CN114526812A - Abnormal sound monitoring method and system for mechanical equipment - Google Patents
Abnormal sound monitoring method and system for mechanical equipment Download PDFInfo
- Publication number
- CN114526812A CN114526812A CN202210008533.9A CN202210008533A CN114526812A CN 114526812 A CN114526812 A CN 114526812A CN 202210008533 A CN202210008533 A CN 202210008533A CN 114526812 A CN114526812 A CN 114526812A
- Authority
- CN
- China
- Prior art keywords
- sound
- abnormal noise
- mechanical equipment
- abnormal
- processing
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E30/00—Energy generation of nuclear origin
- Y02E30/30—Nuclear fission reactors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
Abstract
本发明涉及一种机械设备异音监测方法及系统，包括：S1：采集被测机械设备的声音信号和图像信号；S2：识别所述声音信号是否含有异常噪声；S3：若存在，则对所述异常噪声进行声源定位处理；S4：根据所述异常噪声的产生方向和噪声强度、以及所述图像信号，进行声场成像处理；S5：对所述声音信号以及声场成像处理后的信息进行数据分析及故障诊断。本发明通过监测机械设备的异音，不仅能够确定设备的故障部件，还能有效地诊断其故障原因，使维修有目的地进行，以减少计划外停堆，从而尽可能地避免影响到核电机组的正常运作。
The invention relates to a method and system for monitoring abnormal noise of mechanical equipment, comprising: S1: collecting sound signals and image signals of the mechanical equipment under test; S2: identifying whether the sound signal contains abnormal noise; Perform sound source localization processing on the abnormal noise; S4: Perform sound field imaging processing according to the generation direction and noise intensity of the abnormal noise and the image signal; S5: Perform data processing on the sound signal and the information after the sound field imaging processing Analysis and troubleshooting. By monitoring the abnormal sound of mechanical equipment, the invention can not only determine the faulty parts of the equipment, but also effectively diagnose the cause of the fault, so that maintenance can be carried out in a targeted manner, so as to reduce unplanned shutdowns, thereby avoiding affecting the nuclear power unit as much as possible. of normal operation.
Description
Technical Field
The invention relates to the technical field of equipment detection, in particular to a method and a system for monitoring abnormal sound of mechanical equipment.
Background
In a nuclear power plant, mechanical equipment is various in types, long in continuous working time and difficult to avoid to have faults. Mechanical equipment usually fails, which causes abnormal noise to be generated in the equipment, and the abnormal noise is usually used as a basis for judging the existence of the failure. For example, when the sound is in a clunk state and accompanied by a certain vibration phenomenon, it is easier to judge and recognize that a failure is present. When the abnormal noise emitted by the equipment is very slight or complex, a worker cannot effectively identify whether the abnormal noise exists, difficulty is brought to fault diagnosis, and early faults of the equipment cannot be identified timely and accurately. The abnormal sound problem occurs occasionally, and the operation safety, stability, reliability and load rate of the nuclear power station unit are seriously influenced.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method and a system for monitoring abnormal sounds of mechanical equipment, aiming at least one defect existing in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows: a method for monitoring abnormal sound of mechanical equipment is constructed, and comprises the following steps:
s1: collecting a sound signal and an image signal of a mechanical device to be detected;
s2: identifying whether the sound signal contains abnormal noise;
s3: if the abnormal noise exists, sound source positioning processing is carried out on the abnormal noise so as to confirm the generation direction and the noise intensity of the abnormal noise;
s4: performing sound field imaging processing according to the generation direction and the noise intensity of the abnormal noise and the image signal to realize imaging of the generation direction and the noise intensity of the abnormal noise;
s5: and performing data analysis and fault diagnosis on the sound signals and the information after sound field imaging processing so as to identify the part of the mechanical equipment with mechanical fault.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the S1 further includes:
s11: preprocessing the sound signal;
s12: and carrying out image vectorization processing and pixel angle conversion processing on the image signal.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the S11 includes:
and carrying out one or more processing measures of removing direct current components, normalizing, filtering and pre-emphasis processing on the sound signal.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, after S11, the method further includes: and performing feature extraction on the preprocessed sound signals by referring to the sound characteristics of the parts in the mechanical equipment when the parts work normally or abnormally to extract the feature components of the sound signals, thereby providing a detection basis for subsequently identifying whether the preprocessed sound signals contain abnormal noise.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the S2 includes:
and carrying out a deep neural network detection algorithm or an unsupervised abnormal sound detection algorithm on the characteristic components to identify whether the preprocessed sound signals contain abnormal noise.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the sound source localization processing in S3 includes:
acquiring the sound signals through a sensor array, and performing compensation summation on the phases of the acquired array signals to obtain initial direction information; meanwhile, according to the frequency domain characteristics of the abnormal noise, a wave beam peak value corresponding to the abnormal noise is searched in the sound signal; and performing corresponding phase compensation and weighted summation processing according to the preliminary direction information and the beam peak value to obtain a three-dimensional beam pattern for representing the generation direction of the abnormal noise and the sound field intensity of the corresponding direction of the abnormal noise.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the sound field imaging processing in S4 includes:
converting the three-dimensional beam pattern into a two-dimensional color contour map; the representation points in the two-dimensional color contour map are used for displaying the generation direction of abnormal noise of the pixel points in the video image and displaying the sound field intensity by utilizing the shades of the colors of the representation points; and performing transparency processing on the two-dimensional color contour map, and overlapping the characterization points and the pixel points to realize imaging of the generation direction and the noise intensity of the abnormal noise.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the data analysis and fault diagnosis in S5 includes:
setting a fault simulation test bench for simulating faults of various parts of the mechanical equipment, calibrating and recording sample data generated by different fault modes to serve as a fault sample model, performing time domain analysis, frequency spectrum analysis and envelope analysis on the sound signal, and comparing the analyzed sound signal and data information after imaging processing with the fault sample model to realize fault diagnosis of the tested mechanical equipment.
Preferably, in the abnormal sound monitoring method for mechanical equipment according to the present invention, the time-domain analyzing the sound signal further includes:
whether the component has mechanical faults is diagnosed by judging whether the time domain characteristic parameters after time domain analysis exceed preset values;
wherein the time-domain characteristic parameters comprise one or more combinations of mean, peak, root mean square, crest factor, peak-to-peak, and kurtosis.
The invention also constructs a mechanical equipment abnormal sound monitoring system, which comprises:
the acquisition unit is used for acquiring a sound signal and an image signal of the mechanical equipment to be detected;
an abnormal noise identifying unit for identifying whether the sound signal contains abnormal noise;
the sound positioning unit is used for carrying out sound source positioning processing on the abnormal noise;
an imaging processing unit for performing sound field imaging processing according to the generation direction and noise intensity of the abnormal noise and the image signal;
and the data analysis and fault diagnosis unit is used for carrying out data analysis and fault diagnosis on the sound signals and the information after the sound field imaging processing.
The implementation of the invention has the following beneficial effects: the invention monitors and diagnoses the abnormal sound of the mechanical equipment, senses the abnormal sound of the mechanical equipment through sound source positioning processing, determines the specific fault parts of the mechanical equipment, and can effectively diagnose the fault reasons, so that the maintenance is purposefully carried out, the unplanned shutdown is reduced, and the influence on the normal operation of the nuclear power unit is avoided as much as possible.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a first embodiment of a method for monitoring abnormal noise of a mechanical device according to the present invention;
FIG. 2 is a flowchart illustrating a method for monitoring abnormal sounds in a mechanical device according to a second embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a pixel angle conversion process in the abnormal sound monitoring method for a mechanical device according to the present invention;
FIG. 4 is a three-dimensional beam diagram of the abnormal noise monitoring method for mechanical equipment according to the present invention;
FIG. 5 is a two-dimensional color contour map of the abnormal sound monitoring method for mechanical equipment according to the present invention;
FIG. 6 is a schematic diagram of a mechanical apparatus after sound field imaging processing in the abnormal sound monitoring method provided by the present invention;
fig. 7 is a schematic diagram of a representation point and a pixel point in the abnormal sound monitoring method for mechanical equipment according to the present invention after superposition;
fig. 8 is a comparison diagram of visualization processing in the abnormal sound monitoring method for mechanical equipment provided by the invention.
Detailed Description
For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
Fig. 1 is a first embodiment of the abnormal noise monitoring method for mechanical equipment provided by the invention. As shown in fig. 1, the present invention constructs a method for monitoring abnormal noise of mechanical equipment, including: step S1, step S2, step S3, step S4, and step S5.
S1: and collecting sound signals and image signals of the mechanical equipment to be tested.
Fig. 2 is a flowchart of a second embodiment of the abnormal noise monitoring method for mechanical equipment, which is different from the first embodiment in that step S1 further includes: step S11 and step S12.
S11: the sound signal is preprocessed, wherein the preprocessing comprises one or more of removing direct current components, normalizing, filtering and pre-emphasis processing, so as to ensure the integrity of the sound signal and prevent the sound signal from being distorted. After the sound signal is preprocessed, the feature extraction is carried out on the preprocessed sound signal by referring to the sound feature emitted by a component in the mechanical equipment when the component works normally or works abnormally, so as to extract the feature component of the sound signal, and thus, a detection basis is provided for the subsequent identification of whether the preprocessed sound signal contains abnormal noise.
Feature extraction is a means of analyzing and processing the sound signal after preprocessing. The principle of feature extraction is as follows: because the sound production mechanisms of different devices or parts are different, the different devices or parts have unique characteristic components when working normally or having abnormal faults, and therefore the characteristic components can be extracted to be used as the detection basis for whether the sound signals contain abnormal noise. The feature components extracted by the features in the method comprise STFT, MFCC and HPSS.
S12: and carrying out image vectorization processing and pixel angle conversion processing on the image signals to obtain a video image. The image vectorization processing is carried out on the image signals, so that the image signals are converted into a vector matrix. Fig. 3 is a schematic diagram of pixel angle conversion processing in the abnormal sound monitoring method for mechanical equipment provided by the present invention, and as shown in fig. 3, the pixel angle conversion processing is performed on the image signal, so that each pixel point in the image signal and its respective angle in the sound field form a corresponding relationship, and are mapped onto a vector matrix, and finally a transformation matrix is formed. And carrying out image vectorization processing and pixel angle conversion processing on the image signals to obtain a video image, and preparing for subsequent sound field imaging processing.
S2: the method for identifying whether the sound signal contains abnormal noise comprises the following steps: and carrying out a deep neural network detection algorithm or an unsupervised abnormal sound detection algorithm on the characteristic components to identify whether the preprocessed sound signals contain abnormal noise.
Further, the method can be divided into two detection methods according to whether signal detection needs a large number of positive and negative samples for early learning, wherein the two detection methods comprise a deep neural network detection algorithm based on CRNN and an unsupervised abnormal sound detection algorithm based on AutoEncoder.
The depth neural network detection algorithm based on the CRNN is suitable for a detection environment in which positive and negative samples of abnormal noise of equipment are sufficient and the number of the positive and negative samples is balanced. Model training can be carried out by collecting the characteristic components of the sound signals of the mechanical equipment to be detected, so that one or more abnormal noises can be identified and diagnosed, and rapid early warning can be carried out under the abnormal condition. Because the algorithm needs a large number of positive and negative samples as the basis of deep learning, the accuracy of the final identification of the algorithm is influenced by the comprehensiveness and generalization of the positive and negative samples of abnormal noise of the equipment, and therefore the algorithm has the following characteristics: the method is suitable for the condition that the number of positive and negative samples is sufficient; the detection pertinence is strong, and the abnormal type can be diagnosed directly according to the characteristic component of the sound signal; the early-stage data accumulation time is long, and the requirement on the sample is high.
The unsupervised abnormal sound detection algorithm based on the AutoEncoder is characterized in that an AutoEncoder self-coder is used for coding and decoding feature components of normal sound signals, and a coder only detecting a normal sample is formed through deep neural network training. When the system collects abnormal noise existing in abnormal working state of the mechanical equipment to be detected, the output result of the encoder is obviously deviated, so that the output result of the encoder can be used as a main basis for judging whether the abnormal noise exists in the mechanical equipment to be detected. The algorithm only needs a small number of positive samples to realize the construction of the neural network of the self-encoder and does not depend on the support of a large number of negative samples. Therefore, the algorithm is high in usability, core points influencing the accuracy of the algorithm are the generalization and complexity of the positive sample, and the algorithm has excellent detection performance for equipment with stable working noise state. The algorithm is therefore characterized as follows: the method is suitable for sound detection in an unsupervised state with a small number of samples; the algorithm is not dependent on various negative samples, and the algorithm deployment period is fast; only the change of the noise state can be pre-warned, and the actual fault type cannot be accurately diagnosed.
S3: if the abnormal noise exists, sound source positioning processing is carried out on the abnormal noise so as to confirm the generation direction and the noise intensity of the abnormal noise. The sound positioning processing comprises the steps of collecting sound signals through a sensor array, and performing compensation summation on the phases of the collected array signals to obtain initial direction information; meanwhile, according to the frequency domain characteristics of the abnormal noise, a wave beam peak value corresponding to the abnormal noise is searched in the sound signal; and performing corresponding phase compensation and weighted summation processing according to the preliminary direction information and the beam peak value to obtain a three-dimensional beam pattern for representing the generation direction of the abnormal noise and the sound field intensity of the corresponding direction.
S4: and performing sound field imaging processing according to the generation direction and the noise intensity of the abnormal noise and the image signal so as to realize imaging of the generation direction and the noise intensity of the abnormal noise. As shown in fig. 4, 5, 6, and 7, the sound field imaging process includes: converting the three-dimensional beam pattern into a two-dimensional color contour map; the representation points in the two-dimensional color contour map are used for displaying the generation direction of abnormal noise of the pixel points in the video image and displaying the sound field intensity by utilizing the shades of the colors of the representation points; and performing transparency processing on the two-dimensional color contour map, and overlapping the characterization points and the pixel points to realize imaging of the generation direction and the noise intensity of the abnormal noise.
Further, visualization processing is also included after the sound field imaging processing; and the visualization processing is used for intercepting part of the image with higher color value in the image after the sound field imaging processing. Fig. 8 is a comparison diagram of visualization processing in the abnormal sound monitoring method for mechanical equipment provided by the invention. As shown in fig. 8, the left side is an image of the mechanical device under test in normal operation, and the right side is an image of the mechanical device under test in failure.
S5: and performing data analysis and fault diagnosis on the sound signals and the information after the sound field imaging processing so as to identify the part of the mechanical equipment with mechanical fault. Wherein, data analysis and fault diagnosis include: the method comprises the steps of setting a fault simulation test bench for simulating faults of various parts of the mechanical equipment, calibrating and recording sample data generated by different fault modes, using the sample data as a fault sample model, carrying out time domain analysis, frequency spectrum analysis and envelope analysis on a sound signal, and comparing the analyzed sound signal and data information after imaging processing with the fault sample model to realize fault diagnosis of the mechanical equipment to be tested.
The fault simulation test bench comprises a test bench base, a small driving motor, a motor mounting rack, a motor, a rotating shaft, a coupler, a bearing load external member, a rotating shaft friction external member and a rotating shaft vibration measurement external member. The failure modes of the failure simulation test bench include: simulating misalignment faults, namely simulating working states of various components when shafts are misaligned, and also simulating working states of couplings with different rigidities when the couplings affect the power and vibration of a rotor; rotating shaft collision and grinding simulation of rotating mechanical equipment is carried out, and a rotating shaft collision and grinding kit is used for simulating the working state of collision and grinding of a shaft or a rotor; the resonance simulation is that the rotors with different numbers are arranged at different positions of the rotating shaft to excite the resonance frequency so as to simulate the working state of the rotors and the rotating shaft with resonance phenomenon, and the influence of the mass and the rigidity on the resonance frequency and the modal shape can be researched by adjusting the positions of the rotors and the supports; simulating the fault of the bearing, namely simulating the working state of the inner ring, the outer ring, the rolling body and the retainer of the rolling bearing in fault by using the rolling bearing assembly; and (3) simulating the eccentric fault of the rotating shaft, wherein the working state of the rotating shaft with the eccentric fault is simulated by using a rotating shaft radial vibration sleeve.
In this embodiment, in order to ensure the accuracy of the sound source localization process, the failure simulation test platform may be used to simulate failure modes of various components, so as to verify the accuracy of the sound source localization process in the method. For example, when a fault simulation test bench is used for simulating a bearing outer ring fault, a bearing outer ring fault suite is installed on a vibration test bench to simulate the bearing outer ring fault, the test results of the method at different test points are compared, and if the test results at different test points all show that a problem bearing part is abnormal, the sound source positioning processing function is proved to be normal.
Furthermore, the fault simulation test bench can be used for simulating fault modes of various components, the test accuracy of the method can be verified and debugged, the fault simulation test bench is used for simulating various fault modes, the steps S1-S5 in the method are sequentially executed, and finally the test result of the method is compared with the fault components of the fault simulation test bench to verify whether the method is accurate or not. In the test process of various fault modes, representative data of the test process can be calibrated and stored to serve as a fault sample model for comparison.
Further, performing the time domain analysis on the sound signal further includes: whether the component has mechanical faults is diagnosed by judging whether the time domain characteristic parameters after time domain analysis exceed preset values; wherein the time-domain characteristic parameters comprise one or more combinations of mean, peak, root mean square, crest factor, peak-to-peak, and kurtosis.
The mean value is used to reflect the static part of the sound signal, which does not contribute to the diagnosis, but has a large influence on the calculation of other parameters, so that the mean value is taken from the sound signal at the time of calculation, leaving a dynamic part useful for the diagnosis.
The peak, which is used to characterize the maximum value of the noise intensity over a certain time period, is a time-varying maximum and can therefore be used to detect an impact signal.
The root mean square is an effective value representing the sound intensity, is the most important index in the monitoring process, and can be used as an important index for judging the wear type faults.
The crest factor is the ratio of the peak value to the root mean square, and the fault reason of the mechanical equipment to be tested can be predicted according to the variation and the size of the crest factor. For example, when the crest factor increases, it may be a pitting failure, and when the crest factor decreases, it may be a wear failure. The normal crest factor value of the bearing is about 4-5, and when the crest factor is larger than 10, the impact is caused by the local defect of a part; when the crest factor becomes small, it indicates an abnormal condition such as poor lubrication and wear of the bearing, and the noise becomes large.
The peak value is used for representing the amplitude of the vibration, so that the peak value can be used for describing the displacement value of the component, and whether the component has the peeling fault or not can be judged.
Kurtosis is used for representing waveform kurtosis, is particularly sensitive to suddenly-changed sound signals, normally, the kurtosis index should be below 3, and when the kurtosis index is greater than 3.5, the kurtosis index generally indicates the occurrence of a fault, so that early fault diagnosis can be carried out by using the kurtosis, and the kurtosis index is particularly suitable for surface damage type faults. For example, when a bearing fails early, the kurtosis value increases immediately, but rather does not change or gradually decreases as the bearing failure progresses.
The spectrum analysis is one of the basic methods in the fault characteristic research method, and can obtain more comprehensive and accurate fault information in the spectrum aiming at the monitoring object of the method.
The envelope analysis is used for extracting periodic impact signals sent when the bearing fails, and therefore component failures can be diagnosed.
Furthermore, the method also comprises the step of performing trend analysis processing after the fault diagnosis of the tested mechanical equipment is completed. The purpose of trend analysis processing is to arrange maintenance work according to actual conditions and arrange maintenance at proper time, so that decision is scientific and economic loss is reduced as much as possible. The fault diagnosis is the basis of trend analysis processing, and a time history curve of the detected mechanical equipment is drawn according to the characteristic components of the collected sound signals so as to analyze the near-term and far-term trend states of the detected mechanical equipment.
The invention also constructs a mechanical equipment abnormal sound monitoring system, which comprises: the device comprises a collecting unit, an abnormal noise identification unit, a sound positioning unit, an imaging processing unit and a data analysis and fault diagnosis unit.
And the acquisition unit is used for acquiring the sound signal and the image signal of the mechanical equipment to be detected.
In some embodiments, the system further comprises a preprocessing unit for preprocessing the sound signals and performing image vectorization processing and pixel angle conversion processing on the image signals.
The preprocessing unit preprocesses the sound signal by one or more of removing a direct current component, normalizing, filtering and pre-emphasis processing, so as to ensure the integrity of the sound signal and prevent the sound signal from being distorted. After the sound signal is preprocessed, the feature extraction is carried out on the preprocessed sound signal according to the sound feature of the component in the mechanical equipment in normal work or abnormal work so as to extract the feature component of the sound signal.
The preprocessing unit carries out image vectorization processing on the image signals, namely, the image vectorization processing is carried out on the image signals to enable the image signals to be converted into vector matrixes, and the pixel angle conversion processing is carried out on the image signals to enable all pixel points in the image signals and angles of the pixel points in the sound field to form corresponding relations and map the corresponding relations to the vector matrixes, and finally a transformation matrix is formed. And carrying out image vectorization processing and pixel angle conversion processing on the image signals to obtain a video image.
An abnormal noise identification unit for identifying whether the sound signal contains abnormal noise, the identification method comprises: and carrying out a deep neural network detection algorithm or an unsupervised abnormal sound detection algorithm on the characteristic components to identify whether the preprocessed sound signals contain abnormal noise.
And the sound positioning unit is used for carrying out sound source positioning processing on the abnormal noise so as to confirm the generation direction and the noise intensity of the abnormal noise. The sound positioning unit further comprises a collecting sensor array for sound signals and a positioning analysis processor. The positioning analysis processor carries out compensation summation on the phases of the collected array signals to obtain preliminary direction information; meanwhile, according to the frequency domain characteristics of the abnormal noise, a wave beam peak value corresponding to the abnormal noise is searched in the sound signal; and performing corresponding phase compensation and weighted summation processing according to the preliminary direction information and the beam peak value to obtain a three-dimensional beam pattern for representing the generation direction of the abnormal noise and the sound field intensity of the corresponding direction of the abnormal noise.
And the imaging processing unit is used for carrying out sound field imaging processing according to the generation direction and the noise intensity of the abnormal noise and the image signal so as to realize imaging of the generation direction and the noise intensity of the abnormal noise. The imaging processing unit converts the three-dimensional beam pattern into a two-dimensional color contour map; the representation points in the two-dimensional color contour map are used for displaying the generation direction of abnormal noise of the pixel points in the video image and displaying the sound field intensity by utilizing the shades of the colors of the representation points; and performing transparency processing on the two-dimensional color contour map, and overlapping the characterization points and the pixel points to realize imaging of the generation direction and the noise intensity of the abnormal noise.
Furthermore, the imaging processing unit is also used for performing visualization processing on the data after the sound field imaging processing.
And the data analysis and fault diagnosis unit is used for carrying out data analysis and fault diagnosis on the sound signals and the information after the sound field imaging processing so as to identify the part of the mechanical equipment with mechanical fault. The data analysis and fault diagnosis unit comprises a fault simulation test platform for simulating faults of various components of the mechanical equipment and an upper computer for data analysis and fault diagnosis. The upper computer calibrates and records sample data generated by different fault modes, the sample data is used as a fault sample model, time domain analysis, frequency spectrum analysis and envelope analysis are carried out on the sound signal, and the analyzed sound signal and the data information after imaging processing are compared with the fault sample model, so that fault diagnosis of the tested mechanical equipment is realized.
The implementation of the invention has the following beneficial effects: the invention monitors and diagnoses the abnormal sound of the mechanical equipment, senses the abnormal sound of the mechanical equipment through sound source positioning processing, determines the specific fault parts of the mechanical equipment, and can effectively diagnose the fault reasons, so that the maintenance is purposefully carried out, the unplanned shutdown is reduced, and the influence on the normal operation of the nuclear power unit is avoided as much as possible.
It is to be understood that the foregoing examples, while indicating the preferred embodiments of the invention, are given by way of illustration and description, and are not to be construed as limiting the scope of the invention; it should be noted that, for those skilled in the art, the above technical features can be freely combined, and several changes and modifications can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention; therefore, all equivalent changes and modifications made within the scope of the claims of the present invention should be covered by the claims of the present invention.
Claims (10)
1. A mechanical equipment abnormal sound monitoring method is characterized by comprising the following steps:
s1: collecting a sound signal and an image signal of a mechanical device to be detected;
s2: identifying whether the sound signal contains abnormal noise;
s3: if the abnormal noise exists, sound source positioning processing is carried out on the abnormal noise so as to confirm the generation direction and the noise intensity of the abnormal noise;
s4: performing sound field imaging processing according to the generation direction and the noise intensity of the abnormal noise and the image signal to realize imaging of the generation direction and the noise intensity of the abnormal noise;
s5: and performing data analysis and fault diagnosis on the sound signals and the information after sound field imaging processing so as to identify the part of the mechanical equipment with mechanical fault.
2. The abnormal noise monitoring method for mechanical equipment according to claim 1, wherein said S1 further comprises:
s11: preprocessing the sound signal;
s12: and carrying out image vectorization processing and pixel angle conversion processing on the image signal to obtain a video image.
3. The abnormal noise monitoring method for mechanical equipment according to claim 2, wherein said S11 includes:
and carrying out one or more processing measures of removing direct current components, normalizing, filtering and pre-emphasis processing on the sound signal.
4. The abnormal noise monitoring method for mechanical equipment according to claim 3, wherein after said S11, the method further comprises: and performing feature extraction on the preprocessed sound signals by referring to the sound characteristics of the parts in the mechanical equipment when the parts work normally or abnormally to extract the feature components of the sound signals, thereby providing a detection basis for subsequently identifying whether the preprocessed sound signals contain abnormal noise.
5. The abnormal noise monitoring method for mechanical equipment according to claim 4, wherein said S2 includes:
and carrying out a deep neural network detection algorithm or an unsupervised abnormal sound detection algorithm on the characteristic components to identify whether the preprocessed sound signals contain abnormal noise.
6. The abnormal noise monitoring method for mechanical equipment according to claim 5, wherein the sound source localization processing in S3 includes:
acquiring the sound signals through a sensor array, and performing compensation summation on the phases of the acquired array signals to obtain initial direction information; meanwhile, according to the frequency domain characteristics of the abnormal noise, a wave beam peak value corresponding to the abnormal noise is searched in the sound signal; and performing corresponding phase compensation and weighted summation processing according to the preliminary direction information and the beam peak value to obtain a three-dimensional beam pattern for representing the generation direction of the abnormal noise and the sound field intensity of the corresponding direction of the abnormal noise.
7. The abnormal noise monitoring method for mechanical equipment according to claim 6, wherein the sound field imaging processing in S4 includes:
converting the three-dimensional beam pattern into a two-dimensional color contour map; the representation points in the two-dimensional color contour map are used for displaying the generation direction of abnormal noise of the pixel points in the video image and displaying the sound field intensity by utilizing the shades of the colors of the representation points; and performing transparency processing on the two-dimensional color contour map, and overlapping the characterization points and the pixel points to realize imaging of the generation direction and the noise intensity of the abnormal noise.
8. The abnormal noise monitoring method for mechanical equipment according to claim 7, wherein the data analysis and fault diagnosis in S5 includes:
setting a fault simulation test bench for simulating faults of various parts of the mechanical equipment, calibrating and recording sample data generated by different fault modes to serve as a fault sample model, performing time domain analysis, frequency spectrum analysis and envelope analysis on the sound signal, and comparing the analyzed sound signal and data information after imaging processing with the fault sample model to realize fault diagnosis of the tested mechanical equipment.
9. The mechanical equipment abnormal sound monitoring method according to claim 8, wherein the time domain analysis of the sound signal further comprises:
whether the component has mechanical faults is diagnosed by judging whether the time domain characteristic parameters after time domain analysis exceed preset values;
wherein the time-domain characteristic parameters comprise one or more combinations of mean, peak, root mean square, crest factor, peak-to-peak, and kurtosis.
10. A mechanical equipment abnormal sound monitoring system, characterized in that, monitoring system includes:
the acquisition unit is used for acquiring a sound signal and an image signal of the mechanical equipment to be detected;
an abnormal noise identification unit for identifying whether the sound signal contains abnormal noise;
the sound positioning unit is used for carrying out sound source positioning processing on the abnormal noise;
an imaging processing unit for performing sound field imaging processing according to the generation direction and noise intensity of the abnormal noise and the image signal;
and the data analysis and fault diagnosis unit is used for carrying out data analysis and fault diagnosis on the sound signals and the information after the sound field imaging processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210008533.9A CN114526812B (en) | 2022-01-05 | 2022-01-05 | A method and system for monitoring abnormal sound of mechanical equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210008533.9A CN114526812B (en) | 2022-01-05 | 2022-01-05 | A method and system for monitoring abnormal sound of mechanical equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114526812A true CN114526812A (en) | 2022-05-24 |
| CN114526812B CN114526812B (en) | 2025-02-11 |
Family
ID=81621122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210008533.9A Active CN114526812B (en) | 2022-01-05 | 2022-01-05 | A method and system for monitoring abnormal sound of mechanical equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114526812B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115019813A (en) * | 2022-05-31 | 2022-09-06 | 漳州立达信光电子科技有限公司 | Sound energy detection method, device and medium |
| CN115471827A (en) * | 2022-08-26 | 2022-12-13 | 岱特智能科技（上海）有限公司 | A hemodialysis machine pump status warning method and system based on audio recognition |
| CN115753104A (en) * | 2022-10-12 | 2023-03-07 | 中核武汉核电运行技术股份有限公司 | Rolling bearing detection and fault diagnosis device and method based on ultrasonic signal analysis |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090120842A (en) * | 2008-05-21 | 2009-11-25 | 울산대학교 산학협력단 | Fault diagnosis device and method through video signal conversion of machine operation sound and vibration signal |
| CN103198496A (en) * | 2013-03-29 | 2013-07-10 | 浙江大学 | Method for carrying out vectorization on abstract image |
| CN104748764A (en) * | 2015-04-01 | 2015-07-01 | 清华大学 | Method for calibrating space angle of acoustic image plane in acoustic field visualization system |
| CN111537822A (en) * | 2020-05-22 | 2020-08-14 | 国网上海市电力公司 | A Transformer Fault Diagnosis System Based on Acoustic Imaging |
| CN112193959A (en) * | 2020-09-25 | 2021-01-08 | 浙江新再灵科技股份有限公司 | Method and system for detecting abnormal sound of elevator |
| CN113074967A (en) * | 2020-01-06 | 2021-07-06 | 北京谛声科技有限责任公司 | Abnormal sound detection method and device, storage medium and electronic equipment |
| CN113313146A (en) * | 2021-05-10 | 2021-08-27 | 上海核工程研究设计院有限公司 | Intelligent monitoring method for abnormal operation state of equipment |
-
2022
- 2022-01-05 CN CN202210008533.9A patent/CN114526812B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090120842A (en) * | 2008-05-21 | 2009-11-25 | 울산대학교 산학협력단 | Fault diagnosis device and method through video signal conversion of machine operation sound and vibration signal |
| CN103198496A (en) * | 2013-03-29 | 2013-07-10 | 浙江大学 | Method for carrying out vectorization on abstract image |
| CN104748764A (en) * | 2015-04-01 | 2015-07-01 | 清华大学 | Method for calibrating space angle of acoustic image plane in acoustic field visualization system |
| CN113074967A (en) * | 2020-01-06 | 2021-07-06 | 北京谛声科技有限责任公司 | Abnormal sound detection method and device, storage medium and electronic equipment |
| CN111537822A (en) * | 2020-05-22 | 2020-08-14 | 国网上海市电力公司 | A Transformer Fault Diagnosis System Based on Acoustic Imaging |
| CN112193959A (en) * | 2020-09-25 | 2021-01-08 | 浙江新再灵科技股份有限公司 | Method and system for detecting abnormal sound of elevator |
| CN113313146A (en) * | 2021-05-10 | 2021-08-27 | 上海核工程研究设计院有限公司 | Intelligent monitoring method for abnormal operation state of equipment |
Non-Patent Citations (3)
| Title |
|---|
| 周昌国;陈日林;滕鹏晓;韩宝坤;杨亦春;: "星形立体传声器阵列设计与应用", 声学技术, no. 02, 15 April 2013 (2013-04-15) * |
| 联丰迅声: "声源定位与声场成像有什么区别？", Retrieved from the Internet <URL:https://baijiahao.baidu.com/s?id=1661304876635855222&wfr=spider&for=pc> * |
| 马波等: "基于麦克风阵列的机车车辆转向架跑合试验的故障监测研究", 机车电传动, no. 6, 10 November 2019 (2019-11-10), pages 95 - 99 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115019813A (en) * | 2022-05-31 | 2022-09-06 | 漳州立达信光电子科技有限公司 | Sound energy detection method, device and medium |
| CN115471827A (en) * | 2022-08-26 | 2022-12-13 | 岱特智能科技（上海）有限公司 | A hemodialysis machine pump status warning method and system based on audio recognition |
| CN115753104A (en) * | 2022-10-12 | 2023-03-07 | 中核武汉核电运行技术股份有限公司 | Rolling bearing detection and fault diagnosis device and method based on ultrasonic signal analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114526812B (en) | 2025-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114526812B (en) | A method and system for monitoring abnormal sound of mechanical equipment | |
| Jamaludin et al. | Monitoring extremely slow rolling element bearings: part I | |
| JP2020183939A (en) | Method for detecting abnormality of gear box and information processing device | |
| KR100225470B1 (en) | Motor fault monitoring system with signal analysis of power supply line | |
| EP4321851A1 (en) | Method for providing a physically explainable fault information of a bearing by a fault detection model | |
| CN113686576B (en) | Rotary machine early fault diagnosis method based on nonlinear convolution sparse filtering | |
| CN113656989A (en) | Bolt fault diagnosis method and device | |
| CN111458016A (en) | Method for rapidly detecting vibration of drive axle assembly | |
| CN118503882A (en) | Method for collecting operating state parameters and diagnosing faults of three-phase horizontal decanter centrifuge | |
| CN118936894A (en) | Turbomachinery rotor vibration holographic monitoring and dynamic balancing method, equipment and medium | |
| Enshaei et al. | Icphm’23 benchmark vibration dataset applicable in machine learning for systems’ health monitoring | |
| CN117760545A (en) | Online vibration monitoring system for aero-engine rotor | |
| CN106403880B (en) | Method and device for detecting clearance between compressor rotor and stator | |
| CN112711850A (en) | Unit online monitoring method based on big data | |
| CN119644965B (en) | A real-time monitoring and fault diagnosis method and system for trailer production line | |
| JP7515012B2 (en) | How to diagnose the technical condition of rotating equipment | |
| CN120121300A (en) | A sliding bearing-rotor system stability test system and method under complex alternating loads | |
| CN119043620B (en) | A vibration test system for gas turbine impeller production | |
| CN115293192A (en) | Rotating machinery fault judging method, computer equipment and storage medium | |
| Cheng et al. | Quantitative damage detection for planetary gear sets based on physical models | |
| Jiang et al. | Rolling bearing quality evaluation based on a morphological filter and a Kolmogorov complexity measure | |
| CN118111738A (en) | Industrial robot abnormality detection method based on self-encoder and sound sensor | |
| Pinardi et al. | A Novel Hybrid Acquisition System for Industrial Condition Monitoring and Predictive Maintenance | |
| CN112857798B (en) | Multi-shafting mechanical transmission system quality evaluation method and device based on spectrum analysis | |
| Bachir et al. | Application of principal component analysis approach in gas turbine defect diagnosis |
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 |