CN109738512A - Nondestructive detection system and method based on multiple physical field fusion - Google Patents
Nondestructive detection system and method based on multiple physical field fusion Download PDFInfo
- Publication number
- CN109738512A CN109738512A CN201910015763.6A CN201910015763A CN109738512A CN 109738512 A CN109738512 A CN 109738512A CN 201910015763 A CN201910015763 A CN 201910015763A CN 109738512 A CN109738512 A CN 109738512A
- Authority
- CN
- China
- Prior art keywords
- signal
- multiple physical
- detection system
- metal component
- physical field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The present invention provides a kind of nondestructive detection system and method based on the fusion of " electromagnetism-heat-sound " multiple physical field, belongs to field of non destructive testing.Detection system includes signal generator, power amplifier, thermal infrared imager, excitation coil, electromagnetic sensor, metal component to be measured, sonic transducer and signal processing unit.Electromagnetic sensor, acoustic matrix sensor is arranged on the basis of infrared thermal imaging detection system in the present invention, and composition " electromagnetism-heat-sound " merges detection system.The more multi-modal signals based on fusion can excavate more damage characteristics, and the input of multiple physical field signal is passed through to the deep learning network model of learning training, carry out more comprehensive accurate assessment to metal component damage.Detection system in the present invention is improved on the basis of infra-red thermal imaging system, has good feasibility, easy to operate.Detection method strong interference immunity, detection accuracy in the present invention is high, more comprehensive intelligent can detect to metal component damage.
Description
Technical field
The present invention relates to the technical field of nondestructive testing of metal component, in particular to the lossless inspection based on multiple physical field fusion
Examining system and method.
Background technique
Non-destructive testing can carry out status assessment to it under the premise of not damaging component to be measured, be to ensure equipment safety fortune
Capable key technology is widely used in industrial circle.Metal component in production and life using fairly common, to metal
The damage and failure of component carry out accurate effective detection, are of great significance to life and property security of the country and people.?
In terms of metal component non-destructive testing, conventional non-destructive testing technology (such as Magnetic testing, Liquid penetrant testing, ultrasound detection, ray detection
Deng) low efficiency and metal component surface defect can only be detected mostly, generally detected when being related to internal injury it is cumbersome,
Or it is difficult to carry out internal detection.
Non-contacting detection may be implemented in electromagnetic nondestructive (such as vortex, microwave ultraviolet lamp), and speed is fast, adaptability
By force, when but being related to metal component, it is limited to skin effect, cracks of metal surface can only be detected.On electromagnetic nondestructive basis
On the electromagnetic induction thermal imaging detection technique that develops the advantages of having merged EDDY CURRENT and infrared thermal imaging detection technique, be
One kind is high-efficient, detection range is big, high sensitivity, intuitive detection method, but is related to metal component and still is limited to skin
Effect, not high to deep zone defect detection sensitivity, vulnerable to the influence of component surface state, anti-interference is poor.
According to thermoacoustic theory, during Joule heating, alternation heat source will motivate acoustical signal.Component inside institutional framework
Variation (such as damage) Acoustic Wave Propagation Characteristics (such as acoustic attenuation coefficient, the velocity of sound, bandwidth) can be had an impact.Pass through analysis sound
The available component inside faulted condition of the propagation characteristic of signal, but since acoustical signal is faint, it is based purely on contactless sound field
Detection interference it is big, damage reason location is difficult.Electromagnetic induction thermal imaging detection technique it is intrinsic merged electromagnetic induction, joule adds
Multiple physical processes such as heat, heat transfer, heat radiation, thermoacoustic effect, but do not related to also in electromagnetic induction thermal imaging detection technique at present
And arrive thermoacoustic.
Summary of the invention
For the existing detection method based on single electromagnetic field, thermal field or sound field to metal component damage check precision not
Height, the especially disadvantage of deep layer damage check effect difference, the invention discloses the nondestructive detection systems merged based on multiple physical field
And method, electromagnetic field, thermal field, acoustic field signal in fused metal magnetic history, metal surface and internal injury may be implemented
Solid detection.From electromagnetic field, thermal field, sound field signal characteristic, it is believed that be damaged metal is carried out it is physically related
The different observations of connection, every kind of observation are known as a mode.Although they are from identical materials behavior characteristic, different objects
Reason field form has different susceptibilitys to damage.For the non-destructive tests of metal component, more multi-modal informations are merged convenient for drop
Low interference then realizes more acurrate, comprehensive detection to the damage of metal deep layer.
The present invention is achieved through the following technical solutions:
Based on the nondestructive detection system of multiple physical field fusion, including it is signal generator, power amplifier, excitation coil, red
Outer thermal imaging system, sonic transducer, electromagnetic sensor, signal processing unit and metal component to be measured；The signal generator and power
Amplifier, sonic transducer and thermal infrared imager are separately connected, the excitation coil, electromagnetic sensor, thermal infrared imager, sound sensing
Device is all non-contacting to be fixedly mounted on metal component vertical direction to be measured, the signal processing unit and thermal infrared imager, electromagnetism
Sensor, sonic transducer are connected.
The signal generator generates pumping signal and sends the signal to power amplifier and thermal infrared imager, sound sensing
Device；
Received pumping signal power amplification is generated alternating current and acted on excitation coil by the power amplifier；
The excitation coil generates the electromagnetic field of alternation under the action of alternating excitation electric current, in the electromagnetic field effect of alternation
Under, metal component surface to be measured forms current vortex and heats to metal component to be measured, due to the vortex at different metal position
Density is different, and the Temperature Distribution of metal different parts is different and issues thermoacoustic signal；
After the electromagnetic sensor obtains electromagnetic signal, it is connected through signal conditioning circuit, capture card with computer；
The thermal infrared imager enters working condition under the triggering of exciting signal source, obtains infrared thermal imaging sequence；
The sonic transducer is set as acoustic matrix sensor, works under the triggering of pumping signal, can obtain stable sound
Press signal.
The signal processing unit includes signal conditioning circuit, capture card, cabinet, computer and the signal of computer installation
Processing software, computer receive the damage check letter of the different physical field from thermal infrared imager, electromagnetic sensor, sonic transducer
Number, and feature extraction, convergence analysis processing are carried out to electromagnetism, heat, acoustical signal；
The Signal Pretreatments such as the signal conditioning circuit has the function of to be filtered multiple physical field signal, amplify, noise abatement.
The course of work of the invention is that excitation coil generates the electromagnetic field of alternation, electricity under the pumping signal effect of amplification
Magnetic fields are on metal component to be measured and generate current vortex.Since each regio defect state of metal component to be measured is different, therefore whirlpool
Current density distributional difference is larger, and each position eddy heating for heating metal component generates the temperature field of different temperatures distribution and motivates thermoacoustic
Signal.Vortex in metal component can generate secondary magnetic, and the electromagnetic signal of superposition can be detected in electromagnetic sensor.Structure to be measured
The defect of part and the presence of damage can make Eddy Distribution and thermal diffusion change, Infrared Thermogram, thermoacoustic, electromagnetic signal etc.
Also it changes therewith.Different physical field signal is different with different types of weld defect control degree to different location, utilizes depth
Learning algorithm carries out convergence analysis to the detection signal of different physical field, carries out amount comprehensively, intelligent to the damage of metal component
Change assessment.Multiple physical field detection system based on deep learning signal analysis theory can real-time detection analysis metal component macroscopic view spy
Property (electromagnetic property, thermal characteristic, acoustic characteristic), entire failure procedure (stress concentration, dislocation, phase transformation, micro-crack to component
Formed, crack propagation) it is monitored, the damage at discovery each position of metal component promptly and accurately.The present invention magnetizes for metal
Multiple intrinsic physical processes such as electromagnetic induction, Joule heating, thermoacoustic effect in the process carry out more objects under the conditions of same excitation
Information fusion is managed, establishes the multi-modal deep learning model of multi-source later, more accurate, comprehensive damage is carried out to metal component and is examined
It surveys.Compared to traditional single one physical field lossless detection method, this method strong interference immunity, detection accuracy is high, to metal component deep layer
The phenomenon that damage is more sensitive, is not in missing inspection, erroneous detection.
Based on multiple physical field fusion nondestructive detection system and method the following steps are included:
Step 1: pumping signal being generated by signal generator, pumping signal controls the work of sonic transducer and thermal infrared imager
Make state, excitation coil is applied to after amplifying by power amplifier, generates excitation electromagnetic field；
Step 2: the electromagnetic field effect of alternation generates current vortex, current vortex pair in metal component to be measured, metal component to be measured
Metal component heating, the current vortex of different parts different densities make the Temperature Distribution of metal component to be measured generate variation, produce simultaneously
Heat acoustical signal；
Step 3: surface heat distribution being imaged by infrared heat image instrument measuring metal component infrared emanation to be measured, is led to
Sonic transducer acquisition thermoacoustic signal is crossed, electromagnetic signal is collected by electromagnetic sensor, three kinds of different physical field signals are ultimately delivered
To signal processing unit；
Step 4: signal processing unit receives the detection signal of different physical field by computer, at the signal in computer
It manages software and feature extraction is carried out to different physical field signal, component damage is carried out by deep learning model comprehensive, intelligence
Damage check.
Further, it is not less than 1.2KW, excitation by the amplified pumping signal power of power amplifier in the step 1
Time is 500ms-1.5s；Using power amplifier processing pumping signal make that current vortex intensity is motivated to meet detection demand, generally
Current effective value is not less than 10A.
Further, the lift off of the excitation coil in the step 1 and metal component to be measured is not higher than 4mm, prevents from feeling
It should be vortexed too small.
Further, the electromagnetic sensor in the step 3, thermal infrared imager, sonic transducer and metal component distance to be measured
Not higher than 5mm, guarantee the electromagnetism, the heat, acoustical signal that can receive sufficient intensity.
The present invention has the advantages that compared with the non-destructive testing technology of existing single one physical field
1, contacting between different macroscopic propertieies and metal component damage is established in multiple physical field non-destructive testing, compares single object
The lossless detection method for managing field, can be improved more field data utilization efficiency, strong interference immunity, high sensitivity, spatial resolution are more preferable.
2, other electromagnetic nondestructive methods such as electromagnetic induction thermal imaging, EDDY CURRENT are limited to vortex skin effect
Influence, carry out deep zone defect detection when, detection process is cumbersome or is difficult to carry out internal detection, and detection sensitivity is not high,
Influence vulnerable to metal surface state.The lossless detection method of multiple physical field fusion has incorporated component sound field information to be measured, can be with
More accurate detection is carried out to metal component internal flaw.
3, in multiple physical field fusion detection system, the micro-damage quantitative evaluation based on the multi-modal deep learning of multi-source is utilized
Model keeps nondestructive detection system real-time more preferable, more intelligent, visualization.
Detailed description of the invention
Attached drawing is used to provide and further understand to what the present invention was implemented herein, constitutes a part of the application patent, not
Constitute the restriction implemented to the present invention.
Fig. 1 is multiple physical field nondestructive detection system schematic diagram of the present invention；
Fig. 2 is the damage check flow chart of multi-modal deep learning damaged metal assessment models；
Fig. 3 is metal component transient temperature response curve；
Fig. 4 is deep learning lesion assessment network model signal processing schematic diagram.
Label and corresponding component names in attached drawing 1:
1- signal generator, 2- power amplifier, 3- thermal infrared imager, 4- excitation coil, 5- electromagnetic sensor, 6- are to be measured
Metal component, 7- sonic transducer, 8- information process unit.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
As shown in Figure 1, the nondestructive detection system of the invention based on multiple physical field fusion includes signal generator 1, power
Amplifier 2, thermal infrared imager 3, excitation coil 4, electromagnetic sensor 5, metal specimen to be measured 6, sonic transducer 7, signal processing list
Member 8.The signal generator 1 is connected with power amplifier 2, thermal infrared imager 3, sonic transducer 7, the thermal infrared imager 3, electricity
Magnetic Sensor 5, sonic transducer 7 are connected with signal processing unit 8, the excitation coil 4, thermal infrared imager 3, electromagnetic sensor
5, sonic transducer 7 is fixed at metal component vertical direction to be measured.
The signal generator 1 generates pumping signal and transmits signals to power amplifier 2, while pumping signal triggers
Thermal infrared imager 3, sonic transducer 7 enter working condition；
The thermal infrared imager 3 is the FLIR P30 of high speed, high resolution, is measured under the synchronous triggering of signal generator 1
The infrared emanation of metal component to be measured simultaneously generates infrared thermal imagery sequence input signal processing unit 8；
The excitation coil 4 generates the electromagnetic field of alternation under the action of exciting current, and alternating electromagnetic field acts on to be measured
6 surface of metal component, magnetizable metal simultaneously generate current vortex；
The electromagnetic sensor 5 acquires the electromagnetic signal of metal component 6 to be measured and is transmitted to signal processing unit 8；
The sonic transducer 7 is acoustic matrix sensor, can obtain stable sound pressure signal and be transmitted to signal processing unit
8；
The signal processing unit 8 includes the letter in signal conditioning unit, NI cabinet, capture card, computer and computer
Number processing software has based on the multi-modal deep learning signal processing module of multi-source in computer, can handle from thermal infrared imager
3, " electromagnetism-heat-sound " multilayered structure information of electromagnetic sensor 5, sonic transducer 7, more comprehensive to component damage to be measured progress,
The quantitative detection of intelligence.
Based on multiple physical field fusion nondestructive detection system and method the following steps are included:
Step 1: signal generator 1 generates pumping signal, and pumping signal controls the work of sonic transducer 7 and thermal infrared imager 3
Make state, excitation coil 4 is applied to after amplifying by power amplifier 2, generates excitation electromagnetic field；
Step 2: the electromagnetic field effect of alternation generates current vortex, current vortex in metal component 6 to be measured, metal component 6 to be measured
Metal component is heated, the current vortex of different densities makes the Temperature Distribution of metal component 6 to be measured generate variation, while generating thermoacoustic
Signal；
Step 3: by thermal infrared imager 3 measure the infrared emanation of metal component 6 to be measured to surface heat distribution carry out at
Picture, acquires thermoacoustic signal by sonic transducer 7, collects electromagnetic signal by electromagnetic sensor 5, three kinds of different physical field signals are most
It is sent to analysis and processing unit 8 eventually；
Step 4: analysis and processing unit 8 receives the detection signal of different physical field, the signal in computer by computer
Processing software carries out feature extraction, fusion to different physical field signal, accurate to component damage progress by deep learning model,
The detection of intelligence.
Further, the metal defect non-destructive testing process of the multiple physical field lossless detection method is as shown in Figure 2.
Further, the feature extraction in the step 4 includes thermoacoustic signal entirety amplitude or extracts after carrying out spectrum analysis
Pumping signal where frequency amplitude, the feature extraction in the step 4 includes the Barkhausen of electromagnetic signal high frequency section
The eddy current signal of signal or low frequency, the feature extraction in the step 4 include that the transient temperature of material surface and internal state is rung
It answers and the thermography feature of metal component.
Further, the transient temperature response of the metal component to be measured 6 includes the heating process of current vortex, with metal to be measured
The cooling procedure of component 6, transient temperature response curve such as Fig. 3.
Further, it is not less than 1.2KW, excitation by the amplified pumping signal power of power amplifier in the step 1
Time is 500ms-1.5s；Using power amplifier processing pumping signal make that current vortex intensity is motivated to meet detection demand, generally
Current effective value is not less than 10A.
Further, the lift off of the excitation coil 4 in the step 1 and metal component 6 to be measured prevents within 4mm
Inductive loop is too small；Electromagnetic sensor 5, thermal infrared imager 3, sonic transducer 7 and the vertical range of metal component 6 to be measured are less than
5mm guarantees the electromagnetism, the heat, acoustical signal that can receive sufficient intensity.
Embodiment:
The specific work process of detection method of the invention is as follows:
1., multiple physical field detection system builds
Multiple physical field detection system is built according to Fig. 1, signal generator needs to meet the requirement of driving frequency；Power amplification
Device needs to meet the power requirement of excitation coil；The P30 of FLIR company, the U.S. can be used in thermal infrared imager, the thermal imaging system resolution ratio
Reach 0.08 DEG C, reflective-mode can be used in thermal imagery acquisition mode；Using loop coil as excitation coil, coil is left using 6mm
Right combarloy hollow pipe is made, and electromagnetic sensor can be the detection copper coil of multiturn diameter 0.6mm or so, detection coil with
Signal conditioning circuit, capture card are connected；It is PXR0.3 acoustic emission sensor, composition sound sensing that multiple frequencies, which may be selected, in sonic transducer
Device array guarantees the integrality and stability of acoustical signal acquisition；No. 45 steel steel plates may be selected in metal component to be measured.When eddy heating for heating
Between can be set as 1s, the signal acquisition time is set as 2s.
2., the analysis and treatment process of multiple physical field signal
" electromagnetism-heat-sound " multiple physical field signal fused theoretical basis: metal component magnetizes simultaneously under alternating electromagnetism field excitation
Generate current vortex.According to Joule's law, vortex is converted into thermal energy by electric energy in material internal, in metal surface and internal generation temperature
Degree variation；According to thermoacoustic effect, the heat source of variation inspires sound wave, the propagation characteristic (such as acoustic attenuation, the velocity of sound) and material of sound wave
Characteristic is related.Metal magnetic history is actually the coupling of multiple physical processes such as electromagnetic induction, Joule heating, thermoacoustic effect,
And each physics field characteristic is influenced by damaged metal state.Based on the more physical process coupling principles of electromagnetic induction, each physical process
Under same electromagnetic excitation, by merging sonic transducer, electromagnetic sensor, thermal infrared imager, more of realization " electromagnetism-heat-sound "
Fusion non-destructive testing.The damage of metal component is fundamentally the change due to construction material microstructure (dislocation, phase transformation, crackle)
Caused by change, gradually accumulation can cause the variation in material macroscopic properties for the variation of microstructure.The present invention is based on " electromagnetism-heat-
The detection methods of sound " more fusions are by the variation of electromagnetic property, thermal characteristic, acoustic characteristic at material damage, to metal
The damage of component carries out three-dimensional detection.
Electromagnetic signal feature extraction and characterization: under the action of alternating excitation magnetic field, metal component magnetization generates density point
The current vortex of cloth unevenness.Current vortex generates the secondary magnetic opposite with primary magnetic field center, and primary magnetic can be detected in detection coil
Field is distorted with the magnetic field that is superimposed of secondary magnetic, detection signal, will inspire electromagnetism barkhausen signal in high frequency section.Root
According to the harmonic characterisitic of detection signal and the property of electromagnetism barkhausen signal identification of damage integrality, Spectrum Correction side is utilized
Method, the damage whole life process evaluation mode that may be based on electromagnetic property provide electromagnetic information.
Thermal image based on tensor expression with dynamic bayesian network (Dynamic Bayes Networks, DBNs) fusion
Higher-dimension tensor property is extracted and sparse representation: utilizing the space-time based on tensor-state tensor separation method, can obtain
The corresponding tensor state feature of metal component difference faulted condition.State feature can be described by conductivity, magnetic conductivity, thermal conductivity,
Space-time based on tensor-state tensor separation method can tentatively reflect macroscopical electromagnetism of material, thermal characteristics.DBNs can be very
The dynamic characteristic of good expression stochastic system, causality, precedence relationship and conditional relationship in reflection system.In different priori
Under failure mode, the space merged with DBNs, time, material properties higher dimensional space { s, t, σ, μ, ε } are expressed using based on tensor
(by taking electromagnetic attributes as an example, s: space, t: time, σ: conductivity, μ: magnetic conductivity, ε: dielectric constant) self organizing maps separation side
Thermal image sequence is carried out the extraction of higher-dimension tensor property and sparse representation, for the surface based on electromagnetism, thermal characteristics, sub-surface by method
Faulted condition identificates and evaluates model and provides characteristic information.
Blind source separating, feature extraction and the characterization of thermoacoustic array signal: in actually detected environment, metal component magnetized
The thermoacoustic signal excited in journey can have various noise jammings and multiple thermoacoustic sources, need from these complicated thermoacoustic signals
It accurately identifies thermoacoustic feature and identification positioning is carried out to main sound source (such as damage thermoacoustic source), especially to the knowledge of internal injury
Not.Non-negative Matrix Factorization (Non-negative Matrix Factorization, NMF) is important blind source analysis method.?
On the basis of deep learning, the hidden Markov model (Hidden Markov Model, HMM) based on probability similitude is utilized
State evaluating method can obtain good lesion assessment effect.I.e. at given observable status switch C, accurately find
One group of optimal HMM parameter lambda makes probability P (Ci/ λ) it is maximum.Assuming that feature vector of the structure under certain state is Ci, and normal shape
State reference feature vector collection is combined into C '=[C '1,…,C′i,…,C′k].Gradually deviate it since faulted condition evolution is substantially it
The process of normal condition, therefore under different conditions, feature vector C and reference feature vector sequence C ' similitude with material shape
State changes and changes.Therefore, feature vector C and the similarity of characteristic vector sequence C ' are measured, and constructs quantization characteristic
Index extracts the feature of characterization metal component damage.The present invention utilizes the thermoacoustic array signal blind source separating based on NMF and HMM
With state Feature Extraction Technology, Fusion Features are carried out to propagation characteristics such as frequency, phase and its distributions of vortex electric heating acoustical signal,
Thermoacoustic feature is extracted, provides characteristic information with assessment models for the metal component internal injury state recognition based on acoustic characteristic.
Micro-damage quantitative appraisement model based on the multi-modal deep learning of multi-source: deep learning is the non-linear of data guiding
Learning method can not only extract the nonlinear characteristic in high-dimensional data space, and can be by this kind of Feature Mapping to lower dimensional space reality
Existing dimensionality reduction, input parameter is few, and feature extraction is reproducible.The present invention is according to the sparse tables of more multi-scale damage analysis results
Show, by the deep learning to the more transducing signals such as heat, sound, electromagnetism, carried out under the frame of deep learning data fusion with
Interconnection vector regression estimates carry out quantitative evaluation to damage.
Depth Boltzmann machine (Deep Boltzmann Machines, DBM) is a kind of unsupervised deep learning model,
Ability with the complicated crucial global feature of study limits Boltzmann machine (Restricted Boltzmann by multilayer
Machines, RBM) it is formed by stacking, it is made of a visible layer and multiple hidden layers.The visible layer unit of multi-modal DBM is used to retouch
Observation data, more multi-modal detection datas in the corresponding present invention are stated, and hides layer unit and is used to obtain visible layer unit pair
Dependence between dependent variable, i.e., the relationship of multi- scenarios method feature and damaged metal in the present invention.The present invention can be in depth glass
The graceful machine visible layer of Wurz use Gauss neuron, obtain Gauss Bernoulli Jacob DBM (Gaussian Bernoulli DBM,
GDBM), the more experimental datas damaged using signal simulation data or typical material, are trained GDBM, and will input list
Member is expanded as the real value variable that continuously inputs, establish component damage in real time, comprehensively, model is accurately quantitatively evaluated.
3., the multi-modal deep learning MODEL DAMAGE detection of the multi-source based on multiple physical field
By obtaining electromagnetism, heat, acoustical signal in multiple physical field nondestructive detection system, multiple physical field signal inputs computer
Afterwards, after computer carries out feature extraction and characterization, trained GDBM model is inputted, it can be to each of metal component different parts
Class damage carries out more accurate, more intelligent, the better quantitative detection of real-time.
It is all in spirit of that invention and principle not to limit the present invention the foregoing is merely the available embodiment of the present invention
Within modification, replacement, the improvement etc. made, should be included within the scope of the present invention.
Claims (8)
1. a kind of nondestructive detection system based on multiple physical field fusion, which is characterized in that put including signal generator (1), power
Big device (2), thermal infrared imager (3), excitation coil (4), electromagnetic sensor (5), metal component to be measured (6), sonic transducer (7),
Signal processing unit (8)；
The signal generator (1) is connected with power amplifier (2), thermal infrared imager (3), sonic transducer (7), synchronous triggering function
Rate amplifier (2), thermal infrared imager (3), sonic transducer (7) work；
The power amplifier (2) receives pumping signal and amplifies, and generating on excitation coil (4) can be in metal component to be measured
(6) the alternating excitation electric current of vortex is motivated on；
The excitation coil (4) generates the electromagnetic excitation signal of alternation and acts on metal to be measured under the action of alternating current
On component (6)；
The thermal infrared imager (3), electromagnetic sensor (5), sonic transducer (7) are non-contacting to be fixed at metal structure to be measured
Part (6) vertical range is not higher than in the range of 5mm, and the thermal signal acquired respectively, magnetic signal, acoustical signal are transmitted at signal
It manages unit (8).
2. a kind of nondestructive detection system based on multiple physical field according to claim 1, which is characterized in that the excitation coil
(4) current effective value is not less than 10A, and driving frequency 60KHz-300KHz is not more than with the vertical range of metal component to be measured
4mm。
3. a kind of nondestructive detection system based on multiple physical field according to claim 2, which is characterized in that the sonic transducer
It (7) is condenser type acoustic matrix sensor.
4. a kind of nondestructive detection system based on multiple physical field according to claim 3, which is characterized in that the signal processing
Unit (8) includes signal conditioning circuit, NI cabinet, data collecting card, computer processing module.
5. a kind of nondestructive detection system based on multiple physical field according to claim 4, which is characterized in that at the computer
Reason module includes computer and its installation signal processing software in a computer, computer processing module tool data show,
Storage, analytic function.
6. a kind of lossless detection method based on multiple physical field fusion, which is characterized in that using any one of claim 1-5 institute
The nondestructive detection system based on multiple physical field fusion stated, detecting step are as follows:
Step 1: the work of thermal infrared imager (3) and sonic transducer (7) is controlled by the pumping signal that signal generator (1) generates
The frequency of state and exciting current；
Step 2: the alternating electromagnetic field that excitation coil (4) generates acts on metal component to be measured (6), and metal component generates electric whirlpool
It flows and component is heated；
Step 3: acquiring electromagnetism, heat, acoustical signal by electromagnetic sensor (5), thermal infrared imager (3), sonic transducer (7), and pass
It send to signal processing unit (8)；
Step 4: signal processing unit (8) to receive electromagnetism, heat, acoustical signal carry out signal analysis and processing, to metal to be measured
Component (6) carries out damage check.
7. the lossless detection method of multiple physical field fusion according to claim 6, which is characterized in that swash in the step 2
Coil power is encouraged not less than 1.2KW, actuation duration 500ms-1.5s.
8. the lossless detection method of multiple physical field fusion according to claim 7, which is characterized in that electric in the step 3
Magnetic Sensor (5), thermal infrared imager (3), sonic transducer (7) damaged metal signal acquisition time are not less than 1s.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910015763.6A CN109738512A (en) | 2019-01-08 | 2019-01-08 | Nondestructive detection system and method based on multiple physical field fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910015763.6A CN109738512A (en) | 2019-01-08 | 2019-01-08 | Nondestructive detection system and method based on multiple physical field fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109738512A true CN109738512A (en) | 2019-05-10 |
Family
ID=66363923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910015763.6A Pending CN109738512A (en) | 2019-01-08 | 2019-01-08 | Nondestructive detection system and method based on multiple physical field fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109738512A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287629A (en) * | 2019-07-01 | 2019-09-27 | 国网重庆市电力公司电力科学研究院 | Determine computer installation, the device and method of high pressure basin-type insulator structure parameter |
| CN111398352A (en) * | 2020-04-07 | 2020-07-10 | 四川沐迪圣科技有限公司 | Dynamic nondestructive testing system based on electromagnetic-thermal multi-physical-field fusion coil |
| CN111766305A (en) * | 2020-05-14 | 2020-10-13 | 北京工业大学 | Ultrasonic signal similarity analysis method for interface contact state evaluation |
| CN111948211A (en) * | 2020-06-16 | 2020-11-17 | 南京工业职业技术大学 | Composite type pressure-bearing equipment flaw detection system |
| CN111965246A (en) * | 2020-08-11 | 2020-11-20 | 太原理工大学 | Scraper machine fault detection method and detection system based on multi-information fusion |
| CN111982174A (en) * | 2019-05-23 | 2020-11-24 | 中国科学院沈阳自动化研究所 | Force-magnetic-acoustic three-field data fusion industrial equipment damage identification method |
| CN112365468A (en) * | 2020-11-11 | 2021-02-12 | 南通大学 | AA-gate-Unet-based offshore wind power tower coating defect detection method |
| CN112418324A (en) * | 2020-11-25 | 2021-02-26 | 武汉大学 | Cross-modal data fusion method for electrical equipment state perception |
| CN113218998A (en) * | 2021-04-27 | 2021-08-06 | 电子科技大学 | Eddy current thermal imaging defect identification method based on global Moran index |
| CN113297538A (en) * | 2021-06-08 | 2021-08-24 | 南昌航空大学 | Non-ferromagnetic material stress damage monitoring method and device and computer equipment |
| CN113358697A (en) * | 2021-04-27 | 2021-09-07 | 东南大学 | High-resolution photo-thermal pulse compression thermal imaging detection method based on nonlinear frequency modulation |
| CN116910489A (en) * | 2023-09-07 | 2023-10-20 | 中国水利水电第九工程局有限公司 | Wall seepage prevention detection method based on artificial intelligence and related device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120320372A1 (en) * | 2011-06-14 | 2012-12-20 | The Boeing Company | Autonomous Non-Destructive Evaluation System for Aircraft Structures |
| EP2835187A1 (en) * | 2013-08-09 | 2015-02-11 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Method and device for producing a press-hardened component |
| CN104434101A (en) * | 2014-12-14 | 2015-03-25 | 中国科学院电工研究所 | Magneto-thermoacoustic tomographic method and system |
| CN106950278A (en) * | 2017-03-07 | 2017-07-14 | 重庆大学 | Nondestructive detection system and method based on impulse eddy current thermo-acoustic technology |
| WO2018133179A1 (en) * | 2017-01-18 | 2018-07-26 | 中特检科技发展(北京)有限公司 | Multi-mode electromagnetic ultrasonic and magnetic flux leakage detection method, apparatus and system, and sensor |
-
2019
- 2019-01-08 CN CN201910015763.6A patent/CN109738512A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120320372A1 (en) * | 2011-06-14 | 2012-12-20 | The Boeing Company | Autonomous Non-Destructive Evaluation System for Aircraft Structures |
| EP2835187A1 (en) * | 2013-08-09 | 2015-02-11 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Method and device for producing a press-hardened component |
| CN104434101A (en) * | 2014-12-14 | 2015-03-25 | 中国科学院电工研究所 | Magneto-thermoacoustic tomographic method and system |
| WO2018133179A1 (en) * | 2017-01-18 | 2018-07-26 | 中特检科技发展(北京)有限公司 | Multi-mode electromagnetic ultrasonic and magnetic flux leakage detection method, apparatus and system, and sensor |
| CN106950278A (en) * | 2017-03-07 | 2017-07-14 | 重庆大学 | Nondestructive detection system and method based on impulse eddy current thermo-acoustic technology |
Non-Patent Citations (2)
| Title |
|---|
| AIJUN YIN ET AL.: "A structure state evaluation method based on electric-thermo-acoustic effect for tension materials", 《JAPANESE JOURNAL OF APPLIED PHYSICS》 * |
| 郑凯等: "红外热波无损检测技术的研究现状与进展", 《红外技术》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111982174A (en) * | 2019-05-23 | 2020-11-24 | 中国科学院沈阳自动化研究所 | Force-magnetic-acoustic three-field data fusion industrial equipment damage identification method |
| CN110287629A (en) * | 2019-07-01 | 2019-09-27 | 国网重庆市电力公司电力科学研究院 | Determine computer installation, the device and method of high pressure basin-type insulator structure parameter |
| CN110287629B (en) * | 2019-07-01 | 2022-08-09 | 国网重庆市电力公司电力科学研究院 | Computer device, equipment and method for determining structural parameters of high-voltage basin-type insulator |
| CN111398352A (en) * | 2020-04-07 | 2020-07-10 | 四川沐迪圣科技有限公司 | Dynamic nondestructive testing system based on electromagnetic-thermal multi-physical-field fusion coil |
| CN111766305A (en) * | 2020-05-14 | 2020-10-13 | 北京工业大学 | Ultrasonic signal similarity analysis method for interface contact state evaluation |
| CN111948211A (en) * | 2020-06-16 | 2020-11-17 | 南京工业职业技术大学 | Composite type pressure-bearing equipment flaw detection system |
| CN111965246A (en) * | 2020-08-11 | 2020-11-20 | 太原理工大学 | Scraper machine fault detection method and detection system based on multi-information fusion |
| CN111965246B (en) * | 2020-08-11 | 2023-10-31 | 太原理工大学 | Scraper machine fault detection method and system based on multi-information fusion |
| CN112365468A (en) * | 2020-11-11 | 2021-02-12 | 南通大学 | AA-gate-Unet-based offshore wind power tower coating defect detection method |
| CN112418324B (en) * | 2020-11-25 | 2022-06-24 | 武汉大学 | Cross-modal data fusion method for electrical equipment state perception |
| CN112418324A (en) * | 2020-11-25 | 2021-02-26 | 武汉大学 | Cross-modal data fusion method for electrical equipment state perception |
| CN113218998A (en) * | 2021-04-27 | 2021-08-06 | 电子科技大学 | Eddy current thermal imaging defect identification method based on global Moran index |
| CN113358697B (en) * | 2021-04-27 | 2022-03-29 | 东南大学 | High-resolution photo-thermal pulse compression thermal imaging detection method based on nonlinear frequency modulation |
| CN113218998B (en) * | 2021-04-27 | 2022-03-25 | 电子科技大学 | Eddy current thermal imaging defect identification method based on global Moran index |
| CN113358697A (en) * | 2021-04-27 | 2021-09-07 | 东南大学 | High-resolution photo-thermal pulse compression thermal imaging detection method based on nonlinear frequency modulation |
| CN113297538B (en) * | 2021-06-08 | 2022-05-10 | 南昌航空大学 | Non-ferromagnetic material stress damage monitoring method and device and computer equipment |
| CN113297538A (en) * | 2021-06-08 | 2021-08-24 | 南昌航空大学 | Non-ferromagnetic material stress damage monitoring method and device and computer equipment |
| CN116910489A (en) * | 2023-09-07 | 2023-10-20 | 中国水利水电第九工程局有限公司 | Wall seepage prevention detection method based on artificial intelligence and related device |
| CN116910489B (en) * | 2023-09-07 | 2023-11-21 | 中国水利水电第九工程局有限公司 | Wall seepage prevention detection method based on artificial intelligence and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109738512A (en) | Nondestructive detection system and method based on multiple physical field fusion | |
| AbdAlla et al. | Challenges in improving the performance of eddy current testing | |
| Al-Jumaili et al. | Parameter Correction Technique (PCT): A novel method for acoustic emission characterisation in large-scale composites | |
| Zhang et al. | CFRP impact damage inspection based on manifold learning using ultrasonic induced thermography | |
| Li et al. | Quantitative evaluation of surface crack depth with a scanning laser source based on particle swarm optimization-neural network | |
| Shrifan et al. | Prospect of using artificial intelligence for microwave nondestructive testing technique: A review | |
| Feng et al. | Locating defects in anisotropic CFRP plates using ToF-based probability matrix and neural networks | |
| CN107462599B (en) | Microwave thermoacoustic detection monitoring system and method | |
| Zhang et al. | Deep convolutional neural network probability imaging for plate structural health monitoring using guided waves | |
| Li et al. | Method for detecting damage in carbon-fibre reinforced plastic-steel structures based on eddy current pulsed thermography | |
| Chaupal et al. | A review on recent developments in vibration-based damage identification methods for laminated composite structures: 2010–2022 | |
| Wang et al. | Graph-in-graph convolutional network for ultrasonic guided wave-based damage detection and localization | |
| Chen et al. | A study on depth classification of defects by machine learning based on hyper-parameter search | |
| Ai et al. | A transfer learning approach for acoustic emission zonal localization on steel plate-like structure using numerical simulation and unsupervised domain adaptation | |
| Han et al. | Acoustic emission intelligent identification for initial damage of the engine based on single sensor | |
| Ren et al. | Multi-damage imaging of composite structures under environmental and operational conditions using guided wave and Gaussian mixture model | |
| Yan et al. | Pulse-modulation eddy current imaging for 3D profile reconstruction of subsurface corrosion in metallic structures of aviation | |
| Huang et al. | Lift-off nulling and internal state inspection of multi-layer conductive structures by combined signal features in pulsed eddy current testing | |
| Zhang et al. | Attention-based interpretable prototypical network towards small-sample damage identification using ultrasonic guided waves | |
| Li et al. | Intelligent damage recognition of composite materials based on deep learning and ultrasonic testing | |
| Zeng et al. | Lamb wave-based damage assessment for CFRP composite structures using a CHMM-based damage localization algorithm and a damage quantitative expression | |
| Balasubramaniam et al. | A global-local damage localization and quantification approach in composite structures using ultrasonic guided waves and active infrared thermography | |
| Chen et al. | Localization and imaging of internal hidden defects in concrete slabs based on deep learning of vibration signals | |
| Sun et al. | Cross-correlation inspired residual network for pulsed eddy current imaging and detecting of subsurface defects | |
| Zhou et al. | Real-time online inversion of GA-PSO-BP flux leakage defects based on information fusion: numerical simulation and experimental research |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190510 |
|
| WD01 | Invention patent application deemed withdrawn after publication |