CN108896649A - Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound - Google Patents
Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound Download PDFInfo
- Publication number
- CN108896649A CN108896649A CN201810398872.6A CN201810398872A CN108896649A CN 108896649 A CN108896649 A CN 108896649A CN 201810398872 A CN201810398872 A CN 201810398872A CN 108896649 A CN108896649 A CN 108896649A
- Authority
- CN
- China
- Prior art keywords
- yield strength
- magnetic field
- electromagnetic acoustic
- plate
- eigenvalue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/725—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables by using magneto-acoustical effects or the Barkhausen effect
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
Landscapes
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a method for estimating yield strength of a ferromagnetic material by electromagnetic ultrasound based on a magnetostrictive effect, which comprises the steps of taking N ferromagnetic plates with known yield strength, respectively arranging an electromagnetic ultrasound transmitting end and an electromagnetic ultrasound receiving end on the N ferromagnetic plates, measuring a characteristic value of a magnetostrictive curve of each ferromagnetic plate, then establishing a BP neural network model, taking the characteristic value of the magnetostrictive curve of the ferromagnetic plate and the thickness of the ferromagnetic plate as input, taking the material yield strength of the ferromagnetic plate as output, training the BP neural network model, and finally estimating the material yield strength of the ferromagnetic material by the trained BP neural network model. The invention realizes the nondestructive quantitative detection of the mechanical property of the yield strength of the material.
Description
Technical field
The present invention relates to electromagnetic nondestructive testing fields, more particularly to a kind of electromagnetic acoustic that is based on to bend to ferrimagnet
Take the estimation method of intensity.
Background technique
Ferrimagnet has good adaptability in terms of mechanical property, therefore in railway, transport, the energy, building, boat
It, military and chemical industry etc. by as critical material, be widely used.
However, miscellaneous ferrimagnet, can due to different processing technologys, different phosphorus content, structure and
Different doped alloys composite materials make its mechanical-physical and chemical characteristic have very big difference.Moreover, ferrimagnet is in difference
Under application environment, very big difference is also had using the performance of the conditions such as temperature to load, fatigue, burn into.Therefore, it is necessary to pass through
Detection to its magnetic characteristic obtains material microstructure information, so that it is determined that its load, fatigue and etch state, obtain material
Expect life information.
In addition, ferrimagnet in long service, will form defect, accident is damaged, causes casualties
With biggish economic loss.The formation of defect includes the parts such as incubation period, initial crack generation and expansion phase.Ferrimagnet exists
The early stage of damage, that is, incubation period, performance are usually the change of microstructure caused by the various microcosmic medium factors of stress collection
Change.Ferrimagnet causes the factor of defect to include in use：By carrying excessive caused local stress and stress collection
In, local plastic deformation is ultimately caused, defect is generated；Fatigue damage caused by being used for a long time；Temperature change leads to metal heat-expansion
Shrinkage, to be formed in the temperature stress of material internal accumulation；Internal residual stress caused by the processing technologys such as welding is distributed；Heat
The material parameters such as processing parameter such as hardness, carburized (case) depth, austenite and martensite content, crystallite dimension.Therefore, it is necessary to material
The distribution of material internal residual stress and mechanical property are monitored, to prevent the production of the local plastic deformation and defect of material
It is raw.
The Magnetostrictive Properties of ferrimagnet are damaged by the heterogeneous microstructure of material, mechanical properties strength, hardness, fatigue
Hurt the shadows such as Parameters variation, the temperature of state, the corrosion condition of material surface, internal residual stress distribution and externally-applied magnetic field
It rings.Therefore, the electromagnetic ultrasonic signal based on magnetostrictive effect can be with the fatigue damage of reaction material, deformation size, defect etc.
The variation of heterogeneous microstructure, while can be provided for estimating mechanical property of materials parameter（As hardness, yield strength, tensile strength,
Elongation percentage etc.）.
Currently with the method for electromagnetic acoustic non-destructive testing, mainly for detection of defect suffered by material, and to defect into
Row positioning, while the application of pipe thickness context of detection is widely studied using electromagnetic acoustic.Kang Yihua professor is new with force
There is research achievement outstanding in the team of the Central China University of Science and Technology that army professor leads in terms of studying EMAT steel pipe wall thickness measuring.He
Optimization design quiescent biasing magnetic field, using electromagnet replacement permanent magnet method, analysis demonstrate its dominance.Chinese mine
The Zhu Xiu Red Sect of Lamaism of large project team, which awards, tests electromagnetic acoustic magnetic field strength and receiving end ultrasound using experimental analysis steel tube defect
The relationship trend of signal amplitude.However utilize the electromagnetic acoustic of magnetostrictive effect material mechanical characteristic detection at present also not
Someone has very in-depth study in this respect.Mechanical performance is a set of the common counter of metal material, is mechanical production
Important materials performance indicator used in product design.The quality of metal material service performance, determines its use scope and makes
With the service life, and at present to the detection of material mechanical characteristic by the way of stretching offline, this mode can damage material,
Reusing for material is influenced, is resulted in waste of resources.
Summary of the invention
The technical problem to be solved by the present invention is to be directed to the prior art mentioned in background technique to material mechanical characteristic
（Yield strength）The defect of detection provides a kind of estimation method based on electromagnetic acoustic to ferrimagnet yield strength.
The present invention uses following technical scheme to solve above-mentioned technical problem：
Based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, comprise the steps of：
Step 1）, ferromagnetism plate known to N number of material yield strength is taken, N is the natural number greater than 0, ferromagnetic for each
Property plate：
Step 1.1）, electromagnetic acoustic transmitting terminal and electromagnetic acoustic receiving end are respectively set on ferromagnetism plate, and super to electromagnetism
Sound emission end applies the first bias magnetic field, applies the second bias magnetic field to electromagnetic acoustic receiving end, and first bias magnetic field is adopted
It is formed with adjustable DC electromagnet, second bias magnetic field excites to be formed using permanent magnet, electromagnetic acoustic transmitting terminal transmitting electricity
Magnetic ultrasonic signal, electromagnetic acoustic receiving end acquire the electromagnetic ultrasonic signal of electromagnetic acoustic transmitting terminal transmitting；
Step 1.2）, the intensity of first bias magnetic field is walked from preset first magnetic field strength according to preset magnetic field strength
Length is gradually increased to preset second magnetic field strength, and reads electromagnetic acoustic receiving end electromagnetic ultrasonic signal under each magnetic field strength
Amplitude, the magnetic field strength of the first bias magnetic field of fitting corresponds to the magnetostriction of electromagnetic acoustic receiving end electromagnetic ultrasonic signal amplitude
Curve, the intensity for extracting the first bias magnetic field in Magnetostrictive curve corresponding electromagnetic acoustic when being preset first magnetic field strength
The amplitude of receiving end electromagnetic ultrasonic signal is strong as the magnetic field of the First Eigenvalue, corresponding first bias magnetic field of first valley point
Degree be Second Eigenvalue, corresponding first bias magnetic field of first peak point magnetic field strength be third feature value, second paddy
The magnetic field strength of corresponding first bias magnetic field of value point is fourth feature value；
Step 2）, Neural Network Data library is established according to the attribute data of N number of ferromagnetism plate, the ferromagnetism plate
Attribute data includes material yield strength, thickness, the First Eigenvalue, the Second Eigenvalue, third feature value of the ferromagnetism plate
With fourth feature value；
Step 3）, the data in Neural Network Data library are divided into training sample set and test sample collection, and by training sample
The data normalization that collection and test sample are concentrated is between [- 1,1]；
Step 4）, establish BP neural network model, by training sample concentrate the thickness of each ferromagnetism plate, the First Eigenvalue,
As input, training sample concentrates the material of corresponding each ferromagnetism plate for Second Eigenvalue, third feature value, fourth feature value
Expect that yield strength as output, trains the BP neural network model, when training error is less than preset first error threshold value,
Training terminates, and obtains trained BP neural network model；
Step 5）, test sample is concentrated to the thickness, the First Eigenvalue, Second Eigenvalue, third feature of each ferromagnetism plate
Value, fourth feature value input trained BP neural network model, obtain the material that test sample concentrates each ferromagnetism plate
Yield strength estimated value；
Step 6）, calculate test sample and concentrate between each ferromagnetism plate material yield strength, material yield strength estimated value
Relative error, if relative error between ferromagnetism plate material yield strength, material yield strength estimated value be less than it is default
The second error threshold, then it is assumed that ferromagnetism plate is qualified, and no person thinks that ferromagnetism plate is unqualified；
Step 7）, the qualification rate that test sample concentrates ferromagnetic plate part is calculated, i.e., by qualified ferromagnetic plate number of packages amount divided by survey
The quantity of sample this concentration ferromagnetism plate, if test sample concentrates the qualification rate of ferromagnetic plate part to be less than or equal to preset conjunction
Lattice rate threshold value, gos to step 1）；
Step 8）, will need to carry out the thickness, the First Eigenvalue, Second Eigenvalue, of the ferromagnetism plate of material yield strength
Three characteristic values, fourth feature value input trained BP neural network model, obtain needing to carry out the ferromagnetic of material yield strength
The material yield strength estimated value of property plate.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
Preset first magnetic field strength is 0 tesla.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
The exciting signal frequency of the ultrasonic wave transmitting terminal is 200KHz, and pulse number is 8.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
The preset first error threshold value is 0.01.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
Preset second error threshold is 10%.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
The preset qualification rate threshold value is 80%.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
The BP neural network model includes an input layer, two hidden layers and an output layer, wherein the input layer includes 5
A node, a hidden layer include 9 nodes, another hidden layer includes 3 nodes, and output layer includes 1 node.
As the present invention is based on the further prioritization scheme of estimation method of the electromagnetic acoustic to ferrimagnet yield strength,
The step 3）It is 4 that middle training sample set, test sample, which concentrate the ratio of sample size,:1.
The invention adopts the above technical scheme compared with prior art, has the following technical effects：
1. can be realized the quantitative predication to ferrimagnet yield strength by establishing BP neural network model；
2. higher detection qualification rate can be reached, change it is previous it is destructive stretch online to material mechanical characteristic for example
The detection mode of yield strength.
Detailed description of the invention
Fig. 1 is the received ultrasonic signal figure in electromagnetic acoustic detection system of the present invention receiving end；
Fig. 2（a）,（b）Material Magnetostrictive Properties curve graph, electromagnetic ultrasonic signal curve graph respectively in the present invention；
Fig. 3 is that electromagnetic acoustic of the invention receives signal amplitude with the variation diagram of externally-applied magnetic field；
Fig. 4 is that test sample of the present invention is concentrated between each ferromagnetism plate material yield strength, material yield strength estimated value
Relative error schematic diagram.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawing：
The present invention can be embodied in many different forms, and should not be assumed that be limited to the embodiments described herein.On the contrary, providing
These embodiments are thoroughly and complete to make the disclosure, and will give full expression to the scope of the present invention to those skilled in the art.
In the accompanying drawings, for the sake of clarity it is exaggerated component.
Ferromagnetic material has the structure of similar crystal.Between adjacent atom, first magnetic moment is generated due to electron spin,
There is interaction force between first magnetic moment, it drives adjacent first magnetic moments parallel to arrange in the same direction, forms magnetic domain.In nothing
When external magnetic field, each magnetic domain is balanced mutually, and the total intensity of magnetization of material is equal to zero.When there is external magnetic field, magnetic domain meeting
It rotates, so that minor change occurs therewith for length of material or volume, this phenomenon is known as magnetostrictive effect.Different iron
Magnetic material have different Magnetostrictive Properties, Magnetostrictive Properties by material microstructure, external magnetic field, material stress and
Condition of heat treatment etc. influences, based on the electromagnetic ultrasonic signal of magnetostrictive effect by the Magnetostrictive Properties shadow of ferrimagnet
It rings, therefore can reflect the Magnetostrictive Properties of material according to electromagnetic ultrasonic signal, such as Fig. 2 of relationship between the two.
The invention discloses a kind of based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, illustrates such as
Under：
Step 1）, ferromagnetism plate known to 50 material yield strengths is taken, for each ferromagnetism plate：
Step 1.1）, as shown in Figure 1, electromagnetic acoustic transmitting terminal and electromagnetic acoustic receiving end are respectively set on ferromagnetism plate,
And apply the first bias magnetic field to electromagnetic acoustic transmitting terminal, apply the second bias magnetic field to electromagnetic acoustic receiving end, described first
Bias magnetic field is formed using adjustable DC electromagnet, and second bias magnetic field excites to be formed using permanent magnet, electromagnetic acoustic hair
End transmitting electromagnetic ultrasonic signal is penetrated, electromagnetic acoustic receiving end acquires the electromagnetic ultrasonic signal of electromagnetic acoustic transmitting terminal transmitting；It is described
The exciting signal frequency of ultrasonic wave transmitting terminal is 200KHz, and pulse number is 8；
Step 1.2）, the intensity of first bias magnetic field is gradually increased from 0 tesla according to preset magnetic field strength step-length
Extremely preset second magnetic field strength, and the amplitude of electromagnetic acoustic receiving end electromagnetic ultrasonic signal under each magnetic field strength is read, intend
The magnetic field strength for closing the first bias magnetic field corresponds to the Magnetostrictive curve of electromagnetic acoustic receiving end electromagnetic ultrasonic signal amplitude, such as schemes
Shown in 3；
Corresponding electromagnetic acoustic receiving end electromagnetic ultrasonic signal when the intensity for extracting the first bias magnetic field in Magnetostrictive curve is 0
Amplitude as the magnetic field strength of the First Eigenvalue, corresponding first bias magnetic field of first valley point be Second Eigenvalue, the
The magnetic field strength of corresponding first bias magnetic field of one peak point is third feature value, corresponding first biasing of second valley point
The magnetic field strength in magnetic field is fourth feature value；
Step 2）, Neural Network Data library is established according to the attribute data of N number of ferromagnetism plate, the ferromagnetism plate
Attribute data includes material yield strength, thickness, the First Eigenvalue, the Second Eigenvalue, third feature value of the ferromagnetism plate
With fourth feature value；
Step 3）, the data in Neural Network Data library are divided into training sample set and test sample collection, specifically, are divided
The attribute data of 40 ferromagnetism plates is training sample set, and the attribute data of remaining 10 ferromagnetism plates is divided into test
Sample set, and the data normalization that training sample set and test sample are concentrated is between [- 1,1]；
Step 4）, BP neural network model is established, the BP neural network model includes an input layer, two hidden layers and one
A output layer, wherein the input layer includes 5 nodes, and a hidden layer includes 9 nodes, another hidden layer includes 3
Node, output layer include 1 node；
Training sample is concentrated to the thickness, the First Eigenvalue, Second Eigenvalue, third feature value, the 4th of each ferromagnetism plate
For characteristic value as input, training sample concentrates the material yield strength of corresponding each ferromagnetism plate as output, training institute
BP neural network model is stated, when training error is less than 0.01, training terminates, and obtains trained BP neural network model；
Step 5）, test sample is concentrated to the thickness, the First Eigenvalue, Second Eigenvalue, third feature of each ferromagnetism plate
Value, fourth feature value input trained BP neural network model, obtain the material that test sample concentrates each ferromagnetism plate
Yield strength estimated value；
Step 6）, calculate test sample and concentrate between each ferromagnetism plate material yield strength, material yield strength estimated value
Relative error, if the relative error between ferromagnetism plate material yield strength, material yield strength estimated value less than 10%,
Then think that the ferromagnetism plate is qualified, no person thinks that ferromagnetism plate is unqualified；Fig. 4 is that test sample concentrates each ferromagnetic plate
The schematic diagram of relative error between part material yield strength, material yield strength estimated value；
Step 7）, the qualification rate that test sample concentrates ferromagnetic plate part is calculated, i.e., by qualified ferromagnetic plate number of packages amount divided by survey
The quantity of sample this concentration ferromagnetism plate jumps if test sample concentrates the qualification rate of ferromagnetic plate part to be less than or equal to 80%
To step 1）；
Step 8）, will need to carry out the thickness, the First Eigenvalue, Second Eigenvalue, of the ferromagnetism plate of material yield strength
Three characteristic values, fourth feature value input trained BP neural network model, obtain needing to carry out the ferromagnetic of material yield strength
The material yield strength estimated value of property plate.
The purpose of the present invention is to propose to a kind of based on the electromagnetic acoustic of magnetostrictive effect to ferrimagnet yield strength
Estimation method realized by neural network network model to the lossless fixed of material yield strength this mechanical property
Amount detection.
Those skilled in the art can understand that unless otherwise defined, all terms used herein（Including skill
Art term and scientific term）With meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Also
It should be understood that those terms such as defined in the general dictionary should be understood that have in the context of the prior art
The consistent meaning of meaning will not be explained in an idealized or overly formal meaning and unless defined as here.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not limited to this hair the foregoing is merely a specific embodiment of the invention
Bright, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention
Protection scope within.
Claims (8)
1. based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, which is characterized in that comprise the steps of：
Step 1）, ferromagnetism plate known to N number of material yield strength is taken, N is the natural number greater than 0, ferromagnetic for each
Property plate：
Step 1.1）, electromagnetic acoustic transmitting terminal and electromagnetic acoustic receiving end are respectively set on ferromagnetism plate, and super to electromagnetism
Sound emission end applies the first bias magnetic field, applies the second bias magnetic field to electromagnetic acoustic receiving end, and first bias magnetic field is adopted
It is formed with adjustable DC electromagnet, second bias magnetic field excites to be formed using permanent magnet, electromagnetic acoustic transmitting terminal transmitting electricity
Magnetic ultrasonic signal, electromagnetic acoustic receiving end acquire the electromagnetic ultrasonic signal of electromagnetic acoustic transmitting terminal transmitting；
Step 1.2）, the intensity of first bias magnetic field is walked from preset first magnetic field strength according to preset magnetic field strength
Length is gradually increased to preset second magnetic field strength, and reads electromagnetic acoustic receiving end electromagnetic ultrasonic signal under each magnetic field strength
Amplitude, the magnetic field strength of the first bias magnetic field of fitting corresponds to the magnetostriction of electromagnetic acoustic receiving end electromagnetic ultrasonic signal amplitude
Curve, the intensity for extracting the first bias magnetic field in Magnetostrictive curve corresponding electromagnetic acoustic when being preset first magnetic field strength
The amplitude of receiving end electromagnetic ultrasonic signal is strong as the magnetic field of the First Eigenvalue, corresponding first bias magnetic field of first valley point
Degree be Second Eigenvalue, corresponding first bias magnetic field of first peak point magnetic field strength be third feature value, second paddy
The magnetic field strength of corresponding first bias magnetic field of value point is fourth feature value；
Step 2）, Neural Network Data library is established according to the attribute data of N number of ferromagnetism plate, the ferromagnetism plate
Attribute data includes material yield strength, thickness, the First Eigenvalue, the Second Eigenvalue, third feature value of the ferromagnetism plate
With fourth feature value；
Step 3）, the data in Neural Network Data library are divided into training sample set and test sample collection, and by training sample
The data normalization that collection and test sample are concentrated is between [- 1,1]；
Step 4）, establish BP neural network model, by training sample concentrate the thickness of each ferromagnetism plate, the First Eigenvalue,
As input, training sample concentrates the material of corresponding each ferromagnetism plate for Second Eigenvalue, third feature value, fourth feature value
Expect that yield strength as output, trains the BP neural network model, when training error is less than preset first error threshold value,
Training terminates, and obtains trained BP neural network model；
Step 5）, test sample is concentrated to the thickness, the First Eigenvalue, Second Eigenvalue, third feature of each ferromagnetism plate
Value, fourth feature value input trained BP neural network model, obtain the material that test sample concentrates each ferromagnetism plate
Yield strength estimated value；
Step 6）, calculate test sample and concentrate between each ferromagnetism plate material yield strength, material yield strength estimated value
Relative error, if relative error between ferromagnetism plate material yield strength, material yield strength estimated value be less than it is default
The second error threshold, then it is assumed that ferromagnetism plate is qualified, and no person thinks that ferromagnetism plate is unqualified；
Step 7）, the qualification rate that test sample concentrates ferromagnetic plate part is calculated, i.e., by qualified ferromagnetic plate number of packages amount divided by survey
The quantity of sample this concentration ferromagnetism plate, if test sample concentrates the qualification rate of ferromagnetic plate part to be less than or equal to preset conjunction
Lattice rate threshold value, gos to step 1）；
Step 8）, will need to carry out the thickness, the First Eigenvalue, Second Eigenvalue, of the ferromagnetism plate of material yield strength
Three characteristic values, fourth feature value input trained BP neural network model, obtain needing to carry out the ferromagnetic of material yield strength
The material yield strength estimated value of property plate.
2. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In preset first magnetic field strength is 0 tesla.
3. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In the exciting signal frequency of the ultrasonic wave transmitting terminal is 200KHz, and pulse number is 8.
4. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In the preset first error threshold value is 0.01.
5. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In preset second error threshold is 10%.
6. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In the preset qualification rate threshold value is 80%.
7. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In the BP neural network model includes an input layer, two hidden layers and an output layer, wherein the input layer packet
Containing 5 nodes, a hidden layer includes 9 nodes, another hidden layer includes 3 nodes, and output layer includes 1 node.
8. it is according to claim 1 based on electromagnetic acoustic to the estimation method of ferrimagnet yield strength, feature exists
In the step 3）It is 4 that middle training sample set, test sample, which concentrate the ratio of sample size,:1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810398872.6A CN108896649B (en) | 2018-04-28 | 2018-04-28 | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810398872.6A CN108896649B (en) | 2018-04-28 | 2018-04-28 | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108896649A true CN108896649A (en) | 2018-11-27 |
| CN108896649B CN108896649B (en) | 2021-05-07 |
Family
ID=64342622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810398872.6A Active CN108896649B (en) | 2018-04-28 | 2018-04-28 | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108896649B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110274960A (en) * | 2019-08-02 | 2019-09-24 | 大唐锅炉压力容器检验中心有限公司 | A kind of steel pipe microscopic structure appraisal procedure and device based on non-linear ultrasonic |
| CN113325084A (en) * | 2021-05-25 | 2021-08-31 | 南京航空航天大学 | Method for detecting mechanical property of ferromagnetic material based on sound velocity effect |
| CN114113294A (en) * | 2020-08-28 | 2022-03-01 | 宝山钢铁股份有限公司 | Online measuring device and method for determining yield strength and tensile strength of strip steel |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4307616A (en) * | 1979-12-26 | 1981-12-29 | Rockwell International Corporation | Signal processing technique for ultrasonic inspection |
| JPH10332643A (en) * | 1997-06-05 | 1998-12-18 | Nkk Corp | Fatigue crack detection method |
| CN102590353A (en) * | 2012-01-12 | 2012-07-18 | 天津工业大学 | Acoustic emission excitation device of EMAT (Electro Magnetic Acoustic Transducer) transmitting probe |
| CN104316341A (en) * | 2014-11-17 | 2015-01-28 | 金陵科技学院 | Underground structure damage identification method based on BP neural network |
| CN104330477A (en) * | 2014-09-22 | 2015-02-04 | 中国石油天然气集团公司 | Electromagnetic ultrasonic excitation probe design method based on magnetic induced shrinkage or elongation effect |
| CN106996957A (en) * | 2016-01-25 | 2017-08-01 | 天津工业大学 | A kind of ferromagnetic metal lossless detection method loaded based on electromagnetism |
-
2018
- 2018-04-28 CN CN201810398872.6A patent/CN108896649B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4307616A (en) * | 1979-12-26 | 1981-12-29 | Rockwell International Corporation | Signal processing technique for ultrasonic inspection |
| JPH10332643A (en) * | 1997-06-05 | 1998-12-18 | Nkk Corp | Fatigue crack detection method |
| CN102590353A (en) * | 2012-01-12 | 2012-07-18 | 天津工业大学 | Acoustic emission excitation device of EMAT (Electro Magnetic Acoustic Transducer) transmitting probe |
| CN104330477A (en) * | 2014-09-22 | 2015-02-04 | 中国石油天然气集团公司 | Electromagnetic ultrasonic excitation probe design method based on magnetic induced shrinkage or elongation effect |
| CN104316341A (en) * | 2014-11-17 | 2015-01-28 | 金陵科技学院 | Underground structure damage identification method based on BP neural network |
| CN106996957A (en) * | 2016-01-25 | 2017-08-01 | 天津工业大学 | A kind of ferromagnetic metal lossless detection method loaded based on electromagnetism |
Non-Patent Citations (2)
| Title |
|---|
| PAVEL TSVETKOV: "Non-Destructive Testing for Ageing Management of Nuclear Power Components", 《NUCLEAR POWER-CONTROL》 * |
| 周静 等: "基于磁致伸缩特征参数的构件应力测试方法", 《2017远东无损检测新技术论坛论文集》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110274960A (en) * | 2019-08-02 | 2019-09-24 | 大唐锅炉压力容器检验中心有限公司 | A kind of steel pipe microscopic structure appraisal procedure and device based on non-linear ultrasonic |
| CN110274960B (en) * | 2019-08-02 | 2022-07-26 | 大唐锅炉压力容器检验中心有限公司 | Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound |
| CN114113294A (en) * | 2020-08-28 | 2022-03-01 | 宝山钢铁股份有限公司 | Online measuring device and method for determining yield strength and tensile strength of strip steel |
| CN114113294B (en) * | 2020-08-28 | 2023-12-12 | 宝山钢铁股份有限公司 | Online measuring device and method for determining yield strength and tensile strength of strip steel |
| CN113325084A (en) * | 2021-05-25 | 2021-08-31 | 南京航空航天大学 | Method for detecting mechanical property of ferromagnetic material based on sound velocity effect |
| CN113325084B (en) * | 2021-05-25 | 2022-04-22 | 南京航空航天大学 | Method for detecting mechanical property of ferromagnetic material based on sound velocity effect |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108896649B (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shi et al. | Overview of researches on the nondestructive testing method of metal magnetic memory: Status and challenges | |
| Du et al. | An experimental feasibility study of pipeline corrosion pit detection using a piezoceramic time reversal mirror | |
| US9903840B2 (en) | Method for detecting temporally varying thermomechanical stresses and/or stress gradients over the wall thickness of metal bodies | |
| CN108896649A (en) | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound | |
| Chen et al. | Quantitative nondestructive evaluation of plastic deformation in carbon steel based on electromagnetic methods | |
| US20100244591A1 (en) | Method and apparatus for monitoring wall thinning of a pipe using magnetostrictive transducers and variation of dispersion characteristics of broadband multimode shear horizontal (SH) waves | |
| Eslamlou et al. | A review on non-destructive evaluation of construction materials and structures using magnetic sensors | |
| Yao et al. | Structural health monitoring of multi-spot welded joints using a lead zirconate titanate based active sensing approach | |
| Narayanan et al. | Development of in-bore magnetostrictive transducer for ultrasonic guided wave based-inspection of steam generator tubes of PFBR | |
| Huang et al. | Recent advances in magnetic non-destructive testing and the application of this technique to remanufacturing | |
| Dobmann | Non-destructive testing for ageing management of nuclear power components | |
| Gorkunov et al. | The influence of a preliminary plastic deformation on the behavior of the magnetic characteristics of high-strength controllably rolled pipe steel under an elastic uniaxial tension (compression) | |
| Liu et al. | Micromagnetic characteristic changes and mechanism induced by plastic deformation of 304 austenitic stainless steel | |
| Li et al. | Method of measuring the stress of ferromagnetic materials based on EMAT and magnetic Barkhausen noise characteristic parameters | |
| Jančula et al. | Monitoring of corrosion extent in steel S460MC by the use of magnetic Barkhausen noise emission | |
| Zhang et al. | Application of a back-propagation neural network for mechanical properties prediction of ferromagnetic materials by magnetic Barkhausen noise technique | |
| Zhu et al. | Prediction of the tensile force applied on surface-hardened steel rods based on a CDIF and PSO-optimized neural network | |
| CN109738518A (en) | A kind of method and apparatus of nonlinear electromagnetic ultrasound resonance assessment material thermal effectiveness | |
| Wang et al. | Method of measuring the mechanical properties of ferromagnetic materials based on magnetostriction EMAT and sound velocity | |
| Tomáš et al. | Optimization of fatigue damage indication in ferromagnetic low carbon steel | |
| Qi et al. | The microstructural evolution and ultrasonic guided wave transduction performance of annealed magnetostrictive (Fe83Ga17) 99.9 (NbC) 0.1 thin sheets | |
| Mushnikov et al. | Effect of mechanical stresses on the magnetic characteristics of pipeline steels of different classes | |
| Zhang et al. | Theoretical model of magnetoacoustic emission considering the microstructure of ferromagnetic material | |
| Kashefi et al. | On the combined effect of elastic and plastic strain on magnetic Barkhausen noise signals | |
| Zhang et al. | Research on the mechanism of electromagnetic ultrasonic energy transfer based on dynamic multi-magnetic vector coupling |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220317 Address after: No. 29, Qinhuai District, Qinhuai District, Nanjing, Jiangsu Patentee after: Nanjing University of Aeronautics and Astronautics Asset Management Co.,Ltd. Address before: No. 29, Qinhuai District, Qinhuai District, Nanjing, Jiangsu Patentee before: Nanjing University of Aeronautics and Astronautics |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220427 Address after: 211599 No.8, Huyue East Road, Longchi street, Liuhe District, Nanjing City, Jiangsu Province Patentee after: Jiangsu Jinyu Intelligent Detection System Co.,Ltd. Address before: No. 29, Qinhuai District, Qinhuai District, Nanjing, Jiangsu Patentee before: Nanjing University of Aeronautics and Astronautics Asset Management Co.,Ltd. |