A signal feature extraction method and processing system based on waveform matching

Abstract

The present invention provides a signal feature extraction method and processing system based on waveform matching, which belongs to the field of non-destructive testing technology. Lamb wave is a special ultrasonic wave that propagates in an elastic thin plate without boundary constraints. Therefore, the corrosion degradation of a metal thin plate can be tested by Lamb wave velocity. The present invention solves the problem that the arrival signal of the Lamb wave excitation signal is weak and difficult to extract by considering the attenuation and dispersion of the Lamb wave excitation signal during the propagation process and then comparing the correlation between the detection signal and the Lamb wave excitation signal section by section.

Description

Technical Field

[0001] The invention belongs to the technical field of nondestructive testing, and specifically relates to a signal feature extraction method and a processing system based on waveform matching. Background Art

[0002] Defects such as cracks introduced during the processing or rolling process of metal sheets will threaten the safety of product manufacturing and use. At the same time, due to external loading and changes in the use environment, the expansion of small defects inside the metal sheet may cause damage accidents. In addition to strict process to ensure the occurrence of defects as little as possible, detection has become an important means for the reliable application of metal sheets. Therefore, it is very necessary to detect and identify metal sheets.

[0003] Ultrasonic Lamb wave testing occupies an important position in the field of non-destructive testing of thin metal plates due to its advantages such as high detection sensitivity, good sound beam directivity, high detection rate of hazardous defects such as cracks, and wide applicability. In addition, the ultrasonic Lamb wave detection method directly analyzes and judges the existence and nature of defects by the signal generated by the interaction between Lamb waves and defects. It is an active detection method. The active detection system takes less time than passive detection and does not require continuous detection. It is very convenient for actual detection applications. Therefore, the ultrasonic Lamb wave detection method has more research value in metal sheet detection technology.

[0004] Nondestructive testing of metal sheets often uses a windowed sinusoidal signal as the excitation incident component. The propagation time of the excitation signal is used to detect the component performance. Due to noise interference, the arrival signal of the excitation signal may be weak and difficult to extract. Summary of the invention

[0005] The present invention considers the attenuation and dispersion of the excitation and compares the correlation between the detection signal and the excitation signal section by section, thereby proposing a signal feature extraction method based on waveform matching.

[0006] The technical solution adopted by the present invention to solve the above technical problems is:

[0007] A signal feature processing system based on waveform matching, the system comprising: a metal sheet to be detected, a transmitting transducer, a computer, an arbitrary waveform generator, a non-contact laser vibrometer and an oscilloscope;

[0008] The non-contact laser vibrometer is arranged on one side of the metal sheet to be detected, and is used to collect detection signals; the signal input end of the oscilloscope is connected to the signal output end of the non-contact laser vibrometer, and is used to receive and store the signal collected by the non-contact laser vibrometer; the signal output end of the oscilloscope is connected to the signal input end of the computer; the signal output end of the computer is connected to the signal input end of the arbitrary waveform generator, and the computer is used to control the waveform and frequency of the output signal of the arbitrary waveform generator; a transmitting transducer is arranged on the metal sheet to be detected, and the signal output end of the arbitrary waveform generator is connected to the metal sheet to be detected, and is used to transmit the excitation signal to the metal sheet to be detected.

[0009] As a further preferred embodiment of the present invention, a signal conversion module is also included, including a junction box and a vibrometer controller; the junction box and the vibrometer controller are arranged between the signal output end of the non-contact laser vibrometer and the signal input end of the oscilloscope, and are used to convert the analog signal collected by the non-contact laser vibrometer into a digital signal, and transmit it to the oscilloscope.

[0010] As a further preferred embodiment of the present invention, a high-voltage amplifier is also included, wherein the signal input end of the high-voltage amplifier is connected to an arbitrary waveform generator, and the signal output end is connected to a transmitting transducer. The high-voltage amplifier is used to amplify the energy of the excitation signal and transmit it to the metal sheet to be detected through the transmitting transducer.

[0011] As a further preferred embodiment of the present invention, the transmitting transducer adopts a piezoelectric ceramic piece.

[0012] As a further preferred embodiment of the present invention, the distance between the non-contact laser vibrometer and the metal sheet to be detected is 1.5-2 m, and the distance between the transmitting transducer and the laser measuring point is 60 mm.

[0013] As a further preferred embodiment of the present invention, a trigger is further included, and the trigger is connected to the transmitting transducer. When the transmitting transducer is working, the non-contact laser vibrometer is notified to collect the pulse wave.

[0014] As a further preferred embodiment of the present invention, the signal sequence recording period of the oscilloscope is 1.6 ms and the sampling frequency is 512 MHz.

[0015] A signal feature extraction method based on waveform matching includes the following steps:

[0016] S1. Set piezoelectric ceramics on the surface of the metal sheet to be detected to excite Lamb waves. The excitation signal is:

[0017]

[0018] Where n(t) is the excitation signal, m(t) is the modulation envelope function, w0 is the center frequency of the signal, ω0 = 2πf, f = 100KHz;

[0019] S2. The excitation signal propagates in a metal thin plate without boundary constraints. Without being disturbed by noise, the theoretical waveform after propagating x distance is:

[0020]

[0021] Where M is the Fourier transform of the m function, e is the natural logarithm, e≈2.71828, w is the other frequency components in the excitation signal;

[0022] S3. When there is noise interference, the excitation signal attenuates during the propagation process, and the attenuation coefficient h(x) is y:

[0023]

[0024] Where x is the propagation distance and ai is the material attenuation coefficient. At this time, the theoretical predicted waveform of the excitation signal at the excitation point x is:

[0025] f(x,t)=hg

[0026] S4. Shift f along the time axis and calculate the correlation coefficient between f and the detection signal y. The calculation formula is:

[0027]

[0028] Where Af is the amplitude of f, Ay is the amplitude of the detection signal, cov(f,y) is the covariance of f and y, var[f] is the variance of f, and var[y] is the variance of y;

[0029] S5. When a certain coefficient approaches 1, it is considered that the local part of the signal is the same as the excitation signal, which is the arrival peak of the excitation signal. The time interval dt between the excitation signal peak and the arrival peak is calculated, and the wave velocity of the Lamb wave is 60mm / dt;

[0030] S6. According to the exfoliation corrosion test method for aluminum alloy processed products, the metal sheet is immersed in the test medium, and the wave velocity of the Lamb wave gradually decreases with the immersion time.

[0031] As a further preferred embodiment of the present invention, in S1, an excitation signal is generated by an arbitrary waveform generator, and the excitation signal is transmitted to the test block to be tested by connecting the piezoelectric ceramic via a data line; the arbitrary waveform generator is selected from 33250A wave signal generator, amplitude set to 2Vpp.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] Lamb wave is a special ultrasonic wave that propagates in an elastic thin plate without boundary constraints. Therefore, the corrosion degradation of a metal thin plate can be tested by Lamb wave velocity. The present invention provides a signal feature extraction method based on waveform matching. By considering the attenuation and dispersion of the Lamb wave excitation signal during the propagation process, and then comparing the correlation between the detection signal and the Lamb wave excitation signal section by section, the problem of weak arrival of the Lamb wave excitation signal and difficulty in extraction is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a structural schematic diagram of a signal feature processing system based on waveform matching;

[0035] Figure 2 It is the waveform diagram of Lamb wave excitation signal;

[0036] Figure 3 The theoretical waveform after propagation distance x and the theoretical predicted waveform;

[0037] Figure 4 is the correlation coefficient, i.e. the detection signal waveform;

[0038] Figure 5 It is the waveform diagram of the excitation signal reaching the peak;

[0039] Figure 6 It is a schematic diagram of the time interval for the excitation signal to reach the peak;

[0040] In the figure, 1. metal sheet to be tested, 2. transmitting transducer, 3. computer, 4. arbitrary waveform generator, 5. non-contact laser vibrometer, 6. oscilloscope, 7. junction box, 8. vibrometer controller, 9. high-voltage amplifier, 10. trigger. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] For examples, please refer to the attached Figure 1 A signal feature processing system based on waveform matching, the system comprises: a metal sheet to be detected 1, a transmitting transducer 2, a computer 3, an arbitrary waveform generator 4, a non-contact laser vibrometer 5 and an oscilloscope 6.

[0043] A non-contact laser vibrometer 5 is arranged on one side of the metal sheet 1 to be detected, and is used to collect detection signals; the signal input end of the oscilloscope 6 is connected to the signal output end of the non-contact laser vibrometer 5, and is used to receive and store the signal collected by the non-contact laser vibrometer 5, the signal output end of the oscilloscope 6 is connected to the signal input end of the computer 2, the signal output end of the computer 2 is connected to the signal input end of the arbitrary waveform generator 3, and the computer 3 is used to control the waveform and frequency of the output signal of the arbitrary waveform generator 4; an emitting transducer 2 is provided on the surface of the metal sheet 1 to be detected, and the signal output end of the arbitrary waveform generator 4 is connected to the metal sheet 1 to be detected, and is used to transmit the excitation signal to the metal sheet 1 to be detected.

[0044] It should be noted that a signal conversion module is also included, including a junction box 7 and a vibrometer controller 8; the junction box 7 and the vibrometer controller 8 are arranged between the signal output end of the non-contact laser vibrometer 5 and the signal input end of the oscilloscope 6, and are used to convert the analog signal collected by the non-contact laser vibrometer 5 into a digital signal and transmit it to the oscilloscope 6.

[0045] It should be noted that a high-voltage amplifier 9 is also included. The signal input end of the high-voltage amplifier 9 is connected to the arbitrary waveform generator 4, and the signal output end is connected to the transmitting transducer 2. The high-voltage amplifier 9 is used to amplify the energy of the excitation signal and transmit it to the metal sheet 1 to be detected through the transmitting transducer 2.

[0046] In this embodiment, the transmitting transducer 2 is a piezoelectric ceramic piece, and the model of the arbitrary waveform generator 3 is 33250A, the model of high voltage amplifier 9 is HA-405, non-contact laser vibrometer 5 model is PSV-400, the model of the junction box 7 is PSV-400, and the model of the vibrometer controller 8 is OFV-5000. The distance between the non-contact laser vibrometer 5 and the metal sheet 1 to be detected is 1.5m, and the laser measurement point is 60mm away from the piezoelectric ceramic. The signal sequence recording period of the oscilloscope is 1.6ms, and the sampling frequency is 512MHz.

[0047] It should be noted that a trigger 10 is also included. The trigger 10 is connected to the transmitting transducer 2. When the transmitting transducer 2 is working, it notifies the non-contact laser vibrometer 5 to collect pulse waves.

[0048] A signal feature extraction method based on waveform matching includes the following steps:

[0049] S1. A piezoelectric ceramic is set on the surface of the metal sheet to be detected to excite Lamb waves. The excitation signal (such as Figure 2 shown) is:

[0050]

[0051] Where n(t) is the excitation signal, m(t) is the modulation envelope function, w0 is the center frequency of the signal, ω0 = 2πf, f = 100KHz;

[0052] S2. The excitation signal propagates in a metal plate without boundary constraints. Without being disturbed by noise, the theoretical waveform after propagating x distance (such as Figure 3 shown) is:

[0053]

[0054] Where M is the Fourier transform of the m function, e is the natural logarithm, e≈2.71828, w is the other frequency components in the excitation signal;

[0055] S3. When there is noise interference, the excitation signal attenuates during the propagation process, and the attenuation coefficient h(x) is y:

[0056]

[0057] Where x is the propagation distance and ai is the material attenuation coefficient. At this time, the theoretical predicted waveform of the excitation signal at the excitation point x (such as Figure 3 shown) is;

[0058] f(x,t)=hg

[0059] S4. Shift f along the time axis and calculate the correlation coefficient between f and the detection signal y (e.g. Figure 4 As shown), the calculation formula is:

[0060]

[0061] Where Af is the amplitude of f, Ay is the amplitude of the detection signal, cov(f,y) is the covariance of f and y, var[f] is the variance of f, and var[y] is the variance of y;

[0062] S5. Figure 5 As shown in the figure, when a certain coefficient approaches 1, it is considered that the local part of the signal is the same as the excitation signal and is the arrival peak of the excitation signal. The time interval dt between the excitation signal peak and the arrival peak is calculated (such as Figure 6 As shown), the wave speed of Lamb wave is 60mm / dt;

[0063] S6. According to the exfoliation corrosion test method for aluminum alloy processed products, the metal sheet is immersed in the test medium, and the wave velocity of the Lamb wave gradually decreases with the immersion time.

[0064] It should be noted that, in S1, an excitation signal is generated by an arbitrary waveform generator, and the piezoelectric ceramic piece is connected via an RS-232 data line to transmit the excitation signal to the test block to be tested; the amplitude of the arbitrary waveform generator is set to 2Vpp.

[0065] The method can be used for other ultrasonic waves, including but not limited to longitudinal waves, transverse waves, Rayleigh waves, Stoneley waves, and Scholte waves. Signal characteristics include but are not limited to direct waves, reflected waves, and transmitted waves. Although the embodiments of the present invention have been shown and described, it is understood by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A signal feature processing method based on waveform matching, characterized in that: The system of the method comprises: a metal sheet to be detected, a transmitting transducer, a computer, an arbitrary waveform generator, a non-contact laser vibrometer and an oscilloscope; The non-contact laser vibrometer is arranged on one side of the metal sheet to be detected, and is used to collect detection signals; the signal input end of the oscilloscope is connected to the signal output end of the non-contact laser vibrometer, and is used to receive and store the signal collected by the non-contact laser vibrometer; the signal output end of the oscilloscope is connected to the signal input end of the computer; the signal output end of the computer is connected to the signal input end of the arbitrary waveform generator, and the computer is used to control the waveform and frequency of the output signal of the arbitrary waveform generator; a transmitting transducer is arranged on the metal sheet to be detected, and the signal output end of the arbitrary waveform generator is connected to the metal sheet to be detected, and is used to transmit the excitation signal to the metal sheet to be detected; The method comprises the following steps: S1. A piezoelectric ceramic sheet is placed on the surface of the metal sheet to be detected to excite Lamb waves. The excitation signal is: Where n(t) is the excitation signal, m(t) is the modulation envelope function, ω0 is the center frequency of the signal, ω0 = 2πf, f = 100KHz; S2. The excitation signal propagates in a metal plate without boundary constraints. Without being disturbed by noise, the theoretical waveform after propagating x distance is: Where M is the Fourier transform of the m function, e is the natural logarithm, e≈2.71828, ω is the other frequency components in the excitation signal; S3. When there is noise interference, the excitation signal attenuates during the propagation process, and the attenuation coefficient h(x) is y: Where x is the propagation distance, a i is the material attenuation coefficient; at this time, the theoretical predicted waveform of the excitation signal at the excitation point x is; f(x, t) = hg; S4. Shift f along the time axis and calculate the correlation coefficient between f and the detection signal y. The calculation formula is: Among them, A f is the amplitude of f, A y is the amplitude of the detection signal, cov(f,y) is the covariance of f and y, var[f] is the variance of f, and var[y] is the variance of y; S5. When a certain coefficient approaches 1, it is considered that the local part of the signal is the same as the excitation signal, which is the arrival peak of the excitation signal. The time interval dt between the excitation signal peak and the arrival peak is calculated, and the wave velocity of the Lamb wave is 60mm / dt; S6. According to the exfoliation corrosion test method for aluminum alloy processed products, the metal sheet is immersed in the test medium, and the wave velocity of the Lamb wave gradually decreases with the immersion time.

2. The signal feature processing method based on waveform matching according to claim 1, characterized in that: The system also includes a signal conversion module, including a junction box and a vibration meter controller; The junction box and the vibrometer controller are arranged between the signal output end of the non-contact laser vibrometer and the signal input end of the oscilloscope, and are used to convert the analog signal collected by the non-contact laser vibrometer into a digital signal and transmit it to the oscilloscope.

3. The signal feature processing method based on waveform matching according to claim 1 is characterized in that: The system also includes a high-voltage amplifier, a signal input end of the high-voltage amplifier is connected to an arbitrary waveform generator, and a signal output end is connected to a transmitting transducer. The high-voltage amplifier is used to amplify the energy of the excitation signal and transmit it to the metal sheet to be detected through the transmitting transducer.

4. The signal feature processing method based on waveform matching according to claim 3 is characterized in that: The transmitting transducer adopts a piezoelectric ceramic piece.

5. The signal feature processing method based on waveform matching according to claim 4 is characterized in that: The distance between the non-contact laser vibrometer and the metal sheet to be detected is 1.5-2m, and the distance between the transmitting transducer and the laser measuring point is 60mm.

6. The signal feature processing method based on waveform matching according to claim 1, characterized in that: The system further comprises a trigger, which is connected to the transmitting transducer and notifies the non-contact laser vibrometer to collect pulse waves when the transmitting transducer is working.

7. The signal feature processing method based on waveform matching according to claim 1 is characterized in that: The signal sequence recording period of the oscilloscope is 1.6 ms, and the sampling frequency is 512 MHz.

8. The signal feature processing method based on waveform matching according to claim 1, characterized in that: In S1, an excitation signal is generated by an arbitrary waveform generator, and the piezoelectric ceramic piece is connected via a data line to transmit the excitation signal to the test block to be tested; the arbitrary waveform generator is selected from 33250A wave signal generator, amplitude set to 2Vpp.

Images