Nondestructive evaluation method for structure

Abstract

To provide a novel method capable of evaluating an adhesive layer between thin plate parts.SOLUTION: A nondestructive evaluation method for a structure (M) in which a first thin plate part (1), an adhesive layer (3), and a second thin plate part (2) are sequentially laminated, comprises the following steps: transmitting ultrasonic waves to the first thin plate part in an oblique direction; receiving ultrasonic waves that have propagated through the adhesive layer from either the first or second thin plate part in an oblique direction; performing a Fourier transform on the received ultrasonic waves to obtain a measurement spectrum in a frequency domain; calculating a theoretical spectrum in the frequency domain corresponding to parameters characterizing the adhesive layer in a dispersion relation of Lamb waves in a model based on the structure; and evaluating the adhesive layer of the structure on the basis of a specific value, which is the parameter value when the theoretical spectrum matches the measurement spectrum.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a method for non-destructive evaluation of a structure having an adhesive layer. [Background technology]

[0002] A method has been proposed for non-destructively evaluating the adhesive state of plate materials using plate waves (Lamb waves), a type of guided wave obtained by exciting (resonating) ultrasonic waves. Related information can be found in, for example, the following documents: [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Adams, RD, Drinkwater, BW, NDT&E International,Vol.30, No.2(1997), pp.93-98 [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 63-175762 Summary of the Invention [Problem to be solved by the invention]

[0005] Non-Patent Document 1 uses ultrasonic perpendicular flaw detection to image the bonding area and detect defects. In this case, it is necessary to repeatedly scan the ultrasonic probe (probe) over a wide range of the surface to be inspected, which takes a long time for evaluation.

[0006] In Patent Document 1, ultrasonic waves are emitted from a probe abutting on the flange of a honeycomb sandwich panel in which a flange and a core are bonded, and the bonded portion (bonding length) is inspected based on the echo height of the transmitted wave that passes through the bonded portion between the flange and the core. In this case, the relationship between bond length and echo height, which shows an inverse correlation, (calibration curve, etc.) must be obtained in advance by experiment, etc.

[0007] The present invention has been made in view of the above circumstances, and aims to provide a new method for evaluating adhesive layers between thin plate portions. [Means for solving the problem]

[0008] As a result of intensive research, the inventors have newly discovered that adhesive layers can be evaluated by comparing a measured spectrum in the frequency domain obtained from ultrasonic waves passing through an actual adhesive layer with a theoretical spectrum calculated from the dispersion relation (curve) of ultrasonic waves (Lamb waves) using the characteristic quantities of the adhesive layer as input parameters. By expanding on this result, the present invention, which will be described below, has been completed.

[0009] <<Non-destructive evaluation method for structures>> The present invention is a non-destructive evaluation method for a structure in which a first thin plate portion, an adhesive layer, and a second thin plate portion are stacked in order, and includes a transmitting step of transmitting ultrasonic waves in an oblique direction to the first thin plate portion, a receiving step of receiving the ultrasonic waves propagated through the adhesive layer in an oblique direction from the first thin plate portion or the second thin plate portion, a measuring step of Fourier transforming the received ultrasonic waves to obtain a measured spectrum in the frequency domain, a theoretical calculation step of calculating a theoretical spectrum in the frequency domain corresponding to parameters characterizing the adhesive layer from the dispersion relation of a Lamb wave for a model based on the structure, and an evaluation step of evaluating the adhesive layer of the structure based on specific values ​​that are parameter values ​​when the theoretical spectrum matches the measured spectrum.

[0010] According to the non-destructive evaluation method for structures of the present invention (simply referred to as the "evaluation method"), it is possible to efficiently grasp the shape (size, etc.) or condition (presence or absence of defects, etc.) of the adhesive layer between thin plate portions without relying on time-consuming probe scanning or experimental creation of calibration curves.

[0011] Evaluation Device The present invention can be understood as not only an evaluation method but also an evaluation device. For example, the present invention may be a nondestructive evaluation device for a structure in which a first thin plate portion, an adhesive layer, and a second thin plate portion are stacked in this order, the nondestructive evaluation device for a structure including: a transmitting means for transmitting ultrasonic waves obliquely to the first thin plate portion; a receiving means for receiving the ultrasonic waves propagated via the adhesive layer obliquely from the first thin plate portion or the second thin plate portion; a measuring means for Fourier transforming the received ultrasonic waves to obtain a measured spectrum in the frequency domain; a theoretical calculation means for calculating a theoretical spectrum in the frequency domain corresponding to parameters characterizing the adhesive layer from the dispersion relation of a Lamb wave related to a model based on the structure; and an evaluation means for evaluating the adhesive layer of the structure based on specific values ​​that are parameter values ​​when the theoretical spectrum matches the measured spectrum.

[0012] "system" The present invention can also be understood as a production system that reflects (feeds back) the results obtained by the evaluation method and evaluation device in the production of a structure (adhesion process, joining process, etc.).The present invention can also be understood as a test system in which the above-mentioned evaluation device is incorporated into a test device (tensile tester, fatigue tester, etc.).

[0013] "others" (1) The elements of a method, "steps," and the elements of a product, "means," are interchangeable. Steps or means executable by a computer are also elements of a program (including recording media, data structures, etc.) that can be loaded into a computer.

[0014] (2) In this specification, "adhesion" refers to any adhesive that allows ultrasonic waves (especially Lamb waves) to propagate between the adhesive layer and the first thin plate portion and / or the second thin plate portion. Specifically, the adhesive layer is not limited to one that firmly joins the first thin plate portion and the second thin plate portion with an adhesive or the like, but may be one that only adheres to the first thin plate portion or the second thin plate portion.

[0015] (3) Unless otherwise specified, "x to y" in this specification includes a lower limit value x and an upper limit value y. Any numerical value included in the various numerical values ​​or numerical ranges described in this specification may be used as a new lower limit or upper limit value to create a new range such as "a to b." Unless otherwise specified, "x to y MHz" in this specification means x MHz to y MHz. The same applies to other unit systems (mm, etc.). [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a schematic diagram of a bonded body and an evaluation device. [Figure 2] The FEM analysis model is shown below. [Figure 3] This is a diagram showing the relationship between frequency and wave number obtained by FEM analysis. [Figure 4] This is a dispersion curve based on theoretical calculations. [Figure 5] FIG. 1 is a schematic diagram showing the propagation form of ultrasonic waves (Lamb waves). [Figure 6] The measured spectrum (analytical value) and the theoretical spectrum (calculated value) are shown. [Figure 7] 10 is a flowchart showing a procedure for evaluating the bond length of a bonded body from a measured spectrum and a theoretical spectrum. [Figure 8] FIG. 1 is a schematic diagram of a bonded body used in an experiment. [Figure 9] 1 shows the relationship between the actual length and the evaluation length of the adhesive layer of the bonded body. [Figure 10] FIG. 10 is a schematic diagram showing another embodiment of the bonded body. [Figure 11] This is a formula used to search for the theoretical spectrum corresponding to the measured spectrum. DETAILED DESCRIPTION OF THE INVENTION

[0017] One or more components selected from this specification may be added to the above-described components of the present invention. The contents described in this specification may also be applied to devices, systems, etc. as appropriate. Which embodiment is best depends on the target, required performance, etc.

[0018] 《Structure》 The structure to be evaluated is composed of a first thin plate portion, an adhesive layer, and a second thin plate portion stacked in that order. The specific shape and use of the structure are not important as long as it can be evaluated by the present invention.

[0019] (1) The thin plate (part) is preferably one that, when ultrasonic waves are incident at an oblique angle, causes guided waves (Lamb waves) excited (resonated) within the plate to propagate along the in-plane direction. The specific thickness of the plate is not important, but the plate thickness should be smaller than the wavelength of the longitudinal ultrasonic waves within the material. Specifically, the thickness of the thin plate part (plate thickness) is, for example, 0.5 to 10 mm or 1 to 5 mm. The plate thickness may be constant or may vary between the ultrasonic wave transmission position and the ultrasonic wave reception position.

[0020] The thin plate portion may be made of any material such as metal, resin, or composite material, as long as it allows propagation of Lamb waves. Composite materials are made of, for example, fillers (reinforcing fibers, reinforcing particles, etc.) and matrices (resins, metals, etc.). Typical composite materials include carbon fiber reinforced plastics (CFRP) and glass fiber reinforced plastics (GFRP).

[0021] The thin plate portion is not limited to a flat plate shape, but may also be curved or have a curved surface. In the case of a non-flat thin plate portion, the incident angle and exit angle of the ultrasonic wave are angles relative to the tangent plane or tangent line. The terms "first" and "second" used in this specification are used for convenience, and the first thin plate portion and the second thin plate portion may be the same or different in thickness, material, shape, etc.

[0022] (2) The adhesive layer is preferably interposed between the joined surfaces of the first thin plate portion and the second thin plate portion, and adheres or bonds these joined surfaces together. The adhesive layer is usually made of a material (e.g., adhesive, brazing material, etc.) different from that of the first thin plate portion or the second thin plate portion. For example, an adhesive layer formed with a general adhesive is a resin layer. The adhesive layer is usually thinner than the thin plate portions, and although the specific thickness is not critical, it is, for example, about 0.01 to 2 mm, or 0.1 to 1 mm.

[0023] An example of a parameter (feature, index value) that characterizes the adhesive layer to be evaluated is the length of the adhesive layer (referred to as "adhesion length"). The adhesion length can usually be considered as the range in which a line connecting the ultrasonic incident position (transmission position) and the ultrasonic emission position (reception position) intersects with the adhesive layer (the length of the line segment that is cut off from the line by the adhesive layer). The adhesion length is not limited, but may be, for example, 0 to 200 mm, 5 to 150 mm, or 15 to 100 mm. The direction from the incident position (transmission position) to the emission position (reception position) is referred to as the ultrasonic propagation direction, as appropriate.

[0024] There is no particular purpose for evaluating the adhesive layer, but examples include detecting initial defects during the manufacture of a structure, and understanding changes (deterioration, etc.) in the structure after the fact or over time.

[0025] Ultrasonic / Lamb Wave Ultrasonic waves incident on the thin plate at an oblique angle become Lamb waves excited under certain conditions and propagate in the in-plane direction through the thin plate and adhesive layer. The frequency of the ultrasonic waves is, for example, 0.1 to 5 MHz, 0.3 to 3 MHz, or 0.5 to 1.5 MHz.

[0026] The ultrasonic wave transmission angle (incident angle) and reception angle (emission angle) should be approximately equal, for example, within ±3°, ±1.5°, or even ±1° of the analytical value (theoretical value; calculation method will be described later).

[0027] The mode and order of the ultrasonic waves (Lamb waves) of interest depend on the structure, but examples include the 0th mode (particularly the A0 mode) and the 1st mode, which have large amplitudes.

[0028] The ultrasonic waves are preferably transmitted and received without contacting the thin plate portion. Furthermore, the ultrasonic waves are preferably transmitted and received from one side of the structure (for example, the first thin plate portion side). This allows for easy and efficient evaluation of the adhesive layer. [Example]

[0029] The present invention will be explained in more detail below, showing a specific example in which the bond length (size of the bonded layer) of a bonded body (structure) is evaluated non-destructively using ultrasound.

[0030] <<Joint and Evaluation Device>> An overview of the test subject, the junction M, and its evaluation device D is shown in Figure 1. Unless otherwise specified, the directions indicated by the arrows in Figure 1 are up / down or left / right, and the front and back sides of the paper are front and back.

[0031] (1) Zygote The bonded body M is formed by bonding a thin plate 1 (first thin plate portion) and a thin plate 2 (second thin plate portion) facing each other with an adhesive layer 3. In this example, the thin plates 1 and 2 are made of the same material and have the same shape. Specifically, the thin plates 1 and 2 are made of aluminum (alloy) and have a rectangular shape (thickness: 2 mm, width: 100 mm). Their lengths were changed as appropriate.

[0032] Adhesive layer 3 is made of an adhesive (epoxy resin) that bonds together the approximate centers of thin plate 1 and thin plate 2. Adhesive layer 3 has the same width (length in the front-to-back direction) as thin plate 1 and thin plate 2, and its thickness is 0.6 mm. Its length in the longitudinal direction (left-to-right direction) (adhesion length L) was changed as appropriate.

[0033] (2) Evaluation equipment The evaluation device D includes an ultrasonic wave transmitting probe 61, an ultrasonic wave (Lamb wave) receiving probe 62, an ultrasonic wave transmitting / receiving unit 63, and an evaluation unit 64 for the adhesive layer 3.

[0034] The transmitting probe 61 and the receiving probe 62 are air probes arranged on the same side (upper side) of the bonded body M. By using the air probe used in the airborne ultrasonic method, no contact medium (paste, jelly, etc.) is required, and non-contact inspection can be easily performed.

[0035] The mounting angle θ1 (ultrasonic wave incident angle) of the transmitting probe 61 relative to the thin plate 1 (normal line) was set to be the same as the mounting angle θ2 (ultrasonic wave emission angle) of the receiving probe 62 relative to the thin plate 1 (normal line). This angle was calculated as the optimal angle (critical angle) for exciting Lamb waves in the thin plate 1 using the following equation based on the law of refraction (Snell's law). θ=sin -1 (Cair / CLamb) (A) Cair: longitudinal wave velocity of the surrounding medium CLamb: Phase velocity of Lamb waves determined from the mode and frequency to be excited

[0036] The phase velocity of the Lamb wave was calculated using software (Plate Dispersion ver. 1.0, manufactured by Nagoya Institute of Technology). In this example, unless otherwise specified, attention was focused on the Lamb wave in the A0 mode (zeroth-order anti-symmetric propagation mode).

[0037] Ultrasonic measurement of the bonded body M was performed as follows. A transmission wave W1 (ultrasonic wave) emitted from the transmitter / receiver 63 is incident at an oblique angle (incident angle: θ1) from the transmitting probe 61 (transmitting means) onto the left side of the top surface of the thin plate 1 (transmitting step). The transmission wave W1 is excited within the thin plate 1 and propagates in its in-plane direction (left to right) as a Lamb wave G1. The Lamb wave G1 passes through the adhesive layer 3 near the center of the thin plate 1 (and further through the thin plate 2) and propagates in the in-plane direction of the thin plate 1 as a Lamb wave G2. The Lamb wave G2 leaking from the right side of the top surface of the thin plate 1 is received by the receiving probe 62 (receiving means) as a received wave W2 (ultrasonic wave) (receiving step).

[0038] The evaluation unit 64 (evaluation means) performs a Fourier transform on the received wave W2 obtained through the transmitter / receiver 63 to obtain a measurement spectrum. Furthermore, the evaluation unit 64 compares the measurement spectrum with the theoretical spectrum of Lamb waves in the bonded body M (analysis model) to identify (evaluate) the bond length L of the bonded body M (actual object) (evaluation step). These processes are performed by executing a predetermined program (algorithm) on a dedicated computer or a personal computer. The details of the processing method (algorithm) will be described later.

[0039] <<Derivation of the theoretical spectrum>> The theoretical spectrum to be compared with the measured spectrum is obtained, for example, as follows.

[0040] (1)Finite element method analysis The propagation behavior of Lamb waves in the bonded structure M was analyzed using the finite element method (FEM) for the model shown in Figure 2. The length of thin plates 1 and 2 was 600 mm, the length of adhesive layer 3 was L (200 mm in this case), and the distance between the transmitting and receiving probes was 560 mm. Other specifications were as described above. For convenience, the same reference numerals are used to designate corresponding parts in the actual object and the model, and detailed explanations are omitted (the same applies below).

[0041] The time evolution of the pressure generated on the top surface of thin plate 1 due to the propagation of the excited Lamb waves was analyzed in three regions shown in Figure 2 (A: unbonded area on the transmitting side, B: bonded area, C: unbonded area on the receiving side). Regions A and C were located near the center of the area without adhesive layer 3, and region B was located near the center of the area with adhesive layer 3. The length of each region (propagation direction / horizontal direction) was 64 mm, and the time evolution of pressure was recorded at 128 points spaced 0.5 mm apart. Based on this, the relationship between time and position (waveform) in each region was obtained, and this was then subjected to a two-dimensional Fourier transform to determine the relationship between frequency and wavenumber (waveform). Figure 3 shows the results when an A0 mode Lamb wave (330 kHz) was excited within thin plate 1. The phase velocity of the Lamb wave was 2052 m / s, and θ1 = θ2 = 9.5°.

[0042] (2) Dispersion curve Figure 4 shows the dispersion curve of the bonded body M (model) created using the software (Plate Dispersion) mentioned above. The dispersion curve shows the relationship (dispersion relationship) between frequency and wave number (or phase velocity and group velocity) for Lamb waves. Figure 4 also shows propagation modes other than A0. The numbers in the propagation modes indicate the order, and "S" indicates a symmetric mode. Also, modes 0 to 4 in the dispersion curve of the bonded joint are names for convenience, and the modes are numbered (0 → 1 → 2 → 3 → 4) in ascending order of lowest frequency.

[0043] (3) Discussion As is clear from Figure 3, FEM analysis revealed that the wave propagates mainly in one mode in the unbonded area (areas A and C), but branches into two modes in the bonded area (area B).

[0044] Comparing these propagation modes (Fig. 3) with the theoretically determined dispersion curve (Fig. 4) shows that the propagation mode in the unbonded section is consistent with the A0 mode, while the propagation mode in the bonded section is consistent with modes 0 and 1. In other words, it was found that the A0 mode Lamb wave excited in the unbonded section on the transmitting side branches into modes 0 and 1 in the bonded section, and these are combined again into the A0 mode in the unbonded section on the receiving side. This state is shown schematically in Fig. 5.

[0045] The received waveform obtained from the unbonded part is affected by the wave number (k) of each mode branched at the bonded part and the propagation distance (bonding length L), which is reflected in the measured spectrum. Taking into account the propagation characteristics of Lamb waves in the bonded body M and the wave equation (Rayleigh-Lamb equation), the theoretical spectrum to be compared with the measured spectrum of the received wave W2 obtained from the receiving probe 62 is thought to be expressed by equation (1) (see Figure 11).

[0046] In equation (1), B0 and B1 are the amplitude ratios of mode 0 and mode 1 to the A0 mode (before bifurcation), respectively. The sum of their absolute values ​​(|B0| + |B1|) is greater than or equal to 0 and less than 1. Furthermore, k0(f) and k1(f) are the wave numbers at each frequency of mode 0 and mode 1, respectively. These are theoretically determined from the dispersion relation (curve), and are calculated, for example, using software (Plate Dispersion) as shown in Figure 4. Note that e is the base of the natural logarithm (Napier's constant), and j is the imaginary unit.

[0047] (4) Verification The theoretical spectrum obtained from equation (1) was verified as follows.

[0048] For each bonded body M in which the length (L) of the adhesive layer 3 shown in Figure 2 was changed in 10 mm increments (L = 10 to 50 mm), the received waveform of the Lamb wave obtained by the FEM analysis described above was Fourier transformed to obtain a measurement spectrum (analysis value). The results are shown in Figure 6. The vertical axis of this measurement spectrum (normalized amplitude) is normalized by the measurement spectrum (amplitude value) obtained for the thin plate 1 alone (L = 0 mm).

[0049] Next, the theoretical spectrum (calculated value) was calculated for each bonded body M based on equation (1). In this case, B0 = B1 = 0.5 was set, assuming that the Lamb waves propagating from the unbonded portion branch equally without loss at the bonded portion and then combine again at the unbonded portion. The dispersion curve of the bonded portion described above (see Figure 4) was used for k0(f) and k1(f). The obtained results are shown in Figure 6, in comparison with the measured spectrum.

[0050] As can be seen from FIG. 6, if the adhesive length L is the same, the theoretical spectrum obtained from equation (1) matches the measured spectrum, and it was confirmed that the two are approximately the same.

[0051] Evaluation Procedure Based on the above results, a theoretical spectrum that matches the measured spectrum of the bonded structure M (actual object) is searched for, and the procedure (flowchart) for estimating (identifying) the bond length L of the bonded structure M based on the theoretical spectrum is shown in Figure 7. Specifically, the procedure is as follows.

[0052] In step S1, a receiving wave W2 (ultrasonic wave / Lamb wave) is acquired from the junction M, which is the object to be inspected, using the transmitting probe 61, the receiving probe 62, and the transmitting / receiving unit 63, which are arranged at predetermined positions.

[0053] In step S21, the evaluation unit 64 performs a Fourier transform on the received wave W2 to obtain the measured spectrum Am(f) in the frequency domain.

[0054] In step S22, the amplitude of the measurement spectrum Am(f) is normalized to obtain the measurement spectrum A'm(f). Normalization is performed by dividing Am(f) by the measurement spectrum Am0(f) of a Lamb wave obtained in advance for a reference test piece (for example, a thin plate alone). Normalizing with the measurement spectrum obtained using the same measurement system cancels out the effects of the frequency characteristics of the probe and transmission signal, resulting in a measurement spectrum that reflects only the effects of the adhesive joint.

[0055] In step S31, the theoretical spectrum Ac(f) is calculated based on equation (1) for a given adhesion length L within a predetermined range. The calculation of Ac(f) was performed by the previously described method using software (Plate Dispersion).

[0056] In step S32, the residual R shown in equation (2) (see FIG. 11) is calculated using the theoretical spectrum Ac(f) and the measured spectrum A'm(f), where fi is the i-th frequency and N is the total number of frequencies used in the calculation.

[0057] In step S33, it is determined whether the residuals R have been calculated for all assumed adhesion lengths L. If there is an uncalculated adhesion length L, in step S34, the adhesion length L is changed, and step S31 and subsequent steps are repeated.

[0058] After determining the residual R for all bond lengths L, the bond length L with the smallest residual R is output in step S4. The bond length L determined by such an algorithm is estimated (evaluated) as the bond length L of the bonded body M (actual object). Note that FIG. 7 shows an example of a case where the bond length L with the smallest residual R is searched for in a brute force manner within the set range of L. Alternatively, a search can be performed using an optimization algorithm such as Newton's method. By utilizing an optimization algorithm, L can be evaluated (specified) more quickly.

[0059] "experiment" The bond length L of the bonded body M (actual object) was evaluated as follows.

[0060] (1) Bonded body and evaluation equipment First, an overview of the prepared bonded body M and evaluation device D (arrangement of transmitting probe 61 and receiving probe 62) is shown in Figure 8. The adhesive length L was changed in 10 mm increments (L = 10 to 50 mm). The thickness of the adhesive layer 3 and the adhesive length L were set to the desired values ​​using spacers. Two bonded bodies M (test pieces) were prepared for each adhesive length L.

[0061] The transmitting probe 61 and receiving probe 62 of the evaluation device D were air probes (0.4K14 x 20N) manufactured by Japan Probe Co., Ltd. The transmitting / receiving unit 63 was a pulser receiver (JPR-600C-2CH) manufactured by the same company. The pulser receiver's transmission signal was a 21-frequency chirp wave (center frequency: 400 kHz). Chirp waves are ultrasonic waves whose frequency changes over time. The mounting angle θ1 of the transmitting probe 61 (ultrasonic wave incident angle) and the mounting angle θ2 of the receiving probe 62 (ultrasonic wave exit angle) were both set to 9° to excite A0 mode Lamb waves. This angle was determined by substituting the phase velocity of 2176 m / s of a 400 kHz Lamb wave calculated using software (Plate Dispersion) into the above-mentioned equation (A).

[0062] (2) Evaluation Next, for each bonded body M with a different bond length L, the bond length L was evaluated according to the procedure shown in Figure 7. The actual length and the evaluated length for the adhesive layer 3 (bond length L) are compared in Figure 9. The evaluated length was the value at which the residual error R was minimized when the bond length L was changed in 0.1 mm increments (0 to 100 mm).

[0063] As can be seen from FIG. 9, the evaluation length is approximately equal to the actual length, and it was confirmed that the above-described method can evaluate the adhesive length L (range of the adhesive layer 3) of the bonded body M with high accuracy.

[0064] "supplement" (1) In the bonded structure M (adhesive type) of this example, the Lamb waves were predominantly branched into mode 0 and mode 1 at the adhesive joint and then interfered with each other. However, depending on the plate assembly to be joined, the joining form, the joining conditions, etc., it is possible that the Lamb waves may branch into three or more modes at the adhesive joint. In such cases, it is advisable to use equation (3) instead of equation (1) (M≧3).

[0065] (2) In the above-described example, since the material and shape of thin plate 1 and thin plate 2 were the same, B0 and B1 in formula (1) were both set to 0.5. If the material and shape of thin plate 1 and thin plate 2 are different, B0 and B1 can be adjusted appropriately.

[0066] (3) The bonding form of the bonded body M does not have to be the adhesive type shown in Fig. 1, Fig. 2, or Fig. 8. For example, the evaluation method of the present invention can be applied to a patch type or joint type bonded body shown in Fig. 10. When multiple reflections of Lamb waves can occur between both ends of the bonded joint, evaluation accuracy can be ensured by, for example, obtaining a measurement spectrum by separating only the Lamb waves of the first wave packet excluding the multiple reflections.

[0067] In the case of a joint-type junction, when phase inversion occurs in one of the branched modes, resulting in a phase difference of 180° between the branched modes, one of B0 and B1 in equation (1) should be a negative number and the other a positive number.

[0068] (4) Even when the transmitting probe and the receiving probe are arranged on different sides of the thin plate (front and back sides), it is advisable to set one of B0 and B1 shown in equation (1) to a positive number and the other to a negative number.

[0069] It has been confirmed that the evaluation method of the present invention makes it possible to easily and accurately evaluate the adhesive layer of a bonded body without the need for the commonly used vertical flaw detection method of scanning a probe or the need to create a calibration curve in advance through experiments. [Explanation of symbols]

[0070] 1, 2 Thin plate (1st thin plate part, 2nd thin plate part) 3 Adhesive layer D Evaluation equipment M conjugate W1, W2 Ultrasound G1, G2 Lamb waves

Claims

1. A non-destructive evaluation method for a structure in which a first thin plate portion, an adhesive layer, and a second thin plate portion are stacked in this order, comprising: a transmitting step of transmitting ultrasonic waves obliquely to the first thin plate portion; a receiving step of receiving the ultrasonic waves propagated through the adhesive layer from the first thin plate portion or the second thin plate portion in an oblique direction; a measuring step of performing a Fourier transform on the received ultrasonic waves to obtain a measurement spectrum in the frequency domain; a theoretical calculation step of calculating a theoretical spectrum in the frequency domain corresponding to parameters characterizing the adhesive layer from the dispersion relation of a Lamb wave related to a model based on the structure; an evaluation step of evaluating the adhesive layer of the structure based on specific parameter values ​​when the theoretical spectrum matches the measured spectrum; A non-destructive evaluation method for a structure comprising:

2. The method for nondestructive evaluation of a structure according to claim 1 , wherein the parameter is a length of the adhesive layer in the propagation direction of the ultrasonic wave.

3. 3. The nondestructive evaluation method for a structure according to claim 1, wherein the specific value minimizes the difference in amplitude between the measured spectrum and the theoretical spectrum.

4. The nondestructive evaluation method for a structure according to claim 1 , wherein the measurement spectrum is obtained by synthesizing Lamb waves branched into a plurality of propagation modes in the adhesive layer.

5. 2. The method for nondestructive evaluation of a structure according to claim 1, wherein the transmitting step and the receiving step are performed in a non-contact state with respect to the first thin plate portion and / or the second thin plate portion.

6. 6. The method for nondestructive evaluation of a structure according to claim 5, wherein the transmission angle and reception angle of the ultrasonic waves are substantially equal.

7. the first thin plate portion and / or the second thin plate portion are made of metal, resin, or a composite material, The method for nondestructive evaluation of a structure according to claim 1 , wherein the adhesive layer is made of a resin.

8. The nondestructive evaluation method for a structure according to claim 7 , wherein the adhesive layer is made of an adhesive.

Images