CN116297883B - Structure identification method, device and system based on knocking sound and terminal equipment - Google Patents
Structure identification method, device and system based on knocking sound and terminal equipment Download PDFInfo
- Publication number
- CN116297883B CN116297883B CN202310524150.1A CN202310524150A CN116297883B CN 116297883 B CN116297883 B CN 116297883B CN 202310524150 A CN202310524150 A CN 202310524150A CN 116297883 B CN116297883 B CN 116297883B
- Authority
- CN
- China
- Prior art keywords
- knocking
- sample
- frequency
- sound
- training
- 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 37
- 238000012549 training Methods 0.000 claims abstract description 148
- 238000001228 spectrum Methods 0.000 claims abstract description 96
- 230000009466 transformation Effects 0.000 claims abstract description 20
- 238000003066 decision tree Methods 0.000 claims abstract description 16
- 238000010079 rubber tapping Methods 0.000 claims description 30
- 238000004590 computer program Methods 0.000 claims description 22
- 230000002950 deficient Effects 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 15
- 230000003595 spectral effect Effects 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000013075 data extraction Methods 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000012916 structural analysis Methods 0.000 abstract description 3
- 230000008901 benefit Effects 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 description 14
- 230000006872 improvement Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000036541 health Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000036449 good health Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/045—Analysing solids by imparting shocks to the workpiece and detecting the vibrations or the acoustic waves caused by the shocks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/12—Analysing solids by measuring frequency or resonance of acoustic waves
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The application discloses a structure identification method, device, system and terminal equipment based on knocking sound, which are characterized in that after knocking sound data of a sample to be tested are obtained, fourier transformation is firstly carried out on the knocking sound data to obtain knocking sound frequency spectrum, a frequency spectrum amplitude sequence corresponding to the knocking characteristic frequency is extracted from the knocking sound frequency spectrum according to preset knocking characteristic frequency, the frequency spectrum amplitude sequence is input into a decision tree-based knocking characteristic identification model which is trained in advance through a training sample with a marked structure, the structure identification result of the sample to be tested is obtained, and the structure identification based on the knocking sound is completed. Compared with the prior art, all audios are used as data of structural analysis, the method and the device have the advantages that based on the knocking frequency spectrum of the sample to be detected after the knocking frequency Fourier transformation, the frequency spectrum amplitude sequence corresponding to the characteristic frequency in the knocking frequency spectrum is used as the input of the model, so that the input data quantity of model identification can be reduced, and further the efficiency of the structural identification of the sample to be detected can be improved.
Description
Technical Field
The application relates to the field of rapid and intelligent detection of structural health conditions, in particular to a structural identification method, device and system based on knocking sound and terminal equipment.
Background
The structure health condition detection of the material based on acoustics is a modern common technology, and in the prior art, the structure detection of the material to be detected is carried out by analyzing all audio samples generated by knocking the sample to be detected, the structure of the sample to be detected is judged to be complete or defective, and the nondestructive detection of the sample to be detected is realized. However, the data size of all audio samples generated by knocking the sample to be detected is large, so that the existing acoustic nondestructive detection technology has low structure detection efficiency of the sample to be detected in practical application.
Health conditions based on a knock detection structure are a common detection means, but are generally used as an auxiliary detection means for diagnosis even through human ear hearing. The existing structural health detection method based on the knocking frequency spectrum has 2 problems. First, the acquisition of the tapping sounds is disturbed by the external environment, resulting in an incoherent spectrum in the spectrum. Second, the structure is ignored as a multi-degree-of-freedom system, and the characteristic frequency is not single, and the natural vibration frequency of the structure is not recognized as much as possible.
Therefore, a structural identification strategy based on the knock is needed to solve the problems of low detection efficiency and low accuracy of the current acoustic nondestructive detection technology.
Disclosure of Invention
The embodiment of the application provides a structure identification method, device and system based on knocking sound and terminal equipment, so as to improve the detection efficiency of an acoustic nondestructive detection technology.
In order to solve the above-mentioned problems, an embodiment of the present application provides a structure recognition method based on a knock, including:
acquiring knocking sound data of a sample to be detected;
performing Fourier transformation on the knocking sound data to obtain a knocking sound frequency spectrum of the sample to be detected;
according to a preset knocking characteristic frequency, extracting a spectrum amplitude corresponding to the knocking characteristic frequency value from the knocking frequency spectrum to obtain a spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of training samples of each marked structure;
inputting the frequency spectrum amplitude sequence into a knocking characteristic recognition model to obtain a structure recognition result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training.
As an improvement of the above solution, the step of obtaining the preset tapping characteristic frequency includes:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
As an improvement of the scheme, the training method of the knocking characteristic recognition model comprises the following steps:
acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure;
extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency;
constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one;
inputting the amplitude corresponding to the knocking characteristic frequency of each training sample in the training sample database and the class of the marked structure corresponding to each training sample into the decision tree model for training to obtain a knocking characteristic recognition model.
As an improvement of the above-described aspect, the structure recognition result includes: a complete structure identification result and a defective structure identification structure.
As an improvement of the above solution, the acquiring tapping sound data of the sample to be measured includes:
responding to the starting operation of a user, generating starting data, and transmitting the starting data to electromagnet equipment and radio equipment so that the electromagnet equipment performs knocking control on the sample to be detected according to the starting data; the sound receiving device collects sound generated by the sample to be detected after the electromagnet device knocks the sample to be detected, and knocking sound data of the sample to be detected are obtained;
and receiving the knocking sound data collected by the sound receiving equipment to obtain the knocking sound data of the sample to be detected.
Correspondingly, an embodiment of the present application further provides a structure recognition device based on the knock, including: the system comprises a data acquisition module, a data conversion module, a data extraction module and a model identification module;
the data acquisition module is used for acquiring knocking sound data of the sample to be detected;
the data conversion module is used for carrying out Fourier transformation on the knocking sound data to obtain a knocking sound frequency spectrum of the sample to be detected;
the data extraction module is used for extracting a frequency spectrum amplitude corresponding to the knocking characteristic frequency value from the knocking sound frequency spectrum according to a preset knocking characteristic frequency to obtain a frequency spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of training samples of each marked structure;
the model identification module is used for inputting the frequency spectrum amplitude sequence into a knocking characteristic identification model to obtain a structure identification result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training.
As an improvement of the above solution, the step of obtaining the preset tapping characteristic frequency includes:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
As an improvement of the scheme, the training method of the knocking characteristic recognition model comprises the following steps:
acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure;
extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency;
constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one;
inputting the amplitude corresponding to the knocking characteristic frequency of each training sample in the training sample database and the class of the marked structure corresponding to each training sample into the decision tree model for training to obtain a knocking characteristic recognition model.
As an improvement of the above-described aspect, the structure recognition result includes: a complete structure identification result and a defective structure identification structure.
As an improvement of the above solution, the data acquisition module includes: a response unit and a receiving unit;
the response unit is used for responding to the starting operation of a user, generating starting data, and transmitting the starting data to the electromagnet equipment and the radio equipment so that the electromagnet equipment can perform knocking control on the sample to be detected according to the starting data; the sound receiving device collects sound generated by the sample to be detected after the electromagnet device knocks the sample to be detected, and knocking sound data of the sample to be detected are obtained;
the receiving unit is used for receiving the knocking sound data collected by the radio equipment and obtaining the knocking sound data of the sample to be detected.
Accordingly, an embodiment of the present application further provides a structure recognition system based on a knock, including: the device comprises a sample to be tested, electromagnet equipment, radio equipment, a shell and a control box; the sample to be tested is connected with the electromagnet device, the electromagnet device is connected with the control box, the radio receiving device is connected with the control box, and the sample to be tested, the electromagnet device, the radio receiving device and the control box are positioned in the shell; the control box applies the structural recognition method based on the knocking sound according to the application.
As an improvement of the above-mentioned scheme, the housing is composed of a sound absorbing device and a sound insulating device; and the electromagnet equipment is connected with the sample to be tested through a knocking hammer.
Accordingly, an embodiment of the present application also provides a computer terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, where the processor implements a structure recognition method based on a click sound according to the present application when the computer program is executed.
Accordingly, an embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium includes a stored computer program, and when the computer program runs, controls a device where the computer readable storage medium is located to execute a structure recognition method based on a clicking sound according to the present application.
From the above, the application has the following beneficial effects:
after knocking sound data of a sample to be detected are obtained, fourier transformation is carried out on the knocking sound data to obtain knocking sound frequency spectrum, a frequency spectrum amplitude sequence corresponding to the knocking characteristic frequency is extracted from the knocking sound frequency spectrum according to preset knocking characteristic frequency, the frequency spectrum amplitude sequence is input into a knocking characteristic recognition model which is trained in advance through a training sample with a marked structure, a structure recognition result of the sample to be detected is obtained, and structure recognition based on the knocking sound is completed. Compared with the prior art, all audios are used as data of structural analysis, the method is based on the knocking sound frequency spectrum of the sample to be detected after the knocking sound Fourier transformation, and the frequency spectrum amplitude sequence corresponding to the characteristic frequency in the knocking sound frequency spectrum is used as the input of the model, so that the input data quantity of model identification can be reduced, the excessive fitting risk during intelligent identification model training is reduced, and further the efficiency of the structural identification of the sample to be detected is improved.
Furthermore, the structure recognition system based on the knocking sound can collect knocking audio in a closed and sound-absorbing environment through the arrangement of the suction device and the sound-insulating device, so that noise pollution of the external environment is prevented. The structure identification of the application can be carried out in different environments, which is beneficial to the wide popularization of the products related to the application.
Drawings
FIG. 1 is a flow chart of a method for identifying a structure based on a knock according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a structure recognition device based on a knock according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a structural identification system based on a knock according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a structural identification system based on a knock according to another embodiment of the present application;
FIG. 5 is an example of a spectrum sample provided by an embodiment of the present application;
fig. 6 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1
Referring to fig. 1, fig. 1 is a flow chart of a structure recognition method based on a knock according to an embodiment of the present application, as shown in fig. 1, the embodiment includes steps 101 to 104, and the steps are specifically as follows:
step 101: and acquiring knocking sound data of the sample to be detected.
In this embodiment, the acquiring tapping sound data of the sample to be measured includes:
responding to the starting operation of a user, generating starting data, and transmitting the starting data to electromagnet equipment and radio equipment so that the electromagnet equipment performs knocking control on the sample to be detected according to the starting data; the sound receiving device collects sound generated by the sample to be detected after the electromagnet device knocks the sample to be detected, and knocking sound data of the sample to be detected are obtained;
and receiving the knocking sound data collected by the sound receiving equipment to obtain the knocking sound data of the sample to be detected.
Step 102: and carrying out Fourier transformation on the knocking sound data to obtain a knocking sound frequency spectrum of the sample to be detected.
In this embodiment, fourier transforming tapping sound data is a common technique, and the present application aims to analyze and process the data based on fourier transformed data.
Step 103: according to a preset knocking characteristic frequency, extracting a spectrum amplitude corresponding to the knocking characteristic frequency from the knocking sound spectrum to obtain a spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of the training samples of each marked structure.
In this embodiment, the step of obtaining the preset tapping characteristic frequency includes:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
In a specific embodiment, the predetermined number may be 50 and can be self-adjusted by one skilled in the art.
In a specific embodiment, the tapping characteristic frequency comprises a natural frequency of vibration of the sample structure.
Step 104: inputting the frequency spectrum amplitude sequence into a knocking characteristic recognition model to obtain a structure recognition result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training.
In this embodiment, the method for training the knock feature recognition model includes:
acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure;
extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency;
constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one;
inputting the amplitude corresponding to the knocking characteristic frequency of each training sample in the training sample database and the class of the marked structure corresponding to each training sample into the decision tree model for training to obtain a knocking characteristic recognition model.
In this embodiment, the structure identification result includes: a complete structure identification result and a defective structure identification structure.
In a specific embodiment, the training sample data can be seen in fig. 5, specifically: the left side of fig. 5 is the defective structure and the right side is the complete structure for the training sample data in fig. 5 above: the susceptibility frequency is 100, 200, 400 and … … for the first and 0, 500, 600, 1200, 2000, 2500, 2800, 3200 and 3800 and … … for the second, and the tapping characteristic frequency is 100, 200, 400, … …, 1200 and 3800 and … … for the second, and the corresponding capability of the tapping characteristic frequency is recorded after summarizing. In the latter data analysis (training), the read frequency is required for each sample: 0. 100, 200, 400, … …, 500, 600, 1200, 2000, 2500, 2800, 3200, 3800, … … and a spectral amplitude corresponding to each frequency, while including whether the structure of the sample is defective or complete data.
In this embodiment, the preprocessing the tapping sound data of each training sample to obtain tapping susceptibility data includes:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample;
and extracting spectrum data corresponding to a preset number of peak points from the knocking frequency spectrum of each training sample, and taking the spectrum data as knocking susceptibility data of each training sample.
In a specific embodiment, model training is divided into a number of stages: manufacturing an indoor model, testing and collecting samples, training and verifying the model and using the model; the use process comprises the following steps: and acquiring data on site, importing the data into a model, and obtaining a result.
In a specific embodiment, the steps corresponding to the structure identification method in this embodiment are specifically: firstly, collecting knocking sounds of a structure with good health conditions and collecting knocking sounds of a defect structure; performing Fourier transform on the knocking sound sample to obtain a knocking sound frequency spectrum; the peak corresponding frequencies of the spectrum of all sample tapping sounds are recorded as the tapping susceptibility frequencies of the taps. Each knocking sound segment, and the knocking susceptibility frequency takes 50 frequency values; pooling the tapping susceptibility frequencies of all the tapping sound samples, referred to as the tapping characteristic frequency of the structure; establishing a decision tree training sample library, wherein the sample library comprises the frequency spectrum amplitude corresponding to the knocking susceptibility frequency of each knocking sound and the health condition of the knocking part; training a decision tree, and inputting the knocking susceptibility frequency of the knocking sound to be detected into a decision tree model according to the energy value to obtain a defective identification result.
Accordingly, an embodiment of the present application further provides a structure recognition system based on a knock, including: a sample 301 to be tested, an electromagnet device 302, a radio receiving device 303, a shell 304 and a control box 305; the sample 301 to be tested is connected with the electromagnet device 302, the electromagnet device 302 is connected with the control box 305, the radio receiving device 303 is connected with the control box 305, and the sample 301 to be tested, the electromagnet device 302, the radio receiving device 303 and the control box 305 are positioned in the shell 304; the control box 305 applies the structure recognition method based on the knock according to the present application.
As an improvement of the above-mentioned scheme, the housing is composed of a sound absorbing device and a sound insulating device; and the electromagnet equipment is connected with the sample to be tested through a knocking hammer.
In a specific embodiment, please refer to fig. 4, which includes: the device comprises an electromagnet knocking mechanism 401, radio equipment 402, a control box 403, a shell 404 and an object 405 to be detected; the electromagnet knocking mechanism is used for controlling knocking actions of the knocking hammer; the shell adopts a sound absorbing device to prevent the sound of the knocks from generating echoes; the radio equipment is close to the knocking surface, the position of the radio equipment can be finely adjusted, and the radio equipment is used for diagnosing the nuances of the knocking sounds; the control box is applied to the structure identification method based on the knocking sound, and can realize the functions of knocking control, collecting sound data interception, displaying, fourier integral transformation and machine learning identification.
After the knocking sound data of the sample to be detected is obtained, fourier transformation is firstly carried out on the knocking sound data to obtain a knocking sound frequency spectrum, a frequency spectrum amplitude sequence corresponding to the knocking characteristic frequency is extracted from the knocking sound frequency spectrum according to the preset knocking characteristic frequency, the frequency spectrum amplitude sequence is input into a knocking characteristic recognition model which is trained in advance through a training sample with a marked structure, a structure recognition result of the sample to be detected is obtained, and structure recognition based on knocking sound is completed. According to the embodiment, the peak value under the individual frequency of the knocking sound is extracted instead of the whole piece of data, so that the data volume is greatly reduced, the main characteristics of the original data are found out, and the utilization efficiency of the original data is improved.
Example two
Referring to fig. 2, fig. 2 is a schematic structural diagram of a structure recognition device based on a striking sound according to an embodiment of the present application, including: a data acquisition module 201, a data conversion module 202, a data extraction module 203 and a model identification module 204;
the data acquisition module 201 is configured to acquire tapping sound data of a sample to be tested;
the data conversion module 202 is configured to perform fourier transform on the tapping sound data to obtain a tapping sound spectrum of the sample to be tested;
the data extraction module 203 is configured to extract, according to a preset tapping characteristic frequency, a spectrum amplitude corresponding to the tapping characteristic frequency value from the tapping sound spectrum, so as to obtain a spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of training samples of each marked structure;
the model identification module 204 is configured to input the spectrum amplitude sequence into a knocking feature identification model to obtain a structure identification result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training.
As an improvement of the above solution, the step of obtaining the preset tapping characteristic frequency includes:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
As an improvement of the scheme, the training method of the knocking characteristic recognition model comprises the following steps:
acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure;
extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency;
constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one;
inputting the amplitude corresponding to the knocking characteristic frequency of each training sample in the training sample database and the class of the marked structure corresponding to each training sample into the decision tree model for training to obtain a knocking characteristic recognition model.
As an improvement of the above-described aspect, the structure recognition result includes: a complete structure identification result and a defective structure identification structure.
As an improvement of the above, the data acquisition module 201 includes: a response unit and a receiving unit;
the response unit is used for responding to the starting operation of a user, generating starting data, and transmitting the starting data to the electromagnet equipment and the radio equipment so that the electromagnet equipment can perform knocking control on the sample to be detected according to the starting data; the sound receiving device collects sound generated by the sample to be detected after the electromagnet device knocks the sample to be detected, and knocking sound data of the sample to be detected are obtained;
the receiving unit is used for receiving the knocking sound data collected by the radio equipment and obtaining the knocking sound data of the sample to be detected.
According to the embodiment, knocking sound data of a sample to be detected are obtained through the data obtaining module, fourier transformation is carried out on the knocking sound data through the data conversion module, knocking sound frequency spectrum of the sample to be detected is obtained, the knocking sound frequency spectrum is input into the data extraction module, corresponding frequency spectrum amplitude is extracted according to preset knocking characteristic frequency, a frequency spectrum amplitude sequence is generated, the frequency spectrum amplitude sequence is input into a knocking characteristic identification model in the model identification module for structural identification, a structural identification result of the sample to be detected is obtained, and structural identification of the sample to be detected is completed. Compared with the prior art, all audios are used as data of structural analysis, the method and the device have the advantages that the peak value in the knocking sound frequency spectrum is used as the input of the model based on the knocking sound frequency spectrum after Fourier transformation of the knocking sound of the sample to be detected, so that the input data amount of model identification can be reduced, and further the efficiency of structural identification of the sample to be detected can be improved.
Example III
Referring to fig. 6, fig. 6 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
A terminal device of this embodiment includes: a processor 601, a memory 602 and a computer program stored in said memory 602 and executable on said processor 601. The processor 601, when executing the computer program, implements the steps of the respective structure identification method based on the knocks described above in the embodiment, for example, all the steps of the structure identification method based on the knocks shown in fig. 1. Alternatively, the processor may implement functions of each module in the above-described device embodiments when executing the computer program, for example: all modules of the structure recognition device based on the striking sound shown in fig. 2.
In addition, an embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium includes a stored computer program, where when the computer program runs, the device where the computer readable storage medium is located is controlled to execute the structure identification method based on the clicking sound according to any one of the embodiments.
It will be appreciated by those skilled in the art that the schematic diagram is merely an example of a terminal device and does not constitute a limitation of the terminal device, and may include more or less components than illustrated, or may combine certain components, or different components, e.g., the terminal device may further include an input-output device, a network access device, a bus, etc.
The processor 601 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, and the processor 601 is a control center of the terminal device, and connects various parts of the entire terminal device using various interfaces and lines.
The memory 602 may be used to store the computer programs and/or modules, and the processor 601 may implement various functions of the terminal device by executing or executing the computer programs and/or modules stored in the memory, and invoking data stored in the memory 602. The memory 602 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data (such as audio data, phonebook, etc.) created according to the use of the handset, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
Wherein the terminal device integrated modules/units may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as stand alone products. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth.
It should be noted that the above-described apparatus embodiments are merely illustrative, and the units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the embodiment of the device provided by the application, the connection relation between the modules represents that the modules have communication connection, and can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art will understand and implement the present application without undue burden.
While the foregoing is directed to the preferred embodiments of the present application, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the application, such changes and modifications are also intended to be within the scope of the application.
Claims (8)
1. A structure recognition method based on a click sound, comprising:
acquiring knocking sound data of a sample to be detected;
performing Fourier transformation on the knocking sound data to obtain a knocking sound frequency spectrum of the sample to be detected;
according to a preset knocking characteristic frequency, extracting a spectrum amplitude corresponding to the knocking characteristic frequency from the knocking sound spectrum to obtain a spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of training samples of each marked structure;
inputting the frequency spectrum amplitude sequence into a knocking characteristic recognition model to obtain a structure recognition result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training; the training method of the knocking characteristic recognition model comprises the following steps: acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure; extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency; constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one; inputting the corresponding amplitude of the knocking characteristic frequency of each training sample in the training sample database and the corresponding class of the marked structure of each training sample into the decision tree model for training to obtain a knocking characteristic recognition model;
the step of obtaining the preset knocking characteristic frequency comprises the following steps:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
2. The structure recognition method based on the click sound according to claim 1, wherein the structure recognition result includes: a complete structure identification result and a defective structure identification structure.
3. The method for identifying a structure based on a click sound according to claim 2, wherein the acquiring click sound data of the sample to be tested comprises:
responding to the starting operation of a user, generating starting data, and transmitting the starting data to electromagnet equipment and radio equipment so that the electromagnet equipment performs knocking control on the sample to be detected according to the starting data; the sound receiving device collects sound generated by the sample to be detected after the electromagnet device knocks the sample to be detected, and knocking sound data of the sample to be detected are obtained;
and receiving the knocking sound data collected by the sound receiving equipment to obtain the knocking sound data of the sample to be detected.
4. A structure recognition apparatus based on a click sound, comprising: the system comprises a data acquisition module, a data conversion module, a data extraction module and a model identification module;
the data acquisition module is used for acquiring knocking sound data of the sample to be detected;
the data conversion module is used for carrying out Fourier transformation on the knocking sound data to obtain a knocking sound frequency spectrum of the sample to be detected;
the data extraction module is used for extracting a frequency spectrum amplitude corresponding to the knocking characteristic frequency from the knocking sound frequency spectrum according to a preset knocking characteristic frequency to obtain a frequency spectrum amplitude sequence of the sample to be detected; the preset knocking characteristic frequency comprises frequencies corresponding to a preset number of peak points of training samples of each marked structure;
the model identification module is used for inputting the frequency spectrum amplitude sequence into a knocking characteristic identification model to obtain a structure identification result of the sample to be detected; the knocking characteristic recognition model is obtained by inputting the corresponding magnitudes of the knocking characteristic frequencies of a plurality of training samples and the categories of marked structures into a decision tree model for training; the training method of the knocking characteristic recognition model comprises the following steps: acquiring knocking frequency spectrums of a plurality of training samples; wherein the classes of the noted structures include: complete structure and defective structure; extracting and obtaining a corresponding amplitude value of the knocking characteristic frequency from the knocking sound frequency spectrum of each training sample according to the knocking characteristic frequency; constructing and obtaining a training sample database according to the amplitude corresponding to the knocking characteristic frequency of each training sample and the class of the marked structure corresponding to each training sample; the corresponding amplitude of the knocking characteristic frequency of each training sample corresponds to the type of the marked structure corresponding to each training sample one by one; inputting the corresponding amplitude of the knocking characteristic frequency of each training sample in the training sample database and the corresponding class of the marked structure of each training sample into the decision tree model for training to obtain a knocking characteristic recognition model;
the step of obtaining the preset knocking characteristic frequency comprises the following steps:
performing Fourier transformation on the knocking sound data of each training sample to obtain a knocking sound frequency spectrum of each training sample; wherein the training sample and the sample to be tested are the same type;
extracting and obtaining spectrum data corresponding to all wave crest points in the knocking frequency spectrum of each training sample; wherein the spectral data comprises: tapping frequency and spectral amplitude;
selecting the knocking frequencies corresponding to the preset number of wave peak points from the knocking frequencies corresponding to all wave peak points according to the sequence of the magnitude from the wave peak point corresponding to the minimum value of the frequency, and taking the knocking frequencies as knocking susceptibility frequencies of each training sample;
summarizing different knocking susceptibility frequencies of all the training samples, and sequencing the training samples according to the frequency values from small to large to obtain preset knocking characteristic frequencies.
5. A structural identification system based on a click sound, comprising: the device comprises a sample to be tested, electromagnet equipment, radio equipment, a shell and a control box; the sample to be tested is connected with the electromagnet device, the electromagnet device is connected with the control box, the radio receiving device is connected with the control box, and the sample to be tested, the electromagnet device, the radio receiving device and the control box are positioned in the shell; the control box applies the structure recognition method based on the knock as defined in any one of claims 1 to 3.
6. The structural identification system based on striking sounds according to claim 5, wherein said casing is composed of sound absorbing means and sound insulating means; and the electromagnet equipment is connected with the sample to be tested through a knocking hammer.
7. A computer terminal device comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing a structure identification method based on a click sound as claimed in any one of claims 1 to 3 when the computer program is executed.
8. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored computer program, wherein the computer program, when run, controls a device in which the computer readable storage medium is located to perform a structure recognition method based on a click sound as claimed in any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310524150.1A CN116297883B (en) | 2023-05-11 | 2023-05-11 | Structure identification method, device and system based on knocking sound and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310524150.1A CN116297883B (en) | 2023-05-11 | 2023-05-11 | Structure identification method, device and system based on knocking sound and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116297883A CN116297883A (en) | 2023-06-23 |
| CN116297883B true CN116297883B (en) | 2023-08-25 |
Family
ID=86796176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310524150.1A Active CN116297883B (en) | 2023-05-11 | 2023-05-11 | Structure identification method, device and system based on knocking sound and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116297883B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102999279A (en) * | 2012-11-29 | 2013-03-27 | 四川长虹电器股份有限公司 | Method for controlling intelligent device through knocks |
| CN103776903A (en) * | 2014-01-15 | 2014-05-07 | 南京航空航天大学 | Wind power blade delamination detection method and system |
| CN114056381A (en) * | 2021-11-24 | 2022-02-18 | 西南交通大学 | Railway vehicle wheel flat scar monitoring method |
| CN114354187A (en) * | 2022-01-05 | 2022-04-15 | 上海交通大学 | Gear fault classification detection method and system based on identification of meshing stiffness |
| CN115293192A (en) * | 2022-06-21 | 2022-11-04 | 华中科技大学 | Rotating machinery fault judging method, computer equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220291175A1 (en) * | 2020-08-28 | 2022-09-15 | Industrial Technology Research Institute | Acoustic resonance diagnostic method for detecting structural degradation and system applying the same |
-
2023
- 2023-05-11 CN CN202310524150.1A patent/CN116297883B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102999279A (en) * | 2012-11-29 | 2013-03-27 | 四川长虹电器股份有限公司 | Method for controlling intelligent device through knocks |
| CN103776903A (en) * | 2014-01-15 | 2014-05-07 | 南京航空航天大学 | Wind power blade delamination detection method and system |
| CN114056381A (en) * | 2021-11-24 | 2022-02-18 | 西南交通大学 | Railway vehicle wheel flat scar monitoring method |
| CN114354187A (en) * | 2022-01-05 | 2022-04-15 | 上海交通大学 | Gear fault classification detection method and system based on identification of meshing stiffness |
| CN115293192A (en) * | 2022-06-21 | 2022-11-04 | 华中科技大学 | Rotating machinery fault judging method, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116297883A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109831733B (en) | Method, device and equipment for testing audio playing performance and storage medium | |
| WO2020037555A1 (en) | Method, device, apparatus, and system for evaluating microphone array consistency | |
| CN110444202B (en) | Composite voice recognition method, device, equipment and computer readable storage medium | |
| CN110503960B (en) | Real-time uploading method, device and equipment of voice recognition result and storage medium | |
| CN110797031A (en) | Voice change detection method, system, mobile terminal and storage medium | |
| CN110691314A (en) | Linear microphone array performance test method and clamp | |
| CN108304322B (en) | Pressure testing method and terminal equipment | |
| CN110689885B (en) | Machine synthesized voice recognition method, device, storage medium and electronic equipment | |
| US8108166B2 (en) | Analysis of chirp frequency response using arbitrary resampling filters | |
| CN116297883B (en) | Structure identification method, device and system based on knocking sound and terminal equipment | |
| CN110322587B (en) | Evaluation recording method, device and equipment in driving process and storage medium | |
| CN116705059B (en) | Audio semi-supervised automatic clustering method, device, equipment and medium | |
| US11297430B2 (en) | Terminal device, apparatus and method for simulating the characteristic of an audio processor, and computer-readable storage medium | |
| CN115985331B (en) | Audio automatic analysis method for field observation | |
| CN110070891B (en) | Song identification method and device and storage medium | |
| CN110096612A (en) | The acquisition methods and system of the online audio analysis data of voice log | |
| CN114863943B (en) | Self-adaptive positioning method and device for environmental noise source based on beam forming | |
| CN112687274A (en) | Voice information processing method, device, equipment and medium | |
| CN112233693B (en) | Sound quality evaluation method, device and equipment | |
| CN111292748B (en) | Voice input system adaptable to multiple frequencies | |
| CN111885474A (en) | Microphone testing method and device | |
| CN112926724A (en) | Grading method and device for yield of injection molding product and electronic equipment | |
| CN112951199B (en) | Audio data generation method and device, data set construction method, medium and equipment | |
| CN116861917A (en) | Dialog intention recognition method and device, storage medium and electronic equipment | |
| CN116665660A (en) | Vehicle-mounted audio identification method, device, medium and equipment |
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 |