JP2005156305A - Evaluation method of internal defect - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an evaluation method of internal defects capable of obtaining correlation between a flaw distribution acquired by an ultrasonic inspection method and that acquired by a radiographic test method. <P>SOLUTION: This evaluation method of internal defects has a flaw detection process for acquiring flaw detection data reflecting a spacial distribution of the internal defects existing inside a specimen by ultrasonic inspection, a two-dimensional flaw information acquisition process for projecting the flaw detection data in one or two or more monitoring regions in the Z-direction, and acquiring two-dimensional flaw information in each monitoring region respectively, a determination data conversion process for converting each two-dimensional flaw information into the same or one kind or two or more kinds of different determination data respectively, and a determination process for setting the same or one or two or more different determination thresholds respectively in each monitoring region, comparing each determination data with one or two or more allowable data allowable to each determination threshold, and determining the quality of the specimen. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an internal defect evaluation method, and more particularly to an internal defect evaluation method for evaluating internal defects such as unopened cracks and cast holes existing in a specimen.

  Non-destructive inspection is an inspection method for examining the presence / absence, position, size, or material of a flaw without destroying a material or product. Non-destructive inspection does not require destruction of materials and products, and is used for product inspection before shipment, quality evaluation of an object after installation on the site, life prediction, and the like.

  As a nondestructive inspection method, a radiation transmission test method, an ultrasonic flaw detection method, a penetration flaw detection test method, an eddy current flaw detection test method, and the like are known. Among these, the penetrant flaw detection test method and the eddy current flaw detection test method are methods suitable mainly for inspection of surface defects. On the other hand, the radiation transmission test method and the ultrasonic flaw detection method are mainly suitable for detecting internal defects. Therefore, it is common to use a radiation transmission test method or an ultrasonic flaw detection method for detecting cracks, blowholes, cast holes and the like formed inside welded parts, cast products and the like.

  The radiation transmission test method is an inspection method in which an object is irradiated with radiation and the state of a defect is examined from a change in intensity of the transmitted radiation. The intensity of the radiation that passes through the material varies depending on the type and thickness of the material, but does not depend much on the inner surface shape of the defect. Therefore, the radiation transmission test method uses defects whose inner surface shape is irregular (for example, a cast hole). Even so, it can be detected with high accuracy. In addition, when a plurality of defects are arranged along the radiation transmission direction, the degree appears as a change in the intensity of the transmitted radiation, so that the defect information inside the specimen can be obtained relatively easily. There is.

  On the other hand, the ultrasonic flaw detection method is an inspection method for investigating the presence or absence of a defect by propagating an ultrasonic wave from the surface of the test body to the inside and detecting the ultrasonic wave reflected by the defect. In the ultrasonic flaw detection method, flaw information inside the specimen can be acquired relatively easily by optimizing the direction in which the probe is scanned. However, since ultrasonic waves are straight, there is a problem that it is difficult to detect defects on the downstream side with respect to the traveling direction of the ultrasonic waves when a plurality of defects are arranged along the traveling direction of the ultrasonic waves. there were.

  In order to solve this problem, various proposals have heretofore been made. For example, Patent Document 1 discloses an ultrasonic transducer in which a plurality of piezoelectric layers are independently formed in a matrix, a drive unit capable of driving any piezoelectric layer, and driving any piezoelectric layer. The detection unit that receives the reflected echo reflected from the irradiation target by the plurality of piezoelectric layers and detects the electrical signal generated at that time, and the processing unit that visualizes the state of the inspection target from the detected electrical signal, An ultrasonic inspection apparatus including the above is disclosed. In this document, when barium titanate or lead zirconate titanate having a thickness of 0.1 μm to 100 μm is used as the piezoelectric layer, ultrasonic waves having a high frequency can be generated, and the resolution is improved. Is described.

  Non-Patent Document 1 discloses a 3D ultrasonic inspection apparatus including an ultrasonic camera in which piezoelectric elements are arranged in a matrix. This document describes that the inside of an inspection object can be imaged at high speed three-dimensionally by instantaneously collecting thousands to tens of thousands of waveforms with such an ultrasonic camera and performing image composition processing. ing.

  Furthermore, Non-Patent Document 2 discloses that an ultrasonic wave is transmitted from an array probe in which a large number of minute ultrasonic transducers are arranged at different timings, and a combined beam is transmitted in a specific direction, or A phased array method that converges to a predetermined depth is disclosed. This document describes that the phased array method can generate a beam having an arbitrary variety of beam angles and focal lengths by computer control, so that a flaw detection aiming at a complicated target portion can be accurately performed.

JP 2003-149213 A http://www.toshiba.co.jp/efort/market/camera/example.htm http://rd-tech.co.jp/thchno.html

  Since the radiation transmission test is less affected by the inner surface shape of the defect and the structure of the specimen, it is a defect having an irregular inner surface shape (for example, a cast hole) or a defect inside the specimen having a coarse grain structure. However, detection is possible with relatively high accuracy. Therefore, a radiation transmission test method is generally used for final inspection before shipment of products (for example, castings) in which these are problems.

  However, the radiation transmission test method requires environmental considerations when implemented due to the handling of radiation. In addition, there is a physical restriction that radiation must be emitted from one of the test bodies and a film or a fluorescent plate must be disposed on the opposite side. For this reason, even if it is necessary to inspect non-destructively the product quality degradation, life prediction, etc. after installing the product at the site, radiation will occur at the site due to environmental reasons and / or physical constraints. Permeation tests may not be performed.

  On the other hand, the ultrasonic flaw detection method is not subject to environmental restrictions unlike the radiation transmission test method. Further, the inspection can be performed only by arranging the probe on at least one surface of the test body, and the shape of the probe can be arbitrarily selected, so that there are few physical restrictions. Furthermore, if the various methods described above are used, even three-dimensional information of defects can be acquired.

  However, the signal intensity obtained by the pulse reflection method greatly depends on the reflectance of the reflection source. Therefore, in the conventional method, when the attenuation or scattering of ultrasonic waves is large, it may be difficult to accurately detect the position and size of the defect. Furthermore, the correlation between the result obtained by the ultrasonic flaw detection method and the result obtained by the radiation transmission test method is small, and a method for obtaining the correlation has not been established. Therefore, when a product whose inspection at the time of shipment is standardized by the radiation transmission test method is installed at the site and the product needs to be re-inspected, if the radiation transmission test cannot be used at the site, the product quality There is a problem that accurate evaluation of deterioration, life prediction, etc. is difficult.

  The problem to be solved by the present invention is to provide an internal defect evaluation method capable of accurately evaluating the presence or absence of defects even when the attenuation or scattering of ultrasonic waves is large. Another problem to be solved by the present invention is to provide an internal defect evaluation method capable of obtaining a correlation between a flaw distribution obtained by an ultrasonic flaw detection method and a flaw distribution obtained by a radiation transmission test method. There is.

  In order to solve the above-described problem, the internal defect evaluation method according to the present invention performs ultrasonic flaw detection on a specimen, and obtains flaw detection data reflecting the spatial distribution of the internal defects present in the specimen. A process, a two-dimensional flaw information acquisition step of projecting the flaw detection data in one or more monitoring areas in the Z direction, and acquiring two-dimensional flaw information for each of the monitoring areas, and the two-dimensional flaw information A determination data conversion step for converting flaw information into one or two or more types of determination data that are the same or different, and one or two or more determination threshold values are set for each monitoring area And a determination step of comparing each determination data with one or more allowable data allowed for each determination threshold value and determining the quality of the specimen. The gist. In this case, it is preferable that the test body obtains the flaw detection data by making an ultrasonic wave incident from the XY plane of the test body in the Y direction as a direction in which a radiation transmission test should be performed. .

  When flaw detection data in the monitoring area is projected in the Z direction to obtain two-dimensional flaw information, and this is converted into data for determination such as the indicated length, area ratio, average intensity, etc. It becomes possible to distinguish clearly. For this reason, whether or not the quality of the product is good can be determined even when the attenuation or scattering of the ultrasonic waves is large.

  In addition, when the radiation transmission test of the specimen is performed from the Y direction, two-dimensional flaw information correlated with the radiation transmission test method can be obtained when ultrasonic waves are incident from the XY plane of the specimen. If this is converted into determination data and the determination data is compared with the permissible data corresponding to the determination threshold value, quality determination having a correlation with the radiation transmission test method can be performed.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. The internal defect evaluation method according to the present invention includes a flaw detection process, a two-dimensional flaw information acquisition process, a determination data conversion process, and a determination process.

  First, the flaw detection process will be described. The flaw detection step is a step of performing flaw detection on the specimen and obtaining flaw detection data reflecting the spatial distribution of internal defects existing inside the specimen. In the present invention, the shape, material, etc. of the test specimen are not particularly limited, and can be applied to test specimens having various shapes, materials, etc. In particular, when the test specimen is a high manganese cast steel product, the ultrasonic attenuation is large and a defect with large scattering is likely to occur, but the method according to the present invention can accurately determine the quality of the product. it can.

  Also, the flaw detection method, the probe structure, etc. are not particularly limited, and various flaw detection methods and probes can be used according to the shape of the specimen, the inspection purpose, and the like. For example, as shown in FIG. 1, when the test body 10 is rectangular, the probe 12 may be disposed on the XY plane of the test body 10 and the XY plane may be scanned using the probe 12. . The scanning method is not particularly limited, and various scanning methods such as zigzag scanning, horizontal scanning, and vertical scanning can be used. Any flaw detection method may be used as long as flaw detection data reflecting the spatial distribution of internal defects existing in the specimen can be obtained. Specifically, an optimum method is selected according to the shape of the test body 10, the structure of the probe 12, the inspection purpose, such as the vertical flaw detection method and the oblique flaw detection method.

  For example, when using a probe 12 having one piezoelectric element, a vertical flaw detection method using one set or two or more sets of a probe for transmission and a probe for reception is used. It is preferable to acquire flaw detection data by using. For example, when using a probe 12 having a plurality of piezoelectric elements (for example, a matrix type probe in which piezoelectric elements are arranged in a matrix), a single probe is used. The flaw detection data may be acquired by the vertical flaw detection method, or the flaw detection data may be acquired by the vertical flaw detection method using one set or two or more sets of probes. Further, instead of mechanical scanning of the probe 12, an array probe capable of changing the timing and transmitting ultrasonic waves from a large number of minute piezoelectric elements arranged as the probe 12 is used. You may make it scan electrically.

  In the present invention, the “X direction”, “Y direction”, and “Z direction” are used for convenience in determining the direction of data processing to be described later, and the XY plane does not necessarily exceed the XY plane. The present invention is not limited to the case where the sound wave is incident (that is, the Z direction is the ultrasonic wave incident direction). However, in order to obtain flaw detection data correlated with the radiation transmission test, ultrasonic flaw detection is performed from the direction (or surface) perpendicular to the direction (or surface) where the radiation transmission test should be performed. Is preferred. For example, when the direction in which the radiation transmission test is performed is the Y direction, the ultrasonic incident surface is preferably an XY plane.

  Next, the two-dimensional flaw information acquisition process will be described. The two-dimensional flaw information acquisition step is a step of projecting flaw detection data in one or more monitoring regions in the Z direction and acquiring two-dimensional flaw information for each monitoring region. There are various methods for acquiring such two-dimensional flaw information, but the following method is preferable.

  The first method for acquiring two-dimensional flaw information includes a three-dimensional flaw information acquisition process, a mask process, a sensitivity correction process, and a two-dimensional flaw information conversion process.

  The three-dimensional flaw information acquisition step is a step of processing flaw detection data obtained in the flaw detection step and acquiring three-dimensional flaw information inside the specimen. Here, the “three-dimensional flaw information” refers to information including the size and shape of internal defects existing in the specimen and their positions (ie, spatial distribution). In this case, the method for acquiring the flaw detection data is not particularly limited, and various methods can be used according to the shape of the specimen, the inspection purpose, and the like. Specifically, the above-described vertical flaw detection method, a method using a matrix probe, a method using an array probe, and the like are preferable.

  When flaw detection data is processed using a predetermined calculation program, three-dimensional flaw information reflecting the spatial distribution of defects is obtained. In this case, the flaw detection data processing method is not particularly limited, and various methods can be used. Specific examples of the flaw detection data processing method include an aperture synthesis method and a phased array method.

  2A to 2C are schematic diagrams of the acquired three-dimensional flaw information. When the flaw detection data detected by the probe is processed by a computer using a predetermined method, as shown in FIG. 2, the positions, sizes, and shapes of the defects 10a, 10b, 10c,. 3D flaw information indicating the above is obtained. The defects in the test body 10 are arranged at random, and some of them may be arranged in the Z direction. The defect 10b and the defect 10c in FIG. 2 schematically represent such a state.

  When a single probe having one piezoelectric element is used for such a test body 10 and a vertical flaw detection is performed only from the Z direction, the reflectance of the defect 10b located on the ultrasonic incident side is measured. 2 is lower than the reflectance of the defect 10c (that is, when the scattering of ultrasonic waves at the defect 10b is large), as shown in FIG. 2 (a), only weak reflected echoes due to the defect 10b are detected. A strong reflected echo due to 10c is not detected. However, when the three-dimensional flaw information is acquired using the various methods described above, as shown in FIGS. 2B and 2C, the reflected echo due to the defect 10c is also included in the three-dimensional flaw information. Will be recorded.

  Further, the ultrasonic wave incident on the test body 10 is reflected on the surface and the bottom surface of the test body 10 and is observed as a surface echo and a bottom surface echo, respectively. In this case, it is relatively easy to separate the reflected echo from the defect from the surface echo and the bottom echo at the central portion of the test body 10. However, it is often difficult to separate the outer peripheral portion of the test body 10, and reflected echoes due to surface irregularities, surface shapes, and the like may be observed. In FIG. 2, the cross-hatched region on the outer periphery of the test body 10 schematically represents a state in which a reflection echo caused by a reflection source other than such an internal defect is observed.

  Note that the three-dimensional flaw information acquisition step only needs to acquire at least three-dimensional flaw information having a data structure that can be processed later, and imaging as illustrated in FIG. 2 is not necessarily required. However, when the three-dimensional flaw information is visualized, an unexpected reflection echo detected by force majeure can be discriminated and eliminated with the naked eye, and there is an advantage that the determination accuracy is improved.

  Next, the sensitivity correction process will be described. The sensitivity correction step is a step of correcting the sensitivity of the acquired three-dimensional flaw information. The size of the reflected echo due to the defects 10a, 10b,... Inside the test body 10 mainly depends on the inner surface shape, size, etc. of the defects 10a, 10b, etc., but the intensity of the ultrasonic wave incident on the test body 10 Depends on other factors. These factors also change depending on the intensity of the ultrasonic wave transmitted from the probe and the contact state between the probe and the specimen.

  Therefore, in the sensitivity correction step, fluctuations in the reflected echo caused by factors other than such defects are corrected. The correction method is not particularly limited, and an optimal method may be selected in accordance with the method for acquiring three-dimensional flaw information, the material of the specimen, and the like. Specifically, a method of taking the ratio (= A / B) of the reflected echo intensity (A) caused by the defect to the reflected echo intensity (B) from the defect-free region, the defect with respect to the reflected echo intensity (C) from the bottom surface Suitable examples include a method of taking the ratio of echo intensity (A) (= A / C), a method of applying a constant coefficient according to the propagation distance of ultrasonic waves, and the like.

  Although sensitivity correction has the advantage of improving the accuracy of quality determination, depending on the shape of the specimen, the acquisition method of three-dimensional flaw information, the inspection purpose, etc., the intensity of the reflected echo due to the defect can be directly measured. Even if it is used, it may be possible to determine with sufficient accuracy. In such a case, the sensitivity correction step may be omitted. When performing sensitivity correction, the sensitivity correction process may be performed before a mask process described later, or may be performed after the mask process.

  Next, the mask process will be described. The masking process is a process of cutting out one or more monitoring areas from the three-dimensional flaw information whose sensitivity is corrected as necessary. Here, the “monitoring area” refers to an area to be inspected among the areas inside the specimen. As described above, the acquired three-dimensional flaw information may include reflection echoes caused by reflection sources other than internal defects. If data processing is performed in a subsequent process while including information other than such internal defects, it may cause erroneous determination.

  Therefore, in the mask process, information outside the monitoring area is deleted from the acquired three-dimensional flaw information. Which part of the three-dimensional flaw information is set as the monitoring region can be arbitrarily selected according to the shape of the specimen, the inspection purpose, and the like. If the monitoring area is cut out, it is impossible to detect a defect that exists outside the monitoring area. However, when it is necessary to evaluate defects outside the monitoring area, there is no particular problem because a new monitoring area may be set or another nondestructive inspection method may be used in combination.

  The method for cutting out the monitoring area is not particularly limited, and an optimum method is selected according to the shape of the specimen and the acquisition method of the three-dimensional flaw information. For example, when the appearance pattern of the reflected echo caused by the reflection source other than the shape of the specimen and the internal defect is known, the specimen shape and the monitoring area of the predetermined pattern are input to the evaluation device in advance and acquired. The monitoring area may be automatically cut out from the three-dimensional flaw information thus obtained according to a predetermined pattern.

  Further, for example, when the appearance pattern of the reflected echo caused by the reflection source other than the shape of the specimen and / or the internal defect is unknown, the acquired three-dimensional flaw information is displayed on the monitor. Next, the reflection echo caused by the reflection source other than the internal defect is visually determined, and this can be manually deleted from the three-dimensional defect information.

  FIG. 3A to FIG. 3C are schematic views of monitoring areas cut out from the three-dimensional flaw information shown in FIG. In the example shown in FIG. 3, the monitoring area is an internal area within a certain depth from the surface in the three-dimensional flaw information. In addition, the other area in the vicinity of the surface (that is, the cross-hatched area in FIG. 2) is deleted from the three-dimensional flaw information.

  Also, as shown in FIG. 3, one monitoring area may be cut out from the three-dimensional flaw information, or two or more monitoring areas may be cut out. For example, in a cast product, various defects may occur in various parts depending on the material, shape, and the like. In addition, the structure is often coarse, and even when defects of the same shape and the same type occur, the intensity of the reflected echo varies depending on the depth from the surface. In such a case, it is preferable to set a plurality of monitoring areas. If a plurality of monitoring areas are set, it becomes easy to apply different criteria for each monitoring area, so that highly accurate determination can be performed.

  When a plurality of monitoring areas are set, the setting method is not particularly limited, and can be arbitrarily selected according to the inspection purpose, the shape of the specimen, and the like. For example, a plurality of monitoring regions may be set along the depth direction from the ultrasonic incident surface, or a plurality of monitoring regions may be set along the in-plane direction of the incident surface. Furthermore, these may be combined. In this case, the monitoring areas may overlap each other, or the interval between the monitoring areas may be zero or a fixed finite value. Further, the degree of overlap and / or the size of the interval may be the same for each monitoring area, or may be different for each monitoring area.

  FIG. 4 shows an example of a test body in which a plurality of monitoring areas are set. In FIG. 4, the test body consists of a cast rail 20. Moreover, let the upper surface of the casting rail 20 be an XY plane, and let the height direction be a Z direction.

  From the XY plane toward the Z direction, the surface layer portion and the middle portion of the rail head portion 20a, and the position corresponding to the boundary line between the rail head portion 20a and the upright portion 20b (hereinafter referred to as “neck portion”), Three monitoring areas are set: an upper monitoring area 22a, an intermediate monitoring area 22b, and a lower monitoring area 22c. In this case, the size, position, interval, and the like of each monitoring area can be arbitrarily selected according to the shape of the casting rail 20, the purpose of inspection, and the like as described above. Note that such a method for setting the monitoring region can be applied to a test body other than the cast rail.

Next, the two-dimensional flaw information conversion process will be described. The two-dimensional flaw information conversion step is a step in which the maximum value or integrated value in the Z direction of the three-dimensional flaw information in each monitoring area is projected in the Z direction and converted into two-dimensional flaw information. Here, “projecting the maximum value in the Z direction in the Z direction” means that the maximum value of the reflected echo intensity in the Z direction at a certain coordinate (X ₀ , Y ₀ ) is the coordinate (X ₀ , Y ₀ ). "Projecting the integrated value in the Z direction in the Z direction" means that the integrated value of the reflected echo intensity in the Z direction is a two-dimensional flaw in coordinates (X ₀ , Y ₀ ). To represent as information. Which one is used is selected according to the shape, material, purpose of inspection, etc. of the specimen.

  For example, when detecting a relatively large defect, or when the specimen has a shape and / or material that does not easily generate noise, the maximum value or integrated value in the Z direction is used as two-dimensional flaw information. Whichever of these is used, a highly accurate determination can be made. On the other hand, if a relatively small defect is detected, or if the specimen has a shape and / or material that is likely to generate noise, the maximum value in the Z direction is taken as two-dimensional flaw information. Is preferred. Taking an integrated value in such a case is not preferable because noise is also integrated and the determination accuracy is lowered.

  FIG. 5 is a schematic diagram of two-dimensional flaw information obtained by projecting the maximum value in the monitoring area shown in FIG. 3 in the Z direction. In FIG. 5, since the defects 10a, 10d, and 10e do not have other defects in the Z direction, their reflected echo intensities are directly projected on the XY plane. On the other hand, since the defects 10b and 10c are arranged vertically in the Z direction, the value of the defect 10c having a high reflection echo intensity is not the value of the defect 10b having a low reflection echo intensity in the region where the defect 10b and the defect 10c overlap. Is selected as the maximum value and projected onto the XY plane.

  When a plurality of monitoring areas are set, the maximum value or integrated value in the Z direction may be projected in the Z direction for each monitoring area in accordance with the procedure shown in FIGS. For example, as shown in FIG. 4, when three monitoring areas, an upper monitoring area, an intermediate monitoring area, and a lower monitoring area, are set, the flaw detection data in each monitoring area is projected in the Z direction, What is necessary is just to convert into upper 2D flaw information, intermediate 2D flaw information, and lower 2D flaw information.

  A second method for acquiring two-dimensional flaw information is a method using so-called “C scope data”, and includes a maximum value acquisition step, a sensitivity correction step, and a two-dimensional flaw information conversion step.

  The maximum value acquisition step is a step of acquiring the maximum value in the Z direction of the flaw detection data in one or more monitoring areas. In the second method, only the maximum value in the Z direction in the monitoring area (gate) is acquired from the flaw detection data, and acquisition of the three-dimensional flaw information is not performed. This is different from the first method. Further, since the three-dimensional flaw information is not acquired, the flaw detection data needs to be acquired by projecting the maximum value, that is, by making the ultrasonic wave incident from the XY plane.

  In the second method, one or a plurality of monitoring regions may be set, a plurality of monitoring regions along the depth direction from the incident surface and / or along the in-plane direction of the incident surface. Since the point that can be set and the overlap and / or interval of the monitoring areas can be arbitrarily set are the same as in the first method, the description thereof is omitted.

  The sensitivity correction step is a step of correcting the acquired flaw detection data or the maximum sensitivity. Note that the sensitivity correction step may be performed before or after the maximum value acquisition step. Also, since various methods can be used as the sensitivity correction method and the sensitivity correction step may be omitted depending on the inspection purpose, it is the same as the sensitivity correction step in the first method, and thus will be described. Is omitted.

  The two-dimensional flaw information conversion step is a step of projecting each maximum value in one or two or more monitoring areas in the Z direction and converting each maximum value into two-dimensional flaw information. The two-dimensional information conversion step in the second method is the same as 2 in the first method except that the data to be projected in the Z direction is the maximum value in the Z direction obtained directly from the flaw detection data in the monitoring area. Since this is the same as the dimension flaw information conversion step, description thereof is omitted.

  Next, the determination data conversion step will be described. The determination data conversion step is a step of converting each two-dimensional flaw information acquired by the two-dimensional flaw information acquisition step into the same or different one or more types of determination data. There are various methods for converting the two-dimensional flaw information into determination data, and the following methods are particularly preferable.

  The first method of converting the two-dimensional flaw information into determination data is a method using “instruction length” as the determination data, a binarization step, a one-dimensional flaw information conversion step, and instruction length calculation. Process.

  First, the binarization process will be described. The binarization step is a step of binarizing one or two or more two-dimensional flaw information using the same or different binarization threshold values, and acquiring one or two or more binarized data. Here, the “binarization threshold value” refers to a value for distinguishing between noise and a reflected echo caused by a defect. As the binarization threshold value, an optimum value may be selected according to the magnitude of noise, the size of a defect to be detected, and the like.

  When two or more monitoring areas are set, the binarization threshold value may be the same for each monitoring area, or may be different for each monitoring area. In particular, as shown in FIG. 4, in the case where a plurality of monitoring areas are set in the Z direction, when the ultrasonic incident surface is an XY plane, the scattering of ultrasonic waves and the ultrasonic wave depending on the depth from the incident surface to the monitoring area. Since the degree of attenuation is different, it is preferable to use a different binarization threshold value for each monitoring area.

  Specifically, the two-dimensional flaw information is divided into minute regions having an appropriate size. Next, for each minute area, it is determined whether or not the reflected echo intensity exceeds the binarization threshold value. When the reflected echo intensity does not exceed the binarization threshold value, the first numerical value (for example, “0”) is given to the minute region, and when the reflection echo intensity exceeds the binarization threshold value, A second numerical value (for example, “1”) is given to the minute area.

  FIG. 6 shows a schematic diagram of binarized data obtained by binarizing the two-dimensional flaw information shown in FIG. As shown in FIG. 6, by binarizing the two-dimensional flaw information, noise (hatched areas in FIG. 5) is removed, and only reflected echoes caused by defects remain in the binarized data. Become. Further, by optimizing the binarization threshold value, not only the defect 10c but also a relatively weak reflected echo caused by the defect 10b can be reflected in the binarized data.

  Next, the one-dimensional flaw information conversion process will be described. The one-dimensional flaw information conversion step is a step of projecting the maximum value or integrated value in the Y direction of each binarized data in the Y direction and converting it into one or more one-dimensional flaw information. By projecting the binarized data in the Y direction, the amount of defects that are three-dimensionally dispersed inside the specimen can be expressed as a function of X.

  In this case, as the value of the one-dimensional flaw information (the numerical value on the vertical axis), a projection value may be used directly, or a standardized projection value may be used. The normalization method is not particularly limited, and various methods can be used. For example, the ratio (hereinafter referred to as “instruction ratio”) of the actually measured projection value with respect to the projection value when it is assumed that a defect exceeding the binarization threshold exists over the entire area in the Y direction is expressed as a percentage ( %).

  FIG. 7A shows a schematic diagram of one-dimensional flaw information acquired from the binarized data shown in FIG. In FIG. 7A, the vertical axis represents the projection value in the Y direction of the binarized data and indicates the normalized value (instruction ratio). As shown in FIG. 7A, the value on the vertical axis of the one-dimensional flaw information reflects the three-dimensional flaw information, and as the size of the defect existing in the YZ plane at a certain coordinate X increases, / Or higher values as the amount of defects increases.

  Next, the instruction length calculation step will be described. The instruction length calculation step is a step of calculating an instruction length (determination data) in the X direction in which each one-dimensional flaw information exceeds one or two or more determination threshold values. Here, the “determination threshold value” refers to a reference value for determining pass / fail of the specimen. The determination threshold value can be arbitrarily selected according to the inspection purpose. One determination threshold value may be set for the same monitoring area, or a plurality of determination threshold values may be set. The determination threshold value may be the same for each monitoring area, or may be different for each monitoring area. However, in order to obtain a correlation with the radiation transmission test, it is preferable to determine the type and number of determination threshold values so as to correspond to the determination standard of the radiation transmission test.

  For example, in the radiation transmission test, the specimen is classified into a predetermined grade according to the length and darkness of the defect photographed on the film. In this case, the length of the defect corresponds to the dimension of the defect projected on the plane perpendicular to the radiation irradiation surface, and the density is the length of the defect and / or the amount of the defect extending in the radiation irradiation direction. It corresponds to.

  On the other hand, when ultrasonic flaw detection is performed from the direction perpendicular to the direction in which the radiation transmission test is to be performed, the projection value on the vertical axis of the one-dimensional flaw information indicates the length of the defect distributed in the Y direction and / or The value almost corresponds to the amount of defects. Further, the length in the X direction (indicated length) where the one-dimensional flaw information exceeds a certain value is a value substantially corresponding to the length of the defect projected in the direction perpendicular to the X direction.

  Thus, for example, in a radiation transmission test, when three levels of criteria are provided for the length and darkness of defects, in the method according to the present invention, three levels of determination thresholds are provided. For each, the instruction length may be calculated. At this time, if the determination threshold value is optimized, a determination result correlated with the radiation transmission test can be obtained.

In the example shown in FIG. 7A, three levels of determination threshold values are provided for the projection value on the vertical axis, that is, L line, M line, and H line, respectively. In addition, the lengths in the X direction in which the one-dimensional flaw information exceeds the L line, the M line, and the H line are calculated as the instruction length X _L , the instruction length X _M , and the instruction length X _H , respectively.

  A second method of converting the two-dimensional flaw information into determination data is a method that uses “area ratio” as determination data, and includes a binarization step and an area ratio calculation step. Among these, the binarization process is the same as the first method, and thus the description thereof is omitted.

  The area ratio calculation step is a step of calculating an area ratio (determination data) of a region exceeding the binarization threshold with respect to the total area of each binarized data. When a plurality of monitoring areas are set, the area ratio is calculated for each monitoring area. Further, the area ratio is calculated as it is without considering the determination threshold value. Furthermore, when calculating the area ratio, the “Z direction” for obtaining the two-dimensional flaw information is not necessarily the ultrasonic incident direction, and the “X direction” or the “Y direction” is the ultrasonic incident direction. It may be. The area ratio obtained in this way becomes a value corresponding to the result of the radiation transmission test by optimizing the incident surface of the ultrasonic wave, and even when the scattering and / or attenuation of the ultrasonic wave is large. The presence or absence of a defect in a certain monitoring area can be determined with relatively high accuracy.

  A third method for converting the two-dimensional flaw information into determination data includes an average intensity calculation step for calculating the average intensity of each two-dimensional flaw information. Here, “average intensity” refers to the average value of reflected echo intensity in the two-dimensional flaw information. That is, the average intensity calculation step is characterized in that the average intensity is calculated as it is without binarizing each acquired two-dimensional information. When a plurality of monitoring areas are set, the average intensity is calculated for each monitoring area. Further, the average intensity is calculated as it is without considering the determination threshold value. Furthermore, when calculating the average intensity, the “Z direction” for obtaining the two-dimensional flaw information is not necessarily the incident direction of the ultrasonic wave, and the “X direction” or the “Y direction” is the incident direction of the ultrasonic wave. It may be. The average intensity obtained in this way does not necessarily correspond to the result of the radiation transmission test. However, even if the scattering and / or attenuation of the ultrasonic wave is large, the presence or absence of a defect in a certain monitoring area is confirmed. It can be determined with relatively high accuracy.

  When a plurality of monitoring areas are set, one or two or more types of determination data may be acquired for each monitoring area according to the above-described procedure. For example, as shown in FIG. 4, when three two-dimensional flaw information of upper two-dimensional flaw information, intermediate two-dimensional flaw information, and lower two-dimensional flaw information is acquired, the same or different 1 for each monitoring area. What is necessary is just to calculate seed | species or 2 or more types of determination data (instruction length, an area ratio, and / or average intensity | strength).

  Next, the determination process will be described. In the determination process, one or two or more determination threshold values that are the same or different are set for each monitoring area, and one or two or more allowable values are allowed for each determination data and each determination threshold value. This is a step of comparing the data and determining the quality of the specimen. As the determination method, an optimum one is selected according to the number and / or type of determination data.

  For example, when “instruction length” is used as the determination data, specifically, the determination threshold value and one or more allowable instruction lengths (allowable data) corresponding to the determination threshold value are associated with the quality of the specimen. Reference data (determination table) is determined in advance, and the quality is determined by comparing the reference data with the actually calculated instruction length.

  The relationship between the predetermined judgment threshold value and the permissible instruction length corresponding thereto and the judgment standard for the quality of the specimen can be arbitrarily selected according to the material and shape of the specimen, the purpose of inspection, etc. it can. In general, the shorter the instruction length with respect to a certain determination threshold value, the better the product with fewer defects. In addition, one determination threshold value may be set for one monitoring area, and one permissible instruction length corresponding thereto may be set, or a plurality of permissible instruction lengths may be set. Also good. The same applies when two or more determination thresholds are used. Furthermore, a different determination threshold value and / or a different allowable instruction length may be set for each monitoring area.

FIG. 7B shows an example of the determination table. In the example shown in FIG. 7 (b), H lines, M lines and L lines (determination threshold) the exceeding instruction length of the actually measured values _X H, _{X M,} and _{X L} are each, H lines, M lines And permissible instruction lengths L ₁ (mm), L ₂ (mm), and L ₃ (mm) allowed for the L line (determination threshold value) (however, L ₁ <L ₂ <L ₃ ) If it is short, it is determined as “first grade”.

Similarly, the instruction lengths exceeding the H line, M line, and L line are the allowable instruction lengths L ₂ (mm), L ₃ (mm), and L ₄ (mm) (where L ₂ <L ₃ < If it is shorter than L ₄ ), it is determined as “second class”, and the instruction lengths exceeding the H line, M line, and L line are allowable instruction lengths L ₃ (mm), L ₄ (mm) and L, respectively. If it is shorter than L ₅ (mm) (however, L ₃ <L ₄ <L ₅ ), it is determined as “Class 3”. Furthermore, the instruction lengths exceeding the H line, the M line, and the L line are the permissible instruction lengths L ₄ (mm), L ₅ (mm), and L ₆ (mm) (where L ₄ <L ₅ <L ₆ ) If it is shorter, it is determined as “4th class”.

  The same applies when one or two values are used as the determination threshold value, or when four or more values are used. The given determination threshold value and the permissible instruction length corresponding thereto are used. Depending on the class, it may be classified into a plurality of grades. The same applies when different determination threshold values are used for each monitoring area.

  In addition, for example, when “area ratio” is used as the determination data, specifically, each calculated area ratio and one or more allowable area ratios (allowable) allowed for each determination threshold value are used. Data) to determine the quality of the specimen. In general, the smaller the area ratio, the better the product with fewer defects. Specifically, reference data (determination table) for associating one or two or more determination threshold values and one or two or more allowable area ratios corresponding to these with the quality of the test specimen is determined in advance. What is necessary is just to compare data with the area ratio actually calculated.

  For example, when “average intensity” is used as the determination data, specifically, each calculated average intensity and one or more allowable average intensity (allowable) for each determination threshold value are used. Data) to determine the quality of the specimen. In general, the smaller the average strength, the better the product with fewer defects. Specifically, reference data (determination table) that correlates one or two or more determination threshold values and one or two or more allowable average strengths corresponding to the determination threshold values and the quality of the specimen is determined in advance. The data may be compared with the actually calculated average intensity.

  Note that the determination using the indicated length, area ratio, average intensity, and the like described above may be performed for one monitoring area, or any one of them may be performed by one. You may perform with respect to a monitoring area | region. Also, for example, as shown in FIG. 4, when an upper monitoring area, an intermediate monitoring area, and a lower monitoring area are set, one or two or more determination threshold values are set for these, and each monitoring is set. What is necessary is just to compare each judgment data calculated about the area with one or more permissible data allowed for each judgment threshold. In this case, different types of determination data, determination threshold values and / or allowable data may be used for each monitoring area.

  Next, the operation of the internal defect evaluation method according to the present invention will be described. By processing the flaw detection data obtained using the ultrasonic flaw detection method, three-dimensional flaw information is acquired, and when three-dimensional flaw information in one or more monitoring areas is projected in the Z direction, the spatial distribution of defects is obtained. Reflected 1 or 2 or more two-dimensional flaw information is obtained. Similarly, when the maximum value of flaw detection data in one or two or more monitoring areas is acquired and projected in the Z direction, one or two or more two-dimensional flaw information reflecting the spatial distribution of defects is obtained. Further, in the case where the direction in which the radiation transmission test is to be performed is the Y direction, if the XY plane is an ultrasonic incident surface, two-dimensional flaw information that reflects the spatial distribution of defects and correlates with the radiation transmission test is obtained. can get.

  Next, when the two-dimensional flaw information is projected in the Y direction, one-dimensional flaw information that reflects the spatial distribution of defects and correlates with the radiation transmission test is obtained. Next, if a certain determination threshold value is provided, and the instruction length exceeding the determination threshold value is calculated from the obtained one-dimensional flaw information, the spatial distribution of defects can be indirectly evaluated. Moreover, if the incident direction of ultrasonic waves and the determination threshold are optimized, a determination result correlated with the radiation transmission test can be obtained. Similarly, if the acquired two-dimensional flaw information is converted into determination data such as area ratio and average intensity, and this is compared with allowable data corresponding to the determination threshold, the spatial distribution of defects is indirectly determined. Can be evaluated.

  Further, when two-dimensional flaw information in a predetermined monitoring area is acquired and the indicated length, area ratio and / or average intensity is calculated using this information, even if the reflected echo of the defect is not clearly observed, The degree of defect (size, number, etc.) appears as an increase or decrease in the indicated length, area ratio, and / or average intensity. Therefore, even if the scattering and / or attenuation of ultrasonic waves is large, the quality of the product can be determined with high accuracy.

  Furthermore, according to the material, shape, etc. of the test body, a plurality of monitoring areas are set, one or more determination data most suitable for each monitoring area is calculated, and 1 most suitable for each monitoring area. Alternatively, if two or more judgment thresholds are applied, the quality of the specimen is high even when the type and degree of defects differ depending on the site and / or when the scattering and / or attenuation of ultrasonic waves is large. It can be determined with accuracy.

  For example, as shown in FIG. 4, in the case where the test body is a cast rail, when a cast hole is generated in the surface layer portion of the rail head, only the inner region within a certain depth from the surface of the test body. Is set as the monitoring area, the cast hole in the surface layer portion may be overlooked. Further, in the case where only the surface layer portion has a problem, if a wide monitoring region including a region other than the surface layer portion is set, the amount of the surface layer portion may be underestimated.

  On the other hand, when an upper monitoring area is set near the surface layer portion and two-dimensional flaw information is acquired only for the upper monitoring area, determination data including only defect information generated near the surface layer portion is obtained. Therefore, if an appropriate determination threshold is set for this, the presence or absence of defects near the surface layer portion can be determined with high accuracy.

  Similarly, when an intermediate monitoring area and a lower monitoring area are set in the middle and lower portions of the rail head, respectively, and two-dimensional flaw information is acquired only for these areas, only defect information generated in these areas is included. Judgment data is obtained. Therefore, if an appropriate judgment threshold is set for this, even if the scattering and / or attenuation of the ultrasonic wave is large, it exists inside the rail such as a cast hole in the middle part or a horizontal crack in the lower part. It is possible to determine the quality of the product due to the defect to be performed with high accuracy.

  Since the method according to the present invention does not require environmental considerations, it can be used not only for product inspection before shipment but also for evaluation for quality deterioration and life prediction in the field. Further, at the time of inspection, it is only necessary to arrange at least one probe on one surface of the specimen, so that there are few physical restrictions. Furthermore, by optimizing the incident direction of the ultrasonic wave and the determination threshold, a high correlation with the radiation transmission test can be obtained.

  Therefore, if the present invention is applied to on-site re-inspection of products whose final inspection before shipment is standardized by radiation transmission tests, quality degradation that has occurred since the installation of the object until now Evaluation and life prediction can be performed easily and relatively accurately.

  Next, an internal defect evaluation program used in the method according to the present invention will be described. 8 and 9 show an example of the flowchart. First, in step 1 (hereinafter simply referred to as “S1”), it is determined whether or not flaw detection data has been input. If no flaw detection data is input, the process waits as it is. On the other hand, if flaw detection data is input, the process proceeds to S2.

  In S2, it is determined whether or not to acquire three-dimensional flaw information. If three-dimensional flaw information is not acquired, the process proceeds to S8. On the other hand, when acquiring three-dimensional flaw information, the process proceeds to S3. In S3, flaw detection data acquired by performing ultrasonic flaw detection on the specimen is processed, and three-dimensional flaw information inside the specimen is acquired (three-dimensional flaw information acquisition procedure). In this case, as a method for processing flaw detection data, and in the case where the direction in which the radiation transmission test is performed is the Y direction, the radiation transmission test is performed when the XY plane is an ultrasonic incident surface. As described above, correlated flaw detection data can be obtained.

  Next, in S4, it is determined whether or not to perform sensitivity correction. If no sensitivity correction is required, the process proceeds to S6 as it is. On the other hand, if sensitivity correction is necessary, the process proceeds to S5, where the sensitivity of the three-dimensional flaw information is corrected (sensitivity correction procedure). In this case, as described above, various methods can be used as the sensitivity correction method, and the sensitivity correction procedure and the mask procedure (S6) described later may be interchanged.

  Next, in S6, one or more monitoring regions are cut out from the three-dimensional flaw information subjected to sensitivity correction as necessary (masking procedure). In this case, the monitoring area can be arbitrarily set, and the monitoring area may be cut out manually while looking at the monitor, or automatically based on pre-input data. Is as described above. Further, in S7, the maximum value or integrated value in the Z direction of the three-dimensional flaw information in each monitoring area is projected in the Z direction and converted into two-dimensional flaw information (two-dimensional flaw information conversion procedure).

  On the other hand, when the three-dimensional flaw information is not acquired (S2: NO), the process proceeds to S8, and the maximum value in the Z direction of the flaw detection data in one or more monitoring areas is acquired (maximum value acquisition procedure). Then, if necessary (S9: YES), after correcting the sensitivity of the maximum value in S10 (sensitivity correction procedure), in S11, each maximum value is projected in the Z direction and converted into two-dimensional flaw information, respectively. (Two-dimensional flaw information conversion procedure).

  The “two-dimensional flaw information acquisition procedure” is flaw detection data that reflects the spatial distribution of internal defects existing inside the test specimen, acquired by performing ultrasonic flaw detection on the test specimen. Alternatively, a process in which two or more monitoring areas are projected in the Z direction and two-dimensional flaw information is acquired for each monitoring area, and S3 to S7 and S8 to S11 correspond to this. Moreover, S3-S7 and S8-S11 may be provided with only any one.

  After the two-dimensional flaw information is acquired in S7 or S11, the process proceeds to S21 in FIG. In S21, it is determined whether or not the instruction length is calculated as determination data. When the instruction length is not calculated (S21: NO), the process proceeds to S31. On the other hand, when the instruction length is calculated (S21: YES), the process proceeds to S22, and the obtained two-dimensional flaw information is binarized using the same or different binarization threshold values, one or two or more. The binarized data is acquired (binarization procedure). In this case, as described above, the binarization threshold value of each monitoring area can be arbitrarily set according to the inspection purpose or the like.

  Next, in S23, the maximum value or integrated value in the Y direction of each binarized data is projected in the Y direction and converted into one or more one-dimensional flaw information (one-dimensional flaw information conversion procedure). Next, in S24, the instruction length (determination data) in the X direction in which each one-dimensional flaw information exceeds one or two or more determination threshold values is calculated (instruction length calculation procedure). In this case, the determination threshold can be arbitrarily set according to the inspection purpose, the point where one or a plurality of values can be used as the determination threshold, and for each monitoring area As described above, different determination threshold values may be used.

  Further, in S25, the calculated instruction length is compared with one or more allowable instruction lengths (allowable data) allowed for each determination threshold value, and the quality of the specimen is determined ( Judgment procedure). Specifically, the determination is performed by preparing in advance a determination table that associates the determination threshold value and the allowable instruction length with the quality of the specimen, and comparing this with each actually measured instruction length. In this case, as described above, the determination threshold value and the allowable instruction length corresponding to the determination threshold value can be arbitrarily selected according to the purpose of inspection.

  When the determination by the instruction length is completed and / or when the instruction length is not calculated (S21: NO), the process proceeds to S31. In S31, it is determined whether to calculate the area ratio. When the area ratio is not calculated (S31: NO), the process proceeds to S41. On the other hand, when the area ratio is calculated (S31: YES), the process proceeds to S32.

  In S32, the acquired two-dimensional flaw information is binarized using the same or different binarization threshold values, and one or two or more binarized data is acquired (binarization procedure). Next, in S33, an area ratio (determination data) of a region exceeding the binarization threshold with respect to the total area of each binarized data is calculated (area ratio calculation procedure). Further, in S34, the calculated area ratio is compared with one or more allowable area ratios (allowable data) allowed for each determination threshold value to determine the quality of the specimen (determination) procedure).

  When the determination based on the area ratio is completed and / or when the area ratio is not calculated (S31: NO), the process proceeds to S41. In S41, it is determined whether to calculate the average intensity. When the average intensity is not calculated (S41: NO), the process proceeds to S44. On the other hand, when the average intensity is calculated (S41: YES), the process proceeds to S42.

  In S42, the average intensity (determination data) of each acquired two-dimensional flaw information is calculated (average intensity calculation procedure). Next, in S43, the calculated average strength is compared with one or more allowable average strengths (allowable data) allowed for each determination threshold to determine the quality of the specimen (determination) procedure).

  When the determination based on the average intensity is completed and / or when the average intensity is not calculated (S41: YES), the process proceeds to S44. In S44, it is determined whether or not there is a next monitoring area. If there is a next monitoring area (S44: YES), the process returns to S21, and the above-described procedure is repeated for the new monitoring area. On the other hand, if there is no next monitoring area (S44: YES), the determination ends.

  The “determination data conversion procedure” refers to a procedure for converting each two-dimensional flaw information into the same or different determination data, or two or more types of determination data. S22 to S24, S32 to S33, and S42. Corresponds to this. In addition, the “determination procedure” is to set one or two or more determination threshold values that are the same or different for each monitoring area, and allow 1 for each determination data and each determination threshold value. Or the procedure which contrasts with 2 or more tolerance data and determines the quality of a test body, and S25, S34, and S43 correspond to this. In addition, the determination data conversion procedure and the determination procedure corresponding to the determination data conversion procedure may include any one or two.

  Further, the reference data (determination threshold value and permissible data corresponding thereto) serving as a determination criterion may be individually input for each inspection, or one or two as shown in FIG. More than one determination table may be stored in the storage means in advance, and a table necessary for determination may be automatically called.

  In the examples shown in FIGS. 8 and 9, the instruction length is calculated (S24) and the quality is determined (S25), the area ratio is calculated (S33) and the quality is determined (S34), and the average intensity is calculated (S42). The quality determination (S43) is automatically processed in the computer, but the one-dimensional flaw information converted in S23, the binarized data obtained in S32, or obtained in S7 or S11. The two-dimensional flaw information is output as “determination data” as it is, and based on the output information, the instruction length, the area ratio and the average intensity are calculated by manual operation, visual inspection, and the determination using these May be performed.

  Further, for example, as shown in FIG. 4, when upper 2D flaw information, intermediate 2D flaw information, and lower 2D flaw information are acquired from flaw detection data, the upper 2D The dimension flaw information, the intermediate two-dimensional flaw information, and the lower two-dimensional flaw information may be converted into one or two or more types of determination data that are the same or different. In addition, using the determination procedure, one or two or more determination threshold values are set for the upper monitoring area, the intermediate monitoring area, and the lower monitoring area, respectively, and each determination data and each determination threshold value are set. The quality of the specimen may be determined by comparing with one or more permissible data that is permissible.

  Next, an internal defect evaluation apparatus used in the method according to the present invention will be described. The internal defect evaluation apparatus includes flaw detection means and calculation means. The flaw detection means is for performing ultrasonic flaw detection on the specimen. In the present invention, the flaw detection means is not particularly limited as long as it can acquire flaw detection data reflecting the spatial distribution of internal defects existing in the specimen using ultrasonic flaw detection. .

  Specifically, various kinds of ultrasonic flaw detectors capable of measuring the spatial distribution of defects can be used as they are as flaw detection means. Such an ultrasonic flaw detector generally includes a pulse transmitter for generating ultrasonic waves, a probe for transmitting and receiving ultrasonic waves generated by the pulse transmitter, and a reflected wave received by the probe. Is displayed as an electrical signal, display means (monitor) for displaying the waveform captured by the receiver, and moving means for moving the probe by a predetermined distance in a predetermined direction.

  In this case, a probe having one piezoelectric element, a matrix probe, or an array probe may be used as the probe, and a scanning method, a flaw detection method, an image As described above, various methods can be used as the synthesis method and the like. Further, when the direction in which the radiation transmission test is to be performed is the Y direction, flaw detection data correlated with the radiation transmission test can be obtained if the XY plane of the test body is the ultrasonic incident surface. .

  The calculation means is for evaluating the presence or absence of an internal defect in the specimen based on the flaw detection data obtained by the flaw detection means. Specifically, a computer storing an internal defect determination program capable of executing the flowcharts shown in FIGS. 8 and 9 is used as the calculation means. The flaw detection data (electrical signal of reflected wave) detected by the flaw detection means is output to the calculation means, and predetermined processing is performed by the calculation means.

  That is, the calculation means projects flaw detection data in one or more monitoring areas in the Z direction, and acquires two-dimensional flaw information for each monitoring area, and each two-dimensional flaw information acquisition means. Judgment information conversion means for converting flaw information into one or two or more types of determination data, which are the same or different, is provided. In addition, the calculation means sets one or two or more determination threshold values that are the same or different for each monitoring area, and the determination data and one or more allowable threshold values for each determination threshold value. A judgment means for judging the quality of the specimen may be further provided.

(1) As the two-dimensional flaw information acquisition means, (A) a three-dimensional flaw information acquisition means, a mask means, a two-dimensional flaw information conversion means, and a sensitivity correction means as required, or (B ) Any of the maximum value acquisition means, the two-dimensional flaw information conversion means, and the one provided with sensitivity correction means as required, and (2) (A) 2 as the determination data conversion means A valuation means, a one-dimensional flaw information acquisition means, and an instruction length calculation means; (B) a binarization means and an area ratio calculation means; or (C) an average intensity calculation means. Any of these may be used in the same manner as the internal defect evaluation program described above.

  Also, two-dimensional flaw information acquisition means, determination data conversion means, determination means, three-dimensional flaw information acquisition means, mask means, two-dimensional flaw information conversion means, sensitivity correction means, maximum value acquisition means, binarization means, The dimensional flaw information converting means, the instruction length calculating means, the area rate calculating means, and the average intensity calculating means are respectively a two-dimensional flaw information acquisition procedure, a determination data conversion procedure, a determination procedure in the internal defect determination program described above. Three-dimensional flaw information acquisition procedure, mask procedure, two-dimensional flaw information conversion procedure, sensitivity correction procedure, maximum value acquisition procedure, binarization procedure, one-dimensional flaw information conversion procedure, instruction length calculation procedure, area ratio calculation procedure, and Since it is the same as the function of the average intensity calculation procedure, the description is omitted.

  Further, when the upper two-dimensional flaw information, the intermediate two-dimensional flaw information, and the lower two-dimensional flaw information are acquired from the flaw detection data, the upper two-dimensional flaw information and the intermediate two-dimensional flaw information are obtained using the determination data conversion means. And the lower two-dimensional flaw information may be converted into one or more types of determination data that are the same or different. Also, using the determination means, one or two or more determination threshold values are set for the upper monitoring area, the intermediate monitoring area, and the lower monitoring area, respectively, and each determination data and each determination threshold value are set. The quality of the specimen may be determined by comparing with one or more permissible data that is permissible.

  When the specimen is evaluated using the internal defect evaluation program and the internal defect evaluation apparatus described above, determination data reflecting the spatial distribution of defects is obtained. Therefore, if a certain threshold value is set, and the permissible data corresponding to this threshold value is compared with the data for determination, the presence or absence of defects can be detected with high accuracy even when the scattering and / or attenuation of ultrasonic waves is large. Can be detected. Furthermore, if the ultrasonic incident surface is optimized, determination data having a correlation with the result of the radiation transmission test can be obtained. Therefore, if the determination threshold value is optimized, a determination result correlated with the radiation transmission test can be obtained.

  As shown in FIG. 10, ultrasonic flaw detection was performed on the rail (high manganese cast steel product) from the Z direction to obtain flaw detection data. Next, the flaw detection data was subjected to data processing using the method according to the present invention, and one-dimensional flaw information in the X direction was obtained. The flaw detection test conditions and flaw detection data processing conditions are as follows.

Ultrasonic probe used: 2 MHz, 16 × 16 array Coupling material: Sonicoat Image synthesis method: Aperture synthesis method Monitoring area size: 400 mm × 48 mm × 55 mm

  Next, based on the acquired one-dimensional information, the instruction length exceeding the determination threshold value was calculated. The determination threshold value was set to three levels of 50.0% (H line), 25.0% (M line), and 12.5% (L line) in the indicated ratio.

Next, a radiation transmission test was performed on the same rail. In the test, radiation was applied from the Y direction in FIG. The radiation transmission test conditions are as follows.
X-ray device: Linac X-ray output: 6 MeV
Intensifying screen: Front 0.3mm, Back 0.3mm
Film: Fuji # 80

  FIG. 11 shows the relationship between the grade of each rail (RT grade) obtained by the radiation transmission test and the grade of each rail (UT grade) obtained by the method according to the present invention to which ultrasonic flaw detection is applied. From FIG. 11, it can be seen that the evaluation result using the method according to the present invention has a good correlation with the evaluation result by the radiation transmission test. The correlation coefficient obtained from FIG. 11 was 0.8.

  The embodiment of the present invention has been described in detail above, but the present invention is not limited to the above embodiment, and various modifications can be made without departing from the scope of the present invention.

  The internal defect evaluation method according to the present invention can be used for evaluation of internal defects before shipment and on-site of test specimens (for example, tanks, piping, railroad rails, etc.) made of welded products, cast products, and the like.

It is a schematic diagram which shows an example of the flaw detection method for acquiring three-dimensional flaw information. It is a schematic diagram which shows an example of 3D flaw information. It is a schematic diagram which shows the state from which the monitoring area | region was cut out from the three-dimensional flaw information shown in FIG. It is a schematic diagram which shows an example of the test body from which the several monitoring area | region was cut out. It is a schematic diagram which shows an example of the two-dimensional flaw information obtained by projecting the three-dimensional flaw information in the monitoring area | region of FIG. 3 to a Z direction. It is a schematic diagram which shows an example of the binarized data obtained by binarizing the two-dimensional flaw information of FIG. FIG. 7A is a schematic diagram showing an example of one-dimensional flaw information obtained by projecting the binarized data shown in FIG. 6 in the Y direction, and FIG. 7B is an example of a determination table. is there. It is a flowchart of the program for internal defect evaluation used for the method which concerns on this invention. It is a continuation of the flowchart shown in FIG. 1 is a schematic view of a test body used in Example 1. FIG. It is a figure which shows the correlation with the evaluation result (RT grade) by the radiation transmission test method, and the evaluation result (UT grade) by the method which concerns on this invention.

Explanation of symbols

10 Specimens 10a to 10e Defects 12a to 12c Probe

Claims (10)

A flaw detection process for performing ultrasonic flaw detection on the test specimen and acquiring flaw detection data reflecting the spatial distribution of internal defects present in the test specimen;
Two-dimensional flaw information acquisition step of projecting the flaw detection data in one or two or more monitoring areas in the Z direction and acquiring two-dimensional flaw information for each of the monitoring areas;
A determination data conversion step for converting each of the two-dimensional flaw information into one or two or more types of determination data that are the same or different, respectively.
For each of the monitoring areas, one or two or more determination threshold values that are the same or different are set, and each determination data and one or more allowable data allowed for each of the determination threshold values are set. And a determination step for determining the quality of the test specimen. In the specimen, the direction in which a radiation transmission test is to be performed is the Y direction,
2. The internal defect evaluation method according to claim 1, wherein in the flaw detection step, the flaw detection data is acquired by causing an ultrasonic wave to enter from an XY plane of the specimen. The two-dimensional flaw information acquisition step includes
(A) a three-dimensional flaw information acquisition step of processing the flaw detection data and acquiring three-dimensional flaw information inside the specimen;
(B) a mask process of cutting out one or more of the monitoring areas from the three-dimensional flaw information;
(C) a two-dimensional flaw information conversion step of projecting, in the Z direction, a maximum value or an integrated value in the Z direction of the three-dimensional flaw information in each of the monitoring areas, and converting each into the two-dimensional flaw information. The internal defect evaluation method according to claim 1 or 2.   The internal defect evaluation method according to claim 3, further comprising a sensitivity correction step of correcting the sensitivity of the three-dimensional flaw information before or after the masking step. The two-dimensional flaw information acquisition step includes
(A) a maximum value acquisition step of acquiring a maximum value in the Z direction of the flaw detection data within one or more of the monitoring areas;
(B) The internal defect evaluation method according to any one of claims 1 to 4, further comprising: a two-dimensional flaw information conversion step of projecting each maximum value in the Z direction and converting the maximum value into the two-dimensional flaw information. .   The internal defect evaluation method according to claim 5, further comprising a sensitivity correction step of correcting sensitivity of the flaw detection data or the maximum value before or after the maximum value acquisition step. The determination data conversion step includes:
(A) A binarization step of binarizing each of the two-dimensional flaw information using the same or different binarization threshold values to obtain one or two or more binarized data;
(B) a one-dimensional flaw information conversion step of projecting the maximum value or integrated value in the Y direction of each of the binarized data in the Y direction and converting it into one or more one-dimensional flaw information;
(C) an instruction length calculation step of calculating an instruction length (determination data) in the X direction in which each one-dimensional flaw information exceeds one or two or more of the determination threshold values.
The determination step compares the calculated instruction lengths with one or more allowable instruction lengths (allowable data) allowed for the determination threshold values, and determines the quality of the specimen. The internal defect evaluation method according to claim 1, wherein the internal defect is determined. The determination data conversion step includes:
(A) A binarization step of binarizing each of the two-dimensional flaw information using the same or different binarization threshold values to obtain one or two or more binarized data;
(B) an area ratio calculating step of calculating an area ratio (determination data) of a region exceeding the binarization threshold with respect to the total area of each binarized data;
The determination step compares the calculated area ratio with one or more allowable area ratios (allowable data) allowed for the determination threshold values, and determines the quality of the specimen. The internal defect evaluation method according to claim 1, wherein the internal defect is evaluated. The determination data conversion step includes an average intensity calculation step of calculating an average intensity (determination data) of each of the two-dimensional flaw information,
The determination step compares the calculated average strength with one or more allowable average strengths (allowable data) allowed for the determination threshold values to determine the quality of the specimen. The internal defect evaluation method according to claim 1, wherein the internal defect is evaluated. The internal defect evaluation method according to claim 1, wherein the test body is a high manganese cast steel product.

Images