WO2024016598A1 - 一种多层pcb单激励多接收涡流检测传感器、系统及方法 - Google Patents

一种多层pcb单激励多接收涡流检测传感器、系统及方法 Download PDF

Info

Publication number
WO2024016598A1
WO2024016598A1 PCT/CN2022/141829 CN2022141829W WO2024016598A1 WO 2024016598 A1 WO2024016598 A1 WO 2024016598A1 CN 2022141829 W CN2022141829 W CN 2022141829W WO 2024016598 A1 WO2024016598 A1 WO 2024016598A1
Authority
WO
WIPO (PCT)
Prior art keywords
excitation
coil
signal
eddy current
receiving
Prior art date
Application number
PCT/CN2022/141829
Other languages
English (en)
French (fr)
Inventor
高斌
白宗璞
赵祥禹
谌梁
罗飞
鲁鹏
唐超
张勇
姜世强
Original Assignee
四川德源管道科技股份有限公司
聪明猪检测技术(成都)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 四川德源管道科技股份有限公司, 聪明猪检测技术(成都)有限公司 filed Critical 四川德源管道科技股份有限公司
Publication of WO2024016598A1 publication Critical patent/WO2024016598A1/zh

Links

Images

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
    • G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
    • G01N27/90—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents
    • G01N27/9013—Arrangements for scanning
    • 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
    • G01N27/90—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents
    • G01N27/904—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws using eddy currents with two or more sensors
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E30/00—Energy generation of nuclear origin
    • Y02E30/30—Nuclear fission reactors

Definitions

  • the invention relates to the technical field of non-destructive testing, and in particular to a multi-layer PCB single excitation multi-receiving eddy current detection sensor, system and method.
  • Pipelines are mostly made of ferromagnetic materials, and most of them are buried underground.
  • the complex underground soil environment causes corrosion and damage to the inside of the pipelines, as well as hidden problems that exist in the production environment. As the service time of the pipelines increases, they may deteriorate. Cause serious damage such as oil and gas leakage to pipelines. Therefore, non-destructive testing of the inside of pipelines is crucial.
  • the defects in pipelines are complex and generally include the following defects: metal corrosion, that is, the metal pipe wall deteriorates due to chemical or electrochemical reactions between the metal pipeline and the underground soil environment; cracks, whose length and width differ greatly, have Long and narrow abnormal features, which are further divided into axial cracks and circumferential cracks; debris, irrelevant materials in the pipeline, which are generally removed by pigs; dents, due to surface plastic deformation of the pipeline under the influence of external forces; and some in the pipeline Abnormalities caused by the production process during the manufacturing process, such as the reduction in wall thickness caused by grinding the pipe wall, the elongated grooves caused by mechanical cutting, or the interlayer separation of the inner wall of the metal, showing a multi-layered state.
  • the current three-axis high-definition magnetic flux leakage detector for identifying internal defects in pipelines can identify defects in different directions, but its length is long and large, making it difficult to pass through small-diameter pipelines and ball valve pipelines. Most of the current eddy current internal detectors can only Detecting defects does not identify them.
  • the prior art CN111257410A discloses a multi-layer PCB differential eddy current detection sensor, which uses four layers of receiving coils connected in series and only outputs one detection signal, which has insufficient resolution of defects.
  • the purpose of the present invention is to overcome the problems existing in non-destructive testing technology in the prior art, and provide a multi-layer PCB single excitation multi-receiving eddy current detection sensor, system and method.
  • the excitation coil is a single-layer PCB square coil, on which an external excitation signal input interface input1 and an output interface output1 are provided.
  • the receiving coil is a multi-layer PCB structure.
  • the receiving coil is a PCB differential square coil composed of two triangular coils. Different layers of the receiving coil are connected through vias; each receiving coil is connected individually. There is a signal input and output interface, and a separate detection signal is output.
  • a multi-layer PCB single excitation multi-receiving eddy current detection sensor includes four receiving coils, and the four receiving coils are respectively located at the four corners of the excitation coil.
  • a multi-layer PCB single excitation multi-receiving eddy current detection sensor the wire diameter of the excitation coil is 0.1925mm-0.3625mm, the line spacing is 0.0885mm-0.1825mm, and the side length is 25mm-40mm.
  • a multi-layer PCB single excitation multi-receiving eddy current detection sensor the receiving coil wire diameter is 0.065mm-0.100mm, the line spacing is 0.065mm-0.100mm, and the side length is 12.5mm-20mm.
  • a multi-layer PCB single-excitation multi-reception eddy current detection system is also provided.
  • the system includes the above-mentioned sensor, and also includes a signal generator and an analog adder.
  • the signal generator is connected to the excitation coil and is used to generate an excitation signal;
  • the analog adder is used to receive the detection signal output by each receiving coil and combine the multiple detection signals into one detection signal.
  • the invention also provides a multi-layer PCB single excitation multi-reception eddy current detection method for processing signals collected by the sensor.
  • the method includes:
  • a multi-layer PCB single excitation multi-receiver eddy current detection method the encoder includes three hidden layers and a fully connected layer, and the hidden layer is a one-dimensional convolution layer.
  • a multi-layer PCB single excitation multi-receiver eddy current detection method includes:
  • xi is the data point of the input signal, is the data point corresponding to the output signal;
  • yi is the data point of the actual signal, are the data points of the compressed signal.
  • a multi-layer PCB single excitation multi-receiver eddy current detection method is used to compress the multi-channel detection signals output by the multiple receiving coils into one detection signal through an encoder, including:
  • a multi-layer PCB single excitation multi-receiver eddy current detection method the preprocessing includes:
  • the present invention uses multiple multi-layer receiving coils.
  • the multiple receiving coils are symmetrical with respect to the excitation coil, minimizing the impact of the excitation coil on the receiving coil.
  • Each receiving coil is individually connected to a signal input.
  • the output interface and a separate detection signal are output.
  • Each receiving coil has different response time and waveform shape to defects, which enables more comprehensive detection of defects and improves the ability to resolve defects.
  • the present invention uses a PCB planar coil, which occupies a small volume and is easy to package, making it easy to install on the internal detector for on-site detection.
  • the excitation coil of the present invention is a single-layer PCB square coil, which generates uniform spiral eddy currents. Defects in all directions will affect the eddy current flow, so it is sensitive to defects in all directions.
  • the receiving coil adopts a differential structure.
  • the differential structure greatly reduces the impact of the excitation primary magnetic field on the detection coil and improves detection sensitivity.
  • the receiving coil adopts a multi-layer structure, increasing the number of coil turns to improve detection sensitivity, lowering the optimal excitation frequency and increasing skin depth.
  • the present invention compresses the multi-channel detection signals output by multiple receiving coils into one detection signal, and restores the combined one-channel detection signal into a multi-channel signal through the decoder.
  • multiple receiving coils are used, only An ADC is used to collect signals, which reduces hardware complexity and power consumption.
  • Figure 1 is a schematic structural diagram of a sensor in an example of the present invention
  • Figure 2 is a schematic structural diagram of an excitation coil in an example of the present invention.
  • Figure 3 is a schematic structural diagram of a receiving coil in an example of the present invention.
  • Figure 4 is a schematic structural diagram of the first layer of the receiving coil in an example of the present invention.
  • Figure 5 is a schematic structural diagram of the second layer of the receiving coil in an example of the present invention.
  • Figure 6 is a schematic structural diagram of the third layer of the receiving coil in an example of the present invention.
  • Figure 7 is a schematic structural diagram of the fourth layer of the receiving coil in an example of the present invention.
  • Figure 8 is a schematic structural diagram of a detection system in an example of the present invention.
  • Figure 9 shows a ferromagnetic X80 flat plate with artificial defects of different sizes and shapes in an example of the present invention
  • Figure 10 is an algorithm framework diagram of a post-processing algorithm in an example of the present invention.
  • Figure 11 is a training network diagram of the post-processing algorithm Autoencoder in an example of the present invention.
  • Figure 12 is a diagram of the detection signals of defects at different angles of the ferromagnetic X80 flat plate using the single excitation multi-reception detection sensor of the present invention in an example of the present invention
  • Figure 13 is a diagram of the detection signals of different shapes of defects on the ferromagnetic X80 flat plate using the single excitation multi-reception detection sensor of the present invention in an example of the present invention
  • Figure 14 is a diagram of the detection signals of defects at different depths of the ferromagnetic X80 flat plate using the single excitation multi-reception detection sensor of the present invention in an example of the present invention; among them, Figures 12, 13, and 14 only use the Receiving coil one and receiving coil two;
  • Figure 15 is a diagram of detection signals of different shapes of ferromagnetic X80 flat plates using the single excitation multi-reception detection sensor of the present invention in an example of the present invention
  • Figure 16 is a diagram of detection signals of defects at different depths of a ferromagnetic X80 flat plate using the single excitation multi-reception detection sensor of the present invention in an example of the present invention.
  • connection should be understood in a broad sense.
  • connection or integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be an internal connection between two components.
  • connection or integral connection
  • connection, or integral connection can be a mechanical connection or an electrical connection
  • it can be a direct connection or an indirect connection through an intermediate medium
  • it can be an internal connection between two components.
  • specific meanings of the above terms in the present invention can be understood on a case-by-case basis.
  • This invention mainly adopts the principle of eddy current detection, and generates uniform spiral eddy currents through excitation coils.
  • Multiple differential receiving coils pick up changes in the magnetic field, merge multiple detection signals into one, collect one signal, and use improved automatic
  • the encoder encodes and decodes the signal to restore the four-channel signal. Through the response time of the four-channel signal to the defect and the change of the defect shape, the basic information of the defect is obtained, thereby improving the resolution of the defect.
  • a multi-layer PCB single excitation multi-receiving eddy current detection sensor including a single excitation coil and multiple receiving coils.
  • Figure 1 is a front view, with the excitation coil on the front and multiple receiving coils.
  • a receiving coil is located behind the excitation coil.
  • the size of the excitation coil layer is equivalent to that of the receiving coil layer.
  • the multiple receiving coils are symmetrical with respect to the excitation coil.
  • the excitation coil is a single-layer square PCB.
  • each receiving coil is a multi-layer PCB structure, and the receiving coil is a PCB differential square coil composed of two triangular coils , different layers of the receiving coil are connected through via holes; each receiving coil is independently connected to a signal input and output interface, and independently outputs a detection signal.
  • the signal input and output interfaces of the receiving coil are all set on level one.
  • via holes are provided between different layers of the receiving coil and connected by copper wires.
  • the receiving coil is set at a certain coupling position corresponding to the excitation coil to ensure that the information of the excitation coil can be received, and the arrangement of multiple receiving coils should be completely symmetrical about the excitation coil. For example, one should be set at each of the four corners of the excitation coil. or multiple receiving coils, or receiving coils may be provided at other positions of the excitation coil. The reason why multiple receiving coils remain symmetrical is for differential.
  • the signal of the receiving coil is mainly affected by the eddy current generated in the excitation coil and the test piece, in order to detect defects and improve detection sensitivity, it is necessary to make the signal of the receiving coil only affected by The influence of eddy current is to reduce the effect of the primary magnetic field of the excitation coil.
  • the receiving coil is set to differential and needs to be symmetrical about the excitation coil, so that the signal from the excitation coil to the receiving coil is close to zero. Try to eliminate the influence of the excitation coil on the receiving coil. When the defect is disturbed When eddy current occurs, the changes caused by the receiving coil will be more obvious.
  • the signal generator generates a sinusoidal signal of a certain frequency and amplifies it.
  • This signal is received by the excitation coil as an external excitation signal, and then drives the excitation coil. Since the excitation coil generates an alternating magnetic field, the principle of eddy current detection is adopted. According to Maxwell's equations Theoretically, an eddy current is formed on the test piece, and the eddy current flows in a uniform spiral eddy current. When the excitation coil encounters a defect, the eddy current flow direction changes at the defect. Because the eddy current changes, the secondary magnetic field generated by the eddy current changes, and then The amplitude and phase of the receiving coil change, thereby accurately detecting defect information.
  • the receiving coils are respectively arranged at different positions of the excitation coil, and the receiving coils at different positions respond to the eddy current changes at the corresponding positions, their response times and waveform shapes to the defects are different. According to the response of the multi-channel receiving coils to the defects, Response time and waveform shape can be used to determine defect characteristics and improve the ability to distinguish defects.
  • a multi-layer PCB single excitation multi-reception eddy current detection sensor As shown in Figure 3, it includes four receiving coils.
  • the receiving coils all adopt a four-layer PCB structure.
  • the four receiving coils are respectively located in the Excite the four corners of the coil.
  • the wire diameter of the excitation coil is 0.1925mm-0.3625mm
  • the line spacing is 0.0885mm-0.1825mm
  • the side length of the entire excitation coil is 25mm-40mm.
  • the receiving coil uses a four-layer PCB square coil.
  • Each square coil is composed of two triangular coils.
  • the side length of each square coil is a 2
  • the copper wire diameter is d 21
  • the line spacing is d 22 .
  • the coil wire The diameter d 21 is 0.065mm-0.100mm
  • the line spacing d 22 is 0.065mm-0.100mm
  • the side length a 2 is 12.5mm-20mm.
  • the first layer is provided with the input interface input2 and the output interface output2 of the receiving coil one
  • the input interface input3 and the output interface output3 of the receiving coil two are the input interface input4 and the output interface output4 of the receiving coil three
  • the input interface input5 and the output interface output5 of the receiving coil four are received.
  • the receiving coil is provided with a total of twenty-four vias via1, via2 to via24 to connect the four receiving coils and the four-layer coils.
  • One end of the first-layer receiving coil coil12 is connected to the interface input2, and the other end is connected to the via hole via5.
  • One end of the second-layer receiving coil coil12 is connected to the first-layer receiving coil coil12 through the via hole via5, and the other end is connected to the via hole.
  • one end of the third-layer receiving coil coil12 is connected to the second-layer receiving coil coil12 through the via hole via2, and the other end is connected to the via hole via6; one end of the fourth-layer receiving coil coil12 is connected to the third-layer receiving coil coil12 through the via hole via6.
  • the other end is connected to the via hole via3 of the fourth layer receiving coil coil11; one end of the third layer receiving coil coil11 is connected to the fourth layer receiving coil coil11 through the via hole via3, and the other end is connected to the via hole via1; the second layer receiving coil coil11 is connected to coil11 of the third-layer receiving coil through via1, and the other end is connected to via via4; coil11 of the first-layer receiving coil is connected to coil11 of the second-layer receiving coil through via4, and the other end is connected to the interface output2.
  • the detection coil since the detection coil is located at the four corners of the excitation coil, it will only respond to the eddy current changes in the covered part, so its response time and waveform shape to the defect are different. According to the response time and waveform shape of the detection coil to the defect, the defect characteristics can be judged , improving the ability to distinguish defects.
  • a multi-layer PCB single excitation multi-receiver eddy current detection system is provided, as shown in Figure 8.
  • the system includes the sensor described in the above embodiment, and also includes a signal generator and an analog adder.
  • the generator is connected to the excitation coil and is used to generate an excitation signal;
  • the analog adder is used to receive the detection signal output by each receiving coil and combine multiple detection signals into one detection signal; the output end of the analog adder is used for signal processing
  • the modules are connected.
  • the eight interfaces including the interface input4 and the output interface output4, the input interface input5 and the output interface output5 of the receiving coil four are all connected to the analog adder.
  • the sensor When used, the sensor is set on the test piece, the signal generator generates a sinusoidal signal of a certain frequency and amplifies it, and then drives the excitation coil. Since the excitation coil generates an alternating magnetic field, according to the theory of Maxwell's equations, a magnetic field is formed on the test piece. Eddy current, the eddy current flows in a uniform spiral shape. When there is a defect on the specimen, the eddy current will be disturbed. The secondary magnetic field generated by the eddy current will change, and then the amplitude and phase of the receiving coil will change, and then the defect will be detected. Four The response time and waveform shape of each receiving coil to the defect are different. According to the response time and waveform shape of the receiving coil to the defect, the defect information is obtained and the shape of the defect is judged.
  • the analog adder combines four detection signals into one signal, and restores the synthesized signal through a decoding algorithm. On the one hand, it can detect defects more comprehensively and improve the resolution of defects. On the other hand, only An ADC is needed to collect the synthesized signal, which reduces hardware complexity and power consumption.
  • a multi-layer PCB single excitation multi-receiver eddy current detection method includes:
  • the post-processing algorithm of the present invention first constructs the algorithm training set data.
  • the experimental test phase it is possible to collect the signals of the multi-channel receiving coils that are not added by the analog adder, and the signals of the multi-channel receiving coils that are added by the adder. signal; when it is actually put into use, multiple receiving coils and adder modules are packaged into one module, and only one synthetic signal can be collected. Therefore, in terms of training set data construction, by scanning and collecting different types of defects (including welds, defects at different angles, depths, and shapes) on different laboratory specimens with a single excitation and multiple receiving coils, we can obtain the data of different types of defects. Multiple detection signals and one synthetic detection signal are the basis for the later training of the algorithm to extract the characteristics of defect types.
  • the above-mentioned signal preprocessing and feature extraction are performed.
  • the information that needs to be extracted needs to include defect angle information and defect shape information. Normalize the collected signals and perform smoothing filtering to remove noise caused by jitter and speed effects.
  • Reusing wavelet transform has the characteristics of being able to fully highlight certain aspects of the problem, and can perform localized analysis of time (space) frequency, carry out multi-scale detailed analysis of the signal, and obtain information in different frequency domains.
  • EMD is used to decompose the signal and obtain the signal components of each layer to extract information of different modes.
  • an Autoencoder training network model was built, the collected data was divided into training sets and data sets according to 7:3, and trained using the K-fold cross-validation method.
  • the purpose of the patent of this invention is to restore the original multi-channel signal through one compressed signal in practical applications. Therefore, the signal compressed by the encoder must be similar to the signal actually obtained by the adder. Therefore, in addition to ensuring that the input and output of the network are similar, the loss function of the training network also needs to ensure that the one-dimensional signal compressed by the encoder is similar to the real one-dimensional signal.
  • the loss function of the Autoencoder training network model includes:
  • xi is the data point of the input signal, is the data point corresponding to the output signal;
  • yi is the data point of the actual signal, are the data points of the compressed signal.
  • the Autoencoder training network model is divided into an encoder part and a decoder part.
  • the encoder consists of three hidden layers and a fully connected layer, where the hidden layer is played by a one-dimensional convolutional layer.
  • the decoder function is the inverse function of the encoder and has a similar structure to the encoder. It is responsible for decoding the compressed signal into a signal that is as close as possible to the original input signal.
  • the algorithm of the present invention first decomposes and restores the output of two receiving coils. When there are two signals as input, the algorithm can accurately decompose the defect morphology characteristics of different angles, different depths and different shapes. After decomposing the conditions of the four receiving coils, the algorithm can accurately decompose the defect morphology characteristics of different shapes and different depths. This is because the four-channel signals contain more complex and comprehensive information, which improves the ability to distinguish defects.
  • FIG. 9 is a schematic diagram of the defects of the ferromagnetic X80 flat plate.
  • the lift-off is 3mm and the speed is 30mm/
  • the detection signal amplitude does not change.
  • the detection signal amplitude changes, and the response state and response time of the multi-channel signal to the defect are different.
  • Figures 12-14 are two-way detection signals at different angles, different shapes, and different depths.
  • Figures 15-16 are four-way detection signals of different shapes and different depths.
  • Figures 12 and 13 have added synthesis One signal, the rest of the figures are all signals decomposed by the algorithm and the actual multi-channel signal.
  • the synthesized signal is the signal after the simulated addition of the multi-channel detection signals in the formal test.
  • the signal decomposed by the algorithm is based on the synthesized signal.
  • One channel signal is the signal obtained by algorithm decomposition
  • the actual multi-channel signal is the signal obtained during the experimental test phase. It can be seen from the signal diagram that the signal decomposed by the algorithm is basically consistent with the actual signal.
  • the algorithm can predict the defect shape represented by the signal. Therefore, a synthesized signal can be obtained from the actual situation to identify and classify defects.

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)

Abstract

一种多层PCB单激励多接收涡流检测传感器、系统及方法，属于无损检测技术领域，传感器包括单个激励线圈和多个接收线圈，多个接收线圈相对于激励线圈对称；激励线圈为单层的PCB方形线圈，激励线圈上设置有外部激励信号输入接口input1和输出接口output1；接收线圈为多层PCB结构，接收线圈是由两个三角形线圈组成的PCB差分方形线圈，接收线圈的不同层之间通过过孔连接；每一个接收线圈上均单独连接有信号输入输出的接口，并单独输出一路检测信号。系统通过模拟加法器，将多路检测信号合并为一路，并结合改进的自编码方法将信号复原，通过综合多路信号对缺陷的响应时间和缺陷形态的变化，提高对缺陷的分辨能力。

Description

一种多层PCB单激励多接收涡流检测传感器、系统及方法 技术领域
本发明涉及无损检测技术领域，尤其涉及一种多层PCB单激励多接收涡流检测传感器、系统及方法。
背景技术
我国油气管道运输处于加速发展阶段，管道内部的无损检测对于在役管道安全工作至关重要。管道多由铁磁性材料构成，大多数埋藏在地下，地下复杂的土壤环境对管道内部的腐蚀破坏，以及管道在生产环境中存在的隐性问题等，随着管道服役时间的增加，都可能会对管道造成油气泄露等严重危坏。因此对管道内部进行无损检测至关重要。
而管道中缺陷性状复杂，一般有以下缺陷：金属腐蚀，即由于金属管道与所处地下土壤环境发生化学或电化学反应，使金属管壁发生劣化；裂纹，其长度与宽度相差较大，具有狭长的异常特征，又分为轴向裂纹和环向裂纹；碎片，管道内无关的物质，一般通过清管器清除；凹陷，因为管道在外力的影响下发生的表面塑性变形；以及一些在管道的制造期间由于生产工艺造成的异常，比如在打磨管壁时造成的壁厚减小，在机械切削时造成的细长凹槽，或是金属内壁出现的层间分离，呈多层状态。除了检测管道内部缺陷，对正常管道的必要组成部分如焊缝、三通、弯头等也需要检出并判断其精确位置和数量。目前针对管道内部缺陷识别的三轴高清漏磁检测器可以识别不同方向缺陷，但其长度较长、体积较大，对小径管道及发球阀管道不易通过，而目前的涡流内检测器大多只能检出缺陷并不能识别。
现有一些涡流线圈阵列的设计虽然提高了对缺陷的分辨能力，但由于线圈数量多且激励接收方式多变，造成明显的功耗问题以及硬件复杂度问题。此外，在现有技术CN111257410A中公开了一种多层PCB差分涡流检测传感器，其采用四层串联的接收线圈，只输出一路检测信号，对缺陷的分辨能力不足。
发明内容
本发明的目的在于克服现有技术中无损检测技术存在的问题，提供了一种多层PCB单激励多接收涡流检测传感器、系统及方法。
本发明的目的是通过以下技术方案来实现的：
提供一种多层PCB单激励多接收涡流检测传感器，
包括单个激励线圈和多个接收线圈，所述多个接收线圈相对于所述激励线圈对称；所述激励线圈为单层的PCB方形线圈，其上设置有外部激励信号输入接口input1和输出接口output1；所述接收线圈为多层PCB结构，所述接收线圈是由两个三角形线圈组成的PCB差分方形线圈，所述接收线圈的不同层之间通过过孔连接；每一个接收线圈上均单独连接有信号输入输出的接口，并单独输出一路检测信号。
作为一优选项，一种多层PCB单激励多接收涡流检测传感器，包括四个接收线圈，所述四个接收线圈分别位于所述激励线圈的四个角。
作为一优选项，一种多层PCB单激励多接收涡流检测传感器，所述激励线圈的线径为0.1925mm-0.3625mm，线间距为0.0885mm-0.1825mm，边长为25mm-40mm。
作为一优选项，一种多层PCB单激励多接收涡流检测传感器，所述接收线 圈线径为0.065mm-0.100mm，线间距为0.065mm-0.100mm，边长为12.5mm-20mm。
还提供一种多层PCB单激励多接收涡流检测系统，系统包括上述传感器，还包括信号发生器和模拟加法器，所述信号发生器与所述激励线圈连接，用于产生激励信号；所述模拟加法器用于接收每一个接收线圈输出的检测信号并将多路检测信号合并为一路检测信号。
本发明还提供一种多层PCB单激励多接收涡流检测方法，用于对所述的传感器采集的信号进行处理，所述方法包括：
搭建Autoencoder训练网络模型，通过编码器将所述多层接收线圈输出的多路检测信号压缩成一路检测信号，并通过解码器将所述合并后的一路检测信号复原为多路信号。
作为一优选项，一种多层PCB单激励多接收涡流检测方法，所述编码器包括三个隐藏层和一个全连接层，所述隐藏层为一维卷积层。
作为一优选项，一种多层PCB单激励多接收涡流检测方法，所述Autoencoder训练网络模型的损失函数包括：
Figure PCTCN2022141829-appb-000001
其中，x i是输入信号的数据点，
Figure PCTCN2022141829-appb-000002
是对应输出信号的数据点；y i是实际一路信号的数据点，
Figure PCTCN2022141829-appb-000003
是压缩信号的数据点。
作为一优选项，一种多层PCB单激励多接收涡流检测方法，所述通过编码器将所述多个接收线圈输出的多路检测信号压缩成一路检测信号，包括：
对多路检测信号进行预处理及特征提取，并对多路检测信号进行压缩降维。
作为一优选项，一种多层PCB单激励多接收涡流检测方法，预处理包括：
对采集到的多路检测信号做平滑滤波处理。
需要进一步说明的是，上述各选项对应的技术特征在不冲突的情况下可以 相互组合或替换构成新的技术方案。
与现有技术相比，本发明有益效果是：
(1)本发明使用多个多层接收线圈，所述多个接收线圈相对于所述激励线圈对称，最大限度的降低激励线圈对接收线圈的影响，每一个接收线圈上均单独连接有信号输入输出的接口，并单独输出一路检测信号，每个接收线圈对于缺陷的响应时间和波形形态不同，对缺陷进行更全面的检测，提高了对缺陷的分辨能力。
(2)本发明采用PCB平面型线圈，占用体积小，且易于封装，便于安装在内检测器上进行实地检测。
(3)本发明的激励线圈为单层的PCB方形线圈，其产生均匀螺旋状涡流，各个方向的缺陷都会影响涡流流动，因此其对各个方向缺陷均敏感。
(4)接收线圈采用差分式结构，差分结构极大的降低了激励一次磁场对检测线圈的影响，提高了检测灵敏度。
(5)接收线圈采用多层结构，增加了线圈匝数以提高了检测灵敏度，并降低最佳激励频率，增加趋肤深度。
(6)本发明将多个接收线圈输出的多路检测信号压缩成一路检测信号，并通过解码器将所述合并后的一路检测信号复原为多路信号，虽然采用多接收式线圈，但是只使用了一路ADC来采集信号，降低了硬件复杂度以及功耗。
附图说明
图1为本发明一示例中的传感器的结构示意图；
图2为本发明一示例中的激励线圈的结构示意图；
图3为本发明一示例中的接收线圈的结构示意图；
图4为本发明一示例中的接收线圈第一层的结构示意图；
图5为本发明一示例中的接收线圈第二层的结构示意图；
图6为本发明一示例中的接收线圈第三层的结构示意图；
图7为本发明一示例中的接收线圈第四层的结构示意图；
图8为本发明一示例中的检测系统的结构示意图；
图9为本发明一示例中的带有不同尺寸、不同形状人工缺陷的铁磁性X80平板；
图10为本发明一示例中的后处理算法的算法框架图；
图11为本发明一示例中的后处理算法Autoencoder的训练网络图；
图12为本发明一示例中利用本发明所述单激励多接收检测传感器对铁磁性X80平板不同角度的缺陷的检测信号图；
图13为本发明一示例中利用本发明所述单激励多接收检测传感器对铁磁性X80平板不同形状的缺陷的检测信号图；
图14为本发明一示例中利用本发明所述单激励多接收检测传感器对铁磁性X80平板不同深度的缺陷的检测信号图；其中，图12、图13、图14只使用了接收线圈中的接收线圈一和接收线圈二；
图15为本发明一示例中利用本发明所述单激励多接收检测传感器对铁磁性X80平板不同形状的检测信号图；
图16为本发明一示例中利用本发明所述单激励多接收检测传感器对铁磁性X80平板不同深度的缺陷的检测信号图。
具体实施方式
下面结合附图对本发明的技术方案进行清楚、完整地描述，显然，所描述 的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
在本发明的描述中，需要说明的是，属于“中心”、“上”、“下”、“左”、“右”、“竖直”、“水平”、“内”、“外”等指示的方向或位置关系为基于附图所述的方向或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。此外，属于“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性。
在本发明的描述中，需要说明的是，除非另有明确的规定和限定，属于“安装”、“相连”、“连接”应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或一体地连接；可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通。对于本领域的普通技术人员而言，可以具体情况理解上述术语在本发明中的具体含义。
此外，下面所描述的本发明不同实施方式中所涉及的技术特征只要彼此之间未构成冲突就可以相互结合。
本发明主要采用涡流检测的原理，通过激励线圈产生均匀的螺旋状涡流，多个差分接收线圈来拾取磁场的变化，将多路检测信号合并为一路，对一路信号进行采集，并通过改进的自编码器将信号编码并解码来复原四路信号，通过四路信号对缺陷的响应时间和缺陷形态的变化，得到缺陷的基本信息，进而提高了对缺陷的分辨能力。
实施例1
在一示例性实施例中，提供一种多层PCB单激励多接收涡流检测传感器， 如图1所示，包括单个激励线圈和多个接收线圈，图1为主视图，正面为激励线圈，多个接收线圈位于激励线圈的后面，激励线圈层与接收线圈层的尺寸相当，所述多个接收线圈相对于所述激励线圈对称；如图2所示，所述激励线圈为单层的PCB方形线圈，其上设置有外部激励信号输入接口input1和输出接口output1；如图3所示，每个接收线圈均为多层PCB结构，所述接收线圈是由两个三角形线圈组成的PCB差分方形线圈，所述接收线圈的不同层之间通过过孔连接；每一个接收线圈上均单独连接有信号输入输出的接口，并单独输出一路检测信号，所述接收线圈的信号输入输出的接口均设置在第一层。
具体地，接收线圈的不同层之间之间设置有过孔，并通过铜线连接。接收线圈设置在对应激励线圈的某一耦合位置处，保证能够接收到激励线圈的信息，并且多个接收线圈的排布应该关于激励线圈完全对称，例如，在激励线圈的四个角分别设置一个或多个接收线圈，也可以是激励线圈的其他位置处设置接收线圈。多个接收线圈保持对称的原因是为了差分，由于接收线圈的信号主要是由激励线圈和试件中产生的涡流两部分影响，为了检测缺陷，提高检测灵敏度，需要使接收线圈的信号尽量只受涡流的影响，即降低激励线圈的一次磁场的作用，接收线圈设置为差分，需要关于激励线圈对称，这样激励线圈对接收线圈的信号接近零，尽量消除激励线圈对接收线圈的影响，当缺陷扰动涡流时，引起接收线圈的变化会更加明显。
进一步地，信号发生器产生一定频率的正弦信号并进行放大，该信号作为外部激励信号被激励线圈接收，然后驱动激励线圈，由于激励线圈产生交变磁场，采用涡流检测的原理，根据麦克斯韦方程组理论，在被测试件上形成涡流，涡流流向为均匀螺旋状涡流，当激励线圈遇到缺陷时，涡流在缺陷处流向发生改变，因为涡流发生了变化，涡流生成的次级磁场发生变化，进而接收线圈的 幅值和相位发生变化，从而准确的检测出缺陷信息。
进一步地，由于接收线圈分别设置在激励线圈的不同位置处，不同位置处的接收线圈对应相应位置的涡流变化响应，因此其对于缺陷的响应时间和波形形态不同，根据多路接收线圈对缺陷的响应时间和波形形态，判断缺陷性状，提高了对缺陷的分辨能力。
实施例2
基于实施例1，提供一种多层PCB单激励多接收涡流检测传感器，如图3所示，包括四个接收线圈，接收线圈均采用四层PCB结构，所述四个接收线圈分别位于所述激励线圈的四个角。
进一步地，所述激励线圈的线径为0.1925mm-0.3625mm，线间距为0.0885mm-0.1825mm，整个激励线圈的边长为25mm-40mm。
接收线圈采用四层的PCB方形线圈，每个方形线圈由两个三角形线圈组成，每个方形线圈边长为a 2，铜线线径为d 21，线间距为d 22，具体地，线圈线径d 21为0.065mm-0.100mm，线间距d 22为0.065mm-0.100mm，边长a 2为12.5mm-20mm，其中，第一层设置有接收线圈一的输入接口input2和输出接口output2，接收线圈二的输入接口input3和输出接口output3，接收线圈三的入接口input4和输出接口output4，接收线圈四的入接口input5和输出接口output5。
如图4-7所示，接收线圈共设置有二十四个过孔via1、via2到via24来进行四个接收线圈及四层线圈之间的连接，以接收线圈一来说明，其由线圈coil11和线圈coil12组成，第一层接收线圈coil12两端一端连接接口input2，另一端连接过孔via5；第二层接收线圈coil12一端通过过孔via5与第一层接收线圈coil12连接，另一端连接过孔via2；第三层接收线圈coil12一端通过过孔via2与第二 层接收线圈coil12连接，另一端连接至过孔via6；第四层接收线圈coil12一端通过过孔via6与第三层接收线圈coil12连接，另一端连接至第四层接收线圈coil11的过孔via3；第三层接收线圈的coil11一端通过过孔via3与第四层接收线圈coil11连接，另一端连接至过孔via1；第二层接收线圈的coil11通过via1与第三层接收线圈的coil11连接，另一端连接至过孔via4；第一层接收线圈的coil11通过via4与第二层接收线圈的coil11连接，另一端连接至接口output2。
进一步地，由于检测线圈位于激励线圈的四角，只会对其覆盖部分的涡流变化响应，因此其对于缺陷的响应时间和波形形态不同，根据检测线圈对缺陷的响应时间和波形形态，判断缺陷性状，提高了对缺陷的分辨能力。
实施例3
在该实施例中，提供一种多层PCB单激励多接收涡流检测系统，如图8所示，系统包括以上实施例中所述的传感器，还包括信号发生器和模拟加法器，所述信号发生器与所述激励线圈连接，用于产生激励信号；所述模拟加法器用于接收每一个接收线圈输出的检测信号并将多路检测信号合并为一路检测信号；模拟加法器输出端与信号处理模块连接，信号处理模块中有解码算法，解码算法将所述合并后的一路检测信号复原为多路信号。
具体地，将激励线圈的input1和output1与激励信号源的信号发生器连接，将接收线圈一的输入接口input2和输出接口output2，接收线圈二的输入接口input3和输出接口output3，接收线圈三的输入接口input4和输出接口output4，接收线圈四的输入接口input5和输出接口output5等八个接口均连接至模拟加法器。
使用时，将传感器设置在被测试件上，信号发生器产生一定频率的正弦信号并进行放大，然后驱动激励线圈，由于激励线圈产生交变磁场，根据麦克斯 韦方程组理论，在被测试件上形成涡流，涡流流向为均匀螺旋状，当试件上存在缺陷时，会使涡流发生扰动，涡流产生的次级磁场会发生变化，进而接收线圈的幅值和相位发生变化，进而检测到缺陷，四个接收线圈对于缺陷的响应时间和波形形态不同，根据接收线圈对缺陷的响应时间和波形形态，获取缺陷信息，判断缺陷形状。
进一步地，模拟加法器将四路检测信号合并成一路信号，并通过解码算法将合成的信号复原，一方面能够对缺陷进行更全面的检测，提高了对缺陷的分辨能力，另一方面，只需使用一路ADC来采集该合成信号，降低了硬件复杂度以及功耗。
实施例4
在该实施例中，提供一种多层PCB单激励多接收涡流检测方法，所述方法包括：
搭建Autoencoder训练网络模型，通过编码器将所述多个接收线圈输出的多路检测信号压缩成一路检测信号，并通过解码器将所述合并后的一路检测信号复原为多路信号。
具体地，本发明的后处理算法，首先进行算法训练集数据的构建，在实验测试阶段，可以采集到未通过模拟加法器加起来的多路接收线圈的信号、通过加法器加起来的一路合成信号；而在真正投入使用时，多个接收线圈和加法模块是封装起来的，成为一个模块，只能采集到一路合成信号。所以对于训练集数据构建方面，通过单激励多接收线圈对实验室不同试件上的不同种类缺陷(包括焊缝，不同角度、深度、形态的缺陷)进行扫描采集，可以获取到不同缺陷种类的多路检测信号与一路合成检测信号，这是算法后期训练提取缺陷种类特 征的基础。
接着进行上述信号的预处理及特征提取，为了做到对管道内的缺陷进行分类，需要提取到的信息需要包括缺陷角度信息，缺陷形态信息。对采集到的信号归一化，做平滑滤波去除掉因为抖动、速度效应引起的噪声。再利用小波变换有能够充分突出问题某些方面的特征，能对时间(空间)频率的局部化分析的特点，对信号进行多尺度的细化分析，获取到不同频域的信息。利用EMD对信号分解，得到的各层信号分量，提取不同模态的信息。
进一步地，搭建Autoencoder训练网络模型，将采集的数据按7:3划分为训练集和数据集，采用K折交叉验证法训练。与一般自编码器训练方式不同，本发明专利的目的是实际应用中，能通过一路压缩信号还原出原始的多路信号。所以要求编码器压缩的一路信号也要和实际通过加法器得到的一路类似。所以训练网络的损失函数除了负责保证网络的输入与输出相近外，还需要保证编码器压缩的一维信号和真实一路信号相似。如图10-11所示，所述Autoencoder训练网络模型的损失函数包括：
Figure PCTCN2022141829-appb-000004
其中，x i是输入信号的数据点，
Figure PCTCN2022141829-appb-000005
是对应输出信号的数据点；y i是实际一路信号的数据点，
Figure PCTCN2022141829-appb-000006
是压缩信号的数据点。
进一步地，Autoencoder训练网络模型分为编码器部分和解码器部分。其中编码器由三个隐藏层及一个全连接层构成，其中隐藏层由一维卷积层充当。负责原始数据特征抽取，将原始信号压缩降维。解码器函数是编码器的反函数，和编码器结构类似，负责将压缩信号解码成尽量接近原输入信号的信号。
本发明的算法首先对两个接收线圈一路输出的情况进行了分解还原，当输入有两路信号时，算法能准确分解出不同角度、不同深度、不同形状的缺陷形貌特征。再对四个接收线圈的情况进行分解，算法能准确分解出不同形状和不 同深度的缺陷形貌特征。这是因为四路信号包含的信息更复杂更全面，提高对缺陷的分辨能力。
实施例5
基于实施例4，提供一种利用本发明所述单激励多接收涡流传感器对铁磁性X80平板不同缺陷的检测方法，图9为铁磁性X80平板缺陷示意图，在提离为3mm，速度为30mm/s的情况下进行检测，当传感器经过无缺陷位置时，检测信号幅度不变，当传感器经过缺陷时，检测信号幅度发生变化，并且多路信号对于缺陷的响应状态与响应时间不同。如图12-14依次为不同角度、不同形状、不同深度的两路检测信号图，图15-16分别是不同形状、不同深度的四路检测信号，其中，图12、图13中加了合成的一路信号，其余的图中均是算法分解的信号以及实际的多路信号，合成的一路信号即为在正式测试中多路检测信号经过模拟加法后的信号，算法分解的信号即根据合成的一路信号通过算法分解得到的信号，实际的多路信号即在实验测试阶段得到的信号。从信号图中可以看到，算法分解的信号与实际的信号基本一致，面对不同的信号，算法均能预测出信号所代表的缺陷形态。因此，可以从实际中得到合成的一路信号来对缺陷进行识别并分类。
以上具体实施方式是对本发明的详细说明，不能认定本发明的具体实施方式只局限于这些说明，对于本发明所属技术领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干简单推演和替代，都应当视为属于本发明的保护范围。

Claims (10)

  1. 一种多层PCB单激励多接收涡流检测传感器，其特征在于，包括单个激励线圈和多个接收线圈，所述多个接收线圈相对于所述激励线圈对称；所述激励线圈为单层的PCB方形线圈，其上设置有外部激励信号输入接口input1和输出接口output1；所述接收线圈为多层PCB结构，所述接收线圈是由两个三角形线圈组成的PCB差分方形线圈，所述接收线圈的不同层之间通过过孔连接；每一个接收线圈上均单独连接有信号输入输出的接口，并单独输出一路检测信号。
  2. 根据权利要求1所述的一种多层PCB单激励多接收涡流检测传感器，其特征在于，包括四个接收线圈，所述四个接收线圈分别位于所述激励线圈的四个角。
  3. 根据权利要求1所述的一种多层PCB单激励多接收涡流检测传感器，其特征在于，所述激励线圈的线径为0.1925mm-0.3625mm，线间距为0.0885mm-0.1825mm，边长为25mm-40mm。
  4. 根据权利要求1所述的一种多层PCB单激励多接收涡流检测传感器，其特征在于，所述接收线圈线径为0.065mm-0.100mm，线间距为0.065mm-0.100mm，边长为12.5mm-20mm。
  5. 一种多层PCB单激励多接收涡流检测系统，其特征在于，包括权利要求1-4中任意一项所述的传感器，还包括信号发生器和模拟加法器，所述信号发生器与所述激励线圈连接，用于产生激励信号；所述模拟加法器用于接收每一个接收线圈输出的检测信号并将多路检测信号合并为一路检测信号。
  6. 一种多层PCB单激励多接收涡流检测方法，其特征在于，用于对权利要求1-4中任意一项所述的传感器采集的信号进行处理，所述方法包括：
    搭建Autoencoder训练网络模型，通过编码器将所述多层接收线圈输出的多路检测信号压缩成一路检测信号，并通过解码器将所述合并后的一路检测信号 复原为多路信号。
  7. 根据权利要求6所述的一种多层PCB单激励多接收涡流检测方法，其特征在于，所述编码器包括三个隐藏层和一个全连接层，所述隐藏层为一维卷积层。
  8. 根据权利要求6所述的一种多层PCB单激励多接收涡流检测方法，其特征在于，所述Autoencoder训练网络模型的损失函数包括：
    Figure PCTCN2022141829-appb-100001
    其中，x i是输入信号的数据点，
    Figure PCTCN2022141829-appb-100002
    是对应输出信号的数据点；y i是实际一路信号的数据点，
    Figure PCTCN2022141829-appb-100003
    是压缩信号的数据点。
  9. 根据权利要求6所述的一种多层PCB单激励多接收涡流检测方法，其特征在于，所述通过编码器将所述多个接收线圈输出的多路检测信号压缩成一路检测信号，包括：
    对多路检测信号进行预处理及特征提取，并对多路检测信号进行压缩降维。
  10. 根据权利要求9所述的一种多层PCB单激励多接收涡流检测方法，其特征在于，预处理包括：
    对采集到的多路检测信号做平滑滤波处理。
PCT/CN2022/141829 2022-07-22 2022-12-26 一种多层pcb单激励多接收涡流检测传感器、系统及方法 WO2024016598A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210864417.7 2022-07-22
CN202210864417.7A CN115201325B (zh) 2022-07-22 2022-07-22 一种多层pcb单激励多接收涡流检测传感器、系统及方法

Publications (1)

Publication Number Publication Date
WO2024016598A1 true WO2024016598A1 (zh) 2024-01-25

Family

ID=83584468

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/141829 WO2024016598A1 (zh) 2022-07-22 2022-12-26 一种多层pcb单激励多接收涡流检测传感器、系统及方法

Country Status (2)

Country Link
CN (1) CN115201325B (zh)
WO (1) WO2024016598A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115201325B (zh) * 2022-07-22 2023-07-21 四川德源管道科技股份有限公司 一种多层pcb单激励多接收涡流检测传感器、系统及方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050007108A1 (en) * 2003-07-11 2005-01-13 Teodor Dogaru Probes and methods for detecting defects in metallic structures
CN101231267A (zh) * 2008-01-31 2008-07-30 华南理工大学 一种多相涡流检测系统
CN111257410A (zh) * 2020-03-23 2020-06-09 四川德源石油天然气工程有限公司 一种多层pcb差分涡流检测传感器
CN113077444A (zh) * 2021-04-01 2021-07-06 复旦大学 一种基于cnn的超声无损检测图像缺陷分类方法
CN114460168A (zh) * 2022-01-24 2022-05-10 辽宁红阳检测有限公司 一种脉冲涡流检测系统及方法
CN114498960A (zh) * 2021-12-24 2022-05-13 深圳大学 一种异物检测装置、方法及无线电能传输系统
CN114518406A (zh) * 2022-02-24 2022-05-20 四川德源管道科技股份有限公司 一种差分涡流谐振检测传感器及系统
CN115201325A (zh) * 2022-07-22 2022-10-18 四川德源管道科技股份有限公司 一种多层pcb单激励多接收涡流检测传感器、系统及方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1720135A1 (de) * 2005-05-06 2006-11-08 BEB Industrie-Elektronik AG Einrichtung zum Feststellen von Dicken und Dickenvariationen
JP2012242358A (ja) * 2011-05-24 2012-12-10 Kobe Steel Ltd 渦流探傷装置
DE102017208874A1 (de) * 2017-05-24 2018-11-29 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Wirbelstromsensor für zerstörungsfreies Prüfen eines Substrats
CN112114028B (zh) * 2020-08-26 2022-06-14 厦门大学 基于多场耦合传感器的螺栓孔边裂纹监测方法
CN113109430B (zh) * 2021-04-14 2023-04-25 湖南大学 柔性差分涡流传感器、检测系统及方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050007108A1 (en) * 2003-07-11 2005-01-13 Teodor Dogaru Probes and methods for detecting defects in metallic structures
CN101231267A (zh) * 2008-01-31 2008-07-30 华南理工大学 一种多相涡流检测系统
CN111257410A (zh) * 2020-03-23 2020-06-09 四川德源石油天然气工程有限公司 一种多层pcb差分涡流检测传感器
CN113077444A (zh) * 2021-04-01 2021-07-06 复旦大学 一种基于cnn的超声无损检测图像缺陷分类方法
CN114498960A (zh) * 2021-12-24 2022-05-13 深圳大学 一种异物检测装置、方法及无线电能传输系统
CN114460168A (zh) * 2022-01-24 2022-05-10 辽宁红阳检测有限公司 一种脉冲涡流检测系统及方法
CN114518406A (zh) * 2022-02-24 2022-05-20 四川德源管道科技股份有限公司 一种差分涡流谐振检测传感器及系统
CN115201325A (zh) * 2022-07-22 2022-10-18 四川德源管道科技股份有限公司 一种多层pcb单激励多接收涡流检测传感器、系统及方法

Also Published As

Publication number Publication date
CN115201325B (zh) 2023-07-21
CN115201325A (zh) 2022-10-18

Similar Documents

Publication Publication Date Title
RU2419787C2 (ru) Система и способ контроля трубопроводов импульсными вихревыми токами
CN112232400B (zh) 一种基于深浅特征融合的不锈钢焊缝超声缺陷检测方法
CN106287240B (zh) 一种基于声发射的管道泄漏检测装置及单传感器定位方法
CN104965023B (zh) 多模态导波工业管道诊断方法
WO2024016598A1 (zh) 一种多层pcb单激励多接收涡流检测传感器、系统及方法
CN104297342B (zh) 一种发动机空气导管电子束焊缝的超声波检测方法
CN1743839A (zh) 一种结构缺陷超声在线智能识别系统及识别方法
CN107121501A (zh) 一种汽轮机叶轮缺陷分类方法
CN111257410A (zh) 一种多层pcb差分涡流检测传感器
CN101281166A (zh) 一种永磁磁扰动无损检测方法与装置
Shi et al. Signal extraction using complementary ensemble empirical mode in pipeline magnetic flux leakage nondestructive evaluation
CN102520063B (zh) 连续管在役检测评价方法及其系统
CN103713054A (zh) 一种管道近焊缝区缺陷导波特征信号提取方法
CN2809640Y (zh) 一种结构缺陷超声在线智能识别系统
CN105953081A (zh) 一种判识油管线破漏的预警装置及预警方法
Lei et al. Ultrasonic pig for submarine oil pipeline corrosion inspection
Jie et al. Application of a morphological non-sampling wavelet method to the online detection signal processing of coal mine wire rope
CN106290587B (zh) 基于svd分解的储罐底板超声导波检测信号降噪算法
CN211206377U (zh) 检测非铁磁性换热器管束缺陷的灵敏度标定样管
Deng et al. Simulation analysis of eddy current testing parameters for surface and subsurface defect detection of aviation aluminum alloy plate
CN219245435U (zh) 基于交叉激励的管道涡流内检测器
Jiang et al. Defect signal detection of station process pipelines based on the BP neural network
CN201716298U (zh) 钢管超声波探伤装置
CN100535652C (zh) 管道中缺陷的自适应重构方法
CN110673199B (zh) 一种基于低频声压信号分析的u型管堵塞状态评估方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22951854

Country of ref document: EP

Kind code of ref document: A1