CN116520143B - Voiceprint data-based rotating equipment monitoring method, device, equipment and medium - Google Patents
Voiceprint data-based rotating equipment monitoring method, device, equipment and medium Download PDFInfo
- Publication number
- CN116520143B CN116520143B CN202310799871.3A CN202310799871A CN116520143B CN 116520143 B CN116520143 B CN 116520143B CN 202310799871 A CN202310799871 A CN 202310799871A CN 116520143 B CN116520143 B CN 116520143B
- Authority
- CN
- China
- Prior art keywords
- signal
- sound signal
- noise ratio
- equipment
- rotating
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 230000005236 sound signal Effects 0.000 claims abstract description 181
- 238000003062 neural network model Methods 0.000 claims abstract description 58
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000000737 periodic effect Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 11
- 238000012806 monitoring device Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
-
- 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
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/346—Testing of armature or field windings
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/18—Artificial neural networks; Connectionist approaches
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Control Of Electric Motors In General (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a rotating equipment monitoring method, a device, equipment and a medium based on voiceprint data, wherein the method obtains a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal by acquiring the first sound signal recorded in real time when the rotating equipment operates and preprocessing the first sound signal; inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation; the device status of the rotating device is determined based on the device status tag. Therefore, the technical scheme of the invention realizes real-time monitoring of the rotating equipment so as to timely identify equipment faults of the rotating equipment.
Description
Technical Field
The invention belongs to the technical field of rotating equipment, and particularly relates to a rotating equipment monitoring method, device, equipment and medium based on voiceprint data.
Background
With the continuous development of the mechanical industry, rotating equipment plays a vital role in industries such as petrochemical industry, metallurgy, nuclear energy, coal and electric power.
At present, the running state of the rotating equipment is monitored because the rotating equipment is widely applied to daily life and production, and the real-time understanding of the running state of the rotating equipment is very important. The main structure of the common rotating equipment is a motor stator, a motor rotor and an air gap between the motor stator and the motor rotor, in practical application, the damage of equipment constituent elements or the aging of equipment are unavoidable because the rotating equipment may need to operate for a long time, but if a technician is arranged to conduct periodical fault investigation on the rotating equipment, the rotating equipment is not only subjected to shutdown inspection, but also a certain manpower resource is wasted.
In summary, how to monitor a rotating device in real time to identify a device fault of the rotating device in time has become a technical problem to be solved in the technical field of the rotating device.
Disclosure of Invention
The invention mainly aims to provide a rotating equipment monitoring method, device, equipment and medium based on voiceprint data. The method aims to realize real-time monitoring of the rotating equipment so as to timely identify equipment faults of the rotating equipment.
In order to achieve the above object, the present invention provides a method for monitoring a rotating device based on voiceprint data, the method for monitoring a rotating device based on voiceprint data comprising:
acquiring a first sound signal recorded in real time when the rotating equipment operates, and preprocessing the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal;
inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
and determining the equipment state of the rotating equipment based on the equipment state label.
Optionally, the step of preprocessing the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal includes:
performing Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal;
and denoising and voiceprint recognition processing is carried out on the first sound signal to obtain first voiceprint data in the first sound signal.
Optionally, the method further comprises:
calculating a second signal-to-noise ratio and second voice data corresponding to a second voice signal of the known equipment state label;
inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model;
acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals;
determining a device status tag corresponding to the second sound signal based on the second signal-to-noise ratio, the frequency signal, the amplitude signal, and the harmonic signal;
and adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain the target deep neural network model.
Optionally, the step of determining a device status tag corresponding to the second sound signal based on the second signal-to-noise ratio, the frequency signal, the amplitude signal, and the harmonic signal includes:
if the second signal-to-noise ratio meets a first preset condition, determining that the equipment state label corresponding to the second sound signal is an air gap fault;
If the frequency signal and the harmonic signal meet a second preset condition, determining that the equipment state label corresponding to the second sound signal is a motor stator fault;
and if the amplitude signal and the harmonic signal meet a third preset condition, determining that the equipment state label corresponding to the second sound signal is a motor rotor fault.
Optionally, before the step of determining that the equipment state label corresponding to the second sound signal is an air gap fault if the second signal-to-noise ratio meets the first preset condition, the method further includes:
calculating a difference value between the second signal-to-noise ratio and a standard signal-to-noise ratio, and detecting whether the difference value is larger than a preset difference value, wherein the standard signal-to-noise ratio is the signal-to-noise ratio of a third sound signal recorded in real time under the condition that the equipment state of the rotating equipment is a normal running state;
and when the difference value is larger than the preset difference value and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the second signal-to-noise ratio meets a first preset condition.
Optionally, before the step of determining that the equipment state label corresponding to the second sound signal is a motor stator fault if the frequency signal and the harmonic signal meet a second preset condition, the method further includes:
Detecting whether the frequency signal is in a preset frequency range;
and when the frequency signal is detected not to be in the preset frequency range and the harmonic signal is an aperiodic signal, determining that the frequency signal and the harmonic signal meet a second preset condition.
Optionally, before the step of determining that the equipment state label corresponding to the second sound signal is a motor rotor fault if the amplitude signal and the harmonic signal meet a third preset condition, the method further includes:
detecting whether the amplitude signal is in a preset amplitude range;
and when the amplitude signal is detected to be in a preset amplitude range and the harmonic signal is a periodic signal, determining that the amplitude signal and the harmonic signal meet a third preset condition.
In addition, in order to achieve the above object, the present invention also provides a rotating equipment monitoring device based on voiceprint data, the rotating equipment monitoring device based on voiceprint data comprising:
the preprocessing module is used for acquiring a first sound signal recorded in real time when the rotating equipment operates, preprocessing the first sound signal and acquiring a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal;
The equipment state label module is used for inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target deep neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
and the equipment state module is used for determining the equipment state of the rotating equipment based on the equipment state label.
In addition, to achieve the above object, the present invention also provides a terminal device including: the system comprises a memory, a processor and a voiceprint data-based rotating equipment monitoring program stored in the memory and capable of running on the processor, wherein the voiceprint data-based rotating equipment monitoring program of the terminal equipment realizes the steps of the voiceprint data-based rotating equipment monitoring method when being executed by the processor.
In addition, in order to achieve the above object, the present invention also provides a computer-readable storage medium having stored thereon a voiceprint data based rotating apparatus monitoring program which, when executed by a processor, implements the steps of the voiceprint data based rotating apparatus monitoring method as described above.
The embodiment of the invention provides a rotating equipment monitoring method, device, equipment and medium based on voiceprint data, wherein the method obtains a first signal-to-noise ratio and first voiceprint data corresponding to a first sound signal by acquiring the first sound signal recorded in real time when the rotating equipment operates and preprocessing the first sound signal; inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises motor stator faults, motor rotor faults, air gap faults and normal operation; and determining the equipment state of the rotating equipment based on the equipment state label.
According to the embodiment of the invention, a first sound signal is obtained by recording sound emitted by the rotating equipment in real time when the rotating equipment operates, the first sound signal is preprocessed, the signal-to-noise ratio of the first sound signal and voiceprint data in the first sound signal, namely the first signal-to-noise ratio and the first voiceprint data, are obtained, the first signal-to-noise ratio and the first voiceprint data are input into a pre-trained target deep neural network model, and an equipment state label corresponding to the first sound signal is obtained, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation, and finally, the equipment state of the rotating equipment at the moment is determined based on the obtained equipment state label.
Drawings
FIG. 1 is a schematic device architecture diagram of a hardware operating environment of a terminal device according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating steps of a first embodiment of a method for monitoring a rotating device based on voiceprint data according to the present invention;
FIG. 3 is a schematic diagram of an overall flow chart of monitoring a rotating device according to an embodiment of a method for monitoring a rotating device based on voiceprint data;
fig. 4 is a schematic functional block diagram of an embodiment of a rotating device monitoring apparatus based on voiceprint data according to the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic device structure diagram of a hardware running environment of a terminal device according to an embodiment of the present invention.
The terminal equipment of the embodiment of the invention can be the terminal equipment applied to the technical field of rotating equipment. Specifically, the terminal device may be a smart phone, a PC (PerSonal Computer ), a tablet computer, a portable computer, or the like.
As shown in fig. 1, the terminal device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a DiSplay (diselay), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., wi-Fi interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
It will be appreciated by those skilled in the art that the terminal device structure shown in fig. 1 is not limiting of the terminal device and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a rotating device monitoring program based on voiceprint data may be included in a memory 1005, which is a computer storage medium.
In the terminal shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server and performing data communication with the background server; the user interface 1003 is mainly used for connecting a client and communicating data with the client; and the processor 1001 may be configured to call a voiceprint data based rotating device monitoring program stored in the memory 1005 and perform the following operations:
the method comprises the steps of recording a first sound signal in real time when the rotating equipment operates, and preprocessing the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal;
inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
And determining the equipment state of the rotating equipment based on the equipment state label.
Optionally, the processor 1001 may be further configured to invoke a voiceprint data based rotating device monitoring program stored in the memory 1005 and perform the following operations:
performing Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal;
and denoising and voiceprint recognition processing is carried out on the first sound signal to obtain first voiceprint data in the first sound signal.
Optionally, the processor 1001 may be further configured to invoke a rotating device monitoring program based on voiceprint data stored in the memory 1005, and further perform the following operations:
calculating a second signal-to-noise ratio and second voice data corresponding to a second voice signal of the known equipment state label;
inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model;
acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals;
determining a device status tag corresponding to the second sound signal based on the second signal-to-noise ratio, the frequency signal, the amplitude signal, and the harmonic signal;
And adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain the target deep neural network model.
Optionally, the processor 1001 may be further configured to invoke a voiceprint data based rotating device monitoring program stored in the memory 1005 and perform the following operations:
if the second signal-to-noise ratio meets a first preset condition, determining that the equipment state label corresponding to the second sound signal is an air gap fault;
if the frequency signal and the harmonic signal meet a second preset condition, determining that the equipment state label corresponding to the second sound signal is a motor stator fault;
and if the amplitude signal and the harmonic signal meet a third preset condition, determining that the equipment state label corresponding to the second sound signal is a motor rotor fault.
Optionally, the processor 1001 may be further configured to invoke a rotating device monitoring program based on voiceprint data stored in the memory 1005, and before the step of determining that a device status tag corresponding to the second sound signal is an air gap fault if the second signal-to-noise ratio meets a first preset condition, perform the following operations:
Calculating a difference value between the second signal-to-noise ratio and a standard signal-to-noise ratio, and detecting whether the difference value is larger than a preset difference value, wherein the standard signal-to-noise ratio is the signal-to-noise ratio of a third sound signal recorded in real time under the condition that the equipment state of the rotating equipment is a normal running state;
and when the difference value is larger than the preset difference value and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the second signal-to-noise ratio meets a first preset condition.
Optionally, the processor 1001 may be further configured to invoke a rotating device monitoring program based on voiceprint data stored in the memory 1005, and before the step of determining that a device status tag corresponding to the second sound signal is a motor stator fault if the frequency signal and the harmonic signal meet a second preset condition, further perform the following operations:
detecting whether the frequency signal is in a preset frequency range;
and when the frequency signal is detected not to be in the preset frequency range and the harmonic signal is an aperiodic signal, determining that the frequency signal and the harmonic signal meet a second preset condition.
Optionally, the processor 1001 may be further configured to invoke a rotating device monitoring program based on voiceprint data stored in the memory 1005, and before the step of determining that a device status tag corresponding to the second sound signal is a motor rotor fault if the amplitude signal and the harmonic signal meet a third preset condition, further perform the following operations:
Detecting whether the amplitude signal is in a preset amplitude range;
and when the amplitude signal is detected to be in a preset amplitude range and the harmonic signal is a periodic signal, determining that the amplitude signal and the harmonic signal meet a third preset condition.
Based on the terminal equipment, the embodiments of the rotating equipment monitoring method based on voiceprint data are provided.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for monitoring a rotating device based on voiceprint data according to a first embodiment of the present invention. It should be noted that although a logical sequence is shown in the flowchart, in some cases, the method of monitoring a rotating device based on voiceprint data of the present invention may of course perform the steps shown or described in a different order than that shown. For convenience of description, the following description will be made in such a manner that an execution body is omitted.
In a first embodiment of the method for monitoring a rotating device based on voiceprint data of the present invention, the method for monitoring a rotating device based on voiceprint data of the present invention includes:
step S10, acquiring a first sound signal recorded in real time when the rotating equipment operates, and preprocessing the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal;
In this embodiment, when the rotating device is running, a first sound signal is obtained by recording sound emitted by the rotating device in real time, and the first sound signal is preprocessed to obtain a signal-to-noise ratio (hereinafter referred to as a first signal-to-noise ratio) of the first sound signal and voiceprint data (hereinafter referred to as first voiceprint data) in the first sound signal.
In the method for detecting the rotating equipment based on the voiceprint data, a sound signal acquisition device is arranged in the rotating equipment and is used for following the starting of the rotating equipment when the rotating equipment is started and continuously recording sound signals sent by the rotating equipment in the running process of the rotating equipment.
In an exemplary operation process of the rotating device, a sound signal emitted by the rotating device is continuously recorded through a sound signal collecting device inside the rotating device to obtain a first sound signal, a duration of the first sound signal is set to 10 minutes, that is, the recorded sound signal is obtained once every ten minutes, and then the obtained first sound signal is preprocessed to obtain a signal-to-noise ratio of the first sound signal and voiceprint data in the first sound signal.
It should be noted that, in the method for detecting a rotating device based on voiceprint data according to the present invention, the duration of the first sound signal is set to 10 minutes, but it should be understood that, based on different design requirements of practical applications, the duration of the sound signal may be any duration meeting the practical requirements in different feasible embodiments, and the method for detecting a rotating device based on voiceprint data according to the present invention is not limited to the duration of the sound signal.
Further, in a possible embodiment, in the step S10, the step of "preprocessing the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal" includes:
step S101, carrying out Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal;
in this embodiment, fourier transform is performed on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal.
Illustratively, a fourier transform is performed on the first sound signal based on the power spectrum, and then the power spectral densities of the signal and the noise are calculated, resulting in a signal-to-noise ratio of the first sound signal.
Step S102, performing denoising processing and voiceprint recognition processing on the first sound signal, to obtain first voiceprint data in the first sound signal.
In this embodiment, denoising processing and voiceprint recognition processing are performed on the first sound signal, so as to obtain first voiceprint data in the first sound signal.
The denoising method for the first sound signal may be an average filtering, median filtering and kernel wavelet denoising method, and after obtaining the denoised sound signal, the voiceprint recognition algorithm is used for extracting voiceprint data from the denoised sound signal, where the common voiceprint recognition algorithm is based on methods such as an autocorrelation function, a cross correlation function and a short-time fourier transform of a time domain, methods such as a power spectrum density, a frequency spectrum and a cepstrum of a frequency domain, and methods such as wavelet decomposition and wavelet packet decomposition of a wavelet transform.
Step S20, inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
in this embodiment, the device state label corresponding to the first sound signal is obtained by inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model, where the device state label includes a motor stator fault, a motor rotor fault, an air gap fault, and normal operation.
The first signal-to-noise ratio and the first voiceprint data are input into a pre-trained target depth neural network model, and a device state label corresponding to the first sound signal is output through the target depth neural network model based on the input, wherein the device state label can be a motor stator fault, a motor rotor fault, an air gap fault and normal operation, and the rotating device is indicated to be free of faults when the device state label is in normal operation.
It should be noted that, the stator fault of the motor may be a short circuit, an open circuit, a ground fault, etc. in the stator winding, resulting in abnormal current and increased temperature, thereby causing equipment fault; the rotor of the motor may have unbalanced rotor, worn bearing, overheat bearing and other problems, resulting in unbalanced rotation and increased vibration, thus leading to equipment failure; the air gap fault may be that the air gap between the motor stator and the motor rotor is too large or too small, which results in unstable operation of the device, increased noise and reduced energy efficiency, and thus causes the device to malfunction.
Step S30, determining the device status of the rotating device based on the device status tag.
In this embodiment, the device state of the rotating device is determined based on the device state label output by the target deep neural network model.
For example, after inputting the first sound signal of approximately ten minutes into the pre-trained target deep neural network model, a device status tag representing the rotating device within approximately ten minutes is obtained, and specifically, the device status tag output by the target deep neural network model may be at least one of normal operation or motor stator failure, motor rotor failure, and air gap failure.
As shown in fig. 3, an overall flow chart of monitoring a rotating device is shown, firstly, after the rotating device is started, recording sound emitted by the rotating device in real time to obtain recorded sound signals, then obtaining sound signals of a near preset duration every interval of preset duration, then calculating a signal-to-noise ratio of the sound signals, extracting voiceprint data in the sound signals, taking the signal-to-noise ratio and the voiceprint data as input of a target deep neural network model, obtaining a device state label output by the model, and finally determining a device state of the rotating device within the near preset duration according to the device state label.
In this embodiment, according to the method for monitoring a rotating device based on voiceprint data, a first sound signal is obtained by recording a sound emitted by the rotating device in real time when the rotating device is operated; performing Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal; denoising and voiceprint recognition processing is carried out on the first sound signal, so that first voiceprint data in the first sound signal are obtained; inputting the first signal-to-noise ratio and the first voiceprint data into a pre-trained target depth neural network model to obtain an equipment state label corresponding to a first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation; and determining the device state of the rotating device based on the device state label output by the target deep neural network model.
Thus, according to the embodiment of the invention, through real-time recording of the sound emitted by the rotating equipment when the rotating equipment operates, a first sound signal is obtained, the first sound signal is preprocessed, the signal-to-noise ratio of the first sound signal and the voiceprint data in the first sound signal are obtained, namely, the first signal-to-noise ratio and the first voiceprint data are input into a pre-trained target depth neural network model, and an equipment state label corresponding to the first sound signal is obtained, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation, and finally, the equipment state of the rotating equipment at the moment is determined based on the obtained equipment state label.
Further, based on the first embodiment of the method for monitoring a rotating device based on voiceprint data according to the present invention, a second embodiment of the method for monitoring a rotating device based on voiceprint data according to the present invention is provided.
In this embodiment, the method for monitoring a rotating device based on voiceprint data of the present invention may further include:
step A10, calculating a second signal-to-noise ratio and second voice data corresponding to a second voice signal of the known equipment state label;
in the present embodiment, a signal-to-noise ratio (hereinafter referred to as a second signal-to-noise ratio to show distinction) corresponding to a sound signal (hereinafter referred to as a second sound signal to show distinction) of a known device status tag and voiceprint data (hereinafter referred to as a second voiceprint data to show distinction) are calculated.
Illustratively, various fault conditions are experimentally manufactured, and the occurrence of a single fault and the occurrence of more than one fault are tested respectively, and simultaneously, the sound signal emitted by the rotating device, namely the second sound signal, is recorded under each set of experimental conditions, then fourier transformation is performed on the second sound signal based on a power spectrum, and then the power spectrum densities of the signal and the noise are calculated, so that the signal-to-noise ratio of the second sound signal, namely the second signal-to-noise ratio, is obtained. And denoising the second sound signal, and extracting voiceprint data of the denoised sound signal by using a voiceprint recognition algorithm after obtaining the denoised sound signal to obtain second voiceprint data.
Step A20, inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model;
in this embodiment, the second signal-to-noise ratio and the second voice data are input into the initial deep neural network model constructed in advance.
Step A30, acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals;
in this embodiment, the acoustic features in the second acoustic data are obtained by a feature extraction module of the pre-constructed initial deep neural network model, where the acoustic features include a frequency signal, an amplitude signal, and a harmonic signal.
The voiceprint data includes acoustic features such as a frequency signal, an amplitude signal, and a harmonic signal. Specifically, the frequency signal includes a fundamental frequency, formants, and the like of sound; the amplitude signal refers to the amplitude of sound; harmonic signals refer to the amplitude ratio of different frequency components in sound. Different voiceprint recognition algorithms can extract different acoustic features, and a proper algorithm is specifically required to be selected according to practical application.
Step A40, determining a device state label corresponding to the second sound signal based on the second signal-to-noise ratio, the frequency signal, the amplitude signal and the harmonic signal;
in this embodiment, the device status tag corresponding to the second sound signal is determined based on the signal-to-noise ratio of the second sound signal and the frequency signal, the amplitude signal, and the harmonic signal in the second sound data.
Further, in a possible embodiment, the step a40 includes:
step A401, if the second signal-to-noise ratio meets a first preset condition, determining that a device state label corresponding to the second sound signal is an air gap fault;
step A402, if the frequency signal and the harmonic signal meet a second preset condition, determining that a device state label corresponding to the second sound signal is a motor stator fault;
and step A403, if the amplitude signal and the harmonic signal meet a third preset condition, determining that the equipment state label corresponding to the second sound signal is a motor rotor fault.
In this embodiment, whether the second signal-to-noise ratio satisfies a first preset condition is detected, if the second signal-to-noise ratio satisfies the first preset condition, it is determined that the equipment state label corresponding to the second sound signal is an air gap fault, the frequency signal and the harmonic signal are detected to satisfy the second preset condition, if the frequency signal and the harmonic signal satisfy the second preset condition, it is determined that the equipment state label corresponding to the second sound signal is a motor stator fault, the amplitude signal and the harmonic signal are detected to satisfy a third preset condition, and if the amplitude signal and the harmonic signal satisfy the third preset condition, it is determined that the equipment state label corresponding to the second sound signal is a motor rotor fault.
It should be noted that, when none of the above three faults occurs, it is determined that the device status label of the rotating device is normal operation.
Further, in a possible embodiment, before the step a401, the method may further include:
step B10, calculating a difference value between the second signal-to-noise ratio and a standard signal-to-noise ratio, and detecting whether the difference value is larger than a preset difference value, wherein the standard signal-to-noise ratio is a signal-to-noise ratio of a third sound signal recorded in real time under the condition that the equipment state of the rotating equipment is a normal running state;
in this embodiment, a difference between the second snr and a standard snr is calculated, where the standard snr is a snr of a sound signal (hereinafter referred to as a third sound signal to indicate distinction) sent by the rotating device when the device state of the rotating device is a normal operating state, and then whether the difference is greater than a preset difference is detected.
The signal-to-noise ratio of the sound signal emitted by the rotary device in the normal operating state is calculated in advance, recorded as a standard signal-to-noise ratio, and then the difference between the second signal-to-noise ratio and the standard signal-to-noise ratio is calculated.
And B20, when the difference value is larger than the preset difference value and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the second signal-to-noise ratio meets a first preset condition.
In this embodiment, when the difference is greater than the preset difference and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, it is determined that the second signal-to-noise ratio satisfies the first preset condition.
When the air gap between the motor stator and the motor rotor of the rotary device is too large or too small, the operation of the device is unstable, the noise is increased, the energy efficiency is reduced, and the device is further broken down.
The standard signal-to-noise ratio of the rotating equipment in the normal running state is obtained through experiments in advance, and the acceptable signal-to-noise ratio range is calculated, so that the specific size of the preset difference value is obtained. When the difference between the second signal-to-noise ratio and the standard signal-to-noise ratio exceeds a preset difference, and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the equipment fault label is an air gap fault, specifically, when the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, the noise emitted by the current rotating equipment in the operation state is larger than the noise emitted by the rotating equipment in the normal operation state, and the noise increment is larger than the preset difference, determining that the rotating equipment has a fault and can be positioned as the air gap fault.
It should be noted that, based on different design requirements of practical applications, in different possible embodiments, the preset difference may be any value according to practical requirements, and the present invention is not limited to the magnitude of the preset difference.
Further, in a possible embodiment, before the step a402, the method may further include:
step C10, detecting whether the frequency signal is in a preset frequency range;
in this embodiment, it is detected whether the frequency signal extracted from the second voice data is within a preset frequency range.
It should be noted that, based on different design requirements of practical applications, in different possible embodiments, the preset frequency range may be any value according to practical requirements, and the present invention is not limited to the size of the preset frequency range.
The frequency range of the frequency signal extracted from the voiceprint data in normal operation of the rotating apparatus is calculated in advance as the above-described preset frequency range, and then in actual monitoring, whether the actual frequency signal is within the preset frequency range is detected.
And step C20, when the frequency signal is detected not to be in the preset frequency range and the harmonic signal is a non-periodic signal, determining that the frequency signal and the harmonic signal meet a second preset condition.
In this embodiment, when it is detected that the frequency signal is not within the preset frequency range and the harmonic signal is an aperiodic signal, it is determined that the frequency signal and the harmonic signal satisfy a second preset condition.
It should be noted that, the stator failure generally causes the frequency of the sound to change, and a higher or lower frequency component may occur. At the same time, stator faults can also lead to irregular harmonic components in the sound. Therefore, when it is detected that the actual frequency signal is not within the preset frequency range and that an aperiodic harmonic component occurs, it can be determined that the rotating device may have a motor stator failure.
Further, in a possible embodiment, before the step a403, the method may further include:
step D10, detecting whether the amplitude signal is in a preset amplitude range;
in this embodiment, it is detected whether or not the amplitude signal extracted from the second voice print data is within a preset amplitude range.
It should be noted that, based on different design requirements of practical applications, in different possible embodiments, the preset range may be any value according to practical requirements, and the present invention is not limited to the size of the preset range.
The amplitude range of the amplitude signal extracted from the voiceprint data in normal operation of the rotating apparatus is calculated in advance as the above-described preset amplitude range, and then in actual monitoring, whether the actual amplitude signal is within the preset amplitude range is detected.
And step D20, when the amplitude signal is detected to be in the preset amplitude range and the harmonic signal is a periodic signal, determining that the amplitude signal and the harmonic signal meet a third preset condition.
In this embodiment, when it is detected that the amplitude signal is within the preset amplitude range and the harmonic signal is a periodic signal, it is determined that the amplitude signal and the harmonic signal satisfy a third preset condition.
It should be noted that rotor faults typically cause the amplitude of the sound to change, possibly to be smaller or larger. Meanwhile, the rotor fault also causes periodic harmonic components in the sound, so that when the actual amplitude signal is detected to be not in the preset amplitude range and the periodic harmonic components are generated, the rotary equipment can be determined that the motor rotor fault possibly exists.
And step A50, adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain the target deep neural network model.
In this embodiment, the model parameters of the initial deep neural network model are adjusted based on the device state label corresponding to the second sound signal, so as to obtain the target deep neural network model.
The method includes determining a loss function of the initial deep neural network model, comparing a known device state label with a device state label output by the initial deep neural network model through the loss function to obtain a comparison result, and optimizing model parameters based on the comparison result to obtain the target deep neural network model.
In this embodiment, the method for monitoring a rotating device based on voiceprint data calculates a second signal-to-noise ratio and second voiceprint data corresponding to a second sound signal of a tag with a known device state; inputting the second signal-to-noise ratio and second voice data into a pre-constructed initial deep neural network model; acquiring acoustic features in the second acoustic data through a feature extraction module of a pre-constructed initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals; calculating a difference value between the second signal-to-noise ratio and a standard signal-to-noise ratio, wherein the standard signal-to-noise ratio is the signal-to-noise ratio of a third sound signal sent by the rotating equipment when the equipment state of the rotating equipment is a normal operation state, and then detecting whether the difference value is larger than a preset difference value; when the difference is larger than the preset difference and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the second signal-to-noise ratio meets the first preset condition, and accordingly determining that the equipment state label corresponding to the second sound signal is an air gap fault; detecting whether a frequency signal extracted from second voice data is in a preset frequency range, and determining that the frequency signal and the harmonic signal meet a second preset condition when the frequency signal is detected to be not in the preset frequency range and the harmonic signal is a non-periodic signal, so as to determine that a device state label corresponding to the second voice signal is a motor stator fault; detecting whether an amplitude signal extracted from second voice data is in a preset amplitude range, and determining that the amplitude signal and a harmonic signal meet a third preset condition when the amplitude signal is detected to be in the preset amplitude range and the harmonic signal is a periodic signal, so as to determine that a device state label corresponding to the second voice signal is a motor rotor fault; and adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain the target deep neural network model.
Therefore, the invention builds an initial depth neural network model in advance based on the embodiment of each fault condition on the signal-to-noise ratio and the voiceprint data, and carries out model training on the initial depth neural network model through experimental data to obtain a target depth neural network model, which is used for outputting the equipment state label corresponding to the sound signal according to the signal-to-noise ratio of the input sound signal and the voiceprint data in the sound signal, so that the working state of the rotating equipment can be monitored through the sound sent in the running process of the rotating equipment which is recorded in real time, and further, the equipment fault point can be found in time so as to be convenient for maintenance personnel to maintain the fault point.
In addition, the embodiment of the invention also provides a rotating equipment monitoring device based on the voiceprint data.
Referring to fig. 4, fig. 4 is a schematic functional block diagram of an embodiment of a rotating device monitoring apparatus based on voiceprint data according to the present invention, and as shown in fig. 4, the rotating device monitoring apparatus based on voiceprint data according to the present invention includes:
the preprocessing module 10 is configured to acquire a first sound signal recorded in real time during operation of the rotating device, and preprocess the first sound signal to obtain a first signal-to-noise ratio and first voiceprint data corresponding to the first sound signal;
The device state label module 20 is configured to input the first signal-to-noise ratio and the first voiceprint data into a pre-trained target deep neural network model, and obtain a device state label corresponding to the first sound signal, where the device state label includes a motor stator fault, a motor rotor fault, an air gap fault, and normal operation;
an equipment status module 30 for determining an equipment status of the rotating equipment based on the equipment status tag.
Optionally, the preprocessing module 10 includes:
the first signal-to-noise ratio unit is used for carrying out Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal;
and the first voiceprint data unit is used for carrying out denoising processing and voiceprint recognition processing on the first voice signal to obtain first voiceprint data in the first voice signal.
Optionally, the rotating equipment monitoring device based on voiceprint data of the present invention further comprises:
the second sound signal module is used for calculating a second signal-to-noise ratio and second voice data corresponding to the second sound signal of the known equipment state label;
the initial model module is used for inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model;
The acoustic feature module is used for acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals;
the mapping relation module is used for determining a device state label corresponding to the second sound signal based on the second signal-to-noise ratio, the frequency signal, the amplitude signal and the harmonic signal;
and the target model module is used for adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain the target deep neural network model.
Optionally, the mapping relation module includes:
an air gap fault unit, configured to determine that an equipment state tag corresponding to the second sound signal is an air gap fault if the second signal-to-noise ratio meets a first preset condition;
the motor stator fault unit is used for determining that the equipment state label corresponding to the second sound signal is a motor stator fault if the frequency signal and the harmonic signal meet a second preset condition;
and the motor rotor fault unit is used for determining that the equipment state label corresponding to the second sound signal is a motor rotor fault if the amplitude signal and the harmonic signal meet a third preset condition.
Optionally, the rotating equipment monitoring device based on voiceprint data of the present invention further comprises:
the first detection module is used for calculating the difference value between the second signal-to-noise ratio and the standard signal-to-noise ratio and detecting whether the difference value is larger than a preset difference value, wherein the standard signal-to-noise ratio is the signal-to-noise ratio of a third sound signal recorded in real time under the condition that the equipment state of the rotating equipment is a normal running state;
and the first judging module is used for determining that the second signal-to-noise ratio meets a first preset condition when the difference value is larger than the preset difference value and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio.
Optionally, the rotating equipment monitoring device based on voiceprint data of the present invention further comprises:
the second detection module is used for detecting whether the frequency signal is in a preset frequency range or not;
and the second judging module is used for determining that the frequency signal and the harmonic signal meet a second preset condition when the frequency signal is detected not to be in the preset frequency range and the harmonic signal is an aperiodic signal.
Optionally, the rotating equipment monitoring device based on voiceprint data of the present invention further comprises:
the third detection module is used for detecting whether the amplitude signal is in a preset amplitude range or not;
And the third judging module is used for determining that the amplitude signal and the harmonic signal meet a third preset condition when the amplitude signal is detected to be in the preset amplitude range and the harmonic signal is a periodic signal.
The present invention also provides a computer storage medium, on which a rotating device monitoring program based on voiceprint data is stored, where the rotating device monitoring program based on voiceprint data implements the steps of the rotating device monitoring program method based on voiceprint data according to any one of the embodiments above when the rotating device monitoring program based on voiceprint data is executed by a processor.
The specific embodiment of the computer storage medium of the present invention is substantially the same as the above embodiments of the method for monitoring a rotating device based on voiceprint data according to the present invention, and will not be described herein.
The present invention also provides a computer program product, which comprises a computer program, and the computer program when executed by a processor implements the steps of the method for monitoring a rotating device based on voiceprint data according to any one of the embodiments described above, which is not described herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a TWS headset or the like) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (7)
1. The rotating equipment monitoring method based on the voiceprint data is characterized by comprising the following steps of:
Acquiring a first sound signal recorded by a sound signal acquisition device in real time when the rotating equipment operates, wherein the sound signal acquisition device is arranged inside the rotating equipment;
performing Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal;
denoising and voiceprint recognition processing is carried out on the first sound signal, so that first voiceprint data in the first sound signal are obtained;
calculating a second signal-to-noise ratio and second voice data corresponding to a second voice signal of the known equipment state label;
inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model;
acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals;
if the second signal-to-noise ratio meets a first preset condition, determining that the equipment state label corresponding to the second sound signal is an air gap fault;
if the frequency signal and the harmonic signal meet a second preset condition, determining that the equipment state label corresponding to the second sound signal is a motor stator fault;
If the amplitude signal and the harmonic signal meet a third preset condition, determining that the equipment state label corresponding to the second sound signal is a motor rotor fault;
adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain a target deep neural network model;
inputting the first signal-to-noise ratio and the first voiceprint data into the target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
and determining the equipment state of the rotating equipment based on the equipment state label.
2. The method for monitoring a rotating device based on voiceprint data according to claim 1, wherein before the step of determining that a device status tag corresponding to the second sound signal is an air gap fault if the second signal-to-noise ratio satisfies a first preset condition, the method further comprises:
calculating a difference value between the second signal-to-noise ratio and a standard signal-to-noise ratio, and detecting whether the difference value is larger than a preset difference value, wherein the standard signal-to-noise ratio is the signal-to-noise ratio of a third sound signal recorded in real time under the condition that the equipment state of the rotating equipment is a normal running state;
And when the difference value is larger than the preset difference value and the second signal-to-noise ratio is smaller than the standard signal-to-noise ratio, determining that the second signal-to-noise ratio meets a first preset condition.
3. The method for monitoring a rotating device based on voiceprint data according to claim 1, wherein before the step of determining that a device status tag corresponding to the second sound signal is a motor stator fault if the frequency signal and the harmonic signal satisfy a second preset condition, the method further comprises:
detecting whether the frequency signal is in a preset frequency range;
and when the frequency signal is detected not to be in the preset frequency range and the harmonic signal is an aperiodic signal, determining that the frequency signal and the harmonic signal meet a second preset condition.
4. The method for monitoring a rotating device based on voiceprint data according to claim 1, wherein before the step of determining that a device status tag corresponding to the second sound signal is a motor rotor failure if the amplitude signal and the harmonic signal satisfy a third preset condition, the method further comprises:
detecting whether the amplitude signal is in a preset amplitude range;
And when the amplitude signal is detected to be in a preset amplitude range and the harmonic signal is a periodic signal, determining that the amplitude signal and the harmonic signal meet a third preset condition.
5. A rotating equipment monitoring device based on voiceprint data, the rotating equipment monitoring device based on voiceprint data comprising:
the preprocessing module is used for acquiring a first sound signal recorded by the sound signal acquisition device in real time when the rotating equipment operates, wherein the sound signal acquisition device is arranged in the rotating equipment; performing Fourier transform on the first sound signal to obtain a first signal-to-noise ratio of the first sound signal; denoising and voiceprint recognition processing is carried out on the first sound signal, so that first voiceprint data in the first sound signal are obtained;
the equipment state label module is used for calculating a second signal-to-noise ratio and second voice data corresponding to a second voice signal of the known equipment state label; inputting the second signal-to-noise ratio and the second voice data into a pre-constructed initial deep neural network model; acquiring acoustic features in the second acoustic data through a feature extraction module of the initial deep neural network model, wherein the acoustic features comprise frequency signals, amplitude signals and harmonic signals; if the second signal-to-noise ratio meets a first preset condition, determining that the equipment state label corresponding to the second sound signal is an air gap fault; if the frequency signal and the harmonic signal meet a second preset condition, determining that the equipment state label corresponding to the second sound signal is a motor stator fault; if the amplitude signal and the harmonic signal meet a third preset condition, determining that the equipment state label corresponding to the second sound signal is a motor rotor fault; adjusting model parameters of the initial deep neural network model based on the equipment state label corresponding to the second sound signal to obtain a target deep neural network model; inputting the first signal-to-noise ratio and the first voiceprint data into the target depth neural network model to obtain an equipment state label corresponding to the first sound signal, wherein the equipment state label comprises a motor stator fault, a motor rotor fault, an air gap fault and normal operation;
And the equipment state module is used for determining the equipment state of the rotating equipment based on the equipment state label.
6. A terminal device, characterized in that the terminal device comprises: a memory, a processor and a voiceprint data based rotating device monitoring program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the voiceprint data based rotating device monitoring method of any one of claims 1 to 4.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a voiceprint data based rotating apparatus monitoring program which, when executed by a processor, implements the steps of the voiceprint data based rotating apparatus monitoring method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310799871.3A CN116520143B (en) | 2023-07-03 | 2023-07-03 | Voiceprint data-based rotating equipment monitoring method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310799871.3A CN116520143B (en) | 2023-07-03 | 2023-07-03 | Voiceprint data-based rotating equipment monitoring method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116520143A CN116520143A (en) | 2023-08-01 |
| CN116520143B true CN116520143B (en) | 2023-09-12 |
Family
ID=87392543
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310799871.3A Active CN116520143B (en) | 2023-07-03 | 2023-07-03 | Voiceprint data-based rotating equipment monitoring method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116520143B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3246674A1 (en) * | 2016-05-19 | 2017-11-22 | Monhart Akustik s.r.o. | Method for diagnostic of magnetic circuits in large electric machines |
| CN110910897A (en) * | 2019-12-05 | 2020-03-24 | 四川超影科技有限公司 | Feature extraction method for motor abnormal sound recognition |
| CN112326210A (en) * | 2019-07-17 | 2021-02-05 | 华北电力大学（保定） | Large motor fault diagnosis method combining sound vibration signals with 1D-CNN |
| CN112857767A (en) * | 2021-01-18 | 2021-05-28 | 中国长江三峡集团有限公司 | Hydro-turbo generator set rotor fault acoustic discrimination method based on convolutional neural network |
| CN114371353A (en) * | 2021-12-09 | 2022-04-19 | 国网安徽省电力有限公司怀远县供电公司 | Power equipment abnormity monitoring method and system based on voiceprint recognition |
| CN114636929A (en) * | 2022-03-29 | 2022-06-17 | 安徽理工大学 | Audio characteristic analysis electric monorail crane motor fault prediction system based on MFCC fusion GRU |
| CN115467787A (en) * | 2022-08-23 | 2022-12-13 | 湖北青云优信科技开发有限公司 | Motor state detection system and method based on audio analysis |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7127076B2 (en) * | 2003-03-03 | 2006-10-24 | Phonak Ag | Method for manufacturing acoustical devices and for reducing especially wind disturbances |
| CN109061474A (en) * | 2018-10-15 | 2018-12-21 | 株洲中车时代电气股份有限公司 | A kind of motor bearings trouble-shooter |
| US20200233397A1 (en) * | 2019-01-23 | 2020-07-23 | New York University | System, method and computer-accessible medium for machine condition monitoring |
-
2023
- 2023-07-03 CN CN202310799871.3A patent/CN116520143B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3246674A1 (en) * | 2016-05-19 | 2017-11-22 | Monhart Akustik s.r.o. | Method for diagnostic of magnetic circuits in large electric machines |
| CN112326210A (en) * | 2019-07-17 | 2021-02-05 | 华北电力大学（保定） | Large motor fault diagnosis method combining sound vibration signals with 1D-CNN |
| CN110910897A (en) * | 2019-12-05 | 2020-03-24 | 四川超影科技有限公司 | Feature extraction method for motor abnormal sound recognition |
| CN112857767A (en) * | 2021-01-18 | 2021-05-28 | 中国长江三峡集团有限公司 | Hydro-turbo generator set rotor fault acoustic discrimination method based on convolutional neural network |
| CN114371353A (en) * | 2021-12-09 | 2022-04-19 | 国网安徽省电力有限公司怀远县供电公司 | Power equipment abnormity monitoring method and system based on voiceprint recognition |
| CN114636929A (en) * | 2022-03-29 | 2022-06-17 | 安徽理工大学 | Audio characteristic analysis electric monorail crane motor fault prediction system based on MFCC fusion GRU |
| CN115467787A (en) * | 2022-08-23 | 2022-12-13 | 湖北青云优信科技开发有限公司 | Motor state detection system and method based on audio analysis |
Non-Patent Citations (1)
| Title |
|---|
| 李平川 等.《洗衣机故障维修369例》.人民邮电出版社,1996,第83页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116520143A (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yin et al. | Weak fault feature extraction of rolling bearings based on improved ensemble noise-reconstructed EMD and adaptive threshold denoising | |
| Jiang et al. | Study on Hankel matrix-based SVD and its application in rolling element bearing fault diagnosis | |
| Su et al. | Rolling element bearing faults diagnosis based on optimal Morlet wavelet filter and autocorrelation enhancement | |
| US10520397B2 (en) | Methods and apparatuses for defect diagnosis in a mechanical system | |
| CN110135492B (en) | Equipment fault diagnosis and abnormality detection method and system based on multiple Gaussian models | |
| CN108731921B (en) | Equipment connecting piece fault monitoring method and system | |
| CA2875071A1 (en) | Method and system for testing operational integrity of a drilling rig | |
| Grebenik et al. | Roller element bearing acoustic fault detection using smartphone and consumer microphones comparing with vibration techniques | |
| CN113670434B (en) | Method and device for identifying sound abnormality of substation equipment and computer equipment | |
| Attoui et al. | Novel machinery monitoring strategy based on time–frequency domain similarity measurement with limited labeled data | |
| JP2023513641A (en) | System and method for fault detection based on robust attenuation signal separation | |
| Aburakhia et al. | A hybrid method for condition monitoring and fault diagnosis of rolling bearings with low system delay | |
| US11409873B2 (en) | Detection of cyber machinery attacks | |
| US11506717B1 (en) | System and method for diagnosing stator inter-turn faults in synchronous motors | |
| CN113551765A (en) | Sound spectrum analysis and diagnosis method for equipment fault | |
| CN116520143B (en) | Voiceprint data-based rotating equipment monitoring method, device, equipment and medium | |
| CN114242085A (en) | Fault diagnosis method and device for rotating equipment | |
| CN114639391A (en) | Mechanical failure prompting method and device, electronic equipment and storage medium | |
| Babu et al. | Review on various signal processing techniques for predictive maintenance | |
| CN113505898A (en) | Equipment predictive maintenance method based on AI technology | |
| CN113805105A (en) | Three-phase transformer detection method and system | |
| CN112287775A (en) | Motor fault diagnosis method and device and related components | |
| Thapliyal et al. | Application of DWT and PDD for bearing fault diagnosis using vibration signal | |
| CN116717461B (en) | Intelligent monitoring method and system for operating state of vacuum pump | |
| JP7236572B2 (en) | Abnormal noise inspection method, abnormal noise inspection program, motor with abnormal noise inspection function |
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 |