CN1743839A - Structure defect ultrasonic on-line intelligent identifying system and identifying method - Google Patents
Structure defect ultrasonic on-line intelligent identifying system and identifying method Download PDFInfo
- Publication number
- CN1743839A CN1743839A CNA2005100357880A CN200510035788A CN1743839A CN 1743839 A CN1743839 A CN 1743839A CN A2005100357880 A CNA2005100357880 A CN A2005100357880A CN 200510035788 A CN200510035788 A CN 200510035788A CN 1743839 A CN1743839 A CN 1743839A
- Authority
- CN
- China
- Prior art keywords
- defect
- signal
- ultrasonic
- circuit
- neural network
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/028—Material parameters
- G01N2291/0289—Internal structure, e.g. defects, grain size, texture
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
This invention provides a super acoustic online intelligent system for detecting structure defect, which contains super acoustic defect detector comprising synchronous circuit, emission circuit, probe with transducer and receiving circuit, high speed data sampling card connected with synchronous circuit by gate control circuit, and computer, wherein the the logic control circuit connected with gate control circuit and micro processor of computer, the micro processor having each other connected wavelet analysis module and artificial neural network.
Description
Technical field
The present invention relates to the Ultrasonic Nondestructive system, specifically be meant a kind of structure defect ultrasonic on-line intelligent identifying system and recognition methods.
Background technology
Along with development of modern industry, Dynamic Non-Destruction Measurement and application thereof have seemed more and more important.Metallic element may produce various defectives in material smelting and process, for the carrying parts, particularly high temperature, high pressure, parts at a high speed, its inside exists defective may cause great personal injury, cause heavy losses, in these accidents, weld defects is one of topmost factor.
Because the harm of crack-type defect is bigger in the welded structure, its degree of depth accurately quantitatively in the structural intergrity assessment, seem particularly important.The Ultrasonic Nondestructive technology is used comparatively general in the measurement of depth of defect, its use generally be very short pulse ultrasonic wave of duration, can regard that the harmonic wave by unlimited a plurality of different frequencies is formed by stacking as, after defect reflection is met in ultrasonic pulse, the variation of harmonic wave will cause the variation of its frequency spectrum, thereby provide the information of reflection defect property, it is wide to have the measurand scope, the detection degree of depth is big, defect location is accurate, the detection sensitivity height, cost is low, speed is fast, easy to use, harmless and be convenient to characteristics such as on-the-spot use, but its result is subjected to instrument easily, probe, the influence of factor such as workpiece and personnel.
All develop in the ultrasound examination field both at home and abroad at present towards the digitizing direction.As: the ULTRAPAC of U.S. acoustics company (PAC) research and development is advanced integrated, the digitalized ultrasonic C-scanning imaging system of computing machine, can intactly survey the different shape defective of part, determine defective locations, the detected image report is provided, the permanent recording testing result is convenient to analyze and without the result under the testing conditions relatively again.Creative Company of section subsidizes the emission of high-performance multichannel ultrasonic and the receiving system based on pci bus of research and development in the domestic Wuhan, have the dynamic display defect ripple of A sweep position, according to standard and gate warning automatically, automatic gain, numeral inhibition, defective waveform playback, automatically make the DAC curve, include multiple general and special-purpose functions such as non-destructive testing standard; The PXUT-280 type ultra-sonic defect detector that Nantong friend's connection company produces possesses imaging function, can scan the shape that shows defective in the workpiece by B; The full digital reflectoscope CTS-3600 type of Shantou ultrasound instrument research institute, characteristics such as have sample frequency height, big, the Chinese operation interface of memory capacity, small portable, be easy to learn and use can be widely used in the Ultrasonic Detection field.
In Ultrasonic NDT, analysis to signal is the important step that directly influences the testing result accuracy, existing ultra-sonic defect detector is all also very limited to the Signal Processing ability, quantitatively has very big subjectivity from angle intuitively to what the shown flaw echo of reflectoscope was carried out defective, under some situation, different testing staff may be by notable difference to the evaluation result of same defective, and equipment and working environment etc. also can have influence on final testing result.Simultaneously, the use of numeric type reflectoscope makes that the data that can collect in the Ultrasonic Detection are more and more, relies on fully manually to come data are analyzed and more become very difficult, has reduced the precision and the work efficiency of ultrasound examination.Prior art is to the research and the improvement of ultrasonic detection equipment, emphasis mainly has been placed on the hardware device, its influence to the ultrasound examination result is limited, and has improved the cost of equipment, therefore is necessary to explore the recognition result that the data processing method that makes new advances is obtained defective.
Spectrum analysis technique is the signal processing technology that early is applied to Non-Destructive Testing, main employing fast fourier transform is that FFT (Fast Fourier Transform) algorithm is realized, but because ultrasonic signal has time-varying characteristics, Fourier transform can only picked up signal overall spectrum, and can not carry out partial analysis.Thereby be difficult for obtaining frequency spectrum preferably, and its frequency spectrum can not reflect temporal signatures.And wavelet analysis (Wavelet Analysis) all has analysis ability preferably in time domain and frequency domain, the time domain and the frequency domain character of flaw echo can be shown, be applicable to the analysis to ultrasound echo signal, this technology is widely used in fields such as signal Processing, Flame Image Process, pattern-recognition, speech recognition, seismic prospecting, condition monitoring and fault diagnosises.
Artificial neural network is a kind ofly to simulate the system of the 26S Proteasome Structure and Function of human brain with device, system or the existing computer technology that physically can realize, artificial nerve network model mainly contains two big classes at present: with the Hopfield network model is the feedback-type of representative and based on the feed-forward type of multilayer perceptron.Comparatively common in the Multi-layered Feedforward Networks practical application, and ripe BP (BackPropagation) algorithm that has the tutor to learn, i.e. the error signal back-propagation algorithm of adopting more.Artificial neural network carries out distributed storage and parallel processing to information, have self-organization, self study and adaptation function, can approach the ability of any Nonlinear Mapping arbitrarily, be that the tester only need provide network input and just can obtain required network output valve, obtained application widely in fields such as signal Processing, Based Intelligent Control, pattern-recognition, machine vision, nonlinear optimization, automatic target identification, knowledge processing, sensing technologies.
Therefore, adopt wavelet analysis technology that Ultrasonic Detection is carried out signal Processing, and using artificial neural networks realization Intelligent Recognition, be effective means and the development trend that improves ultrasonic detection precision.At present, domestic and international research mechanism is applied to Ultrasonic Detection with signal Processing and Artificial Neural Network and has also done a large amount of research, but mostly its testing result is the qualitative identification to defect type, still crack depth is not provided the equipment of detection by quantitative.
Summary of the invention
Purpose of the present invention provides a kind of accuracy of detection height in shortcoming that overcomes above-mentioned prior art and weak point, and the deal with data amount is big, high efficiency, but the butt welded seam defective is carried out the structure defect ultrasonic on-line intelligent identifying system of Intelligent Recognition.
The present invention also aims to provide the recognition methods of said structure defective ultrasonic on-line intelligent identifying system.
Purpose of the present invention realizes by following proposal: this structure defect ultrasonic on-line intelligent identifying system comprises reflectoscope, high-speed data acquisition card, computing machine, described ultrasonic sound probing device comprises synchronizing circuit, radiating circuit, probe, receiving circuit interconnects successively and forms, the described probe internally provided transducer that has, described high-speed data acquisition card is connected with synchronizing circuit by gating circuit, be connected with probe by broad band amplifier, also be connected with logic control circuit between the microprocessor of described gating circuit and computing machine, described microprocessor is built-in with interconnective wavelet analysis module, the artificial neural network module.
Described transducer is used for changing the high-frequency electrical oscillation energy into acoustic energy by mechanical vibration by the inverse piezoelectric effect.Described gating circuit is mainly used to select flaw echoes and gets rid of and transmit and the Bottom echo signal.Described logic control circuit is used for the collection situation of signal is controlled.
Described microprocessor is connected with keyboard, mouse, display.
The recognition methods of structure defect ultrasonic on-line intelligent identifying system of the present invention, its step comprises:
(1) starts ultra-sonic defect detector, synchronizing circuit is the pulse signal of 200～30ns by frequency 0.5～3MHz emission duration, and triggering radiating circuit, radiating circuit produces a high-frequency emission pulse signal excitation transducer, this transponder pulse signal adds to receiving circuit simultaneously, top at timebase line forms an initial pulse signal, by the inverse piezoelectric effect, transducer changes the high-frequency electrical oscillation energy into acoustic energy by mechanical vibration, is coupled into ultrasound wave and propagates in medium in test specimen;
(2) reflected signal of defective and test specimen is received the circuit reception after transmiting sample, carries out signal through broad band amplifier then and amplifies, and is carried out the collection of ultrasound echo signal by high-speed data acquisition card;
(3) enter the microprocessor of computing machine after the signals collecting, carry out signal Processing by the wavelet analysis module, carry out to obtain after the wavelet compression sign amount of depth of defect, then the sign amount of depth of defect is carried out the defective Intelligent Recognition as neural network input in the artificial neural network module, obtain the depth of defect recognition result.
The relative prior art of the present invention has following advantage and effect:
(1) the present invention handles signal by the wavelet analysis module, and the more convenient needed information of defect recognition that obtains effectively by the Intelligent Recognition of artificial neural network module realization to defective, has improved the precision that detects then greatly.
(2) the present invention's null position of having avoided prior art to need to carry out in the Ultrasonic Detection process is determined the problem in harmony velocity modulation school, has significantly reduced testing staff's workload, has improved work efficiency.
(3) present device transformation input cost is low, the profile of equipment and the appearance similar of laptop computer, easy to carry, system adopts window interface, visual pattern, the user can be suitable for the commercial production scene to the online detection at labour equipment by clicking the functions such as sampling, signal Processing, feature extraction and Intelligent Recognition in the different button realization Ultrasonic Detection processes, has market popularization value preferably.
Description of drawings
Fig. 1 is a structure defect ultrasonic on-line intelligent identifying system structural representation of the present invention;
Fig. 2 is the workflow diagram of structure defect ultrasonic on-line intelligent identifying system of the present invention;
Fig. 3 is the workflow diagram of wavelet analysis module shown in Figure 1;
Fig. 4 is the workflow diagram of artificial neural network module shown in Figure 1;
Fig. 5 is the construction process process flow diagram of artificial neural network module shown in Figure 1;
Fig. 6 is the training process process flow diagram of artificial neural network module shown in Figure 1;
Embodiment
Below in conjunction with embodiment and accompanying drawing the present invention is described in further detail, but embodiments of the present invention are not limited thereto.
Embodiment
As shown in Figure 1, this structure defect ultrasonic on-line intelligent identifying system comprises reflectoscope, high-speed data acquisition card, computing machine, ultrasonic sound probing device comprises that synchronizing circuit, radiating circuit, probe, receiving circuit interconnect successively and forms, probe internally provided have a transducer, high-speed data acquisition card is connected with synchronizing circuit by gating circuit, be connected with probe by broad band amplifier, also be connected with logic control circuit between the microprocessor of gating circuit and computing machine, microprocessor is built-in with interconnective wavelet analysis module, artificial neural network module.
Transducer is used for changing the high-frequency electrical oscillation energy into acoustic energy by mechanical vibration by the inverse piezoelectric effect.
Gating circuit is mainly used to select flaw echoes and gets rid of and transmit and the Bottom echo signal.
Logic control circuit is mainly used to the collection situation of signal is controlled.
The microprocessor of computing machine is connected with keyboard, mouse, display.
As shown in Figure 1, 2, the recognition methods of structure defect ultrasonic on-line intelligent identifying system of the present invention, its step comprises:
(1) starts ultra-sonic defect detector, synchronizing circuit is the pulse signal of 200～30ns by frequency 0.5～3MHz emission duration, and triggering radiating circuit, radiating circuit produces a high-frequency emission pulse signal excitation transducer, this transponder pulse signal adds to receiving circuit simultaneously, top at timebase line forms an initial pulse signal, by the inverse piezoelectric effect, transducer changes the high-frequency electrical oscillation energy into acoustic energy by mechanical vibration, is coupled into ultrasound wave and propagates in medium in test specimen;
(2) reflected signal of defective and test specimen is received the circuit reception after transmiting sample, carries out signal through broad band amplifier then and amplifies, and is carried out the collection of ultrasound echo signal by high-speed data acquisition card;
(3) enter the microprocessor of computing machine after the signals collecting, carry out signal Processing by the wavelet analysis module, carry out to obtain after the wavelet compression sign amount of depth of defect, then the sign amount of depth of defect is carried out the defective Intelligent Recognition as neural network input in the artificial neural network module, obtain the depth of defect recognition result.
Small echo is a kind of special limited length, and mean value is zero waveform.It has two characteristics, and the one, " little " promptly has tight support or approximate tight support in time domain; The 2nd, have the positive and negative undulatory property that replaces.It is defined as follows:
If ψ (t) ∈ is L
2(R), L
2(R) represent square-integrable real number space, i.e. the limited signal space of energy, its Fourier transform is that ψ (ω) satisfies enabled condition:
Claim that ψ (t) is a basic small echo or female small echo.Generating function ψ (t) after flexible and translation, just can be obtained a little wave train.
For continuous situation, little wave train is:
Wherein a is a contraction-expansion factor, and b is a shift factor.
For arbitrary function f (t) ∈ L
2(R) continuous wavelet transform is:
It is inversely transformed into:
As shown in Figure 3, the wavelet analysis module is carried out the step of signal Processing and is:
Selected wavelet function;
(2) determine maximum decomposition scale;
(3) signal that collects is carried out wavelet decomposition, obtain the low frequency part and the HFS of signal;
(4) wavelet coefficient of high-frequency signal is carried out thresholding and handle, can use the method for global threshold that the method for threshold value also can be set by layering in the threshold process process;
(5) wavelet coefficient after the thresholding processing is reconstructed, can obtains the compression result of ultrasonic signal, from the compressed signal of having rejected noise, can extract the sign amount of depth of defect.
As shown in Figure 4, the artificial neural network module is carried out the step of defective Intelligent Recognition and is:
(1) extracts the sign amount that the wavelet analysis module is carried out the depth of defect that obtains after the signal Processing;
(2) because each variable has been represented different physical quantitys, their span possibility difference is very big, so need carry out normalized.Here in order to guarantee that the data after the normalization all are positive numbers, select the extreme difference method that the sign amount of depth of defect is carried out normalized;
(3) the structure artificial neural network carries out Intelligent Recognition.
As shown in Figure 5, the step of structure artificial neural network is:
(1) input and output of determining neural network according to the sign amount and the result to be detected of depth of defect, the defective sign amount here is 1, defects detection result is the degree of depth, also is one, so the input and output node is 1;
(2) selection of the network number of plies, the BP artificial neural network of selecting for use here as the feed-forward type neural network, must have input layer and output layer, so the research of the network number of plies is mainly considered the number of hidden layer.For the BP network, following theorem is arranged: given any ε＞0 and arbitrary function f:[0,1]
n→ R
m, there are three layers of BP network, it can approach f in the ε square error precision arbitrarily.Therefore, the network number of plies is chosen for three layers here;
(3) hidden layer nodal point number purpose deterministic process needs at first rule of thumb that formula carries out tentative calculation to nodal point number, then at concrete application by test and Selection, make network have the hidden layer node number of enough generalization abilities and enough output accuracies;
(4) the connection weights between each node of neural network and the initial value of node threshold value are set, it has been generally acknowledged that the initial value of each weights and threshold value should be set to equally distributed random number between [1,1].
In the type of selecting artificial neural network with after configuring the structure of artificial neural network, can adopt mathematical analysis software matlab6.5, use the activation function of matlab language construct typical case neural network, as shown in Figure 6, carry out BP artificial neural network training process.
As mentioned above, can realize the present invention preferably.
Claims (7)
1, a kind of structure defect ultrasonic on-line intelligent identifying system, it is characterized in that: comprise reflectoscope, high-speed data acquisition card, computing machine, described ultrasonic sound probing device comprises synchronizing circuit, radiating circuit, probe, receiving circuit interconnects successively and forms, the described probe internally provided transducer that has, described high-speed data acquisition card is connected with synchronizing circuit by gating circuit, be connected with probe by broad band amplifier, also be connected with logic control circuit between the microprocessor of described gating circuit and computing machine, described microprocessor is built-in with interconnective wavelet analysis module, the artificial neural network module.
2, by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 1, it is characterized in that, may further comprise the steps:
(1) starts ultra-sonic defect detector, synchronizing circuit is the pulse signal of 200～30ns by frequency 0.5～3MHz emission duration, and triggering radiating circuit, radiating circuit produces a high-frequency emission pulse signal excitation transducer, this transponder pulse signal adds to receiving circuit simultaneously, top at timebase line forms an initial pulse signal, by the inverse piezoelectric effect, transducer changes the high-frequency electrical oscillation energy into acoustic energy by mechanical vibration, is coupled into ultrasound wave and propagates in medium in test specimen;
(2) reflected signal of defective and test specimen is received the circuit reception after transmiting sample, carries out signal through broad band amplifier then and amplifies, and is carried out the collection of ultrasound echo signal by high-speed data acquisition card;
(3) enter the microprocessor of computing machine after the signals collecting, carry out signal Processing by the wavelet analysis module, carry out to obtain after the wavelet compression sign amount of depth of defect, then depth of defect sign amount is carried out the defective Intelligent Recognition as neural network input in the artificial neural network module, obtain the depth of defect recognition result.
3,, it is characterized in that described wavelet analysis module carries out the step of signal Processing and be by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 2:
(1) selected wavelet function;
(2) determine maximum decomposition scale;
(3) signal that collects is carried out wavelet decomposition, obtain the low frequency part and the HFS of signal;
(4) wavelet coefficient to high-frequency signal carries out the thresholding processing;
(5) wavelet coefficient after the thresholding processing is reconstructed, can obtains the compression result of ultrasonic signal, from the compressed signal of having rejected noise, can extract the sign amount of depth of defect.
4, by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 3, it is characterized in that: described thresholding is handled method that adopts global threshold or the method that layering is provided with threshold value.
5,, it is characterized in that described artificial neural network module carries out the step of defective Intelligent Recognition and be by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 2:
(1) extracts the sign amount that the wavelet analysis module is carried out the depth of defect that obtains after the signal Processing;
(2) adopt the extreme difference method that the sign amount of depth of defect is carried out normalized;
(3) the structure artificial neural network carries out Intelligent Recognition.
6, by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 5, it is characterized in that the step of described structure artificial neural network is:
(1) the sign amount according to depth of defect is 1, and result to be detected is the degree of depth, is 1 also, determines that the input and output node of neural network is 1;
(2) choosing the network number of plies is three layers;
(3) the hidden layer nodal point number is carried out tentative calculation, pass through test and Selection at concrete application then, determine hidden layer node number;
(4) the connection weights between each node of neural network and the initial value of node threshold value are set.
7, by the recognition methods of the described a kind of structure defect ultrasonic on-line intelligent identifying system of claim 6, it is characterized in that: the initial value of described each weights and threshold value is set to equally distributed random number between [1,1].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005100357880A CN1743839A (en) | 2005-07-15 | 2005-07-15 | Structure defect ultrasonic on-line intelligent identifying system and identifying method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005100357880A CN1743839A (en) | 2005-07-15 | 2005-07-15 | Structure defect ultrasonic on-line intelligent identifying system and identifying method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1743839A true CN1743839A (en) | 2006-03-08 |
Family
ID=36139299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2005100357880A Pending CN1743839A (en) | 2005-07-15 | 2005-07-15 | Structure defect ultrasonic on-line intelligent identifying system and identifying method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1743839A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187649B (en) * | 2007-12-12 | 2010-04-07 | 哈尔滨工业大学 | Heterogeneous material diffusion welding interface defect automatic identification method |
| CN101975825A (en) * | 2010-09-10 | 2011-02-16 | 江苏大学 | Method for extracting characteristic parameters of acoustic emission signals of drawing part cracks |
| CN101620203B (en) * | 2009-05-13 | 2011-08-10 | 中华人民共和国南通出入境检验检疫局 | Deficiency ultrasonic detecting equipment of mechanical equipment based on wavelet theory |
| CN102279223A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | Identifying and assessing system for deformation and damage of 16Mn steel welded structure on basis of SNF (Strongest Neighbor Filter) and DSD (Deformation Sensitive Decimation) policies |
| CN102279222A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | 16Mn steel load carrying piece fatigue damage state identification system based on spot noise figure (SNF) strategy and direct stream digital (DSD) strategy |
| CN102288679A (en) * | 2011-05-16 | 2011-12-21 | 北京航空航天大学 | System for identifying fatigue damage state of 16Mn steel welding position based on strongest neighbor filter (SNF) strategy and direct stream digital (DSD) strategy |
| CN102495140A (en) * | 2011-11-02 | 2012-06-13 | 清华大学 | Wavelet denoising method and wavelet denoising system both oriented to ultrasonic signals |
| CN102636577A (en) * | 2012-03-23 | 2012-08-15 | 南京航空航天大学 | Method for extracting rim ultrasonic flaw detection signal on the basis of wavelet transformation |
| CN101509604B (en) * | 2009-03-20 | 2012-11-28 | 武汉大学 | Method and device for detecting and assessing deposit in metal pipe |
| CN103048098A (en) * | 2011-10-17 | 2013-04-17 | 中国石油天然气集团公司 | Oil tubing string connector metal-metal sealing surface ultrasonic imaging examination method |
| CN105122009A (en) * | 2013-03-19 | 2015-12-02 | 巴鲁夫公司 | Inductive displacement measuring sensor and method for operating the latter |
| CN105388212A (en) * | 2015-10-23 | 2016-03-09 | 北京大学 | Ultrasonic detecting method of defects in thick-wall composite tubular structure |
| CN106124623A (en) * | 2016-06-20 | 2016-11-16 | 哈尔滨理工大学 | Sheet metal micro-crack identification and alignment system and detection method based on this system |
| CN106932478A (en) * | 2015-12-30 | 2017-07-07 | 中国石油天然气股份有限公司 | Ultrasonic detection method and device |
| CN108226293A (en) * | 2017-12-26 | 2018-06-29 | 中国航空工业集团公司基础技术研究院 | A kind of signal pickup assembly for the transmission parallel detection of sandwich structure ultrasonic reflection |
| CN108318582A (en) * | 2017-12-26 | 2018-07-24 | 中国航空工业集团公司基础技术研究院 | A kind of signal acquisition method for the transmission parallel detection of sandwich structure ultrasonic reflection |
| CN109239206A (en) * | 2018-06-20 | 2019-01-18 | 诸暨市逍遥管道科技有限公司 | The supersonic detection method of defect inspection auxiliary electric fusion joint intelligence phased array |
| CN111855810A (en) * | 2020-07-20 | 2020-10-30 | 济南大学 | Rail foot damage identification method and system based on recurrent neural network |
-
2005
- 2005-07-15 CN CNA2005100357880A patent/CN1743839A/en active Pending
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187649B (en) * | 2007-12-12 | 2010-04-07 | 哈尔滨工业大学 | Heterogeneous material diffusion welding interface defect automatic identification method |
| CN101509604B (en) * | 2009-03-20 | 2012-11-28 | 武汉大学 | Method and device for detecting and assessing deposit in metal pipe |
| CN101620203B (en) * | 2009-05-13 | 2011-08-10 | 中华人民共和国南通出入境检验检疫局 | Deficiency ultrasonic detecting equipment of mechanical equipment based on wavelet theory |
| CN101975825B (en) * | 2010-09-10 | 2012-03-21 | 江苏大学 | Method for extracting characteristic parameters of acoustic emission signals of drawing part cracks |
| CN101975825A (en) * | 2010-09-10 | 2011-02-16 | 江苏大学 | Method for extracting characteristic parameters of acoustic emission signals of drawing part cracks |
| CN102279222A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | 16Mn steel load carrying piece fatigue damage state identification system based on spot noise figure (SNF) strategy and direct stream digital (DSD) strategy |
| CN102288679A (en) * | 2011-05-16 | 2011-12-21 | 北京航空航天大学 | System for identifying fatigue damage state of 16Mn steel welding position based on strongest neighbor filter (SNF) strategy and direct stream digital (DSD) strategy |
| CN102279223A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | Identifying and assessing system for deformation and damage of 16Mn steel welded structure on basis of SNF (Strongest Neighbor Filter) and DSD (Deformation Sensitive Decimation) policies |
| CN103048098A (en) * | 2011-10-17 | 2013-04-17 | 中国石油天然气集团公司 | Oil tubing string connector metal-metal sealing surface ultrasonic imaging examination method |
| CN102495140A (en) * | 2011-11-02 | 2012-06-13 | 清华大学 | Wavelet denoising method and wavelet denoising system both oriented to ultrasonic signals |
| CN102636577A (en) * | 2012-03-23 | 2012-08-15 | 南京航空航天大学 | Method for extracting rim ultrasonic flaw detection signal on the basis of wavelet transformation |
| CN105122009B (en) * | 2013-03-19 | 2017-09-15 | 巴鲁夫公司 | Inductance type force-displacement measurement sensor and the method for operating the sensor |
| CN105122009A (en) * | 2013-03-19 | 2015-12-02 | 巴鲁夫公司 | Inductive displacement measuring sensor and method for operating the latter |
| US10209097B2 (en) | 2013-03-19 | 2019-02-19 | Balluff Gmbh | Inductive displacement measuring sensor and method for operating the latter |
| CN105388212A (en) * | 2015-10-23 | 2016-03-09 | 北京大学 | Ultrasonic detecting method of defects in thick-wall composite tubular structure |
| CN105388212B (en) * | 2015-10-23 | 2018-08-10 | 北京大学 | The supersonic detection method of defect in a kind of thick walled composite tubular shape structure |
| CN106932478A (en) * | 2015-12-30 | 2017-07-07 | 中国石油天然气股份有限公司 | Ultrasonic detection method and device |
| CN106124623A (en) * | 2016-06-20 | 2016-11-16 | 哈尔滨理工大学 | Sheet metal micro-crack identification and alignment system and detection method based on this system |
| CN108226293A (en) * | 2017-12-26 | 2018-06-29 | 中国航空工业集团公司基础技术研究院 | A kind of signal pickup assembly for the transmission parallel detection of sandwich structure ultrasonic reflection |
| CN108318582A (en) * | 2017-12-26 | 2018-07-24 | 中国航空工业集团公司基础技术研究院 | A kind of signal acquisition method for the transmission parallel detection of sandwich structure ultrasonic reflection |
| CN108318582B (en) * | 2017-12-26 | 2021-04-20 | 中国航空工业集团公司基础技术研究院 | Signal acquisition method for ultrasonic reflection and transmission parallel detection of sandwich structure |
| CN108226293B (en) * | 2017-12-26 | 2021-05-07 | 中国航空工业集团公司基础技术研究院 | Signal acquisition device for ultrasonic reflection and transmission parallel detection of sandwich structure |
| CN109239206A (en) * | 2018-06-20 | 2019-01-18 | 诸暨市逍遥管道科技有限公司 | The supersonic detection method of defect inspection auxiliary electric fusion joint intelligence phased array |
| CN109239206B (en) * | 2018-06-20 | 2023-05-23 | 诸暨市逍遥管道科技有限公司 | Ultrasonic detection method for intelligent phased array of auxiliary electric welding joint for automatic defect identification |
| CN111855810A (en) * | 2020-07-20 | 2020-10-30 | 济南大学 | Rail foot damage identification method and system based on recurrent neural network |
| CN111855810B (en) * | 2020-07-20 | 2023-02-03 | 济南大学 | Rail foot damage identification method and system based on recurrent neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1743839A (en) | Structure defect ultrasonic on-line intelligent identifying system and identifying method | |
| Zhang et al. | Deep convolutional neural network probability imaging for plate structural health monitoring using guided waves | |
| CN112232400A (en) | Stainless steel weld ultrasonic defect detection method based on depth feature fusion | |
| Cheng et al. | Automatic defect depth estimation for ultrasonic testing in carbon fiber reinforced composites using deep learning | |
| Legendre et al. | Neural classification of Lamb wave ultrasonic weld testing signals using wavelet coefficients | |
| CN101571513A (en) | Ultrasonic guided wave detection device for quality evaluation of composite laminated plate | |
| CN104502451A (en) | Method for identifying flaw of steel plate | |
| Zhao et al. | Automated quantification of small defects in ultrasonic phased array imaging using AWGA-gcForest algorithm | |
| CN117825516A (en) | Ultrasonic defect detection system and method with accurate positioning | |
| CN107153095A (en) | The health detecting system and its method of work of a kind of composite plate structure | |
| CN2809640Y (en) | Ultrasonic online intelligent recognition system for structural defect | |
| CN106645399B (en) | A kind of damage Detection of Smart Composite Structure appraisal procedure and system | |
| Lei et al. | Ultrasonic pig for submarine oil pipeline corrosion inspection | |
| Queiroz et al. | Damage detection in composite materials using tap test technique and neural networks | |
| CN107727749A (en) | A kind of ultrasonic quantitative detection method based on wavelet packet fusion feature extraction algorithm | |
| Bouden et al. | Signal processing methods for materials defects detection | |
| CN112285651A (en) | Method and system for identifying position of composite impact source based on feature extraction matrix similarity | |
| Cammarata et al. | Application of principal component analysis and wavelet transform to fatigue crack detection in waveguides | |
| CN117349718A (en) | Anti-noise damage classification method for composite material plate structure based on Lamb waves | |
| Selim et al. | Wavelet transform applied to internal defect detection by means of laser ultrasound | |
| Flores Cuenca et al. | A novel flaw detection approach in carbon steel pipes through ultrasonic guided waves and optimized transformer neural networks | |
| Wang et al. | Automated ultrasonic testing for near-surface flaws in CFRP | |
| Hinders et al. | Intelligent structural health monitoring with ultrasonic Lamb waves | |
| CN106442744A (en) | Ultrasonic signal processing method for removing response characteristics of ultrasonic probe | |
| Kim et al. | Acoustic emission reflection signal classification of PVDF-type AE sensor using convolutional neural network-transfer learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |