JP7091646B2 - Surface scratch depth determination device - Google Patents

Description

The present invention relates to a surface flaw depth determination device. In particular, the present invention relates to a surface flaw depth determination device capable of accurately determining the depth of a surface flaw such as a surface crack accompanied by a relatively steep shape change generated in a steel piece.

Conventionally, an ultrasonic flaw detection method has been widely used to guarantee the quality of steel pieces.
In recent years, among the ultrasonic flaw detection methods, a two-dimensional imaging method using a one-dimensional array type ultrasonic probe equipped with a plurality of oscillators has been used as a method capable of accurately determining the position of a flaw that has occurred. The ultrasonic flaw detection method is attracting attention.

Specifically, a one-dimensional array type ultrasonic probe is placed along the side surface of the steel piece, and the one-dimensional array type ultrasonic probe receives an echo from the steel piece to output a flaw detection signal. By performing signal processing (opening synthesis processing, TFM (Total Focusing Method), etc.), a two-dimensional image of the cross section in the direction orthogonal to the longitudinal direction of the steel piece is generated, and the waste is detected using this two-dimensional image. An ultrasonic flaw detection method has been proposed (see, for example, Patent Documents 1 and 2). Amplitude synthesis processing and TFM, which are used as two-dimensional imaging methods, divide the coordinate space of the cross section of the scratched material such as a piece of steel in the direction facing the one-dimensional array type ultrasonic probe into a mesh, and the one-dimensional array. The propagation time and amplitude value of the flaw detection signal from the reflection source measured at multiple points at the same time by the type ultrasonic probe are determined based on the positional relationship between the flawed material and the one-dimensional array type ultrasonic probe and the sound velocity information. This is a method of reconstructing an image of a reflection source by integrating them in a predetermined mesh.
According to the ultrasonic flaw detection method using the above two-dimensional imaging method, a two-dimensional image having high spatial resolution can be generated. Therefore, the position of the flaw in the flaw detection region of each one-dimensional array type ultrasonic probe. Can be accurately determined.

However, in the ultrasonic flaw detection method using the above two-dimensional imaging method, due to the influence of noise (bottom echo, etc.) generated on the surface layer (side surface of the steel piece), the surface layer to the depth direction (from the surface layer) in the two-dimensional image. Generally, a predetermined range (in the vertical direction) is a dead zone according to the flaw detection frequency. For example, when the flaw detection frequency is 3 MHz, the dead zone is about 2 mm from the surface layer. For this reason, for example, when a surface crack that accompanies a shape change from the surface layer to the depth direction occurs in the steel piece, the depth of the surface crack is evaluated to be smaller than the actual value, or it occurs inside the steel piece instead of the surface layer. There is a risk of erroneous determination as a flaw. In addition, surface cracks may not be detected.

In order to detect flaws that change the shape from the surface layer to the depth direction, such as surface cracks, the shape of the side surface of the steel piece is measured by the optical cutting method with a laser light source and imaging means instead of the ultrasonic flaw detection method. It is conceivable to detect a flaw based on the shape change measured by this shape measuring device using the shape measuring device.

However, surface cracks that occur in steel pieces are often accompanied by relatively steep shape changes. Therefore, as shown in FIG. 1 (a), the shape of the flaw may be accurately measured depending on the relationship between the shape of the surface flaw (surface crack) and the orientation of the imaging means, while FIG. 1 ( As shown in b), it may not be possible to measure the shape of the deep part of the flaw. That is, there may be a place where the shape measurement data is missing (a place indicated by a white circle in FIG. 1B) due to the deviation from the field of view of the image pickup means.
If the shape measurement data is missing, even if surface flaws can be detected, the depth may be determined to be smaller than the actual depth. That is, if the shape measuring device is used alone, it may be difficult to accurately determine the depth of surface flaws generated on the steel pieces.
In FIG. 1, the steel piece is designated by a reference numeral B, and the steel piece B is designated by a reference numeral Y in the longitudinal direction. The same applies to FIG. 2 described later.

FIG. 2 illustrates the problems in determining the depth of surface flaws by the ultrasonic flaw detection method using the conventional two-dimensional imaging method described above and the problems in determining the depth of surface flaws by the optical cutting method. It is a figure. FIG. 2A shows a problem of the optical cutting method when the surface scratch depth is shallow, and FIG. 2B shows a problem of the optical cutting method when the surface scratch depth is medium. (C) shows a problem of the optical cutting method when the depth of the surface flaw is deep. FIG. 2 (d) shows the problem of the ultrasonic flaw detection method when the surface flaw depth is shallow, and FIG. 2 (e) shows the problem of the ultrasonic flaw detection method when the surface flaw depth is medium. FIG. 2 (f) shows a problem of the ultrasonic flaw detection method when the depth of the surface flaw is deep. Specifically, FIGS. 2 (d) to 2 (f) show an example of a two-dimensional image of a cross section of a steel piece generated by an ultrasonic flaw detection method. Further, the hatched region in FIG. 2 means a dead zone of the surface layer portion (for example, when the flaw detection frequency is 3 MHz, a range of about 2 mm from the surface layer (side surface of the steel piece)) in the ultrasonic flaw detection method.

As shown in FIGS. 2 (a) to 2 (c), according to the optical cutting method, the portion where the shape measurement data is missing due to being out of the field of view of the imaging means (in FIGS. 2 (a) to 2 (c)). (Points indicated by white circles) occur, and even if surface flaws can be detected, the depth may be determined to be smaller than the actual depth, and any surface flaw may be erroneously determined to be the same depth. be.
Further, in the example shown in FIG. 2D, since the entire surface flaw is located in the dead zone of the surface layer portion, the surface flaw cannot be detected by the ultrasonic flaw detection method.
Further, as shown in FIGS. 2 (e) and 2 (f), even if the deep part of the surface flaw located inside the dead zone of the surface layer can be detected by the ultrasonic flaw detection method, the dead zone is formed. Since the located surface layer portion cannot be detected, the depth of the surface flaw may be determined to be smaller than the actual depth, or the flaw may be erroneously determined to be a flaw generated inside the steel piece instead of the surface layer portion.

Japanese Unexamined Patent Publication No. 2011-203037 Japanese Unexamined Patent Publication No. 2009-236668

Therefore, the present invention provides a surface flaw depth determination device capable of accurately determining the depth of a surface flaw such as a surface crack accompanied by a relatively steep shape change generated in a steel piece. Is the subject.

As a result of diligent studies to solve the above problems, the present inventors can execute both the ultrasonic flaw detection method and the optical cutting method using the two-dimensional imaging method, and the ultrasonic waves using the two-dimensional imaging method. By combining the advantages of the flaw detection method (which can accurately determine the position of the deep part of the surface flaw) and the advantage of the optical cutting method (which can accurately determine the position and shape of the surface layer of the surface flaw), the depth of the surface flaw. The present invention was completed with the idea that the above can be accurately determined.

That is, in order to solve the above-mentioned problems, the present invention has a one-dimensional array type ultrasonic probe arranged along the side surface of the steel piece transported in the longitudinal direction, and is arranged along the side surface of the steel piece. , A shape measuring device that measures the shape of the side surface of the steel piece by the optical cutting method, a speed sensor that measures the transport speed of the steel piece, and a flaw detection signal output from the one-dimensional array type ultrasonic probe. By applying signal processing to the steel piece, a two-dimensional image of a cross section in a direction orthogonal to the longitudinal direction of the steel piece was generated, and the shape of the side surface of the steel piece was measured by the shape measuring device. The alignment between the shape measurement data and the generated two-dimensional image was measured by the geometrical positional relationship between the one-dimensional array type ultrasonic probe and the shape measuring device and the steel by the speed sensor. The signal processing device includes a signal processing device that performs based on the transport speed of the piece, and the signal processing device detects surface flaws generated on the side surface of the steel piece based on the shape measurement data, and the shape measurement data and the above 2 Provided is a surface flaw depth determination device characterized in that the depth of the detected surface flaw is determined based on a dimensional image.

According to the surface flaw depth determination apparatus according to the present invention, the signal processing apparatus is based on the shape measurement data obtained by the shape measuring apparatus using the optical cutting method, and the surface layer of the surface flaw generated on the side surface of the steel piece. The portion can be detected accurately (the position and shape of the surface layer portion of the surface flaw can be accurately determined). In addition, the signal processing device can accurately detect the deep part of the surface flaw generated on the side surface of the steel piece based on the two-dimensional image generated by using the flaw detection signal output from the one-dimensional array type ultrasonic probe ( The position of the deep part of the surface flaw can be determined accurately). Therefore, the signal processing device has shape measurement data in the vicinity of the surface flaw (surface layer portion of the surface flaw) detected based on the shape measurement data (data of the portion of the shape measurement data located near the surface layer portion of the surface flaw). Based on the above, it is possible to determine the depth of the surface layer portion of the surface flaw. Further, the depth of the deep part of the surface flaw is determined based on the density of the two-dimensional image in the vicinity of the detected surface flaw (the density of the pixel located near the surface layer portion of the surface flaw among the pixels constituting the two-dimensional image). It is possible to do. That is, since a pixel having a density equal to or higher than a predetermined threshold value can be considered to be a pixel corresponding to the surface flaw, the depth of the deep portion of the surface flaw is determined based on the pixel corresponding to the surface flaw. It is possible. Therefore, the depth of the surface flaw can be accurately determined by taking into account the depth of the surface layer of the surface flaw determined based on the shape measurement data and the depth of the surface flaw determined based on the density of the two-dimensional image. be.
In the surface flaw depth determination device according to the present invention, the shape measurement data is used as a trigger for surface flaw detection (that is, the surface layer portion of the surface flaw is first detected based on the shape measurement data, and then this detection is performed. Since the depth of the surface flaw is determined using the density of the two-dimensional image near the position), it is not necessary to make the vicinity of the surface layer of the two-dimensional image a dead zone as in the conventional case.

Further, in order to solve the above-mentioned problems, the present invention has a one-dimensional array type ultrasonic probe arranged along the side surface of the steel piece and the above-mentioned by an optical cutting method, which is arranged along the side surface of the steel piece. By performing signal processing on the flaw detection signal output from the shape measuring device that measures the shape of the side surface of the steel piece and the one-dimensional array type ultrasonic probe, the direction orthogonal to the longitudinal direction of the steel piece. Based on the shape measurement data obtained by generating a two-dimensional image of the cross section of the steel piece and measuring the shape of the side surface of the steel piece with the shape measuring device, and the generated two-dimensional image, the steel piece The signal processing device includes a signal processing device for determining the depth of surface flaws generated on the side surface, and the signal processing device detects a portion where the shape measurement data is missing and uses the first step as a starting point, and the second step. A second step of specifying a pixel that is located in the depth direction of the steel piece with respect to the starting point and has a density equal to or higher than a predetermined threshold among the pixels constituting the dimensional image, and the second step. The depth of the surface flaw, characterized in that the third step of determining the depth of the surface flaw is executed based on the position of the pixel most distant from the side surface of the steel piece among the pixels identified in. It is also provided as a determination device .

According to the present invention , by the first step executed by the signal processing apparatus, it is possible to detect surface scratches such as surface cracks accompanied by a steep shape change so that the shape measurement data is lost (the position of the surface layer portion of the surface scratches). The starting point can be detected). Next, since surface flaws such as surface cracks generated in the steel pieces are propagated in the depth direction of the steel pieces, the second step among the pixels constituting the two-dimensional image is performed by the second step executed by the signal processing device. It is possible to identify the pixel corresponding to the surface flaw detected in one step. Finally, by the third step executed by the signal processing device, the position of the most distant pixel, that is, the pixel corresponding to the deepest part of the surface flaw is specified, so that the depth of the surface flaw can be accurately determined. It is possible.

Preferably, in the second step, a plurality of threshold values are set according to the distance from the side surface of the steel piece, and the plurality of threshold values are small enough to be separated from the side surface of the steel piece.

When a flaw detection signal for surface flaws such as surface cracks generated in a steel piece is imaged in two dimensions, the strength of the flaw detection signal tends to decrease as the depth becomes deeper. Therefore, by setting a threshold value small enough to be separated from the side surface of the steel piece as in the above preferable configuration, the pixel corresponding to the surface flaw can be accurately identified, and the depth of the surface flaw can be accurately identified. It is possible to judge.

According to the present invention, it is possible to accurately determine the depth of a surface flaw such as a surface crack that occurs in a steel piece and is accompanied by a relatively steep shape change.

It is explanatory drawing explaining the problem in the depth determination of the surface flaw by the optical cutting method. It is explanatory drawing explaining the problem in the depth determination of the surface flaw by the ultrasonic flaw detection method using the conventional 2D imaging method, and the problem in the depth determination of a surface flaw by a light cutting method. It is a schematic block diagram of the surface flaw depth determination apparatus which concerns on one Embodiment of this invention. It is explanatory drawing explaining each step performed by the signal processing apparatus shown in FIG.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.
FIG. 3 is a schematic configuration diagram of a surface flaw depth determination device according to an embodiment of the present invention. FIG. 3A is an overall configuration diagram. FIG. 3B is an enlarged view of the one-dimensional array type ultrasonic probe seen from the arrow Y direction of FIG. 3A. FIG. 3 (c) is an enlarged view of the shape measuring device seen from the arrow Y direction of FIG. 3 (a).
As shown in FIG. 3, the surface flaw depth determination device (hereinafter, appropriately referred to as “depth determination device”) 100 according to the present embodiment includes a one-dimensional array type ultrasonic probe 1 and a shape measuring device 2. And a signal processing device 3.

The one-dimensional array type ultrasonic probe (hereinafter, appropriately referred to as “ultrasonic probe”) 1 of the present embodiment is a steel piece (hereinafter, referred to as a billet) having a substantially rectangular cross section, which is conveyed by the conveying roll R. Four are arranged along each side of the. The ultrasonic probe 1 includes a plurality of (for example, 64) oscillators 11 arranged in one direction. The ultrasonic probe 1 is arranged so that the arrangement direction of the vibrator 11 is substantially parallel to each side surface of the billet B.
Further, a resin shoe (not shown) is attached to the ultrasonic probe 1, and a contact medium such as water is introduced between the shoe and the side surface of the billet B. The ultrasonic wave transmitted from the ultrasonic probe 1 is incident on the inside of the billet B from the side surface of the billet B facing the ultrasonic probe 1 via the shoe and the contact medium, and is a defect existing in the billet B. The echo reflected by is received by the ultrasonic probe 1 via the contact medium and the shoe.

Four shape measuring devices 2 of the present embodiment are arranged along each side surface of the billet B. The shape measuring device 2 includes a laser light source (not shown) that irradiates a laser slit light 2L extending in a direction (X direction or Z direction) orthogonal to the longitudinal direction (Y direction) of the billet B along the side surface of the billet B. It is a device provided with an image pickup means (not shown) for imaging the laser slit light 2L, and measures the shape of the side surface of the billet B by an optical cutting method.

The signal processing device 3 of the present embodiment controls transmission / reception of ultrasonic waves from each of the four ultrasonic probes 1 and irradiation and imaging of laser slit light 2L from each of the four shape measuring devices 2. Further, the signal processing device 3 performs signal processing on the flaw detection signals output from each of the four ultrasonic probes 1 to obtain a cross section in a direction orthogonal to the longitudinal direction (Y direction) of the billet B. Generate a two-dimensional image. The signal processing device 3 synthesizes the generated two-dimensional images (four two-dimensional images in total) for each ultrasonic probe 1 as necessary, and generates an entire two-dimensional image of the entire cross section of the billet B. It is also possible to do.
The signal processing device 3 of the present embodiment performs aperture synthesis processing as signal processing and generates an aperture synthesis image as a two-dimensional image. Further, the signal processing device 3 of the present embodiment does not generate an entire two-dimensional image (aperture synthesis image), but uses four aperture synthesis images for each ultrasonic probe 1 individually to form an aperture synthesis image. The judgment described later is performed for each.
The signal processing device 3 generates a plurality of aperture synthesis images in the longitudinal direction in a state where the billet B is conveyed in the longitudinal direction (direction of the arrow Y) by the roll R. Specifically, a plurality of aperture synthetic images are generated in the longitudinal direction of the billet B, for example, at a pitch of 10 mm.

The signal processing device 3 controls the transmission and reception of the above-mentioned ultrasonic waves by providing known means provided by the so-called flaw detector, such as a pulsar, a receiver, an amplifier, an A / D converter, and a waveform memory. Further, the signal processing device 3 generates an aperture synthesis image by performing an aperture synthesis process on the flaw detection signal stored in the waveform memory. Then, the signal processing device 3 has surface scratches generated on the side surface of the billet B based on the shape measurement data obtained by measuring the shape of the side surface of the billet B by the shape measuring device 2 and the generated aperture synthesis image. A predetermined program is installed to perform each step described below in order to determine the depth of. As for the aperture synthesis process, for example, known contents as described in the above-mentioned Patent Document 2 can be applied, and therefore the description of the specific contents is omitted here.

The signal processing device 3 executes the first to third steps described below in order to determine the depth of surface flaws generated on the side surface of the billet B. Hereinafter, each step executed by the signal processing apparatus 3 will be described in order with reference to FIG. 4 as appropriate.

<First step>
In the first step, the signal processing device 3 detects a portion where the shape measurement data obtained by measuring the shape of each side surface of the billet B by each of the four shape measuring devices 2 is missing and uses it as a starting point. ..
Specifically, as shown in FIG. 4A, the signal processing device 3 has a standard curvature of the corner portion according to the size of the billet B based on the information input from the process computer (not shown). Specify the radius r. Next, the signal processing device 3 minimizes a circle having a radius of curvature r (a circle indicated by reference numeral r in FIG. 4A) in the shape measurement data obtained by synthesizing the four shape measurement data obtained by the four shape measurement devices 3. Shape measurement data (flat part measurement data) of the part corresponding to the flat side surface (flat side surface if there is no scratch) of billet B located between the contact points between each circle r and each shape measurement data after fitting by the square method or the like. ). Finally, the signal processing device 3 detects a portion where the shape measurement data does not exist in the allowable region in the depth direction (the hatched region shown in FIG. 4A) based on the flat portion measurement data. Let this be the starting point S0.

The signal processing device 3 resamples the shape measurement data obtained by the shape measuring device 2 according to the resolution of the aperture synthesis image, and aligns the shape measurement data with the aperture synthesis image. In the present embodiment, as described above, in a state where the billet B is conveyed in the longitudinal direction (Y direction), a plurality of aperture synthesis images are generated at a pitch of, for example, 10 mm. Therefore, the resolution of the aperture synthesis image is any. The ultrasonic probe 1 also has a Y direction of 10 mm / voxel. Further, the resolution of the aperture synthesis image generated for the pair of ultrasonic probes 1 facing in the X direction in FIG. 3 is 2.5 mm / boxel in the arrangement direction of the vibrator 11 (Z direction in FIG. 3). The resolution in the ultrasonic transmission direction (X direction) is 0.15 mm / boxel. Similarly, the resolution of the aperture synthesis image generated for the other pair of ultrasonic probes 1 facing in the Z direction is 2.5 mm / boxel in the arrangement direction (X direction) of the vibrator 11, and the ultrasonic waves. The resolution in the transmission direction (Z direction) is 0.15 mm / boxel. The signal processing device 3 resamples the shape measurement data with the same resolution according to the resolution of the aperture synthesis image, and aligns the shape measurement data with the boxels constituting the aperture synthesis image (alignment in the X direction, the Y direction, and the Z direction). )I do. For the alignment in the X direction and the Z direction, the geometrical positional relationship between the ultrasonic probe 1 and the shape measuring device 2 in the XZ plane is used. Further, for the alignment in the Y direction, the geometrical positional relationship between the ultrasonic probe 1 and the shape measuring device 2 in the Y direction and the transport speed of the billet B measured by the speed sensor (not shown). Is used. In general, the resolution of the shape measurement data obtained by the shape measurement device 23 is higher than the resolution of the aperture synthesis image. Therefore, at the time of the above resampling, a series expansion is applied to the shape measurement data as necessary. Data interpolation is desirable.

<Second step>
In the second step, the signal processing device 3 constitutes an aperture synthesis image because the pixels constituting the aperture synthesis image (in the present embodiment, the aperture synthesis image is generated in a state where the billet B is conveyed in the longitudinal direction). Pixels are voxels. Hereinafter, a case where the pixels constituting the aperture synthesis image are voxels will be described.) Of the above, the pixels are located in the depth direction of the billet B with respect to the starting point S0, and are predetermined. Identify voxels with concentrations above the threshold.
Specifically, as shown in FIG. 4B, the signal processing unit device 3 determines the density of each voxel Bi (i ≧ 1) in the aperture synthesis image located immediately below the starting point S0 with respect to the starting point S0. The preset threshold value is compared with the voxel B1 in order, and the voxel Bi having a concentration equal to or higher than the set threshold value is specified. At this time, in the present embodiment, a plurality of threshold values are set according to the distance from the side surface of the billet B (the side surface where the starting point S0 exists), and the plurality of threshold values are separated from the side surface of the billet B. It is said to be as small as possible.

More specifically, in this embodiment, two threshold values Th1 and Th2 are set, and Th2 <Th1. The signal processing device 3 compares the concentration of each voxel Bi with the threshold value Th1 in order from the voxel B1 closest to the starting point S0, and identifies the voxel Bi having a concentration of the threshold value Th1 or more. Then, when the voxel Bi having a concentration equal to or higher than the threshold value Th1 does not exist, the threshold value Th1 is switched to the threshold value Th2. The signal processing device 3 has a concentration of the threshold value Th1 or higher, and for the voxel Bi located below the voxel Bi located at the position farthest from the side surface of the billet B, the concentration of each voxel Bi and the threshold value Th2. To identify voxel Bi having a concentration of the threshold Th2 or higher.
In the example shown in FIG. 4B, voxels B1 and B2 are specified as voxels having a concentration of the threshold value Th1 or more, and voxels B3 and B4 are specified as voxels having a concentration of the threshold value Th2 or more.

<Third step>
In the third step, the signal processing device 3 determines the depth of surface flaws based on the position of the voxel Bi most distant from the side surface of the billet B among the voxel Bi identified in the second step.
Specifically, in the example shown in FIG. 4C, the signal processing device 3 is based on the position of the voxel B4 farthest from the side surface of the billet B among the voxels B1 to B4 identified in the second step. Determine the depth of surface flaws. More specifically, by substituting the coordinates of the starting point S0 with the coordinates of the voxel B4, the shape measurement data is corrected as shown by the thick line in FIG. 4 (c), and the surface is based on the corrected shape measurement data. The depth H of the flaw is determined.

According to the depth determination device 100 according to the present embodiment described above, the signal processing device 3 is generated on the side surface of the billet B based on the shape measurement data obtained by the shape measurement device 2 using the optical cutting method. The surface layer portion of the surface flaw can be detected accurately (the position and shape of the surface layer portion of the surface flaw can be accurately determined). Further, the signal processing device 3 can accurately detect the deep part of the surface flaw generated on the side surface of the billet B based on the aperture synthesis image generated by using the flaw detection signal output from the ultrasonic probe 1 (surface flaw). The position of the deep part of the can be determined accurately). Therefore, the signal processing device 3 has shape measurement data in the vicinity of the surface flaw (surface layer portion of the surface flaw) detected based on the shape measurement data (data of the portion of the shape measurement data located near the surface layer portion of the surface flaw). ), It is possible to determine the depth of the surface layer portion of the surface flaw. In addition, the depth of the deep part of the surface flaw is determined based on the density of the aperture synthesis image in the vicinity of the detected surface flaw (the density of the pixels located near the surface layer portion of the surface flaw among the voxels constituting the aperture synthesis image). It is possible to do. That is, since a voxel having a concentration of the predetermined threshold values Th1 and Th2 or more can be considered to be a pixel corresponding to the surface flaw, the depth of the deep part of the surface flaw is based on the voxel corresponding to the surface flaw. Can be determined.
Therefore, based on the shape measurement data corrected as described above (in other words, the depth of the surface layer portion of the surface flaw determined based on the shape measurement data, and the depth of the surface flaw determined based on the density of the aperture synthesis image). The depth H of the surface flaw can be accurately determined.

1 ... One-dimensional array type ultrasonic probe 2 ... Shape measuring device 3 ... Signal processing device 100 ... Surface flaw depth determination device B ... Billet (steel piece)

Claims (3)

A one-dimensional array type ultrasonic probe placed along the side surface of a piece of steel transported in the longitudinal direction ,
A shape measuring device arranged along the side surface of the steel piece and measuring the shape of the side surface of the steel piece by an optical cutting method.
A speed sensor that measures the transport speed of the steel pieces and
By applying signal processing to the flaw detection signal output from the one-dimensional array type ultrasonic probe, a two-dimensional image of a cross section in a direction orthogonal to the longitudinal direction of the steel piece is generated, and the shape measurement is performed. The one-dimensional array type ultrasonic probe and the shape measuring device are used to align the shape measurement data obtained by measuring the shape of the side surface of the steel piece with the device and the generated two-dimensional image. A signal processing device based on the geometrical positional relationship of the steel piece and the transport speed of the steel piece measured by the speed sensor.
The signal processing device detects surface flaws generated on the side surface of the steel piece based on the shape measurement data, and determines the depth of the detected surface flaws based on the shape measurement data and the two-dimensional image. do,
A device for determining the depth of surface flaws. A one-dimensional array type ultrasonic probe placed along the side surface of the steel piece,
A shape measuring device arranged along the side surface of the steel piece and measuring the shape of the side surface of the steel piece by an optical cutting method.
By applying signal processing to the flaw detection signal output from the one-dimensional array type ultrasonic probe, a two-dimensional image of a cross section in a direction orthogonal to the longitudinal direction of the steel piece is generated, and the shape measurement is performed. Signal processing to determine the depth of surface flaws on the side surface of the steel piece based on the shape measurement data obtained by measuring the shape of the side surface of the steel piece with an apparatus and the generated two-dimensional image. Equipped with equipment,
The signal processing device is
The first step of detecting a place where the shape measurement data is missing and using it as a starting point,
A second step of identifying pixels constituting the two-dimensional image, which are located in the depth direction of the steel piece with respect to the starting point and have a density equal to or higher than a predetermined threshold value.
Among the pixels specified in the second step, the third step of determining the depth of the surface flaw is executed based on the position of the pixel farthest from the side surface of the steel piece.
A device for determining the depth of surface flaws. In the second step, a plurality of threshold values are set according to the distance from the side surface of the steel piece, and the plurality of threshold values are small enough to be separated from the side surface of the steel piece.
The device for determining the depth of surface flaws according to claim 2.

Images