2 min/time, and the sound duration is set to be 15s, and the sound signal data comprise a sound signal X (t) and a field real-time wind noise signal Y (t) when the blades rotate;
step 2: based on a Local Mean Decomposition (LMD) method, decomposing the sound signal X (t) acquired in the step 1 during blade rotation to obtain k amplitude modulation and frequency modulation components (PF), wherein the method specifically comprises the following steps:
1) First, find all extreme points n of the sound signal X (t) during blade rotation i Including maximum points and minimum points; calculating local mean value point m of two adjacent extreme points i The formula is as follows:
2) Calculating local envelope a of two adjacent extreme points i The formula is as follows:
3) Point of local mean m i All points of (2) are connected in a broken line form and are smoothed to obtain a local mean function m 11 (t)；
4) Will local envelope a i All points of (a) are connected in a broken line form and are smoothed to obtain a local envelope estimation function a 11 (t)；
5) Will local mean function m 11 (t) extracting from the original blade-rotating sound signal X (t)Is taken out and recorded as zero mean signal h 11 (t) the formula is as follows:
h 11 (t)＝X(t)-m 11 (t)
6) Zero mean value signal h 11 (t) demodulating to obtain s 11 The formula (t) is as follows:
7) Repeating the steps (1) - (6) until a 1(n+1) (t) =1, and pure frequency modulation signal s is obtained 1n (t)；
8) All local envelope estimation functions a obtained by the iteration are used for 11 (t) multiplying to obtain the envelope signal a 1 (t), specifically:
a 1 (t)＝a 11 (t)a 12 (t)a 13 (t)…a 1n (t)
at this time, the first amplitude modulation component PF of the sound signal X (t) during the rotation of the blade is extracted 1 The specific formula is as follows:
PF 1 ＝a 1 (t)s 1n (t)
9) Cutting the first amplitude-modulated frequency-modulated component PF from the original blade-rotated sound signal X (t) 1 And repeating the steps (1) - (8) until the remaining signal is monotone, at which time the acoustic signal X (t) during blade rotation is decomposed into k PF components and a monotone remaining signal u k (t), which can be expressed specifically as:
step 3: based on the k amplitude modulation frequency modulation components PF and the on-site real-time wind noise signals Y (t) obtained by decomposition in the step 2, the Pearson coefficients of the two components are calculatedObtaining k real-time wind noise similarity, and a specific formulaIs that;
wherein cov is covariance and σ is standard deviation;
step 4: setting a threshold value to be 0.7, comparing the k real-time wind noise similarities obtained in the step 3 with the set threshold value, discarding the corresponding frequency modulation and amplitude modulation components PF as wind noise signals when the real-time wind noise similarities are higher than the threshold value of 0.7, and adding the corresponding frequency modulation and amplitude modulation components PF to reconstruct the signals when the real-time wind noise similarities are lower than the threshold value of 0.7 to obtain a blade sound signal Z (t) after noise reduction;
step 5: and (3) performing visualization processing on the blade sound signal Z (t) subjected to noise reduction in the step (4), reading by using an audioread function in MATLAB to obtain a time sequence diagram, and monitoring and analyzing the fan blade state by observing the periodicity of the blade sound signal Z (t) subjected to noise reduction.
A fan blade anomaly sound monitoring system comprising:
the sound collection module is used for: the device is used for collecting a sound signal X (t) and a field real-time wind noise signal Y (t) when the blades rotate;
and a signal processing module: the method comprises the steps of decomposing a sound signal X (t) when a blade rotates to obtain an amplitude modulation frequency modulation component PF; the real-time wind noise similarity of the amplitude modulation component PF and the on-site real-time wind noise signal Y (t) is calculated, and the real-time wind noise similarity is compared with a set threshold value to obtain a blade sound signal Z (t) after noise reduction;
and a data transmission module: the blade sound signal Z (t) after noise reduction in the signal processing module is transmitted to the abnormal sound detection module;
abnormal sound detection module: the method is used for processing the blade sound signal Z (t) after noise reduction to obtain a time sequence diagram, and judging whether the fan blade is abnormal or not according to the time sequence diagram.
A fan blade abnormal sound monitoring apparatus comprising:
a memory: the computer program is used for storing and realizing a fan blade abnormal sound monitoring method;
a processor: the method is used for realizing abnormal sound monitoring of the fan blade when the computer program is executed.
A computer-readable storage medium: the computer readable storage medium stores a computer program which when executed by a processor is capable of implementing a fan blade abnormal sound monitoring method.
The sound collection module collects sound data of the fan blade by adopting a VS1053 audio decoding chip.
The signal processing module adopts an STM32F103ZET6 chip, and the working temperature of the chip is as follows: the temperature is between 40 ℃ below zero and 60 ℃ below zero, and the working environment of the fan can be adapted.
The data transmission module adopts an L768H-E18V communication module to realize online data transmission, and the module supports a conventional TCP/IP communication protocol and supports 4G network full communication.
The outside of sound collection module has wrapped up in the shell, considers the condition that the fan drifts, installs this shell on the fan tower section of thick bamboo that is located 3m apart from ground, guarantees to gather the sound signal of fan blade when the blade rotates to the below in real time.
Fig. 2 is a timing chart before processing a sound signal of a fault blade, fig. 3 is a timing chart after processing the sound signal of the fault blade, and as can be seen by comparing fig. 2 and 3, the sound signal of the fault blade before processing is interfered by stronger wind noise, the actual running state of the blade cannot be accurately reflected, and the sound signal of the fault blade after processing can clearly show the health condition of a fan;
fig. 4 is a timing chart before normal blade sound signal processing, and fig. 5 is a timing chart after normal blade sound signal processing. By comparing the fig. 4 and fig. 5, it can be known that the sound signal of the normal blade before processing is interfered by stronger wind noise, so that the actual running state of the blade cannot be accurately reflected, and the sound signal of the normal blade after processing can clearly show the health condition of the fan.
By comparing the figures 3 and 5, the time sequence diagram of the sound signal of the fault blade shows obvious fluctuation change, the fluctuation change trend has certain periodicity, and the time sequence diagram of the sound signal of the normal blade shows relatively stable fluctuation trend, so that the on-line monitoring of the health condition of the fan blade is realized by on-line monitoring the fluctuation change of the time sequence diagram of the processed sound signal of the blade, the fault blade is found in time, the unnecessary loss caused by sudden accidents is reduced, and the workload of human judgment is reduced.
Claims (10)
1. The fan blade abnormal sound monitoring method is characterized by comprising the following steps of:
step 1: setting sound collection frequency and duration parameters, and collecting fan blade sound signal data, wherein the sound signal data comprise a sound signal X (t) and a field real-time wind noise signal Y (t) when blades rotate;
step 2: decomposing the sound signal X (t) acquired in the step 1 during rotation of the blade based on a Local Mean Decomposition (LMD) method to obtain k amplitude modulation and frequency modulation components (PF);
step 3: based on the k amplitude modulation frequency modulation components PF and the on-site real-time wind noise signals Y (t) obtained by decomposition in the step 2, the Pearson coefficients of the two components are calculatedObtaining k real-time wind noise similarities;
step 4: comparing the k real-time wind noise similarities obtained in the step 3 with a set threshold, discarding the corresponding amplitude modulation and frequency modulation components PF as wind noise signals when the real-time wind noise similarities are higher than the set threshold, and reconstructing the corresponding amplitude modulation and frequency modulation components PF to obtain a blade sound signal Z (t) after noise reduction when the real-time wind noise similarities are lower than the set threshold;
step 5: and (3) carrying out visual processing on the blade sound signal Z (t) subjected to noise reduction in the step (4) to obtain a time sequence diagram, and carrying out monitoring analysis on the state of the fan blade.
2. The fan blade abnormal sound monitoring method according to claim 1, wherein: the k amplitude modulation and frequency modulation components PF in the step 3 are combined with the fieldPearson coefficient of real-time wind noise signal Y (t)The calculation formula of (2) is as follows:
wherein cov is covariance and σ is standard deviation.
3. The fan blade abnormal sound monitoring method according to claim 1, wherein: and in the step 4, setting the threshold value to be 0.7.
4. A fan blade abnormal sound monitoring system is characterized by comprising
The sound collection module is used for: the device is used for collecting a sound signal X (t) and a field real-time wind noise signal Y (t) when the blades rotate;
and a signal processing module: the method comprises the steps of decomposing a sound signal X (t) when a blade rotates to obtain an amplitude modulation frequency modulation component PF; the real-time wind noise similarity of the amplitude modulation component PF and the on-site real-time wind noise signal Y (t) is calculated, and the real-time wind noise similarity is compared with a set threshold value to obtain a blade sound signal Z (t) after noise reduction;
and a data transmission module: the blade sound signal Z (t) after noise reduction in the signal processing module is transmitted to the abnormal sound detection module;
abnormal sound detection module: the method is used for processing the blade sound signal Z (t) after noise reduction to obtain a time sequence diagram, and judging whether the fan blade is abnormal or not according to the time sequence diagram.
5. The fan blade anomaly sound monitoring system of claim 4, wherein: the sound collection module adopts a VS1053 audio decoding chip.
6. The fan blade anomaly sound monitoring system of claim 4, wherein: the signal processing module adopts an STM32F103ZET6 chip.
7. The fan blade anomaly sound monitoring system of claim 4, wherein: the data transmission module adopts an L768H-E18V communication module.
8. The fan blade anomaly sound monitoring device of claim 4, wherein: the outside of the sound collection module is wrapped with a shell, and the shell is arranged on the fan tower.
9. A fan blade abnormal sound monitoring apparatus, comprising:
a memory: a computer program for storing a method for implementing the fan blade abnormal sound monitoring method of claims 1-3;
a processor: a fan blade anomaly sound monitoring method for implementing the method of claims 1-3 when executing the computer program.
10. A computer-readable storage medium, characterized by: the computer readable storage medium stores a computer program which, when executed by a processor, enables a fan blade abnormal sound monitoring method of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410019518.3A CN117789768A (en) | 2024-01-05 | 2024-01-05 | Method, system, equipment and medium for monitoring abnormal sound of fan blade |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410019518.3A CN117789768A (en) | 2024-01-05 | 2024-01-05 | Method, system, equipment and medium for monitoring abnormal sound of fan blade |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117789768A true CN117789768A (en) | 2024-03-29 |
Family
ID=90398380
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410019518.3A Pending CN117789768A (en) | 2024-01-05 | 2024-01-05 | Method, system, equipment and medium for monitoring abnormal sound of fan blade |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117789768A (en) |
-
2024
- 2024-01-05 CN CN202410019518.3A patent/CN117789768A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7199608B2 (en) | Methods and apparatus for inspecting wind turbine blades, and equipment and storage media therefor | |
| Yu et al. | An adaptive sensitive frequency band selection method for empirical wavelet transform and its application in bearing fault diagnosis | |
| Guo et al. | Faulty bearing signal recovery from large noise using a hybrid method based on spectral kurtosis and ensemble empirical mode decomposition | |
| CN108489719B (en) | A kind of combined failure of rotating machinery diagnostic method based on the unusual spectral factorization of G-P | |
| CN107796611A (en) | A kind of warning system for being used to detect wind-driven generator abnormal work | |
| CN110160767B (en) | Impact period automatic identification and extraction method and system based on envelope analysis | |
| Wang et al. | Weak fault detection for wind turbine bearing based on ACYCBD and IESB | |
| CN115467787A (en) | Motor state detection system and method based on audio analysis | |
| CN115078912A (en) | Method and system for detecting abnormity of roof high-voltage cable connector in real time and train | |
| CN114611551A (en) | Electromechanical fault classification method based on wavelet packet energy spectrum entropy | |
| CN116593829B (en) | Transmission line hidden danger on-line monitoring system based on data analysis | |
| CN105987809A (en) | Centrifugal-compressor semi-open-type impeller crack detection method based on random resonance | |
| CN112782421B (en) | Audio-based rotating speed identification method | |
| CN114263621A (en) | Test method and system for diagnosing and simulating cavitation fault of centrifugal pump | |
| CN111456915A (en) | Fault diagnosis device and method for internal components of fan engine room | |
| CN117789768A (en) | Method, system, equipment and medium for monitoring abnormal sound of fan blade | |
| US20240052810A1 (en) | Blade fault diagnosis method, apparatus and system, and storage medium | |
| CN110703080B (en) | GIS spike discharge diagnosis method, discharge degree identification method and device | |
| CN114609453A (en) | Robot joint current anomaly detection method and device based on statistical process control | |
| CN113436645A (en) | Electromechanical system fault on-line monitoring acoustic processing method under complex noise environment | |
| CN117723303B (en) | Acoustic monitoring method for wind generating set bearing | |
| Altankhuyag et al. | Noise Signal Analysis for Fault Detection | |
| CN117192310A (en) | Online monitoring method, system and medium for partial discharge of high-voltage sleeve | |
| McSwiggan et al. | Analysis of fixed-speed wind farm low-frequency power pulsations using a wavelet-prony method | |
| CN117421587A (en) | Online evaluation method for health state of rotary equipment based on typical correlation coefficient |
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 |