Crystal structure evaluation method and crystal structure evaluation device

Abstract

To provide a crystal structure evaluation method and a crystal structure evaluation device capable of evaluating the uniformity of a crystal structure without depending on measurement conditions when evaluating the uniformity of the crystal structure using ultrasound in a material containing crystal grains.SOLUTION: A crystal structure evaluation method comprises: a measurement step of radiating ultrasound to a test material from a plurality of incident points on a surface, measuring intensity of the ultrasound as a function of time in a reflection direction or a transmission direction of the incident ultrasound, and acquiring ultrasound signals for the respective incident points; an envelope processing step of obtaining an envelope for each of the ultrasound signals; a correlation coefficient acquisition step of creating a plurality of pairs by selecting two envelopes from a group of envelopes, and obtaining a correlation coefficient between the two envelopes in at least a partial time domain for each pair; an averaging step of averaging the plurality of correlation coefficients; and a structure evaluation step of determining that a higher average value obtained indicates that a test material has a crystal structure having highly uniform crystal grains providing lower ultrasonic scattering intensity.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a crystalline structure evaluation method and a crystalline structure evaluation device, and more particularly to a crystalline structure evaluation method and a crystalline structure evaluation device for evaluating the uniformity of the crystalline structure of a material by utilizing ultrasonic waves. [Background technology]

[0002] In materials containing crystalline grains, such as metals, it is important to determine the state of the crystalline structure through inspection in order to ensure the desired material properties. Ultrasonic inspection is used as a non-destructive method for determining the state of the crystalline grains.

[0003] For example, Patent Document 1 discloses a method for determining the anisotropy of crystal grains using ultrasonic waves. In this method, ultrasonic waves are emitted into a material to be measured, and reflected waves within a predetermined time range are extracted from the received reflected waves. The extracted reflected waves are subjected to frequency analysis, and a determination is made as to whether the crystal grains of the material to be measured are deformed based on the ratio of the total reflected wave power of frequencies above a predetermined frequency to the total reflected wave power of all frequencies. By utilizing the fact that anisotropy in the crystal grains inside the material to be measured occurs, harmonics and waveform distortion are generated, and when this ratio exceeds a certain value, it is determined that the crystal grains are anisotropic.

[0004] The evaluation method disclosed in Patent Document 1 uses ultrasound to determine the state of crystal grains, such as anisotropy. However, methods such as the method disclosed in Patent Document 1, which analyze reflected waves in a single measurement, can only detect the average state of the material structure present in the area through which the ultrasound passes. Therefore, it is difficult to detect localized regions where the state of the crystal structure changes unevenly. However, in materials such as metals, even if the majority of the crystal grains in the structure are well-aligned in terms of properties such as grain size and anisotropy, localized regions with grain size and anisotropy different from those of the majority of the crystal grains may occur, resulting in an uneven structure. Therefore, the inventors developed a crystal structure evaluation method and crystal structure evaluation device disclosed in Patent Document 2, which enable evaluation of the unevenness of the crystal structure in materials containing crystal grains using ultrasound, even if only a portion of the material contains uneven crystal structure. In Patent Document 2, in the measurement process, ultrasound is incident on a test piece made of a material containing crystal grains from multiple incident points on the surface, and the intensity of the detected scattered waves is measured as the scattering intensity for each incident point. Then, in the subsequent statistical processing step, the number of incident points at which a predetermined scattering intensity was obtained in the measurement step is organized as a function of scattering intensity to create a scattering intensity function. Furthermore, in the evaluation step, the larger the peak width of the scattering intensity function obtained in the statistical processing step, the higher the nonuniformity of the crystalline structure of the test material is deemed to be. When the nonuniformity of the crystalline structure contributing to scattering is low (high uniformity), the variance in the scattering intensity obtained at each incident point is small. On the other hand, when the nonuniformity of the crystalline structure contributing to scattering is high, the variance in the scattering intensity obtained depending on the incident point is large. Therefore, the higher the nonuniformity of the crystalline structure, the larger the peak width in the scattering intensity function. In this way, by focusing on the peak width of the scattering intensity function, the nonuniformity of the crystalline structure can be evaluated from the results of ultrasonic measurements, even if the area with nonuniformity in the crystalline structure is only present in a portion of the entire material. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-257146 [Patent Document 2] Japanese Patent Publication No. 2020-085888 Summary of the Invention [Problem to be solved by the invention]

[0006] The crystalline structure evaluation method of Patent Document 2 evaluates the uniformity of a material's structure by statistically processing the distribution of scattered wave intensities obtained corresponding to each incident point. This evaluation method statistically processes the absolute values ​​of scattered intensity, so it is assumed that regions of the test material that can be considered to have the same crystalline structure, such as grain size, chemical composition, and anisotropy, provide similar scattering intensities. This assumption holds true when the evaluation is performed under the same conditions for ultrasonic irradiation of the test material and the same detection conditions for the ultrasonic sensor, making this an excellent method for easily quantitatively evaluating the uniformity of crystalline structure. However, when using different inspection equipment, for example, the conditions for ultrasonic irradiation of the test material and the detection conditions for the ultrasonic sensor change, making it difficult to directly compare evaluation results based on data measured under these different conditions.

[0007] For example, Figure 6 shows a schematic diagram of the scattering intensity function for different ultrasonic sensor sensitivities. (a) shows the case of low detection sensitivity, and (b) shows the case of high detection sensitivity. In both cases, the scattering intensity function can be divided into a main peak with a large peak height and a sub-peak with a larger scattering intensity (value on the horizontal axis) and a smaller peak height. The sub-peaks are shown in gray in the figure. The main peak can be attributed to a crystalline structure with a small crystal grain size and occupying a large area, i.e., a structure composed of fine, highly uniform crystal grains. On the other hand, the sub-peak can be attributed to a crystalline structure with a large crystal grain size and occupying a narrow area, i.e., a structure composed of coarse crystal grains that are unevenly formed. In both cases (a) and (b), the presence of two structures with different crystal grain sizes can be detected from the scattering intensity function.

[0008] However, the proportion of the sub-peaks relative to the main peak differs between (a) and (b). As the detection sensitivity of the detection sensor increases, the scattering intensity obtained corresponding to each incident point increases. Furthermore, even weak ultrasonic signals can be detected. As a result, as shown in (b), compared to the case of low detection sensitivity (a), the scattering intensity function shifts toward the high scattering intensity side (right side) and is distributed over a wider range of scattering intensity (wider width). Therefore, even when inspecting the same test material, the peak height and peak position of the sub-peak relative to the main peak change depending on the detection sensitivity. In other words, in the configuration shown in Figure 6, (b) includes larger crystal grains than (a), and the proportion of these large crystal grains is also higher, resulting in an evaluation result that indicates a low uniformity of the crystalline structure. Conversely, when the detection sensitivity of the detection sensor is low, the evaluation result indicates a high uniformity of the crystalline structure.

[0009] When using such a method that statistically processes the distribution of ultrasonic scattered wave intensity, changes in the ultrasonic inspection conditions due to changes in the inspection equipment, changes in the parameter settings of the inspection equipment, fluctuations in the adjustment and calibration of each part of the equipment, such as the detection sensor, or changes over time, may result in an inaccurate evaluation of the uniformity of the crystalline structure. To evaluate the uniformity of the crystalline structure of multiple materials using data measured with different equipment and under different conditions, an indirect comparison, such as comparison with a common standard sample, is necessary. However, when evaluating the uniformity of the crystalline structure of multiple materials based on data measured under various conditions, it is desirable to be able to evaluate the uniformity of the crystalline structure of the materials independently of the specific measurement conditions, such as the type of equipment used, the parameters used for measurement, and the state of each part of the equipment.

[0010] The problem to be solved by the present invention is to provide a crystalline structure evaluation method and a crystalline structure evaluation device that can evaluate the uniformity of the crystalline structure using ultrasound in a material containing crystal grains, without depending on the measurement conditions. [Means for solving the problem]

[0011] In order to solve the above problems, a crystalline structure evaluation method and a crystalline structure evaluation device according to the present invention have the following configuration.

[0012] [1] The crystalline structure evaluation method of the present invention includes a measurement step of irradiating ultrasonic waves onto a test piece made of a material containing crystalline grains from multiple incident points on the surface, measuring the intensity of the ultrasonic waves as a function of time in the reflection direction or transmission direction of the incident ultrasonic waves, and acquiring the ultrasonic signals for each of the incident points as an ultrasonic signal; an envelope processing step of obtaining an envelope for each ultrasonic signal obtained in the measurement step; a correlation coefficient acquisition step of creating multiple sets by selecting two envelopes from the group of envelopes obtained in the envelope processing step, and for each set, obtaining a correlation coefficient between the two envelopes in at least a portion of the time domain; an averaging step of taking the average value of the multiple correlation coefficients obtained in the correlation coefficient acquisition step; and a structure evaluation step of determining that the larger the average value obtained in the averaging step, the more uniformly formed the crystalline structure of the test piece is, with crystal grains that give low ultrasonic scattering intensity.

[0013] [2] In the aspect of [1] above, in the structure evaluation step, it is preferable that the larger the average value, the more the test material is considered to have a crystalline structure in which small crystal grains are formed with high uniformity.

[0014] [3] In the aspect of [1] or [2] above, in the envelope processing step, an envelope is acquired for all of the ultrasonic signals obtained in the measurement step, and in the correlation coefficient acquisition step, a correlation coefficient is evaluated for all combinations of the envelopes obtained in the envelope processing step, and in the averaging step, an average value is taken for all of the correlation coefficients obtained in the correlation coefficient acquisition step.

[0015] [4] In any one of the above aspects [1] to [3], in the measurement step, the ultrasonic signal is acquired in the reflection direction, and in the correlation coefficient acquisition step, the correlation coefficient of the envelope is calculated in a time domain on the ultrasonic signal between a time domain corresponding to a reflected wave reflected by the surface of the test material and a time domain corresponding to a reflected wave reflected by a bottom surface opposite the surface.

[0016] [5] In any one of the above aspects [1] to [4], in the envelope processing step, the envelope may be generated using a Hilbert transform.

[0017] [6] The crystal structure evaluation device of the present invention includes an ultrasonic inspection device that generates and detects ultrasonic waves, and performs any one of the crystal structure evaluation methods [1] to [5] above on a test piece made of a material containing crystal grains. [Effects of the Invention]

[0018] In the crystalline structure evaluation method according to the present invention, which has the configuration described above in [1], ultrasonic waves are incident on a test piece made of a material containing crystalline grains from multiple incident points on the surface, and ultrasonic signals containing the contribution of scattering by the crystalline structure are acquired for each incident point. In this case, if the crystalline structure contributing to the scattering of ultrasonic waves is highly uniform, including small grains that produce low ultrasonic scattering intensity, the ultrasonic signals acquired at each incident point will have similar waveforms. On the other hand, if the crystalline structure contributing to scattering is less uniform or contains many crystal grains that produce high ultrasonic scattering intensity, the waveforms of the ultrasonic signals acquired at each incident point will differ significantly. Therefore, envelopes are created for each ultrasonic signal to reduce the effects of phase shift, and the correlation coefficient between these envelopes is calculated. The correlation coefficient obtained increases with the uniformity of the crystalline structure constituting the test piece, which produces low ultrasonic scattering intensity. In this way, the uniformity of the crystalline structure can be evaluated by focusing on the value of the correlation coefficient between the envelopes of ultrasonic signals acquired at multiple incident points. In this case, the correlation coefficient obtained from the measurement results can be correlated with the uniformity of the crystalline structure by simple calculations, without performing complex calculations or simulations on the measurement results. Furthermore, by aligning the intensity of the incident ultrasonic waves and the sensitivity of the detection sensor, and simply comparing multiple ultrasonic signals acquired at multiple incidence points, the correlation coefficient can be obtained as a numerical value correlating with the uniformity of the crystalline structure. Therefore, the uniformity of the crystalline structure can be evaluated based on the value of the correlation coefficient, regardless of the specific measurement conditions. Therefore, even if the ultrasonic inspection conditions change due to changes in the inspection device, changes in the parameter settings of the inspection device, fluctuations in the adjustment and calibration of various parts of the device, including the detection sensor, or changes over time, by comparing the correlation coefficient values ​​obtained between multiple samples, it is possible to quantitatively compare how closely the samples resemble a crystalline structure containing highly uniform crystal grains, such as fine crystal grains, that provide small ultrasonic scattering intensity.

[0019] In the above aspect [2], in the structure evaluation step, the larger the average correlation coefficient, the more uniformly the test material is considered to have a crystalline structure in which small grains are formed. In the crystalline structure of metals and other materials, characteristics of the crystals that can cause different ultrasonic scattering intensities depending on the state of the crystal grains include the grain size, anisotropy, and chemical composition of the crystal grains. Among these, grain size is a characteristic that can vary significantly depending on the manufacturing conditions of the test material, and differences in grain size are sensitively reflected in the ultrasonic scattering intensity. The smaller the grain size, the lower the ultrasonic scattering intensity at the grain boundaries. Therefore, by applying the crystalline structure evaluation method of the present invention, the uniformity of the grain size within the crystalline structure can be suitably evaluated.

[0020] In the above aspect [3], in the envelope processing step, envelopes are acquired for all ultrasonic signals, and in the correlation coefficient acquisition step, correlation coefficients are evaluated for all combinations of envelopes, and in the averaging step, the average value is calculated for all of these correlation coefficients. In this way, by covering all combinations of envelopes obtained for ultrasonic signals obtained at all incident points and calculating correlation coefficients and their average values, it is possible to evaluate the uniformity of the crystalline structure over a wide area within the test material with high accuracy. If the number of envelopes to be analyzed in the correlation coefficient acquisition step is N, N(N-1) / 2 correlation coefficients will be obtained.

[0021] In the above aspect [4], ultrasonic signals are acquired in the reflection direction, and the correlation coefficient of the envelope of the ultrasonic signals is calculated in the time domain between the time domain corresponding to the reflected waves reflected from the surface of the test material and the time domain corresponding to the reflected waves reflected from the bottom surface. This makes it possible to calculate the correlation coefficient between the envelopes of ultrasonic signals that include the contribution of scattering, distinguishing them from the reflected waves reflected from the surface and bottom surface of the test material. This reduces the influence of reflected waves, allowing for accurate evaluation of the uniformity of the crystal structure.

[0022] In the above aspect [5], the envelope is generated using the Hilbert transform, which allows the envelope to be generated easily and accurately.

[0023] In the crystalline structure evaluation device of the present invention having the configuration [6] above, ultrasonic waves are incident on each incident point on the test material using an ultrasonic inspection device, and ultrasonic signals including the contribution of scattering can be obtained. Then, by executing the crystalline structure evaluation method of the present invention based on the obtained ultrasonic signals, the uniformity of the crystalline structure of the test material can be evaluated simply and independently of the measurement conditions. Ultrasonic inspection devices are widely used as flaw detection devices, etc., and such devices can be repurposed to construct a crystalline structure evaluation device. [Brief explanation of the drawings]

[0024] [Figure 1] 1 is a diagram showing the configuration of a crystal structure evaluation device according to one embodiment of the present invention. [Figure 2] This figure explains the behavior of ultrasonic waves within the test material, where (a) shows the entire test material and (b) shows an enlarged view of the structure. [Figure 3] 1 is an example of an ultrasonic signal detected by an ultrasonic inspection device. [Figure 4] Examples of ultrasonic signals detected by an ultrasonic inspection device are shown, along with their envelopes, for (a) a case where the crystal structure is highly uniform and (b) a case where the crystal structure is less uniform. [Figure 5] 1A and 1B are diagrams illustrating data processing in each step of a crystalline structure evaluation method according to one embodiment of the present invention. (a) shows a group of ultrasonic signals obtained in a measurement step. (b) shows a group of data that has undergone an envelope processing step. (c) is a matrix illustrating a set of data for determining a correlation coefficient in a correlation coefficient acquisition step. [Figure 6] FIG. 1 is a diagram for explaining the crystalline structure evaluation method disclosed in Patent Document 2, and schematically shows the scattering intensity function and peak separation results obtained when (a) the sensitivity of the detection sensor is low and (b) the sensitivity of the detection sensor is high. [Figure 7](a) shows the results of evaluating the uniformity of the crystal structure by the method of the present invention, and (b) by the method of Patent Document 2. Each figure uses measurement data obtained at two different detection sensitivities for three types of samples with different degrees of uniformity of the crystal structure. DETAILED DESCRIPTION OF THE INVENTION

[0025] The crystalline structure evaluation method and the crystalline structure evaluation device according to the embodiment of the present invention are described below. The crystalline structure evaluation method according to the embodiment of the present invention is a method for evaluating the uniformity of the crystalline structure of a test piece made of a material containing crystal grains, and can be executed using the crystalline structure evaluation device according to the embodiment of the present invention.

[0026] [Crystalline structure evaluation device] First, a crystal structure evaluation device according to one embodiment of the present invention will be described. Fig. 1 shows an outline of a crystal structure evaluation device 1 according to this embodiment.

[0027] The crystal structure evaluation device 1 has an ultrasonic inspection device 11, a motion device 12, and an arithmetic and control device 13, and evaluates the uniformity of the crystal structure that constitutes the test material S.

[0028] First, the test material S to be inspected by the crystalline structure evaluation apparatus 1 can be any material containing crystalline grains, such as a polycrystalline material, and is not limited in terms of material or shape. The material may be either a metal or a nonmetal, or may contain both. However, the following discussion will primarily focus on a metallic material. The data presented as examples was acquired for a test material S made of a titanium-based alloy. Titanium-based alloys are metals capable of forming α and β phases. Because the state of the formed phases is easily affected by heat treatment and forging conditions, evaluating the state of the crystalline structure is important to achieve desired material properties. In particular, in α+β-type titanium-based alloys, α-phase grains and β-phase grains coexist stably, making evaluation of the crystalline structure particularly important. The crystalline structure evaluation apparatus 1 can also be suitably used to evaluate the state of the crystalline structure of two-phase SUS alloys, iron-based alloys, and other materials. The following discussion will assume the shape of the test material S to be a bar, particularly a round bar.

[0029] The ultrasonic inspection device 11 is a device that generates ultrasonic waves, irradiates them onto an object, and detects the ultrasonic waves reflected, scattered, etc. by the object. Here, the ultrasonic inspection device 11 irradiates pulsed ultrasonic waves U substantially perpendicularly from a position away from the surface of the test material S placed on the motion device 12 onto the incident point P set on the surface of the test material S, and causes the ultrasonic waves to enter the test material. Then, in the reflection direction of the ultrasonic waves U, it detects the ultrasonic wave signals that have been reflected or scattered by the test material S and returned. This type of ultrasonic inspection device 11 is widely used as an ultrasonic probe in flaw detection devices for detecting flaws in materials. Also in this embodiment, such an ultrasonic probe can be used as the ultrasonic inspection device 11. As the frequency of the ultrasonic waves U generated by the ultrasonic inspection device 11, the frequencies used in general ultrasonic probes can be applied. However, from the viewpoint of increasing the intensity of scattering by crystal grains, it is preferable to use frequencies of about 1 to 30 MHz. More specifically, as the relationship between the average crystal grain diameter d of the material and the wavelength λ of the ultrasonic waves, the inspection according to this embodiment can be preferably applied under the condition that (1 / 30)λ < d < (1 / 3)λ, and it is preferable to select a frequency that satisfies this condition.

[0030] The motion device 12 moves the test material S with respect to the ultrasonic inspection device 11. Here, the motion device 12 is configured as a pedestal-shaped device that can perform axial rotation R and translational motion T. The test material S is placed on the upper surface of the motion device 12. When the test material S is rod-shaped (columnar) such as a round bar, the test material S is placed upright on the upper surface of the motion device 12 at one of its end faces. The motion device 12 can rotate the test material S with respect to the ultrasonic inspection device 11 by performing axial rotation R. It is preferable that the motion device 12 can rotate the test material S one full turn. Also, the motion device 12 can move the test material S in the translational direction with respect to the ultrasonic inspection device 11 by performing translational motion T. When the rod-shaped test material S is placed upright on the upper surface of the motion device 12, the translational motion T will move the test material S along its longitudinal direction.

[0031] The arithmetic and control device 13 controls the ultrasonic inspection device 11 and the motion device 12, and performs computational analysis on the data obtained by the ultrasonic inspection device 11, and can be configured, for example, by a computer. Specifically, the arithmetic and control device 13 controls the ultrasonic measurement by the ultrasonic inspection device 11 and the motion of the test material S by the motion device 12. The arithmetic and control device 13 receives an input of an ultrasonic signal output by the ultrasonic inspection device 11 as an electrical signal corresponding to the intensity of the detected ultrasonic waves, and analyzes the ultrasonic signal. The content of the analysis will be explained in detail later as a crystalline structure evaluation method, but for the multiple obtained ultrasonic signals, envelopes are created and the correlation coefficient between the obtained envelopes is evaluated, and based on the results, the uniformity of the crystalline structure is evaluated.

[0032] When ultrasonic measurement is performed on the material S to be inspected to evaluate the uniformity of the crystal structure, the arithmetic and control device 13 drives the axial rotation R and translational motion T of the motion device 12 to adjust and stop the rotational and translational positions of the material S to allow ultrasonic waves U to be incident from the ultrasonic inspection device 11 at a certain incident point P set on the surface (side) of the material S to be inspected. Then, the ultrasonic inspection device 11 is controlled to cause ultrasonic waves U to be incident approximately perpendicularly from the incident point P into the interior of the material S to be inspected, and ultrasonic waves reflected by the material S to be detected. At this time, an ultrasonic signal indicating the intensity of the detected ultrasonic waves is input to the arithmetic and control device 13 in the form of a function of time and stored.

[0033] After completing the measurement for one incident point P, the arithmetic and control device 13 drives the axial rotation R of the motion device 12 to rotate the test piece S so that ultrasonic waves U can be incident on another incident point P, and then stops the rotation. The same measurement as for the previous incident point P is then performed, and ultrasonic signals are acquired. After performing similar measurements for incident points P set around the entire circumference of the test piece S at predetermined angular intervals, such as 1° intervals, the arithmetic and control device 13 then drives the translational motion T of the motion device 12. Similar measurements are then repeated while rotating at predetermined angular intervals. In this manner, measurements are repeated at predetermined angular intervals around the entire circumference, such as 0.5 mm, along the longitudinal direction of the test piece S. As a result, ultrasonic signals are acquired for each of the numerous incident points P set around the entire surface (side surface) of the test piece S at predetermined angular and length intervals. The ultrasonic signals acquired correspond to the number of incident points P, forming a data group consisting of ultrasonic signals equal to the number of incident points P, as shown in FIG. 5(a). In Fig. 5(a), each graph constituting the data group represents the ultrasonic wave intensity detected at one incident point P as a function of time. The analysis described below is performed on such a data group.

[0034] In the crystalline structure evaluation apparatus 1 described above, the ultrasonic inspection device 11 is fixed, and the test piece S is moved by the movement device 12 to perform measurements at multiple incident points P. However, as long as the ultrasonic inspection device 11 can be moved relative to the test piece S, the configuration is not limited to this. The test piece S may be fixed, and the movement device 12 may be moved around the outer periphery of the test piece S. By moving the ultrasonic inspection device 11 relative to the test piece S in both the circumferential and translational directions, measurements at multiple incident points P can be performed using a single ultrasonic inspection device 11. Alternatively, if the ultrasonic inspection device 11 is arranged to surround the entire circumference of the test piece S, as in an inspection device with multiple channels, measurements can be performed at multiple incident points P set around the entire circumference of the test piece S at once, without requiring rotational movement.

[0035] Furthermore, in the crystalline structure evaluation device 1 described above, ultrasonic waves U incident perpendicularly to the test material S are detected in the reflection direction, but transmitted waves may also be detected in the transmission direction. Furthermore, the incident direction of the ultrasonic waves U does not have to be perpendicular, and can be, for example, an oblique direction.

[0036] As described above, the ultrasonic inspection device 11 is also used for detecting flaws in materials, and the crystalline structure evaluation device 1 according to this embodiment can also be used as a flaw detection device. In other words, when the crystalline structure evaluation device 1 is used to evaluate the uniformity of the crystalline structure of the test material S, flaw detection of the test material S can be performed simultaneously or as an independent process.

[0037] [Crystal structure evaluation method] Next, a method for evaluating the uniformity of the crystal structure of the test material S by using the crystal structure evaluation device 1 as described above and executing a crystal structure evaluation method according to one embodiment of the present invention will be described.

[0038] (Crystal structure and ultrasonic scattering) Before describing each step in the crystalline structure evaluation method, the relationship between the state of the crystalline structure and scattering of ultrasonic waves will be described.

[0039] FIG. 2(a) shows a schematic diagram of the behavior of ultrasonic waves U incident on the test material S. FIG. 2(b) shows an enlarged view of the internal structure of the test material S. FIG. 3 shows an example of an ultrasonic signal obtained by the ultrasonic inspection device 11. FIG. 4(a) and (b) show examples of ultrasonic signals when the crystalline structure is highly uniform and when the crystalline structure is less uniform, respectively. Note that FIG. 3 corresponds to one ultrasonic signal extracted from a data group such as that shown in FIG. 5(a).

[0040] As shown in Figure 2(a), most of the incident ultrasonic wave U is reflected by the surface S1 of the material S to be inspected, becoming a surface reflected wave U1. The surface reflected wave U1 is detected with a very large amplitude at the beginning of the time when the ultrasonic signal is detected, as shown as component A1 in Figure 3. Furthermore, most of the ultrasonic wave U that reaches the bottom surface S2 opposite the surface S1 of the material S to be inspected is reflected by the bottom surface S2, becoming a bottom reflected wave U2. The bottom reflected wave U2 is detected with a relatively large amplitude at the end of the time when the ultrasonic signal is detected, as shown as component A2 in Figure 3.

[0041] In the ultrasonic signal, in addition to the component A1 due to the surface reflected wave U1 and the component A2 due to the bottom reflected wave U2, a small amplitude signal is observed continuously for a long period of time in the time domain between them, as shown as component A3 in Figure 3. This component A3 corresponds to the scattered wave U3 generated when the ultrasonic wave U is scattered inside the test material S.

[0042] As shown in Figure 2(b), which is a schematic enlarged view of the area indicated by the dashed line in Figure 2(a), the metal material constituting the test piece S contains numerous crystal grains S3. The outer edges of the crystal grains S3 are grain boundaries S4. Ultrasonic waves U incident on the crystalline structure are scattered by the crystal grain boundaries S4. The size of the crystal grains S3 in polycrystalline metal materials is typically on the order of submicrons to microns, and backscattering (Rayleigh scattering) contributes significantly to the scattering of the ultrasonic waves U. At least a portion of the scattered waves U3 generated by the crystal grain boundaries S4 reaches the ultrasonic inspection device 11 and is detected. The scattered waves U3 that reach the ultrasonic inspection device 11 become component A3 in the ultrasonic signal, which is observed in the time interval (evaluation region G) between the surface reflected wave U1 and the bottom reflected wave U2. Because scattering by the crystal grain boundaries S4 occurs throughout the entire path from the surface S1 to the bottom S2 of the test piece S, the signal component A3 due to the scattered waves U3 is observed over a long period of time.

[0043] It is known that the scattering intensity of ultrasonic waves U at the grain boundary S4 is proportional to the sixth power of the grain size of the crystal grains S3. In other words, when the grain size of the crystal grains S3 is large overall, scattering at the grain boundary S4 is strong. Figure 4(a) shows the ultrasonic signal detected by the ultrasonic inspection device 11 when the crystal structure is highly uniform, i.e., when the entire crystal structure is composed of small crystal grains S3, with the black line. In this case, the contribution of the scattered wave U3 to the ultrasonic signal is small, and a waveform with small amplitude and time variation of the signal intensity in the evaluation region G is obtained. On the other hand, Figure 4(b) shows the ultrasonic signal detected by the ultrasonic inspection device 11 when the crystal structure is less uniform, i.e., when at least a portion of the crystal structure contains large crystal grains S3, with the black line. In this case, the contribution of the scattered wave U3 in the evaluation region G causes the signal intensity to fluctuate up and down with large amplitude, and a waveform with large time variation of the amplitude is obtained. In this way, the grain size of the crystal grains S3 that make up the crystalline structure is reflected in the waveform of the ultrasonic signal, and when the crystalline structure contains coarse crystal grains S3, as in the case of Figure 4(b), the waveform of the ultrasonic signal will have large undulations.

[0044] Here, we consider comparing the waveforms of multiple ultrasonic signals obtained by injecting ultrasonic waves U from different incident points P, as in the series of data shown in Figure 5(a). Different ultrasonic signals are prone to slight phase shifts and differences in signal strength due to accidental noise. However, one method for comparing the waveforms of multiple ultrasonic signals while minimizing the effects of these non-essential shifts and differences in the relationship between the crystal structure state and the ultrasonic signal waveform is to create the envelope of each ultrasonic signal. Figures 4(a) and (b) show the waveforms of the ultrasonic signals (black lines) along with the envelopes of each waveform (gray lines). When the amplitude and temporal change of the ultrasonic signal waveform are small, as in Figure 4(a), the envelope in the evaluation region G is flat, moving only gradually around zero intensity. On the other hand, when the amplitude and temporal change of the ultrasonic signal waveform are large, as in Figure 4(b), the envelope in the evaluation region G also becomes highly turbulent, with large up and down fluctuations. The envelope can be set on both the positive and negative sides of the vertical axis, and either one or both can be used for analysis, but in the illustrated form, only the positive envelope is used.

[0045] As described above, if only fine crystal grains S3, which provide low ultrasonic scattering intensity, are present along the path of the ultrasonic wave U incident on the test material S until it is detected, the ultrasonic signal will have a small amplitude and its variation in amplitude in the evaluation region G, as shown in Figure 4(a). The envelope will also be flat. If the entire test material S has a highly uniform crystalline structure consisting of a collection of such fine crystal grains S3, a waveform with a small amplitude and its variation, and a flat envelope will be obtained in the evaluation region G, regardless of the incident point P of the ultrasonic wave U and the path determined by it. In this case, the envelopes of two ultrasonic signals obtained by incident ultrasonic waves U on the test material S from any two different incident points P will both be similar, flat, and move near zero intensity in the evaluation region G. Considering the correlation coefficient as a parameter for evaluating the similarity between the two envelopes, if the envelopes of the two ultrasonic signals both have flat shapes, the correlation coefficient will be large. The correlation coefficient (ρ) can be evaluated using the following equation (1): ρ=cov[X,Y] / (σ X σ Y ) (1) where X,Y denote the two envelopes, cov[X,Y] is the covariance between them, σ X ,σ Y indicates the standard deviation of each envelope. ρ in equation (1) can take values ​​between -1 and 1 (-100% to 100%), but since only positive envelopes are considered here, the correlation coefficient also takes values ​​between 0 and 1 (0% to 100%).

[0046] On the other hand, if coarse crystal grains S3 that cause a large ultrasonic scattering intensity are present along the path of the ultrasonic wave U incident on the test material S until it is detected, the ultrasonic signal will have a large amplitude and a large variation in amplitude in the evaluation region G, as shown in Figure 4(b). The envelope will also be highly wavy and disordered. If the test material S has a crystalline structure with a low uniformity that includes such coarse crystal grains S3, or if it has a crystalline structure that is exclusively composed of coarse crystal grains S3, when the ultrasonic wave U passes through the region where the coarse crystal grains S3 are present at the incident point P of the ultrasonic wave U, a waveform with a large amplitude and a large variation in amplitude, and a highly wavy and disordered envelope will be obtained in the evaluation region G. However, if the incident point P of the ultrasonic wave U changes, the specific distribution of the coarse crystal grains S3 along the path of the ultrasonic wave U, i.e., the position, size, and number of coarse crystal grains present along the path, does not necessarily match. As a result, the envelopes of two ultrasonic signals obtained by irradiating ultrasonic waves U onto the test material S from any two different incident points P may differ in the time at which waving occurs (position on the horizontal axis) and the detailed pattern of the waving (specific shape of the waving curve).Thus, although the shapes of the two envelopes have in common the fact that they are largely wavy and disordered, the two envelopes have little similarity in terms of their specific shapes.As a result, the correlation coefficient between the two envelopes is smaller than when the entire test material S is made up of fine crystal grains S3 as described above.

[0047] Furthermore, in a test material S in which regions of low crystalline structure uniformity containing coarse crystal grains S3 coexist with regions of high crystalline structure uniformity consisting of fine crystal grains S3, depending on the incident point P of the ultrasonic wave U, the ultrasonic wave U may pass only through the regions of high crystalline structure uniformity, or it may pass through the regions of low crystalline structure uniformity along its path. In the former case, an ultrasonic signal waveform with a small amplitude and a flat envelope is obtained, as shown in Figure 4(a). In the latter case, an ultrasonic signal waveform with a large amplitude and a distorted envelope is obtained in at least a portion of the evaluation region G, as shown in Figure 4(b). When the correlation coefficient between the envelopes of these two types of ultrasonic signals is estimated, the correlation coefficient value is also smaller in this case than when the entire test material S is composed of fine crystal grains S3.

[0048] In this way, when the envelopes of ultrasonic signals obtained by irradiating ultrasonic waves U from different incident points P are taken and the correlation coefficient between the envelopes is calculated, the correlation coefficient shows a high correlation with the uniformity of the crystalline structure of the test material S. In other words, the more uniform the crystalline structure is, with small crystal grains S3 that give low ultrasonic scattering intensity, the larger the correlation coefficient will be, while the smaller the correlation coefficient will be if the test material S has a non-uniform structure that includes coarse crystal grains S3 in whole or in part. In this way, the uniformity of the crystalline structure can be evaluated by calculating the correlation coefficient of the envelopes between multiple ultrasonic signals obtained at different incident points P.

[0049] (Each step of the crystal structure evaluation method) Next, the steps performed in the crystal structure evaluation method according to the embodiment of the present invention, utilizing the correlation between the correlation coefficient of the envelope and the uniformity of the crystal structure as explained above, will be described. In the crystal structure evaluation method according to the present embodiment, a measurement step, an envelope processing step, a correlation coefficient acquisition step, an averaging step, and an evaluation step are performed in this order.

[0050] (1) Measurement process In the measurement process, ultrasonic waves U are incident from multiple incident points P set on the surface (side) of the material to be inspected S, and the intensity of the detected ultrasonic waves U is measured as a function of time, thereby obtaining an ultrasonic signal for each incident point P. In the measurement process, first, as described above for the crystalline structure evaluation device 1, the calculation and control device 13 controls the motion device 12 and the ultrasonic inspection device 11 to incident ultrasonic waves U to each of multiple incident points P set on the surface of the material to be inspected S at predetermined angular and length intervals, and obtain ultrasonic signals. Through the measurement, a data group such as that shown in Figure 5(a) is obtained. While obtaining ultrasonic signals from the multiple incident points P, the measurement conditions of the ultrasonic inspection device 11, such as the intensity of the ultrasonic waves U incident on the material to be inspected S and the sensitivity of the detection sensor that detects the ultrasonic waves U, are not changed.

[0051] Here, the range of measurement on the surface of the test material S by setting the incident point P may be set in any way as long as the ultrasonic wave U incident from any incident point P can reach a region of the test material S with a volume sufficiently larger than the region where a different crystalline structure from the surrounding area may be formed. From the viewpoint of thoroughly inspecting the entire test material S for the presence of a region with a non-uniform crystalline structure, it is preferable to set the range of the incident point P so that the ultrasonic wave U can reach almost the entire area of ​​the test material S. When the test material S is a rod-shaped material, it is advisable to set the incident point P along the entire circumference and along at least a partial region, preferably the entire region, along the longitudinal direction. Note that the incident point P does not necessarily have to be set along the entire circumference; as long as it is set over at least 180°, it is possible to inspect the structure of the entire cross section. In particular, when the test material S has a shape that is mirror-symmetrical across the central axis, such as a round bar, it is sufficient to set the incident point P over 180° (half the circumference).

[0052] Ultrasonic measurements are performed sequentially at each set incident point P, and ultrasonic signals are obtained as a data group as shown in Figure 5(a). At this time, frequency analysis is performed on the signals recorded as time variations in ultrasonic intensity, and signals of specific frequencies are extracted and used as ultrasonic signals for subsequent analysis. For example, a Fourier transform is performed on each time variation signal, and the time variations of components with the same frequency as the fundamental wave of the incident ultrasonic wave U are extracted and used as ultrasonic signals. The ultrasonic signals shown in Figures 3, 4, and 5(a) are thus obtained by extracting the time variations of components with the same frequency as the fundamental wave through Fourier transform.

[0053] Note that each of the subsequent steps is performed based on a signal of the same frequency component as the fundamental frequency extracted by frequency analysis. This is because the scattered wave U3 is most strongly observed at the fundamental frequency, ensuring detection sensitivity and making it easy to distinguish it from the contributions of other phenomena. However, the method of processing the ultrasonic signal is not limited to the above, as long as it is a method that can obtain an ultrasonic signal that sufficiently includes the contribution of the scattered wave. For example, if the contribution of the scattered wave U3 is sufficiently large, the frequency analysis may be omitted, and the time-varying signal of the total intensity of the detected ultrasonic wave may be used as the ultrasonic signal for subsequent analysis. Alternatively, the time-varying signal of frequencies other than the fundamental frequency obtained by frequency analysis may be used as the ultrasonic signal.

[0054] (2) Envelope processing process In the envelope processing step, an envelope is obtained for each ultrasonic signal obtained in the measurement step in association with each incident point P. By creating an envelope for the ultrasonic signal and using it as the subject of analysis in the next correlation coefficient acquisition step and subsequent steps, the contribution of non-essential elements such as phase shifts and accidental noise in the ultrasonic signal can be reduced. Figure 5(b) shows the ultrasonic signals that make up the data group in Figure 5(a) after envelope processing.

[0055] Although the method for creating the envelope is not particularly limited, it is preferable to use the Hilbert transform. By using the Hilbert transform, the envelope can be created easily and accurately even when there are many ultrasonic signals. Alternatively, as an even simpler method, the envelope may be created by connecting the maximum points of ultrasonic vibration in the ultrasonic signal (the peak tops of vibration at each phase).

[0056] As described above, the envelope curve needs to be generated in at least one of the positive and negative directions of the ultrasonic signal, but it is sufficient to generate it only in the positive direction. The envelope curve may be generated over the entire time domain of the ultrasonic signal, but it is preferable to generate the envelope curve by selecting a domain that includes the component A3 due to the scattered wave U3 (evaluation domain G, which will be set in the next correlation coefficient acquisition process). From the viewpoint of evaluating the uniformity of the crystalline structure over a wide domain inside the test material S with high accuracy, it is preferable to generate the envelope curve for each of the ultrasonic signals obtained in the measurement process. However, it is preferable to exclude ultrasonic signals that are clearly abnormal due to poor condition of the ultrasonic inspection device 11, etc., from the targets for generating the envelope curve.

[0057] (3) Correlation coefficient acquisition process In the correlation coefficient acquisition step, two envelopes are selected from the group of envelopes obtained in the previous envelope processing step to create multiple pairs, and for each pair, the correlation coefficient between the two envelopes is calculated for at least a portion of the time domain. As explained above with reference to FIG. 4 regarding the relationship between the state of the crystalline structure and the ultrasonic signal, the correlation coefficient between the envelopes created for any two ultrasonic signals obtained corresponding to different incident points P reflects the uniformity of the crystalline structure, and a large correlation coefficient is an indicator that the fine crystal grains S3 that give low ultrasonic scattering intensity are highly uniformly formed in the crystalline structure. By calculating the correlation coefficient based on the envelopes rather than the ultrasonic signals themselves, the correlation coefficient can be prevented from becoming excessively small compared to the essential similarity of the waveforms due to factors such as slight phase shifts and accidental noise, and the essential similarity of the two ultrasonic signals can be accurately reflected in the correlation coefficient.

[0058] As explained above, the correlation coefficient can be calculated using Equation (1). In this case, it is preferable to set an evaluation region G in the ultrasonic signal so that the similarity of the waveforms can be selectively evaluated with respect to the contribution of the scattered wave U3, and to calculate the correlation coefficient between the two envelopes within the evaluation region G. As explained above with reference to FIG. 3 , in the ultrasonic signal, the large-amplitude component A1 observed in the early time domain corresponds to the surface-reflected wave U1, and the relatively large-amplitude component A2 observed in the late time domain corresponds to the bottom-reflected wave U2. The small-amplitude component A3 observed between the components A1 and A2 is due to the scattered wave U3. Therefore, to sufficiently exclude the influence of the signals due to the surface-reflected wave U1 and the bottom-reflected wave U2, the evaluation region G is set in the time domain between the two components A1 and A2, and the correlation coefficient between the two envelopes is calculated within the evaluation region G.

[0059] The correlation coefficient can be calculated for a pair of any two envelopes selected from the group of envelopes obtained in the envelope processing step. If multiple pairs are created, the number of pairs is not limited. However, from the perspective of accurately evaluating the uniformity of the crystalline structure over a wide region within the test material S, it is preferable to comprehensively calculate the correlation coefficient for all combinations of envelopes. Here, let N be the number of envelopes created in the envelope processing step, and consider calculating the correlation coefficient between the i-th envelope and the j-th envelope (i≦N, j≦N). The possible pairs of envelopes, i.e., pairs i and j, can be expressed as an N×N matrix, as shown in Figure 5(c). The diagonal elements of this matrix are meaningless because they analyze the same envelope. Furthermore, elements in symmetrical cells on either side of the diagonal element (i.e., pairs ij and ji) are mutually identical, so it is sufficient to consider only one of them. In other words, it is sufficient to calculate the correlation coefficient for pairs located on one side of the diagonal of the matrix, as shown in gray in Figure 5(c). In other words, the number of pairs of envelopes for which the correlation coefficient should be calculated is N(N-1) / 2.

[0060] (4) Average process In the averaging step, the multiple correlation coefficients obtained in the previous correlation coefficient acquisition step are averaged. This allows an average correlation coefficient to be obtained as a representative value that reflects the uniformity of the crystalline structure of the entire area inspected in the test material S. Like the correlation coefficient between two individual envelopes, the average correlation coefficient also reflects the uniformity of the crystalline structure, and a large average correlation coefficient is an indicator that the test material S has a crystalline structure in which fine crystal grains S3 that give low ultrasonic scattering intensity are formed with high uniformity.

[0061] The average value may be calculated as a simple average value. If the correlation coefficient is a negative value, it may be included in the average value as an absolute value. In the averaging step, it is preferable to calculate the average value of all the correlation coefficients obtained in the correlation coefficient acquisition step. Alternatively, if the test material S is divided into multiple parts, such as multiple sections along the longitudinal direction, and it is desired to evaluate the uniformity of the crystalline structure for each part, the average correlation coefficient may be calculated for each part.

[0062] (5) Evaluation process In the evaluation process, the uniformity of the crystalline structure of the test material S is evaluated based on the average correlation coefficient obtained through each of the above processes. As already explained, a large average correlation coefficient is an indicator that the crystal grains S3 that give low ultrasonic scattering intensity, i.e., the crystal grains S3 with small diameters, are formed with high uniformity in the crystalline structure. Therefore, the larger the average correlation coefficient, the closer the crystalline structure obtained is to an ideal crystalline structure in which fine crystal grains are uniformly generated. On the other hand, a small average correlation coefficient can be determined to indicate that a less uniform crystalline structure has been formed, including coarse crystal grains S3 that give high ultrasonic scattering intensity, or that a region including such coarse crystal grains S3 coexists with a region composed of fine crystal grains S3.

[0063] Characteristics of the crystal grains S3 whose degree of uniformity is reflected in the average correlation coefficient include crystal grain size, elemental composition, anisotropy, etc. Each of these characteristics can cause differences in the ultrasonic scattering intensity provided by the crystal grains S3. The larger the value of the average correlation coefficient, the more likely it is that the test material S has a highly uniform crystalline structure in which crystal grains that provide low ultrasonic scattering intensity due to the contribution of one of the characteristics are formed. A small value of the average correlation coefficient indicates that a less uniform crystalline structure including crystal grains S3 that provide high ultrasonic scattering intensity has been formed, or that a region including such crystal grains S3 that provide high ultrasonic scattering intensity coexists with a region composed of crystal grains S3 that provide low ultrasonic scattering intensity.

[0064] If it is known through prior testing or other means which of the various characteristics, such as grain size, chemical composition, or anisotropy, of the material constituting the target test material S may be nonuniform and thus the degree of uniformity for which evaluation is required, the degree of uniformity of that characteristic can be evaluated from the value of the average correlation coefficient. For example, if changes in grain size uniformity are most likely to occur as a result of changes in the uniformity of the crystalline structure due to fluctuations in material manufacturing conditions, the degree of uniformity in grain size can be evaluated based on the average correlation coefficient, as exemplified above. In metallic materials, the uniformity of grain size and its distribution is susceptible to significant changes due to differences in manufacturing conditions, and the ultrasonic scattering intensity is prone to change significantly depending on the grain size. Therefore, in the crystalline structure evaluation method according to this embodiment, the uniformity of grain size is a particularly suitable evaluation target among the various characteristics described above.

[0065] When evaluating the uniformity of the crystalline structure based on the average correlation coefficient, if reference data showing the correspondence between the degree of uniformity of the crystalline structure and the average correlation coefficient are collected through prior testing or the like, the average correlation coefficient obtained in the test of the test material S can be specifically correlated to the uniformity of the crystalline structure. For example, multiple reference samples manufactured under various conditions and with varying degrees of uniformity of crystal grain size can be prepared in advance, and the degree of uniformity of the crystalline structure of each can be confirmed and the average correlation coefficient estimated for each by, for example, observing the cross section with a scanning electron microscope (SEM). The average correlation coefficient actually obtained for the test material S to be tested is then compared with the average correlation coefficients obtained for each reference sample. If there is a reference sample with a similar average correlation coefficient, it can be evaluated that the test material S has the same degree of uniformity of grain size as the reference sample, i.e., that the crystal grains S3 of the same size as the reference sample are formed with the same degree of uniformity. On the other hand, if the average correlation coefficient obtained for the test material S to be inspected is smaller than the average correlation coefficient of a reference sample, the test material S can be evaluated as having a lower grain size uniformity than the reference sample, i.e., as containing, as at least part of its crystalline structure, crystal grains S3 having a larger grain size than those constituting the crystalline structure of the reference sample.

[0066] Alternatively, when determining whether the uniformity of the crystalline structure of each individual product in a metal material manufacturing process is unacceptable from the viewpoint of material properties, etc., it is not necessary to evaluate in detail the degree of grain size uniformity of the test material S. In such cases, it is not necessary to collect reference data using multiple reference samples. Instead, it is sufficient to prepare one reference sample that represents the lower limit of the acceptable uniformity and estimate the average correlation coefficient for that reference sample as a threshold value in advance. The estimated threshold value is then compared with the average correlation coefficient actually obtained for the test material S to be inspected. For example, if the obtained average correlation coefficient is equal to or greater than the set threshold value, it can be evaluated that a crystalline structure with sufficiently fine crystal grains S3 and sufficiently high uniformity has been obtained relative to the acceptable level. On the other hand, if the obtained average correlation coefficient is smaller than the threshold value, it can be evaluated that a crystalline structure with unacceptably low uniformity or a crystalline structure containing unacceptably large crystal grains S3 has been obtained.

[0067] In this embodiment, the average correlation coefficient used to evaluate the uniformity of the crystalline structure is estimated using data on multiple ultrasonic signals measured at multiple incidence points P under the same conditions, based on a correlation coefficient that reflects the mutual similarity between the envelopes of those ultrasonic signals. In other words, the evaluation of the uniformity of the crystalline structure is based solely on the relative relationship between the waveforms of the multiple ultrasonic signals, and does not depend on the characteristics of the ultrasonic signals as absolute values, i.e., signal intensity or specific waveforms. Therefore, even if the measurement conditions, such as the incidence conditions of the ultrasonic waves U and the sensitivity of the detection sensor, are different between the reference sample and the test material S, it does not matter. By comparing the average correlation coefficient values ​​between the reference sample and the test material S, the uniformity of the crystalline structure of the test material S can be evaluated. Therefore, even if the measurement conditions for inspecting the test material S change, the same reference data can be used unchanged to evaluate the uniformity of the crystalline structure of the test material S. Furthermore, by directly comparing the average correlation coefficients obtained for multiple test materials S, the test material S with a higher average correlation coefficient can be considered to have a crystalline structure in which the crystal grains S3 that give a smaller ultrasonic scattering intensity are more uniformly formed. This is also true even if the measurement conditions for the multiple test materials S being compared are different from one another. By comparing the average correlation coefficient values ​​between multiple test materials S, it is possible to compare the uniformity of the crystalline structure using the average correlation coefficient values ​​themselves, without using reference data to compare with the actual crystalline structure state obtained by SEM observation, etc.

[0068] As described above, in the crystalline structure evaluation method according to this embodiment, the scattering intensity of the ultrasonic waves U varies depending on the state of the crystal grains S3, and this is utilized to easily and non-destructively evaluate the uniformity of the crystalline structure based on the ultrasonic signals obtained when the ultrasonic waves U are incident from multiple incident points P on the surface of the test material S. The grain size of the crystal grains S3 of a polycrystalline metal material is typically on the order of submicrons to microns, which is smaller than the wavelength of the ultrasonic waves U. However, by performing evaluation using the scattering phenomenon, particularly backscattering, the uniformity of the properties of the crystalline structure, such as grain size, can be inspected with high sensitivity.

[0069] Unlike inspections performed by irradiating ultrasonic waves U at only one point or by irradiating ultrasonic waves U all at once over a wide area, by calculating the correlation coefficient of the envelopes between multiple ultrasonic signals acquired individually from multiple incident points P, it is possible to sensitively detect the presence of regions with non-uniform structures, even if such regions occupy only a small portion of the entire crystalline structure. Also, as shown in Figure 3, the intensity of the scattered wave U3 is weaker than the surface reflected wave U1 and the bottom reflected wave U2, but by using the envelopes rather than the waveforms themselves for analysis and by averaging the correlation coefficients obtained for sets of multiple envelopes, the influence of noise can be reduced, and highly accurate analysis can be performed based on the ultrasonic signals derived from the scattered wave U3. In this way, by reducing the effects of noise and increasing the sensitivity and accuracy of measurements, when a large number of individual pieces are continuously produced in a manufacturing process for metal materials, etc., even if some of the individual pieces have areas with non-uniform crystal structure in very limited areas, it is possible to detect such non-uniform crystal structure with high sensitivity and precision, and accurately select individuals with a uniform crystal structure that gives them the desired material properties.

[0070] In the crystalline structure evaluation method according to this embodiment, information about the uniformity of the crystalline structure can be obtained simply by creating an envelope for the ultrasonic signal obtained for each incident point P, calculating the correlation coefficient between the multiple envelopes, and obtaining the average value. There is no need to perform complex arithmetic processing or simulations based on a crystalline structure model on the detected ultrasonic signals. The uniformity of the crystalline structure can be easily evaluated using only the simple measurement of the incidence and detection of ultrasonic waves U at each incident point P and simple arithmetic processing. As a result, even when inspecting a large number of individual objects, the inspection of each individual object can be completed in a short time.

[0071] In the crystalline structure evaluation method according to this embodiment, the crystalline structure uniformity is evaluated based on a correlation coefficient indicating the degree of similarity between ultrasonic signals acquired as a series of data groups. The evaluation of crystalline structure uniformity is not related to the intensity of the ultrasonic signals themselves. This is completely different from the crystalline structure evaluation method of Patent Document 2, in which the evaluation of crystalline structure uniformity is based on the scattering intensity of ultrasonic waves U, i.e., the intensity value of the ultrasonic signals themselves. As described above with reference to FIG. 6 , when evaluating crystalline structure uniformity based on the intensity of ultrasonic signals, the evaluation results may change if the measurement conditions in the ultrasonic inspection device 11, such as the incidence conditions of the ultrasonic waves U, the sensitivity of the detection sensor, and the calibration status of each device, change. In contrast, when evaluating crystalline structure uniformity based only on the relative relationship of ultrasonic signals acquired under the same measurement conditions, as in the crystalline structure evaluation method according to this embodiment, the absolute intensity of the ultrasonic signals does not affect the results, and therefore the evaluation results are independent of the measurement conditions. Therefore, in the ultrasonic inspection device 11, even if the ultrasonic inspection conditions change due to changes in the incidence conditions of the ultrasonic waves U or the sensitivity of the detection sensor, fluctuations in the adjustment and calibration of each part of the device including the detection sensor, changes over time, etc., the uniformity of the crystal structure can be stably evaluated. Furthermore, even if the adjustment and calibration of each part of the device including the detection sensor are not performed very strictly, the uniformity of the crystal structure can be evaluated with high reliability.

[0072] (4) Other processes The crystalline structure evaluation device 1 used in the crystalline structure evaluation method according to this embodiment is equipped with an ultrasonic inspection device 11, which can also be used to detect flaws in materials. Therefore, in the crystalline structure evaluation method according to this embodiment, the test material S can be inspected for scratches and defects in addition to evaluating the uniformity of the crystalline structure. Here, scratches and defects refer to those large in size that are not on the order of crystal grains, generally those of submillimeter size or larger.

[0073] If a flaw exists within the test material S, a component due to the reflected wave from the flaw should be observed between the component A1 due to the surface reflected wave U1 and the component A2 due to the bottom reflected wave U2 in the ultrasonic signal obtained corresponding to each incident point P during the measurement process, as shown in Figure 3. If such a component is observed, it can be determined that a flaw exists in the test material S at a position inside the test material S, away from the incident point P. By setting an evaluation region G excluding the signal component due to the flaw and calculating the correlation coefficient of the envelope, both flaw detection and evaluation of the uniformity of the crystalline structure can be performed using the same data. Furthermore, the obtained ultrasonic signal may be analyzed for the bottom reflected wave U2, as well as for sound velocity, velocity dispersion, attenuation, etc., to simultaneously analyze various properties of the material constituting the test material S. [Example]

[0074] Examples of the present invention are shown below. However, the present invention is not limited to these examples. Here, a titanium-based alloy was used as a sample to verify the validity of the crystalline structure evaluation method according to the above embodiment. In addition, the behavior of the evaluation results when the sensitivity of the detection sensor of the ultrasonic inspection device is changed was compared between the crystalline structure evaluation method of the present invention and the crystalline structure evaluation method of Patent Document 2.

[0075] (Test Method) In the test, a titanium-based α+β type alloy (an alloy that forms both α and β phases) was used as the sample. Specifically, a φ16 mm round bar of the alloy was forged using different forging ratios to produce three samples, designated Samples 1 to 3, with different degrees of uniformity in the crystalline structure. SEM observations revealed that the uniformity of the crystalline structure decreased in the order of Samples 1, 2, and 3 (Sample 3 was the least uniform), and that many coarse grains were mixed in with fine grains.

[0076] The prepared samples were subjected to the measurement process, envelope processing process, correlation coefficient acquisition process, and averaging process in the same manner as described above for the crystalline structure evaluation method according to an embodiment of the present invention, and the average correlation coefficient was calculated. Here, ultrasonic pulses with a frequency of 20 MHz were used, and incident points were set at 1° intervals around the entire circumference of the sample and at 0.5 mm intervals over a 10 mm area along the longitudinal direction to perform the measurement process. The ultrasonic signal recorded was the time change of ultrasonic waves with the same frequency component as the fundamental wave in the reflection direction. Each process after the envelope processing process was performed using all ultrasonic signal data obtained in the measurement process, and the number of data (N) was 7141. Measurement and evaluation were performed for each sample under two measurement conditions. That is, the sensitivity of the detection sensor of the ultrasonic inspection device was set to two values: "low sensitivity" and "high sensitivity," and the data was analyzed for each value. Here, the gain of the detection sensor was set to 0 dB and 6 dB for "low sensitivity" and "high sensitivity," respectively.

[0077] In addition, the same data obtained in the measurement process was also evaluated using the crystalline structure evaluation method described in Patent Document 2. Here, the data was analyzed using the procedure described in Patent Document 2. In summary, the scattering intensity was calculated by integrating the signal intensity within the evaluation region (G) for each ultrasonic signal obtained in the measurement process. The number of incident points at which a given scattering intensity was obtained was then organized as a function of scattering intensity to create a scattering intensity function. The graph of this scattering intensity function was separated into a main peak and a sub-peak located on the higher scattering intensity side of the main peak (see Figure 6). The sub-peak (f(x)) was then weighted by its distance from the mode (m) of the scattering intensity function (f(x) |xm|), and this value was integrated over the entire scattering intensity region to obtain a heterogeneity evaluation value.

[0078] (result) FIG. 7 shows the results of the crystal structure evaluation. (a) shows the results when the method of the present invention was used, and (b) shows the results when the method of Patent Document 2 was used. The graphs show the results when low-sensitivity measurement results and high-sensitivity measurement results were used for three samples, respectively. In (a), the low-sensitivity case of Sample No. 1 is set to 100%, and the average correlation coefficient of each evaluation result is displayed as a relative value, which is shown as the "normalized uniformity evaluation value" on the vertical axis. A larger normalized uniformity evaluation value indicates higher uniformity of the crystal structure. In (b), the non-uniformity evaluation value obtained by analysis increases with increasing non-uniformity of the crystal structure. Therefore, the reciprocal of the non-uniformity evaluation value is taken, and the "normalized uniformity evaluation value" is displayed as a relative value, with the low-sensitivity case of Sample No. 1 set to 100%.

[0079] The evaluation results of the crystalline structure evaluation method of the present invention shown in Figure 7(a) show a positive correlation between the uniformity of the crystalline structure and the normalized uniformity evaluation value, i.e., the average correlation coefficient. In other words, as the uniformity of the crystalline structure decreases from Sample No. 1 to Sample No. 2 to Sample No. 3, the normalized uniformity evaluation value decreases. This shows that the crystalline structure evaluation method of the present invention calculates correlation coefficients between envelopes of ultrasonic signals acquired at multiple incidence points and then averages the correlation coefficients. This average correlation coefficient can be effectively used as a parameter that quantitatively represents how close the crystalline structure formed is to an ideal structure containing fine, uniform crystal grains.

[0080] Whether the detection sensor sensitivity is low or high, a common trend is observed in the normalized uniformity evaluation value, decreasing as the uniformity of the crystal structure decreases, from sample No. 1 to sample No. 2 to sample No. 3. The normalized uniformity evaluation value itself is also nearly the same for each sample, whether it is high or low sensitivity. In other words, the crystal structure evaluation method of the present invention can evaluate the uniformity of the crystal structure based on the value of the average correlation coefficient, regardless of the sensitivity of the detection sensor. This is because the average correlation coefficient is a quantity based on the relative similarity of multiple ultrasonic signals, rather than the intensity of the ultrasonic signal itself.

[0081] Next, the evaluation results using the crystalline structure evaluation method of Patent Document 2 (Fig. 7(b)) show a tendency for the normalized uniformity evaluation value to decrease as the crystalline structure uniformity decreases from sample No. 1 to sample No. 2 to sample No. 3, both in the low-sensitivity and high-sensitivity cases. This trend is the same as when the evaluation method of the present invention (Fig. 7(a)) is applied, confirming that the crystalline structure evaluation method of Patent Document 2 can also evaluate the degree of crystalline structure uniformity through ultrasonic signal analysis. However, while the normalized uniformity evaluation values ​​are nearly the same for low and high sensitivity cases in Fig. 7(a), the normalized uniformity evaluation values ​​for all samples in Fig. 7(b) are significantly smaller for high sensitivity cases than for low sensitivity cases. This is because the method of Patent Document 2 evaluates the scattering intensity, which is the integration of the absolute value of the ultrasonic signal intensity. As shown in Fig. 6(b), as the sensitivity of the detection sensor increases, the scattering intensity function shifts toward higher scattering intensity and becomes distributed over a wider range of scattering intensity.

[0082] Even when analysis is performed based on the intensity of ultrasonic signals as absolute values, as in the embodiment of Patent Document 2, the uniformity of the crystalline structure can be quantitatively evaluated as long as the measurement conditions do not change. Furthermore, the intensity value of the ultrasonic signals can be actively used to analyze the state of the crystalline structure. However, changes in measurement conditions, such as the sensitivity of the detection sensor, make it difficult to uniformly interpret the evaluation results regarding the uniformity of the crystalline structure. In contrast, when analysis is performed based on the relative relationship between multiple ultrasonic signals, as in the embodiment of the present invention, it is possible to evaluate the uniformity of the crystalline structure and compare the evaluation results even if the measurement conditions change. Note that in Figure 7, the method of the present invention and the method of Patent Document 2 use completely different types of parameters to calculate the normalized uniformity evaluation value, so comparing the magnitude of the normalized uniformity evaluation value itself between Figures 7(a) and 7(b) is meaningless.

[0083] The embodiments of the present invention have been described above. The present invention is not particularly limited to these embodiments, and various modifications are possible. As described above, the crystalline structure evaluation device can be configured to detect transmitted waves in the transmission direction, but the intensity of the transmitted waves decreases as the intensity of the scattered waves increases. [Explanation of symbols]

[0084] 1. Crystal structure evaluation device 11 Ultrasound inspection equipment 12 Exercise equipment 13 Arithmetic and control equipment P incidence point S Test material S1 surface S2 bottom S3 grain S4 grain boundary U Incident ultrasound U1 surface reflected wave U2 bottom reflected wave U3 scattered wave

Claims

1. a measuring step of irradiating ultrasonic waves from a plurality of incident points on the surface of a test piece made of a material containing crystal grains, measuring the intensity of the ultrasonic waves as a function of time in the reflection direction or transmission direction of the incident ultrasonic waves, and acquiring the ultrasonic signals for each of the incident points; an envelope processing step of obtaining an envelope for each ultrasonic signal obtained in the measuring step; a correlation coefficient acquisition step of creating a plurality of pairs of envelopes by selecting two envelopes from the group of envelopes obtained in the envelope processing step, and calculating a correlation coefficient between the two envelopes in at least a portion of a time domain for each pair; an averaging step of calculating an average value of the plurality of correlation coefficients obtained in the correlation coefficient obtaining step; a structure evaluation step in which the larger the average value obtained in the averaging step, the more the test material is deemed to have a crystalline structure in which crystal grains that give small ultrasonic scattering intensity are formed with high uniformity.

2. 2. The crystalline structure evaluation method according to claim 1, wherein in the structure evaluation step, the test material is deemed to have a crystalline structure in which crystal grains having small grain sizes are formed with high uniformity as the average value increases.

3. In the envelope processing step, an envelope is obtained for all of the ultrasonic signals obtained in the measuring step; In the correlation coefficient acquisition step, correlation coefficients are evaluated for all combinations of the envelopes obtained in the envelope processing step; 2. The crystalline structure evaluation method according to claim 1, wherein in the averaging step, an average value is calculated for all of the correlation coefficients obtained in the correlation coefficient obtaining step.

4. In the measuring step, the ultrasonic signal is acquired in the reflection direction; 2. The crystalline structure evaluation method according to claim 1, wherein in the correlation coefficient acquisition step, the correlation coefficient of the envelope is obtained in a time domain between a time domain on the ultrasonic signal corresponding to a reflected wave reflected by a surface of the test material and a time domain corresponding to a reflected wave reflected by a bottom surface opposite the surface.

5. 2. The crystalline structure evaluation method according to claim 1, wherein in the envelope processing step, the envelope is created using a Hilbert transform.

6. Equipped with an ultrasonic inspection device that generates and detects ultrasonic waves, A crystal structure evaluation device that performs the crystal structure evaluation method according to any one of claims 1 to 5 on a test piece made of a material containing crystal grains.

Images