KR101281273B1 - Method and system for determining material properties using ultrasonic attenuation - Google Patents

Abstract

Methods and systems for determining physical properties by ultrasonic attenuation measurements are disclosed. According to the proposed method, the ultrasonic interaction signal measured on the object is compared with the reference signal generated using the reference object but using the same generation and detection setup. The reference ultrasonic signal has low attenuation and exhibits the same diffraction characteristics as the object for the broadband ultrasonic pulse. The difference is due to the damping of the object. The attenuation and attenuation spectra, a function of frequency, are matched to one model to obtain useful parameters for identifying one of the many properties of an object that is changed by ultrasonic attenuation.

Properties, Ultrasonic Attenuation, Interaction Signals, Wideband Ultrasonic Pulses, Diffraction

Description

Property determination system and method using ultrasonic attenuation {METHOD AND SYSTEM FOR DETERMINING MATERIAL PROPERTIES USING ULTRASONIC ATTENUATION}

FIELD OF THE INVENTION The present invention generally relates to nondestructive ultrasound testing, and more particularly to systems and methods for measuring physical properties of objects using ultrasonic attenuation.

Ultrasonic attenuation is a measure of the attenuation of the ultrasonic intensity during propagation through a material and can be used to evaluate physical properties. Ultrasonic attenuation, for example, is directly related to the size of the crystals in polycrystalline solids, for example most metals. Crystal size has a strong influence on the important physical properties of polycrystalline solids. Ultrasonic attenuation can be used to measure the size and concentration of particles contained in a medium in a solid or liquid state, or to measure the pore distribution of a composite material. Another example is to characterize the viscoelastic and relaxation behavior of polymeric materials by combining and utilizing the attenuation and velocity of ultrasonic waves. Physical mechanisms for generating ultrasonic attenuation include scattering and absorption, and both scattering and absorption can be used to characterize physical properties. This physical mechanism (scattering and absorption) is frequency related, that is to say that different attenuation rates are observed at different frequencies. It is well known to make ultrasonic attenuation measurements using a narrowband system (measurement in this case is generally reported at the center frequency), or using a wideband system including frequency domain analysis.

One common technique used for attenuation measurements is known as pulse echo configuration, in which a test object is applied to a test object by a coupling plant or a solid or liquid buffer (ie, a binding medium). Ultrasound is generated and detected by a coupled piezoelectric transducer. Another technique is through-transmission configuration (or transmission mode), where two transducers facing each other on the opposite side of the subject are used for ultrasound reception and transmission. The through transmission structure must be accessible to both sides of the object. In addition, depending on the through delivery structure, the transducer pair must be perfectly matched or fully characterized and preferably aligned with the subject using a binding medium on both sides. The third technique (known as the pitch catch structure) includes a pair of transducers separated from one another on substantially the same side of the subject. The pitch-catch structure is used for measuring the Rayleigh surface wave, the ultrasonic attenuation of the Lamb wave and the ultrasonic attenuation of the bulk wave.

This structure for attenuation measurement can also be used as a non-contact ultrasonic generation and detection technique using electromagnetic acoustic transducer (EMAT), air coupled transducer or laser ultrasound. Laser ultrasound uses one laser with short pulses for the generation of ultrasound. Energy transfer from the laser to ultrasound occurs in thermoelastic systems where thermal expansion on the surface due to abrupt laser heating results in the generation of ultrasound pulses, or ablation where laser energy removes the thin film on the surface and produces a plasma that induces ultrasound Occurs in the system.

A second laser with a long pulse (continuous wave) is generally used for detection. The second laser illuminates the detection position on the surface of the subject and the phase or frequency shift of the reflected light due to the arrival of the attenuated ultrasonic pulse at the detection position is measured using an optical interferometer system. Ultrasonic detection interferometer systems known in the art include time delay interferometry and nonlinear optical based systems for wavefront adaptation. Thompson and D.E. Chimenti AIP Conf. Monchalin J.P, published by Proc., "Laser-ultrasonic: from the laboratory to industry", Review of Progress in Quantitative Nondestructive evaluation Vol. It is described on page 3-31 of 23A. Ultrasonic generation and detection is performed spaced apart, eliminating the need for alignment of conventional ultrasounds and the need for bonding liquids.

Using any of the structures described above, a general method of measuring ultrasonic attenuation is, for example, ultrasonic pulses (amplitude) detected for two transmission distances of an object using two echo signals reflected between the surfaces of the subject. Determining the attenuation of. Attenuation can be calculated by comparing the amplitudes of the two echoes at each frequency, as described in A. Vary in Nondestructive Testing Handbook, V.7, 2nd pp. 383-431 ASNT (1991).

The calculated attenuation is unfortunately affected by the noise of both sounds, i.e. the uncertainty of the two measurements reduces the accuracy of the attenuation value. If the subject is composed of thick and / or high attenuation material, the second echo has a poor signal-to-noise ratio (SNR). In this case, the two-echo attenuation method may not allow accurate measurement, and the SNR of the first echo may not be used sufficiently. Given the limitations of two conventional echo methods, it is highly desirable to use a single echo approach to determine ultrasonic attenuation. However, the amplitude of the echo also depends on generation strength, coupling efficiency, detection efficiency, and so on. In two conventional echo methods, comparison with one echo potentially accounts for all of these factors providing a standardized reading.

Another difficulty with the two conventional echo methods is that it is necessary to modify the diffraction effects of the ultrasonic pulses in order to obtain the inherent ultrasonic attenuation due to the subject. Although a simplified theoretical model has been used to calculate diffraction corrections for simple shaped subjects, the diffraction pattern in practice can be more complicated.

In order to generate a well generalized inherent ultrasonic attenuation spectrum, eliminating variations due to generation intensity, coupling efficiency, detection efficiency and diffraction are the biggest challenges in using a single echo to determine the properties of an object. In some embodiments of the pulse echo structure, normalization may be performed using ultrasonic pulses (ie, incident echoes) reflected by the surface of the subject to characterize the intensity of the generated ultrasonic pulses, Allows the amplitude of the single echo interaction signal to be used. Here, the interaction refers to the interaction between the ultrasonic pulse and the subject subjected to the ultrasonic pulse. Standardization in the transmission structure is often done by comparison with pulses delivered through a coupling medium without the subject. This structure requires access to both sides of the object, which may not be possible in some industrial applications. In addition, the use of incident echo or sampled readout in this case does not eliminate the diffraction correction requirement to obtain inherent object attenuation. There is still a need for a model with accurate information about the characteristics of the system used.

The use of a single echo in laser ultrasound technology to measure the properties of an object compared to the attenuation of a reference object is disclosed in US Pat. No. 6,684,701 to Duboa. Duboa discloses a method for ultrasound that approaches only one side of a sample composite to measure the voids in the sample composite. The method includes measuring a sample ultrasonic signal from a sample composite object, normalizing the sample ultrasonic signal for surface displacement upon generation on the sample composite object, and separating a sample back-wall from the sample ultrasonic signal. do. The sample frequency spectrum of the sample backwall echo is then determined. Next, the method includes measuring a reference ultrasonic signal from the reference composite, normalizing the reference ultrasonic signal for surface displacement upon creation on the reference composite object, and separating the reference backwall echo from the sample ultrasonic signal. . At this time, the reference frequency spectrum of the reference backwall echo is determined. The invention includes obtaining ultrasonic attenuation of the sample composite object as the ratio of the reference frequency spectrum to the sample frequency spectrum over a predetermined frequency range. Comparing the obtained ultrasonic attenuation to a predetermined attenuation standard allows for a porosity evaluation of the sampled composite object.

Duboa's method is limited to the pulse echo structure described above. According to Duboa, in order to standardize each echo, a correction for the change in the generation intensity and the detection efficiency is required by comparing the amplitude and the surface displacement during generation on the surface of the subject. This approach can only be applied to laser-generated ultrasound in thermoelastic systems. In addition, this method of Dubois does not account for the change caused by the transmission of light through the surface of the subject. Unfortunately surface displacement is not an accurate measure of the ultrasonic pulse energy and does not account for the contribution from the transmitted light. For example, in a carbon epoxy compound, the light transmission of the ultrasonic wave generating laser depends on the thickness of the epoxy layer on the surface, which is in fact significantly changed and difficult to determine. In addition, standardization using surface displacement is not applicable to laser generated ultrasound in a removal scheme where the generated laser generates plasma. Although a strong signal is detected at the time of generation, a strong signal, including the contribution of surface displacement, is mainly caused by the refractive index fluctuations of the plasma.

Thus, there is a need for a method and system for obtaining ultrasonic attenuation measurements using a single echo that corrects for diffraction.

According to the present invention there is provided a method and system for determining the physical properties of an object from a single detected ultrasonic pulse transmitted through the object using ultrasonic attenuation.

The present invention provides for any ultrasonic technology (eg piezoelectric transducer, laser ultrasound and EMAT), any waveform (eg longitudinal wave, shear wave and surface wave) and any generation / detection structure (eg pulse echo). , Transmission, pitch-catch). For a given technique, wave modes, structures, and measurements are performed on the object under investigation and on the reference object. The reference object is used to specify the response of the entire system of the measurement (bandwidth, diffraction, etc.) excluding the subject's inherent attenuation and possible changes in signal strength.

Thus, a method of using ultrasonic attenuation to determine the physical properties of an object is provided. The method includes receiving an interaction signal from an ultrasonic detector. Interaction signal refers to one or more types of wideband ultrasonic pulses that hit the detection location of an object after the ultrasonic pulses have been delivered from the object. When delivered through an object, ultrasonic pulses are attenuated by one or more physical mechanisms.

The portion of the interaction signal corresponding to the attenuated ultrasonic pulse is converted from the time domain to the frequency domain to obtain an amplitude spectrum. The portion may be determined by identifying a portion of the interaction signal corresponding to a single form of wideband ultrasonic pulse acting on the detection location. If there are multiple echoes in the interaction signal, the strongest one may be selected. Determining a portion of the interaction signal may include applying a window selection function to selecting a portion of the interaction signal corresponding to a single type of wideband ultrasound pulse acting on the detection location. For example, cross-correlation of the interaction signal and the standard profile of the ultrasonic pulse may be used to focus the shape within the window. Conventional knowledge of the subject, that is, an estimate of the thickness and ultrasonic velocity for the subject in order to estimate the time of arrival of the ultrasound echo, is used to determine the coarse time window of the interaction signal to narrow the partial search within the interaction signal. May be used.

Once the amplitude spectrum is obtained, the amplitude spectrum is compared with the reference amplitude spectrum to obtain an attenuation spectrum. The reference amplitude spectrum is generated in a similar manner to the amplitude spectrum using the reference object. The reference object has the same diffraction characteristics as the test object with respect to the broadband ultrasonic pulse, but the reference amplitude spectrum has little attenuation. This is because, for the reference object, the attenuation parameters used to correct the attenuation of the amplitude spectrum of the reference object are known, or the reference object is selected so that the attenuation is neglected. For example, the reference object may be selected to have a shape corresponding to the subject, and the reference object may be configured of a material having an ultrasonic speed corresponding to the ultrasonic speed of the subject but having little attenuation.

The attenuation spectrum is fitted to a frequency dependent attenuation model to obtain the attenuation parameter. The damping parameter can be used to calculate the properties of the object depending on the damping. Determining the best fit of the attenuation spectra may include using a model that describes a change in signal strength independent of frequency between the subject and the reference object. In particular, the model can account for changes in signal strength independent of frequency using the derivative of the attenuation spectrum to obtain the attenuation parameter.

의 곡선에 비교하는 것을 포함할 수도 있는데,

는 신호 강도 변화를 설명하는 임의의 진폭 오프셋이고, m 및 n은 감쇠의 각각의 근본적인 메커니즘과 관련된 주파수 차수이고, 파라미터 a 및 b는 물성과 관련되어 포함된 메커니즘으로부터의 기여도를 나타내는 값이다. 감쇠에 대한 단일의 근본적인 메커니즘이 오직 하나 존재할 경우 파라미터 a는 0이다. 적합화하는 것은, m 및 n은 고정되고 감쇠 파라미터는 변하는 것을 포함할 수도 있다.For example, the best fit decision can be expressed by measuring the measured attenuation spectrum.

It may also include comparing to the curve of

Is any amplitude offset describing the change in signal strength, m and n are frequency orders associated with each underlying mechanism of attenuation, and parameters a and b are values representing contribution from the included mechanism in relation to physical properties. Parameter a is zero if there is only one fundamental mechanism for attenuation. Adapting may include that m and n are fixed and the attenuation parameter changes.

The receiving and converting step may be applied repeatedly to a plurality of interacting signals from an object, and the method may further include calculating an average of the plurality of signals acting as attenuation spectra.

The method may further include authenticating each of the determined portions and amplitude spectra of the interaction signal using rejection criteria in both the time and frequency domains.

The method may further include using a calibration curve to associate the attenuation parameter to the physical property. This physical property may be, for example, the particle size of the polycrystalline solid, or the porosity of the composite material.

A system for measuring ultrasonic attenuation is also provided to determine the physical properties of the object. The system includes ultrasonic pulse generation and detection setup for injecting wideband ultrasonic pulses into an object, and after passing through the object, a receiver and a signal processor for digitizing the interaction signal of the detected ultrasonic pulses. . The signal processor is designed to convert a portion of the interactive signal corresponding to a single form of pulse detected from the time domain to the frequency domain to obtain an amplitude spectrum.

The signal processor calculates the attenuation spectrum by comparing the amplitude spectrum with the reference amplitude spectrum. The reference amplitude spectrum is obtained from a reference object having the same diffraction characteristics as that of the object. The reference amplitude spectrum has little attenuation. The signal processor matches the attenuation spectrum to a model that determines the attenuation parameter used to determine the physical properties.

The signal processor is adapted to calculate the average of a number of interactive signals from an object at one or more locations to generate an amplitude spectrum. The signal processor may be applied to perform a time or frequency domain averaging process of multiple interacting signals to generate an amplitude spectrum.

The system may further include a controlled motion system that scans the surface of the object. The signal processor may further use the attenuation parameter to generate an image of the physical properties of the object.

Ultrasonic pulse generation and detection setup may include a generation laser for wideband ultrasonic pulse generation and a detection laser coupled to a light detection system for detecting the shape of the wideband ultrasonic pulse. The production laser and detection laser may be directed to the same side of the subject.

A better understanding and benefit of the present invention is provided in the following description with reference to the accompanying drawings, in which like components have the same reference numerals.

1 is a schematic diagram of a system in accordance with one embodiment of the present invention;

2A is a block diagram of a method for determining voids in an object using ultrasonic attenuation.

2B is a block diagram of a method for obtaining a reference spectrum used in the method of FIG. 2A.

3A is an example of a time domain frequency interaction signal obtained using a laser ultrasound system.

3B is an example of a window and a corresponding amplitude spectrum of a laser ultrasonic interaction signal.

4A is an ultrasonic interaction signal obtained from a single measurement at the detection position of an object.

4B is an average ultrasonic interaction signal obtained from measurements at twelve positions of the object.

5A is an example of an object and reference amplitude spectrum.

5B is an attenuation spectrum measured and matched to one model.

6 illustrates an interaction signal obtained in a particle size measurement application.

7A is an example of a calibration curve for the matched attenuation parameter 'b'.

FIG. 7B is a schematic comparison of particle size measured online with a laser ultrasound system against particle size obtained by metallography. FIG.

FIG. 8A is a C scan image showing locations of porosity of a composite sample. FIG.

8B shows attenuation spectra in the void regions of the composite sample.

The present invention provides a method and system for measuring ultrasonic attenuation of a subject using detected, attenuated broadband ultrasonic pulses. This method allows measurements based on attenuated broadband ultrasonic pulses without considering additional echoes, surface displacements, etc. Indeed, a single wideband ultrasound pulse that has been detected and attenuated can be effectively normalized by comparison with a reference broadband ultrasound pulse that corrects the geometrical parameters and diffraction of the subject. The reference wideband ultrasound pulse may be obtained from a reference object having the same diffraction characteristics as the test subject with respect to the wideband ultrasound pulse. Comparison with the reference can be made in the frequency domain by dividing the detected spectrum by the reference spectrum. The important thing is that the reference spectrum is rarely attenuated, because the attenuation through the reference object is negligibly small or is sufficiently characteristic that the attenuation in the reference object is removed from the reference spectrum.

The presented method is particularly effective for wideband ultrasound systems with good response at low frequencies. In addition, this method requires that the attenuation to be measured varies with frequency, which is the case for most mechanisms of ultrasonic interaction of an object.

1 shows an apparatus according to one embodiment when applying the presented ultrasonic attenuation method. In general terms, the apparatus includes a system for generating wideband ultrasonic pulses on a subject 10, a system for detecting wideband ultrasonic pulses after delivery through a portion of the subject 10, and to measure attenuation of the subject 10. A signal processor coupled. The subject 10 being investigated may be of any geometry and, depending on the structure used, only one side of the subject 10 needs to be accessible.

Although other generating means, including those identified above, may be used in other embodiments, in the illustrated system, a generating laser 12 for generating wideband ultrasonic pulses is included. In a presently more preferred embodiment, the generating laser 12 is a pulsed laser that operates in an ablation or thermoelastic regime to include ultrasonic pulses of the subject 10. Appropriate wavelengths and power intensities of the pulsed laser and the location of generation of the surface of the subject 10 may be selected according to the structure and material for generating the ultrasonic pulse in the subject 10 having certain characteristics.

The broadband ultrasonic pulse detection system may include a detection laser 14 and an optical detection system 16. The detection laser 14 may be a long pulse laser or a continuous laser that irradiates a beam over a detection location on the surface of the subject 10. As described above, the detection position and the generation position are the same in the pulse-echo configuration, on opposite sides in the through transmission configuration, and in the pitch-catch configuration. According to them, they are spaced apart (usually on the same side). The detection laser 14 may be a laser of relatively high power to provide a suitable optical power for detection while compensating for the relatively low reflectance of the subject 10 at the detection position.

In some embodiments, generation laser 12 and / or detection laser 14 and photodetection system 16 may move relative to subject 10 to perform ultrasonic measurements at multiple generation and detection location pairs. It is preferable. Multiple measurements can be used individually for detecting the physical properties of the subject 10 at different locations and / or combined against the spatial mean to produce an average measurement of the physical properties of the subject with high accuracy. This can be done, for example, by movement of the subject 10, movement of the device or movement of the optical device.

The detection laser 14 is a time delay interferometer, such as a stabilized confocal Fabry-Perot interferometer, or an optical detection system that may include a nonlinear light component for wavefront applications as previously introduced ( 16). The light detection system 16 outputs an electrical interaction signal to the signal capture and processor unit 18. For example, the signal capture and processor unit 18 may include an analog to digital signal converter (A / D) to digitize the electrical interaction signal.

The signal capture and processor 18 is adapted to transform the interactive signal from the time domain to the frequency domain, for example by applying a Discrete Fourier Transform to the interactive signal to generate an amplitude spectrum of the detected ultrasonic pulses. Is adapted to. Preferably a fast Fourier transform algorithm well known in the art is used. In a preferred embodiment, only a portion of the interaction signal is transformed, which portion is selected by a window selection function such that only relevant ultrasonic echoes appear in the amplitude spectrum and multiple reflections or echoes are not synthesized.

The signal capture and processor 18 is also configured to compare the amplitude spectra of the reference object and the subject 10 to generate an attenuation spectrum. The reference amplitude spectrum is generated in the same manner as the amplitude spectrum of the subject 10. In some embodiments, the reference body is identical in shape to the subject 10 and has the same diffraction characteristics, but the broadband ultrasound pulses appear to be generally not attenuated at all upon delivery through the reference object. Alternatively, an equivalent reference amplitude spectrum may be obtained by using a reference object having a different shape but having equivalent diffraction characteristics. The attenuation reference amplitude spectrum can be generated using a pseudo object with attenuation that is well characterized by correcting the detected amplitude spectrum using equations known in the ultrasonic sciences.

The signal capture and processor 18 also matches the attenuation spectrum to one model. The optimal sum of the curve and the subject attenuation spectrum of the model provides one or more attenuation parameters of the subject. One or more attenuation parameters are then used to derive the physical properties of the subject that change with ultrasonic attenuation. Preferably, the signal capture and processor 18 must access memory for storing interactive signals, amplitude spectra, attenuation spectra, attenuation parameters, and program instructions for performing processing, as well as derived physical characteristics. At least the derived physical characteristic is usually an output 19. One or more stored information may be displayed graphically or numerically, for example by a display unit or other output device. Signal capture and processor 18 may include one general purpose computer.

Figure 2a shows an embodiment of the ultrasonic attenuation measurement method. For clarity, Figures 3A, 3B, 4A, 4B, 5A and 5B are shown in the description of the steps involved. In step 100, an interactive signal indicative of the attenuated wideband ultrasonic pulse is received, for example, via signal receiving hardware from the light detection system 16 of FIG. The interaction signal is a time varying amplitude signal, which amplitude is related to the surface transition due to the arrival of the ultrasonic wave at the detection position of the subject surface.

An example of the interaction signal detected using the apparatus of FIG. 1 of the pitch catch structure is illustrated in FIG. 3A. The interaction signal begins with a disturbance occurrence, in which the light detection system 16 is attacked by the pseudo reflected light from the generating laser, or is strongly modulated by the generated plasma. The production structure can be worsened by using a common wavelength for both the production and detection lasers. The interaction signal then represents the first, second and third echoes of the wideband ultrasound pulse. In the illustrated interaction signal of FIG. 3A, a pitch catch structure was used for the low damping object. The strong diffraction effect in this structure results in a first echo having a lower amplitude than the second and third echoes. Because of the strong initial diffraction of this embodiment, a second echo of a high quality ultrasonic pulse may be used for the attenuation measurement. It also mentions that shear waves are evident in the interaction signal. For purposes of explanation, longitudinal waves are used for the measurement, but it will be understood that other ultrasonic waves may alternatively be used in other embodiments. It is also mentioned that this method can be applied to any ultrasonic mode (long wave, shear wave, surface wave, etc.) of any structure (pulse echo, through transmission, pitch catch, etc.).

Therefore, a mechanism is needed for identifying certain portions of the interaction signal to be analyzed. The present invention contemplates using the window selection function to identify some of the interactive signals related to a single form (eco) of a given ultrasonic pulse. According to the through transmission structure, even if the ultrasonic pulse is not reflected from any wall, the portion of the interaction signal constituting a single form of the ultrasonic pulse is called an echo. If the thickness and ultrasonic velocity of an object are approximately known, it is desirable to narrow the search by preselection of the coarse time window of the echo. Then, a cross-correlation method can be used that includes comparing the echo pulses with a standard profile to help identify echoes from nearby noise and to concentrate on a narrow time window. To determine if an echo can be used, the noise level determined in a conventional manner may be compared with the peak amplitude of the echo. The interaction signal may be rejected if the quality of the echo is too low.

The echo is represented in the frequency domain by the application of a Discrete Fourier Transform (DFT) (step 104) to derive the amplitude spectrum. For example, a fast Fourier transform (FFT) algorithm may be used. Rejection criteria are used in both the time domain and the frequency domain to ensure that echoes can be identified for noise, and that similarity to the expected spectral shape (eg Gaussian) will accommodate the amplitude spectrum. It can be used to decide whether to reject. It is apparent to those skilled in the art that interactive signals are subject to signal processing techniques to prevent aliasing and minimize spectral leakage. 3B schematically illustrates a time domain interaction signal, in which an echo represented by echo A is selected and concentrated into a time window, and an amplitude spectrum is obtained. Pulsed echo structures were used to generate this interaction signal.

For example, if the amplitude spectrum of an ultrasonic pulse echo has insufficient SNR to provide an attenuation spectrum of the quality required to reliably determine the physical properties of the subject, in some embodiments, a plurality of amplitude spectral measurements It is desirable to calculate the average to provide a high accuracy measurement. While averaging is performed in the time domain using the interaction signal, it is desirable to add in the frequency domain. By averaging multiple spectra, higher quality amplitude spectra may be produced. Since the window selection function ensures that the echoes are substantially synchronized, the accuracy of the window selection function becomes more important when averaging multiple spectra. That is, the window selection function temporarily aligns the different echoes so that correctly gated echoes are used to generate amplitude spectra, minimizing errors due to any overlapping offset.

For example, the window selection function has an advantage when correction for the thickness change of the subject at different generation and detection positions is required. The difference in the thickness of the subject at the detection and generation positions causes a difference in time between the generation of the ultrasonic pulse and the appearance of the pulse at the detection position. If there is such a thickness change, each echo may not be temporarily aligned, and the sum of the digital signals does not produce a collective digital signal of good quality. The window selection function is one mechanism for removing the error caused by the temporal offset of the spectrum.

It will be appreciated that many different forms of average calculation can be applied. The present invention addresses frequency domain averaging using complex spectra including amplitude and phase information from the FFT output. The resulting smoothness of the amplitude spectrum indicates an improvement. It will be appreciated that averaging measurements at various locations has several strengths under certain conditions. For example, if there is coherent noise at a phase offset that changes with the location of the detection point on the subject surface, the average at multiple detection locations will statistically improve the SNR. One example of such coherent noise is coherent backscattered grain noise present in the interaction signal of a polycrystalline object.

Thus, optional steps 106-108 may be applied to average. In step 106, it is determined whether there are other amplitude spectra to be averaged. If there are other amplitude spectra to be averaged, the process returns to step 100. The number of amplitude spectra that must provide an acceptable attenuation spectrum can be empirically determined and requires precise requirements for different applications. Once the last amplitude spectrum is obtained, the amplitude spectra are averaged in step 108 and the average amplitude spectrum is processed in step 110.

4A and 4B schematically show the effect of averaging for noise reduction of the interaction signal. In FIG. 4A, one single interaction signal appears to have been received, but twelve interaction signals are averaged in the time domain to produce the average interaction signal shown in FIG. 4B. These twelve signals are added in the time domain without windowing, and are temporarily aligned with respect to the ultrasonic generation, for example when the averages are measured at the same generation and detection positions or the thickness will not change with each measurement. It is suitable for time. Those skilled in the art will appreciate that FIG. 4B illustrates the improved SNR over FIG. 4A.

In step 110, the amplitude spectrum of the subject is compared with the amplitude spectrum of the reference object. As mentioned above, the reference amplitude spectrum may be generated by applying steps 100 to 108 to the reference object instead of the subject, which has the same diffraction characteristics (for ultrasound) as the diffraction characteristic of the subject (usually for ultrasound), Does not have Thus, the reference amplitude spectrum can be derived using the method described below with reference to FIG. 2B.

More precisely, the ultrasonic attenuation (ie, attenuation spectrum) of the subject as a function of frequency is calculated as the ratio between the spectra of the reference object and the spectra of the subject. In general, the attenuation is a function of the distance traveled by the wave, so the attenuation spectrum can be divided by the penetration distance traveled by the ultrasonic pulse in the subject. It will be understood by those skilled in the art that this distance can be predetermined, measured mechanically, or calculated by the travel time between generation and detection or the time between echoes.

For example, the following equation can be used to calculate the attenuation spectrum in decibels:

f is frequency, α is the attenuation spectrum obtained, d is the distance traveled by the ultrasonic pulse, Aref (f) is the amplitude spectrum of the signal obtained from the reference object with low attenuation, and A (f) is the test The amplitude spectrum of an object. It will be appreciated that the attenuation spectrum α is a relative measure of attenuation relative to the reference object. If the attenuation spectrum is matched to the model as described below, no absolute attenuation spectrum is needed to evaluate the physical properties. If the reference object exhibits some known attenuation, this attenuation is easily removed from Aref in equation (1).

5A illustrates a general example of an amplitude spectrum 20 and a reference amplitude spectrum 21 for a subject. It is noted that the reference amplitude spectrum 21 has a larger amplitude than the amplitude spectrum 20. It will be appreciated within the context of the present invention that this occurs whenever the same conditions apply to both the subject and the reference, but this is not necessary. If the sensitivity of the ultrasonic detector or the intensity of the ultrasonic generator are different, the system of the present invention may use a model to calibrate for any amplitude offset. The attenuation spectrum is matched to the attenuation model in step 112. Since the attenuation spectrum is a relative measurement rather than an absolute attenuation measurement, the attenuation spectrum does not account for amplitude variations that are independent of frequency. Thus, matching is preferably performed using a model that describes the pulse intensity at the time of generation and the possible variation in detection sensitivity between the subject and the reference (ie, amplitude offset).

Many physical mechanisms responsible for attenuation can be modeled by the frequency dependent power law. If one such mechanism exists, the measured attenuation can be matched to the following model:

는 주파수에 무관한 신호 강도의 변이를 설명하는 파라미터이고, b는 물성에 관련된 물리적 감쇠 메커니즘을 나타낸다. 이 모델의 임의의 두 파라미터

및 b는 곡선을 한정한다는 것을 알 수 있다. 감쇠 스펙트럼과 곡선과의 일치 측정은 상이한 곡선에 대해 비교하는 것이다. 일치의 측정이 최적인 곡선은 당업계에 잘 알려진 방식으로, 모델에 가장 일치하는 감쇠 스펙트럼의 최적 일치로 결정된다. 일치화의 결과로, 감쇠 파라미터

및 b가 결정된다.n is the frequency order,

Is a parameter describing the variation of signal strength regardless of frequency, and b represents a physical attenuation mechanism related to physical properties. Any two parameters of this model

And b defines a curve. The coincidence measure between the attenuation spectrum and the curve is to compare against different curves. The curve for which the measure of agreement is optimal is determined by the best agreement of the attenuation spectra that best matches the model, in a manner well known in the art. The attenuation parameter as a result of the matching

And b are determined.

는 피험물체의 물리적 특성을 결정하기 위해서는 필요하지 않으며, 이 오프셋에 대한 기여가 확인되지 않으면

는 시스템의 임의의 관련된 파라 미터에 대응되지 않는다. 따라서, 감쇠 스펙트럼의 감쇠 값 오프셋이 무시될 수도 있다. 앞서 정의된 모델은 신호 강도 변화에 대한 보정을 요구하지 않는 주파수 변화 감쇠의 확인을 허용한다. 견고성을 위해, 0 내지 4의 값을 갖는 파라미터 n은 일치과정 동안 바람직하게 고정되어 있다. 예를 들어, 포함된 주된 메커니즘이 레일리 체계에서 산란될 경우, 주파수의 4 제곱에 의존하는 감쇠에 대응하는 n의 값은 4이다.parameter

Is not necessary to determine the physical properties of the subject and if contribution to this offset is not

Does not correspond to any relevant parameters of the system. Thus, the attenuation value offset of the attenuation spectrum may be ignored. The previously defined model allows the identification of frequency change attenuation that does not require correction for signal strength changes. For robustness, the parameter n with a value of 0 to 4 is preferably fixed during the matching process. For example, if the main mechanism involved is scattered in the Rayleigh system, then the value of n corresponding to the attenuation that depends on four squares of frequency is four.

If two physical mechanisms contribute to the attenuation, such as scattering and absorption, the measured attenuation can be matched to the function

m and n are the frequency orders for absorption and scattering, respectively. Similarly, parameters a and b are coefficients of the two mechanisms involved in relation to the physical properties. According to a well accepted model of absorption and scattering, m is about 0.2 to about 1.5 and n is about 1.5 to about 4. Again, for robustness, the parameters m and n are preferably fixed during the matching process and should not be too close. If two frequency orders m and n are close, equation (2) with a frequency order n effective for the two mechanisms involved should be used, in which case the contribution of each mechanism cannot be identified. It will be appreciated that only one or more parameters may be used to calculate the physical properties of the subject.

는 일치 공정이 적용되는 동안 고정되어 있어야 하고 검사에 의해 쉽게 평가될 수 있다.In the function described above, a parameter describing the variation in signal strength

Should be fixed while the matching process is in place and can be easily assessed by inspection.

5B illustrates the measured attenuation spectrum 25 consistent with the model 26. The measured attenuation spectrum 25 is the attenuation measurement relative to the reference attenuation spectrum. It will be appreciated that the curve 26 according to the model that best matches the attenuation measure 25 is one of many curves in the model. Congruence with curve 26 provides both parameter and offset measurements that allow evaluation of one or more physical properties of the subject.

를 제거한다. 감쇠 스펙트럼의 미분계수는 당업계에 잘 알려진 방법을 사용하여 수치적으로 획득될 수 있다.In some embodiments, properties that vary by the first derivative of the attenuation spectrum are required. This is a constant related to the variation in signal strength between the subject and the reference.

Remove it. The derivative of the attenuation spectrum can be obtained numerically using methods well known in the art.

2B illustrates one embodiment of a method for generating a reference amplitude spectrum. In step 120, an interaction signal representing the ultrasonic pulse is obtained after transmission through the path of the reference object. The reference object has sufficiently characterized attenuation characteristics (preferably negligible attenuation), and the path of the reference object has the same diffraction characteristics as the path through the test object used in FIG. 2A. The reference amplitude spectrum is used to correct the response of the entire system (bandwidth, diffraction, etc.). By proper selection of the reference object, the diffraction effect can be made almost identical to both signals and thus can be automatically removed by comparison with Aref (divided by Aref). In order to have the same diffraction operation, the reference object may be selected to have the same shape and ultrasonic velocity, or may be selected to have a combination of these parameters to provide the same diffraction conditions, which will be understood by those skilled in the art. Preferably the reference amplitude spectrum uses exactly the same generation and detection procedures as applied to the subject.

In step 122, the process as described above is applied to identify the time window (ie, the portion of the interaction signal corresponding to a given echo or similar shape of the broadband ultrasound pulse). The interactive signal in the time window is then transformed into the frequency domain (step 124) to produce a reference amplitude spectrum. If the reference object exhibits negligible ultrasonic attenuation, the reference amplitude spectrum is calculated by applying DFT or FFT to the signal. Otherwise, the characterized attenuation is used to correct the generated amplitude spectrum in a manner familiar to those skilled in the art and in the ultrasonic sciences to construct a non-attenuated reference amplitude spectrum. It will be appreciated that the range of attenuation correction is reduced by the use of low damping reference objects.

In a manner similar to that described above for the subject, the spectrum of the reference object may be averaged by multiple measurements made at one or more locations.

Application example

There are many applications of the presented methods and systems.

In a first example, laser ultrasonic attenuation is used to determine the particle size of the steel on the production line. Very often, austenite particle size associated with austenitic decomposition during cooling is the most important metallurgical parameter for determining the mechanical properties of steel. In order to correctly apply the controlled thermal mechanical processing of the steel piece, the austenitic particle size of the steel piece must first be determined. Therefore, the ability to determine the austenite particle size of a production line is what technology seeks.

The method described above is used to quantitatively determine austenite particle size from ultrasonic attenuation for a wide range of particle sizes (20 to 300 μm) and relatively thick objects (up to 30 mm) in seamless steel tibes. It became. The system uses a long pulsed Nd: YAG laser for detection in pitch catch structures and a Q-Switched Neodymium: Yttrium Aluminum Garnet (Nd) for generating ultrasonic pulses. : YAG) laser. The detection light modulated by the ultrasonic pulse incident at the detection position is demodulated by a Fabry-Perot interferometer.

For calibration, different grades of steel samples were heated in a Gleeble thermomechanical simulator in the range of 900-1250 ° C. and held for about 10 minutes to saturate particle growth. Throughout the entire thermal cycle, laser ultrasonic measurements were performed. After quenching for a suitable time (depending on steel grade), the previous austenite particles were revealed by etching and quantitatively characterized by image analysis. Also in order to obtain a reference amplitude spectrum, measurements were performed on a reference object of steel with the same shape and ultrasonic velocity and low attenuation.

Figure 6 illustrates the interaction signal corresponding to the detected ultrasonic pulses on the production line implementation for particle size determination. The structure of the generation and detection equipment is the pitch catch structure described above.

FIG. 7A is a graph of an example of a calibration curve for interpreting the average particle size as a result of the ultrasonic attenuation parameter b, where b is obtained by matching the attenuation spectrum to the model. Particle size is obtained using standard metallographic techniques.

There are many obstacles to accurate online measurement for real-world applications using real products. Since the tube surface reflects the detection laser light weakly, SNR has become an important problem. To improve the attenuation spectrum, amplitude spectra were calculated by averaging the ultrasonic interaction signals obtained at many locations along the tube, so particle size was evaluated throughout the tube segment or the tube as a whole. It has been found that the above-described frequency domain averaging procedure using complex spectra is suitable.

FIG. 7B shows a comparison between the austenitic particle size measured online by a laser ultrasound system and interpreted using the calibration curve of FIG. 7A and the austenitic particle size obtained by metallography after proper quenching of the same tube. . Because of the conditions of the production line, ultrasonic measurements were expected to be less accurate than those performed under controlled laboratory conditions. In a production environment, due to the inherent difficulty in applying an appropriate cooling process that allows for 'decoration' of the previous austenite particle size, the accuracy of particle size measurement using metallography has been reduced. While the accuracy of metallographic particle size is estimated between 0.5 and 1 ASTM, statistical analysis shows that the laser ultrasonic particle sizes found online have at least the same accuracy as the particle sizes obtained from metallography.

A second application of the presented methods and systems is in the field of composite material testing. Test samples of carbon fiber-reinforced plastics (CFRP) were prepared with flat plates having a rectangular face of 100 mm x 120 mm and a thickness of 6.3 mm. A thin layer of paint was applied to the surface of the composite to obtain good ultrasonic generation of the thermoelastic system.

The generation and detection set up used to demonstrate this disclosure included a multimode CO ₂ laser to generate bulk ultrasonic waves. A neodymium: yttrium aluminum garnet (Nd: YAG) laser coupled to a Fabry Parrot interferometer was used for ultrasonic detection for phase demodulation. The generation laser spot size was about 6.5 mm and the detection laser spot size was about 5 mm. The generation and detection spots almost overlapped with the pulse-echo structure. The frequency sensitivity of the system ranged from 1 to 10 MHz. Laser ultrasound measurements were performed to pinpoint the voids inside the subject. Two-dimensional scans were performed on the surface by sweeping parallel lines across the surface. Each measurement position was about 1.47 mm from the adjacent measurement position.

8A is a greyscale C-scan image of the amplitude of the backwall echo. Circular spots S1-S3 are clearly observed to be approximately 18 mm in size and confirm the location of the voids in the test sample.

FIG. 8B illustrates the attenuation spectra in the pore positions S1, S2, S3 identified in FIG. 8A. The reference spectrum used to obtain these attenuation spectra is obtained from a backwall echo at the location at which the object is known to be normal. Those skilled in the art will appreciate that in CFRP materials, there is little ultrasonic attenuation at locations where apparent voids or other defects are apparent. For composites, the attenuation spectrum is expected to be nearly linear over a specific frequency range of 1 to 8 MHz. As is well known in materials science, the pore / volume fraction Pv changes in direct proportion to the first derivative of the attenuation spectrum of the linear region, where the attenuation is in principle caused by the scattering effect. do. therefore,

c is a specific proportionality constant depending on the shape of the object and the void.

Using this model to determine the amount of voids, it will be seen that the variation correction of signal strength is eliminated, and the derivative removes a constant coefficient independent of frequency. The value of c for this sample is 7.5 vol% mm MHz / dB, and the pore amounts estimated in the regions S1, S2, S3 are estimated to be 1.0, 0.8 and 1.5%, respectively. This evaluation is consistent with the slight values associated with the preparation of test samples.

Of course, many other applications of the method may be considered without departing from the spirit and scope of the invention. In addition to the above-described embodiments using laser ultrasound, the presented method provides piezoelectric transducers or EMATs for any wave shape (long wave, shear wave, surface wave, etc.) and for any structure (pulse echo, through transmission, pitch catch, etc.). It can be applied using different ultrasonic techniques such as. For a given technique, wave mode and structure, the test should be applied to the reference object and the subject to produce a low attenuation reference spectrum.

Claims (27)

In the method of determining the physical properties of a test object using ultrasonic attenuation measurement, (a) receiving an interaction signal from an ultrasonic detector at the detected position of the subject in response to the broadband ultrasonic pulse striking the detected position of the subject, wherein the wideband ultrasonic pulse passes through the subject. Attenuated, and the interaction signal is an interaction signal between the ultrasonic pulse and the subject-; (b) converting a portion of the interactive signal corresponding to the attenuated pulse from a time domain to a frequency domain to obtain an amplitude spectrum; (c) comparing the amplitude spectrum with a reference amplitude spectrum to obtain an attenuation spectrum, wherein the reference amplitude spectrum is applied to a reference object having the same diffraction characteristics as the subject for the broadband ultrasound pulse; a comparison step generated by applying (b), wherein the reference amplitude spectrum is considered to have no attenuation; (d) determining a fitting of the frequency dependent attenuation model and the attenuation spectrum to obtain an attenuation parameter; And (e) using the attenuation parameter to calculate the physical properties of the subject, depending on the attenuation; Method for determining physical properties of a subject using ultrasonic attenuation measurement. The method of claim 1, Applying (a) and (b) to the reference object, Selecting a reference object having a known attenuation parameter; And Using the known attenuation parameter to correct for attenuation of the amplitude spectrum of the reference object; Method for determining the physical properties of a subject using ultrasonic attenuation measurements. The method of claim 1, Applying (a) and (b) to the reference object, Selecting a reference object whose attenuation can be ignored; Method for determining the physical properties of a subject using ultrasonic attenuation measurements. 3. The method of claim 2, Selecting the reference object, Selecting a reference object having a shape corresponding to the test object and made of a material having an ultrasonic speed corresponding to the test object; Method for determining the physical properties of a subject using ultrasonic attenuation measurements. The method of claim 1, And determining a portion of the interaction signal by identifying a portion of the interaction signal corresponding to a single form of the broadband ultrasonic pulse encountered at the detection location. . The method of claim 5, Determining a portion of the interaction signal further includes applying a window selection function to select a portion of the interaction signal corresponding to a single type of broadband ultrasonic pulse encountered at the detection location. Method for determining the physical properties of a subject to be used. The method of claim 6, Applying the window selection function further comprises using a cross-correlation of the interaction signal and the ultrasound pulse standard profile to focus the shape within the window. How to determine the physical properties of an object. The method of claim 5, Determining a portion of the interaction signal may include a thickness for the subject to determine a coarse time window of the interaction signal that includes a single echo of a pulse to narrow the partial search within the interaction signal. And using an estimation of the ultrasonic velocity, the method of determining physical properties of the subject using the ultrasonic attenuation measurement. The method of claim 1, The receiving and converting steps are repeatedly applied to a plurality of interacting signals from the subject, and the method further comprises calculating an average of the plurality of signals. How to determine the physical properties of an object. 9. The method of claim 8, Authenticating each of the determined portions and amplitude spectra of the interaction signal using rejection criteria in both the time and frequency domains. The method of claim 1, Determining the optimal sum of the attenuation spectra further includes ascertaining that the attenuation spectra are optimal for a model that accounts for variations in signal strength independent of frequency between the subject and the reference object. Method for determining the physical properties of a subject to be used. 12. The method of claim 11, The steps to determine the best fit are

Means for comparing the degree of fitting of the attenuation spectrum to the curves of

Is an arbitrary amplitude offset describing the signal strength variation, m and n are the frequency orders associated with each potential attenuation mechanism, and parameters a and b represent the contributions made by the mechanism involved in relation to the physical properties. Method for determining the physical properties of a subject using a measurement. The method of claim 12, The parameter a is a method of determining the physical properties of the subject using the ultrasonic attenuation measurement, 0. The method of claim 12, Optimizing the attenuation spectra includes varying the attenuation parameters, m and n being fixed, wherein the physical properties of the subject using ultrasonic attenuation measurements. 12. The method of claim 11, Determining that the attenuation spectrum is optimal for one model comprises using the derivative of the attenuation spectrum to obtain the attenuation parameter. The method of claim 1, A method of determining physical properties of a subject using ultrasonic attenuation measurements, further comprising using a calibration curve that correlates the attenuation parameters to the physical properties of the subject. The method of claim 1, The physical property is a particle size of the polycrystalline solid, the physical property determination method of the subject using ultrasonic attenuation measurement. The method of claim 1, The physical property of the porous material of the composite material, the method of determining the physical properties of the subject using ultrasonic attenuation measurement. delete delete In the ultrasonic attenuation measurement system for determining the physical properties of the subject, (a) ultrasonic pulse generation and detection set up by injecting a broadband ultrasonic pulse into the subject and detecting the pulse after delivery through the subject; (b) a receiver for digitizing the interaction signal of the detected ultrasonic pulse, the interaction signal being an interaction signal between the ultrasonic pulse and the subject; And (c) includes a signal processor, The signal processor, (Iii) convert a portion of the interaction signal corresponding to a single form of the detected pulse from the time domain to the frequency domain to obtain an amplitude spectrum; (Ii) calculating the attenuation spectrum by comparing the amplitude spectrum with a reference amplitude spectrum derived from a reference object having a diffraction characteristic equal to the diffraction characteristic of the subject and having a known attenuation; (Iii) matching the attenuation spectrum to one model to determine the attenuation parameters used to determine the physical properties of the subject. Ultrasonic attenuation measurement system characterized in that. 22. The method of claim 21, And the signal processor is further adapted to calculate an average of the plurality of interacting signals from the subject at one or various locations to generate an amplitude spectrum. 23. The method of claim 22, And the signal processor is further adapted to perform a time or frequency domain averaging process of the plurality of interacting signals to generate an amplitude spectrum. 22. The method of claim 21, The ultrasonic pulse generation and detection setup includes a generation laser for generating wideband ultrasonic pulses and a detection laser coupled to an optical detection system for detecting the shape of the wideband ultrasonic pulses. Ultrasonic attenuation measurement system comprising a. 25. The method of claim 24, And the generating laser and detecting laser are irradiated onto the same side of the subject. 22. The method of claim 21, And a controlled motion system for scanning the subject surface to obtain attenuation spectra corresponding to attenuation along different paths through the subject. 27. The method of claim 26, Wherein said attenuation spectra are used to generate a physical property image of a subject.

Images