JP7013796B2 - Ultrasonic flaw detection method - Google Patents

Description

INDUSTRIAL APPLICABILITY The present invention uses an ultrasonic probe such as a one-dimensional array type ultrasonic probe to perform ultrasonic flaw detection on a flaw-detected material having a substantially circular cross section such as a tube or a plate material having a rectangular plan view. Regarding the method. In particular, the present invention relates to an ultrasonic flaw detection method capable of detecting inclined flaws extending in a direction inclined with respect to a predetermined reference direction, such as the axial direction of a flawed material having a substantially circular cross section, with high accuracy.

Conventionally, as an ultrasonic flaw detection method using a one-dimensional array type ultrasonic probe in which a plurality of oscillators are arranged in a row, a signal is sent to a flaw detection signal obtained by receiving an echo from a material to be detected by a plurality of oscillators. By applying the treatment, a cross-sectional image which is a two-dimensional image of the flaw detection signal for the cross section in the direction facing the one-dimensional array type ultrasonic probe of the flaw-detected material is generated, and the flaw is detected by using the cross-sectional image. An ultrasonic flaw detection method is known. As a method for generating this cross-sectional image, for example, an aperture synthesis method, a zone focus method, a dynamic depth focus method, and the like are known. In this cross-sectional image, the intensity (amplitude) of the flaw detection signal is expressed as the pixel density, and it is general that the higher the intensity of the flaw detection signal, the higher the pixel density.

On the other hand, in Patent Document 1, ultrasonic waves are transmitted and received from each oscillator (element in Patent Document 1) included in the one-dimensional array type ultrasonic probe (phased array probe in Patent Document 1). An ultrasonic flaw detection method that controls and detects tilted flaws has been proposed.
In order to detect tilted flaws with high accuracy, it is conceivable to apply a cross-sectional image generation method such as the above-mentioned aperture synthesis method to the ultrasonic flaw detection method described in Patent Document 1. Specifically, for example, when the scratched material has a substantially circular cross section and an inclined flaw extending in a direction inclined with respect to the axial direction of the scratched material is detected, a one-dimensional array type ultrasonic probe is used. A one-dimensional array type ultrasonic probe is placed facing the outer surface of the flaw-detected material so that the arrangement direction of the plurality of vibrators is along the axial direction of the flaw-detected material, and the flaw-detected material is arranged from the plurality of vibrators. By transmitting ultrasonic waves to the target and performing signal processing on the flaw detection signal obtained by receiving echoes from the flawed material by multiple oscillators, a cross-sectional image of the flaw detection signal for the cross section of the flaw-detected material is generated. It is conceivable to detect the inclined flaw of the scratched material by using the cross-sectional image.

However, as a result of the study by the present inventors, the echo from the inclined flaw becomes the maximum intensity according to the inclined angle of the inclined flaw (the angle formed by the extending direction of the inclined flaw and the predetermined reference direction of the scratched material). Due to the different propagation paths of ultrasonic waves and the directivity of the one-dimensional array type ultrasonic probe, the energy in the spreading direction (arrangement direction of the transducer) of the transmitted ultrasonic waves is non-uniform. For some reason, even if the tilted flaws have the same dimensions (length, width, depth), the tilt angle of the tilted flaws and the positional relationship between the one-dimensional array type ultrasonic probe and the tilted flaws It was found that the echo intensity from the tilted flaw was different, and the density of the flaw pixel region corresponding to the tilted flaw in the cross-sectional image was different. This is not limited to the case where the scratched material has a substantially circular cross section and an inclined flaw extending in a direction inclined with respect to the axial direction of the scratched material is detected as in the above-mentioned example, and the scratched material has a substantially circular cross section. The same applies to the case where an inclined flaw that is circular and extends in an inclined direction with respect to the radial direction of the scratched material is detected. Further, when the material to be detected is a plate material having a rectangular shape in a plan view and an inclined flaw extending in a direction inclined with respect to the longitudinal direction of the material to be detected is detected, or when the material to be detected is a plate material having a rectangular shape in a plan view and is injured. The same applies to the case of detecting an inclined flaw extending in an inclined direction with respect to the thickness direction of the material.
In the quality assurance of the scratched material, it is often required to judge the quality of the scratched material according to the degree of harmfulness of the scratch (scratch size), but as described above, the dimensions of the inclined scratches are the same. However, if the echo intensity from the tilted flaw differs or the density of the flaw pixel area differs depending on the tilt angle of the tilted flaw, the intensity set in the flaw detection signal to detect the tilted flaw is high. It becomes difficult to appropriately set the threshold value and the threshold value of the density set in the cross-sectional image.

As a method for correcting the non-uniformity of the density of the scratch pixel region in the cross-sectional image, for example, the method described in Patent Document 2 has been proposed.
In the method described in Patent Document 2, a comparison test piece provided with artificial flaws at a plurality of positions along the oscillator arrangement direction is detected to generate a cross-sectional image, and a pixel corresponding to each artificial flaw in the cross-sectional image is generated. By performing interpolation calculation using the density, the flaw pixel density distribution in the oscillator array direction is estimated, and the correction coefficient distribution is obtained from the estimated flaw pixel density distribution. Then, it is a method of correcting the cross-sectional image obtained about the actual scratched material by using the acquired correction coefficient distribution.

When the method described in Patent Document 2 is applied to ultrasonic flaw detection of tilted flaws using a cross-sectional image, a plurality of artificial flaws each having a different tilt angle within a range of at least up to the maximum tilt angle of the tilted flaw to be detected. It is necessary to provide. Since the inclination angle of the artificial flaw that can be provided is finite, the interpolation calculation must be performed for the inclination angle without the artificial flaw. However, as a result of the study by the present inventors, it is difficult to predict in advance the tendency of the density change of the scratch pixel region according to the tilt angle of the tilt flaw, so that it is also difficult to perform the interpolation calculation with high accuracy. It is conceivable to provide a large number of artificial flaws with different inclination angles, but this is not realistic considering the increase in the time required for calibration and the increase in the cost of providing the artificial flaws.

Japanese Patent Application Laid-Open No. 2002-228640 Japanese Unexamined Patent Publication No. 2014-55580

The present invention has been made in view of the above-mentioned problems of the prior art, and uses an ultrasonic probe such as a one-dimensional array type ultrasonic probe with respect to a predetermined reference direction of the scratched material. An object of the present invention is to provide an ultrasonic flaw detection method capable of detecting an inclined flaw extending in an inclined direction with high accuracy.

In order to solve the above problems, in the present invention, the ultrasonic probe is arranged to face the outer surface of the flawed material having a substantially circular cross section, and the ultrasonic probe is placed along the axial direction and the circumferential direction of the flawed material. This is an ultrasonic flaw detection method that detects tilted flaws extending in a direction inclined with respect to the axial direction of the flawed material by moving them relative to each other. The preparatory step of preparing a first test material provided with an omnidirectional artificial flaw having a hole having a substantially circular cross section extending in the direction, and the ultrasonic probe being arranged facing the outer surface of the first test material, the above-mentioned By relatively moving the ultrasonic probe along the axial direction and the circumferential direction of the first test material to detect the first test material, a plurality of first test materials obtained from the flaw detection signals of the first test material . By synthesizing the first test material cross- sectional image generation step of generating one test material cross-sectional image and the plurality of first test material cross-sectional images, a synthetic flaw pixel region which is a pixel region corresponding to the omnidirectional artificial flaw. An omnidirectional artificial flaw image generation step of generating an omnidirectional artificial flaw image including the above, and a correction coefficient calculation step of calculating a correction coefficient based on the following equation (1) for each pixel in the composite flaw pixel region. , The flaw-detected material cross- sectional image generation step of generating a cross- sectional image of the flaw-detected material obtained from the flaw-detecting signal of the flaw-detected material by detecting the flaw-detected material, and the synthetic flaw pixel in the flaw -detected material cross-sectional image. A correction step of generating a corrected cross- sectional image of the scratched material by applying a correction based on the following equation (2) to the corrected pixel region located at the same coordinates as the region, and a cross-sectional image of the damaged material after the correction. Provided is an ultrasonic flaw detecting method comprising an inclined flaw detecting step of detecting an inclined flaw of the flaw-to-be-detected material.
X = 20 ・ log (A / B) ・ ・ ・ (1)
D = C ・ 10 ^{(X / 20)} ・ ・ ・ (2)
In the above equations (1) and (2), X is the correction coefficient. In the above formula (1), A is the maximum density of the pixels in the synthetic flaw pixel region, and B is the density of each pixel in the synthetic flaw pixel region. In the above equation (2), C is the density of each pixel in the correction pixel region before correction, and D is the density of each pixel in the correction pixel region after correction.

According to the present invention, in the preparation step, a first test material in which an omnidirectional artificial flaw is provided on a material having the same material and cross-sectional dimensions as the scratch-detected material is prepared. This omnidirectional artificial flaw is a hole having a substantially circular cross section extending in the radial direction of the first test material (in the wall thickness direction when the first test material is a tube). The tangential direction of the peripheral edge of the hole having a substantially circular cross section continuously changes by 360 °. Therefore, regardless of the maximum inclination angle of the inclination flaw to be detected in the inclination flaw detection step, the tangential direction of the peripheral edge of the hole having a substantially circular cross section and the reference direction (axial direction) of the first test material are formed. The angle continuously changes in the range up to the maximum inclination angle of the inclination flaw to be detected in the inclination flaw detection step. A hole having a substantially circular cross section is preferable as an omnidirectional artificial flaw provided in the first test material because it can be easily machined.
Therefore, in the process of generating a cross- sectional image of the first test material, when the first test material is flawed with an ultrasonic probe to generate a plurality of cross-sectional images of the first test material for the cross section of the first test material, a single cross-sectional image is generated . Even if it is an omnidirectional artificial flaw, the cross section of the first test material is equivalent to the case where a plurality of artificial flaws having different tilt angles are provided and flaws are detected at least in the range up to the maximum tilt angle of the tilted flaw to be detected. An image will be obtained. Further, in each cross-sectional image of the first test material, the ultrasonic probe is relatively moved along the axial direction and the circumferential direction of the first test material in the same manner as in the case of detecting the flawed material, to obtain the first test material. Generated by detecting flaws. Therefore, the position of the flaw pixel region corresponding to the omnidirectional artificial flaw in each first test material cross-sectional image also changes, and each first test material cross-sectional image is synthesized (for example, in the omnidirectional artificial flaw image generation step). The omnidirectional artificial flaw image generated by (processing of adding the density of each corresponding pixel in each first test material cross-sectional image) has artificial flaws having the same dimensions at a plurality of different positions as in Patent Document 2. Similar to the case where a flaw is provided and flaw detection is performed, the flaw pixel region when flaw detection is performed by changing the positional relationship between the ultrasonic probe and the omnidirectional artificial flaw (the flaw pixel region of each first test material cross-sectional image is The combined composite flaw pixel area) will be included.

Next, according to the present invention, in the correction coefficient calculation step, the correction coefficient is calculated based on the equation (1) for each pixel in the composite flaw pixel region of the omnidirectional artificial flaw image . In other words, when the density of each pixel in the composite flaw pixel region is corrected by using the correction coefficient, the correction coefficient in which the density of each pixel in the composite flaw pixel region is equal to each other is set in the composite flaw pixel region. Calculate for each pixel of. That is, if the density of each pixel in the composite flaw pixel region is corrected using this correction coefficient, even if the artificial flaw has various tilt angles (omnidirectional artificial flaw), the ultrasonic probe can also be used. Regardless of the positional relationship between the omnidirectional artificial flaw and the omnidirectional artificial flaw, if the omnidirectional artificial flaws are the same as each other, each pixel in the composite flaw pixel region will have the same density.
As described above, the correction coefficient is calculated by using the first test material cross- sectional image equivalent to the case where a plurality of artificial flaws, each having a different inclination angle, are provided and flaw detection is performed . Therefore , in the correction step, the flaw-detected material cross- sectional image generated from the flaw-finding signal of the flaw-detected material is located at the same coordinates as the composite flaw pixel region in the flaw-detected material cross-sectional image. If the correction pixel area is corrected based on the equation (2), that is, the density of each pixel in the correction pixel area is corrected by using the correction coefficient of each pixel calculated in the correction coefficient calculation step, the correction is performed. If there is a pixel region corresponding to the tilted flaw in the corrected pixel region in the scratched material cross-sectional image, regardless of the tilt angle of the tilted flaw, and regardless of the positional relationship between the ultrasonic probe and the tilted flaw. (Regardless of the position of the pixel area corresponding to the tilted flaw in the corrected cross-sectional image of the scratched material), if the tilted flaws have the same dimensions, it is expected that the cross- sectional image of the damaged material with the same density can be corrected. can.
Therefore, in the tilted flaw detection step, it is possible to appropriately set the threshold value when detecting the tilted flaw of the flawed material using the corrected cross- sectional image of the flawed material.
According to the present invention, for example, when the density B of one pixel in the composite flaw pixel region is equal to the maximum density A (that is, B = A), the correction coefficient X = 0 according to the equation (1). , D = C according to the equation (2). That is, the density of the pixel corresponding to the one pixel in the correction pixel area does not change before and after the correction. On the other hand, when the density B of the other pixels in the composite flaw pixel region is 1/10 of the maximum density A (that is, B = 0.1A), the correction coefficient X = 20 according to the equation (1). , D = 10 · C according to the equation (2). That is, the correction is performed to increase the density before the correction by 10 times. In other words, a correction is applied to each pixel in the corrected pixel region in the cross-sectional image of the scratched material so as to unify the densities of all the pixels in the composite flaw pixel region to the maximum density A. Therefore, if the tilted flaws existing in the scratched material have the same dimensions, the density of the pixel region corresponding to the tilted flaw in the correction pixel region becomes the same, and the tilted flaw can be detected with high accuracy. You can expect it.

As described above, according to the present invention, the ultrasonic probe can be used to detect tilted flaws extending in a direction tilted with respect to a predetermined reference direction (axial direction) of the scratched material with high accuracy. be.
In the process of generating a cross- sectional image of the first test material of the present invention, it is preferable to detect the first test material under the same flaw detection conditions as when detecting the flawed material. The flaw detection conditions are as follows. Includes content.
(1) Ultrasonic probe such as the distance between the ultrasonic probe and the outer surface of the material (damaged material, first test material) and the refraction angle of ultrasonic waves seen from the reference direction (axial direction ) of the material. Arrangement relationship between tentacles and materials.
(2) A control method for each vibrator provided by the ultrasonic probe (control of ultrasonic transmission / reception timing of each vibrator, etc.).
(3) Types of cross-sectional image generation methods such as aperture synthesis method, zone focus method, and dynamic depth focus method when generating a cross-sectional image that is a two-dimensional image of a flaw detection signal.
In the present invention, in the first test material cross- sectional image generation step, the first test material is detected by making the above-mentioned flaw detection conditions equivalent to those when the flaw-detected material is detected, and the first test material cross- sectional image is generated. Therefore, it is possible to improve the accuracy of the correction.
The "tilt flaw" in the present invention also includes a flaw having an inclination angle of 0 ° (that is, a flaw extending parallel to the reference direction (axial direction) of the material to be detected).

Here, according to the results examined by the present inventors, the flaw detection sensitivity (amplification degree of the flaw detection signal) of the ultrasonic probe required for flaw detection of a hole having a substantially circular cross section as an omnidirectional artificial flaw. It was found that there is a large difference between the flaw detection sensitivity of the ultrasonic probe and the flaw detection required when detecting the actual tilted flaws to be detected by the flaw-detected material. Specifically, in order to obtain a flaw detection signal of the same strength as a hole having a substantially circular cross section and an actual tilted flaw, the flaw detection sensitivity of the ultrasonic probe when detecting the actual tilted flaw is omnidirectional artificial. It was found that it was necessary to make it smaller than the flaw detection sensitivity of the ultrasonic probe when detecting flaws. Alternatively, it was found that when the flaw detection sensitivity is set to be the same, the threshold value for detecting an actual tilted flaw needs to be higher than the threshold value for detecting an omnidirectional artificial flaw.
Therefore, using an ultrasonic probe set to the same flaw detection sensitivity as when detecting flaws in the first test material (detecting omnidirectional artificial flaws) in the process of generating a cross- sectional image of the first test material, When the flawed material is detected in the process of generating a cross- sectional image of the flawed material, there is a possibility that noise other than shallow inclined flaws and inclined flaws that do not need to be detected may be over-detected. Similarly, over-detection may occur when the tilted flaw is detected at the threshold value for detecting the omnidirectional artificial flaw in the tilted flaw detecting step.

In order to reduce the risk of over-detection as described above, in the preparation step, the first test material has an inclination angle within the range of the maximum inclination angle of the inclination flaw to be detected of the scratched material. Then , a notch simulating the inclined flaw is provided, and in the first test material cross- sectional image generation step, the notch cross- sectional image obtained from the flaw detection signal of the notch is obtained by detecting the first test material provided with the notch. Is generated, and the flaw detection sensitivity of the ultrasonic probe in the flaw-detected material cross-section image generation step is set based on the notch cross-section image , or the tilt flaw is detected in the tilt flaw detection step. It is preferable to set a threshold value for this.

According to the above preferred method, in addition to providing an omnidirectional artificial flaw, a notch simulating an inclined flaw is provided in the first test material, and the first test material is flaw-detected to obtain a flaw detection signal of the notch. Since the flaw detection sensitivity of the ultrasonic probe is set or the threshold value for detecting the tilted flaw is set by using the notch cross- sectional image , the risk of over-detection can be reduced.

In the above preferred method, the first test material is provided with a notch simulating an omnidirectional artificial flaw and an inclined flaw, but the present invention is not limited to this, and the first test material provided with the omnidirectional artificial flaw is provided. Separately, prepare a second test material with a notch, and use this second test material to set the flaw detection sensitivity of the ultrasonic probe and set the threshold value for detecting tilted flaws. Is also possible.
That is, in the preparatory step, the material having the same material and cross-sectional dimensions as the scratched material has an tilt angle within the range of the maximum tilt angle of the tilted flaw to be detected by the flawed material, and the tilted flaw. A second test material having a notch simulating the above is prepared, the ultrasonic probe is placed facing the outer surface of the second test material, and the ultrasonic probe is placed in the axial direction of the second test material. The second test material flaw detection data generation step of generating the second test material flaw detection data obtained from the flaw detection signal for the second test material by relatively moving along the circumferential direction to detect the second test material. Including
Based on the second test material flaw detection data, the flaw detection sensitivity of the ultrasonic probe in the flaw detected material cross- sectional image generation step is set, or the tilt flaw is detected in the tilt flaw detection step. It is also possible to set a threshold.

According to the present invention, an ultrasonic probe such as a one-dimensional array type ultrasonic probe is used to create an inclined flaw extending in a direction inclined with respect to a predetermined reference direction (axial direction) of the material to be detected. It can be detected with high accuracy.

It is a schematic diagram which shows the schematic structure of the ultrasonic flaw detection apparatus used in the ultrasonic flaw detection method which concerns on 1st Embodiment of this invention. An example of a cross-sectional image generated when each inclined flaw provided on the inner surface is detected in an evaluation test is shown. It is a figure which shows the relationship between the maximum density | concentration of the pixel in the flaw pixel area F corresponding to the tilted flaw in each cross-sectional image shown in FIG. 2 and the tilt angle θ of the tilted flaw. It is explanatory drawing explaining the procedure of generating the cross-sectional image of a plurality of 1st test material in the 1st test material flaw detection data generation process. An example of a plurality of first test material cross-sectional images generated in the first test material flaw detection data generation step is shown. An example of an omnidirectional artificial flaw image is shown. It is explanatory drawing explaining the procedure of the correction coefficient calculation process. It is explanatory drawing explaining the procedure of a correction process. It is explanatory drawing explaining the condition and method of the test which confirms the effect of the correction by the ultrasonic flaw detection method which concerns on 1st Embodiment. It is a figure which shows the result of the test shown in FIG. It is a figure which shows the condition and method of the test which confirms the effect of the correction by the ultrasonic flaw detection method which concerns on 2nd Embodiment, and the example of the result of the 1st test material flaw detection data generation process. It is a figure which shows the result of the test shown in FIG.

Hereinafter, the ultrasonic flaw detection method according to the embodiment of the present invention will be described with reference to the accompanying drawings as appropriate. In the first embodiment, the tilted flaw is detected by using the cross-sectional image obtained from the flaw detection signal, and in the second embodiment, the tilted flaw is detected by using the flaw detection signal itself without using the cross-sectional image obtained from the flaw detection signal. Each of the cases of detecting the above will be described.

<First Embodiment>
First, the configuration of the ultrasonic flaw detector used in the ultrasonic flaw detection method according to the first embodiment will be described.
[Structure of ultrasonic flaw detector according to the first embodiment]
FIG. 1 is a schematic diagram showing a schematic configuration of an ultrasonic flaw detector used in the ultrasonic flaw detection method according to the first embodiment of the present invention. In FIG. 1, the scratched material and the first test material are tubes having a substantially circular cross section, and an inclined flaw extending in a direction inclined with respect to the axial direction is detected (the reference direction is the axis of the scratched material and the first test material). The case (direction) is shown as an example. FIG. 1A shows a view seen from the axial direction of the pipe, and FIG. 1B shows a perspective front view seen from a direction substantially orthogonal to the axial direction of the pipe.
As shown in FIG. 1, the ultrasonic flaw detector 100 according to the present embodiment is a one-dimensional array type ultrasonic probe having a plurality of oscillators 10 arranged in a row (hereinafter, appropriately, “ultrasonic detector”. (Abbreviated as "touching element") 1 and a control / signal processing means 2 are provided.

The ultrasonic probe 1 is arranged to face the outer surface of the scratched material P1 so that the arrangement direction of the plurality of oscillators 10 is along the axial direction of the scratched material P1 (X direction in FIG. 1). Specifically, the ultrasonic probe 1 is arranged in a tank in which a contact medium such as water is stored together with the scratched material P1, for example. As a result, the ultrasonic waves transmitted from each oscillator 10 included in the ultrasonic probe 1 are incident on the flawed material P1 via the contact medium, and the echo from the flawed material P1 is transmitted through the contact medium. Is received by each oscillator 10.

The ultrasonic probe 1 relatively moves along the axial direction and the circumferential direction of the scratched material P1 by an appropriate drive mechanism (not shown), whereby the entire surface of the scratched material P1 is detected. Specifically, the ultrasonic probe 1 is attached to, for example, a uniaxial stage, and when the uniaxial stage is driven, the ultrasonic probe 1 moves along the axial direction of the scratched material P1. On the other hand, the scratched material P1 is supported by, for example, a rotating roller arranged below the scratched material P1, and the rotating roller rotates to rotate in the circumferential direction. The ultrasonic probe 1 moves along the axial direction of the flawed material P1 and the flawed material P1 rotates in the circumferential direction, so that the flaw detection range of the ultrasonic probe 1 is the axis of the flawed material P1. It will move spirally along the direction. At this time, the moving speed of the ultrasonic probe 1 and the rotation speed of the flawed material P1 are set so that the flaw detection range of the ultrasonic probe 1 moving in a spiral shape can cover the entire outer surface of the flawed material P1 ( It is set so that an undetected area does not occur), so that the entire surface of the detected material P1 can be detected.
In the above example, the case where the ultrasonic probe 1 moves along the axial direction of the flawed material P1 and the flawed material P1 rotates in the circumferential direction has been described, but the present invention is limited to this. Instead, the ultrasonic probe 1 rotates along the circumferential direction of the flawed material P1 and the flawed material P1 moves in the axial direction, and the position of the ultrasonic probe 1 is fixed to detect the flaw. The mode in which the material P1 moves in the axial direction while rotating in the circumferential direction, and the position of the flaw-detected material P1 is fixed, and the ultrasonic probe 1 rotates along the circumferential direction of the flaw-detected material P1 to detect the flaw-detected material. Various modes are adopted as long as the ultrasonic probe 1 moves relative to the scratched material P1 along the axial direction and the circumferential direction of the scratched material P1, such as a mode of moving along the axial direction of P1. It is possible.

The control / signal processing means 2 controls transmission / reception of ultrasonic waves from each oscillator 10 included in the ultrasonic probe 1, and each oscillator 10 receives an echo from the flaw-detected material P1 to obtain a flaw detector. By applying signal processing to the signal, a cross-sectional image (for example, an aperture composite image) of the flaw-detecting signal for the cross-section of the flaw-detected material P1 is generated, and the cross-sectional image is used to incline the flaw-detected material P1 in the axial direction. Detects tilted flaws extending in the desired direction.

The control / signal processing means 2 selects and switches the vibrator 10 for transmitting and receiving ultrasonic waves, which is commonly used in a general ultrasonic flaw detector using a one-dimensional array type ultrasonic probe (at once). A switching circuit (which selects and switches the oscillator 10 to be used), a pulsar for transmitting ultrasonic waves from the selected oscillator 10, and for the selected oscillator 10 to receive echoes from the damaged material P1. The same known components as general ultrasonic flaw detectors, such as the receiver of, an amplifier that amplifies the flaw detection signal output from the receiver, and an A / D converter that converts the flaw detection signal output from the amplifier into digital data. Equipped. Further, the control / signal processing means 2 includes a general-purpose personal computer connected to a pulsar, a receiver, and an A / D converter. Digital data is input to this personal computer from the A / D converter, and the pulsar and receiver are controlled (control of the transmission / reception timing of ultrasonic waves of each transducer 10), and the input digital data is input to 2. A predetermined program for performing an operation for converting to a dimensional image (such as an aperture synthesis operation) is installed. Further, this personal computer is provided with a monitor for displaying a cross-sectional image which is a two-dimensional image.

In the ultrasonic flaw detector 100 according to the first embodiment, as shown in FIG. 1, the virtual focusing point C of ultrasonic waves is located on the side of the ultrasonic probe 1 facing the flawed material P1 and on the opposite side. By controlling the transmission timing of each oscillator 10 by the control / signal processing means 2, the ultrasonic probe 1 transmits ultrasonic waves having a fan-shaped wave surface to the flawed material P1 at once. Take the case as an example. Further, a case where the two-dimensional imaging method performed by the control / signal processing means 2 is an aperture synthesis method will be given as an example.

The ultrasonic flaw detection method according to the first embodiment is executed by using the ultrasonic flaw detector 100 having the configuration described above. Hereinafter, the ultrasonic flaw detection method according to the first embodiment will be described.

[Details of ultrasonic flaw detection method according to the first embodiment]
First, the result of evaluating a cross-sectional image obtained by detecting a plurality of inclined flaws having different inclination angles using an ultrasonic flaw detector 100 without correction will be described.
Notches (lengths) with inclination angles θ = 0 °, 15 °, 30 °, and 45 ° on the inner and outer surfaces of the flaw-detected material P1 (outer diameter: 193.7 mm, wall thickness: 12.7 mm) as inclined flaws. S: 12.7 mm, depth: 5% of the wall thickness of the flawed material P1), and the ultrasonic flaw detector 100 is used to detect each inclined flaw of the flawed material P1 in a stationary state, and a cross-sectional image. (Aperture synthesis image) was generated.

FIG. 2 shows an example of a cross-sectional image generated when each inclined flaw provided on the inner surface is detected in the above evaluation test. FIG. 2 (a) is a cross-sectional image generated when a tilted flaw with an inclination angle θ = 0 ° is detected, and FIG. 2 (b) is generated when a flaw is detected with an inclination angle θ = 15 °. A cross-sectional image is shown in FIG. 2 (c), a cross-sectional image generated when an inclined flaw with an inclination angle of θ = 30 ° is detected, and FIG. 2 (d) is an inclining flaw with an inclination angle of θ = 45 °. The cross-sectional image generated in is shown in. The horizontal axis of each cross-sectional image shows the axial position of the scratched material P1, and the vertical axis shows the wall thickness direction position of the scratched material P1. The cross-sectional image shown in FIG. 2 is actually displayed in a monitor provided with the control / signal processing means 2 with different colors depending on the density of each pixel. Although it is difficult to understand because FIG. 2 attached to the present specification is a monochrome display, the density of the pixels in the scratch pixel region F corresponding to the tilted scratches is higher than the density of the pixels in the other pixel regions. The color corresponding to the high density is colored. The same applies to FIGS. 5, 6, 7, and 8 described later.
FIG. 3 is a diagram showing the relationship between the maximum density of pixels in the flaw pixel region F corresponding to the tilted flaw in each cross-sectional image shown in FIG. 2 and the tilt angle θ of the tilted flaw. The maximum density of the pixels in the scratch pixel region F on the vertical axis of FIG. 3 is converted into dB units based on the maximum density when the tilt angle θ = 0 ° of the tilted flaw. Note that FIG. 3 also shows the same result when each inclined flaw provided on the outer surface is detected.

As shown in FIG. 3, it can be seen that the density of the flaw pixel region F corresponding to the tilted flaw in the cross-sectional image differs depending on the tilted angle θ of the tilted flaw even if the tilted flaws have the same dimensions. Further, it can be seen that it is difficult to predict in advance the tendency of the density change of the scratch pixel region F according to the tilt angle θ of the tilt flaw. Therefore, it is difficult to appropriately set the density threshold value set in the cross-sectional image in order to detect the inclined flaw.
Therefore, the ultrasonic flaw detection method according to the first embodiment is characterized by including the steps described below so that the flaw pixel regions in the cross-sectional image have the same density if the tilted flaws have the same dimensions. It is supposed to be.

The ultrasonic flaw detection method according to the first embodiment includes at least a preparation step, a first test material flaw detection data generation step (in the first embodiment, a first test material cross-sectional image generation step) , and omnidirectional artificial flaw data. A generation step (in the first embodiment, an omnidirectional artificial flaw image generation step) , a correction coefficient calculation step, a flaw detection data generation step (in the first embodiment, a flaw detection material cross-section image generation step) , and It has a correction step and a tilt flaw detection step. Hereinafter, each step will be described in sequence.

(Preparation process)
In the preparatory step, a first test material, which is a tube formed by providing an omnidirectional artificial flaw to a material having the same material and cross-sectional dimensions (outer diameter, inner diameter) as the scratched material P1, is prepared. In the omnidirectional artificial flaw provided on the first test material, the angle formed by the tangential direction of the peripheral edge thereof and the reference direction (axial direction) of the first test material is at least a detection target in the inclined flaw detection step described later. It is a flaw that changes continuously in the range up to the maximum tilt angle θmax of the tilt flaw. More specifically, when the first test material is detected by using the ultrasonic probe 1, the peripheral edge of the flaw seen from the direction orthogonal to the wavefront of the ultrasonic wave incident on the flaw is tangential to the above. It is a flaw that changes like this. For example, assuming that the maximum inclination angle θmax (absolute value) of the inclined flaw is 45 °, in the preparation step, the tangential direction of the peripheral edge of the omnidirectional artificial flaw and the reference direction (axial direction) of the first test material are set. This means that a flaw is provided such that the angle θf formed is continuously changed in the range of at least −45 ° ≦ θf ≦ 45 °.

As the omnidirectional artificial flaw as described above, for example, a hole having a substantially circular cross section extending in the radial direction (thickness direction) of the first test material is preferably used. The tangential direction of the peripheral edge of the hole having a substantially circular cross section continuously changes by 360 °. Therefore, regardless of the value of the maximum inclination angle θmax of the inclination flaw to be detected in the inclination flaw detection step described later, the tangential direction of the peripheral edge of the hole having a substantially circular cross section and the reference direction (axial direction) of the first test material. ) Will change continuously in the range up to the maximum tilt angle θmax. A hole having a substantially circular cross section is preferable as an omnidirectional artificial flaw provided in the first test material because it can be easily machined. However, the omnidirectional artificial flaw provided in the first test material in the preparation step of the present invention is not limited to this, and various artificial flaws such as a hole having a substantially semicircular cross section shall be provided as the omnidirectional artificial flaw. Is possible.

In the present embodiment, as a preferred method, in the preparatory step, a notch simulating an inclined flaw to be detected by the flaw-detected material P1 is further provided on the first test material provided with the omnidirectional artificial flaw. Specifically, for example, the inclination angle θ (for example, θ =) in the range of the maximum inclination angle θmax of the inclination flaw to be detected at a portion different from the portion where the omnidirectional artificial flaw of the first test material is provided. At least one notch with (0 °, 45 °, etc.) is provided.

(First test material flaw detection data generation process)
FIG. 4 is an explanatory diagram illustrating a plurality of first test material cross-sectional image generation procedures in the first test material flaw detection data generation step. FIG. 4 is a view seen from the opposite direction (vertical direction in this embodiment) between the ultrasonic probe 1 and the first test material P2.
In the first test material flaw detection data generation step, as shown in FIG. 4, the ultrasonic probe 1 is arranged to face the outer surface of the first test material P2 along the axial direction and the circumferential direction of the first test material P2. By relatively moving the first test material P2 for flaw detection, a plurality of first test material cross-sectional images as first test material flaw detection data obtained from the flaw detection signal for the cross section of the first test material P2 are generated. Specifically, a flaw pixel region corresponding to an omnidirectional artificial flaw (in this embodiment, a hole having a substantially circular cross section extending from the inner surface of the first test material P2 in the wall thickness direction) is present in any cross-sectional image. As described above, the ultrasonic probe 1 is relatively moved along the axial direction and the circumferential direction of the first test material P2 to detect the first test material P2. As a result, the control / signal processing means 2 generates a first test material cross-sectional image which is a plurality of two-dimensional images of the cross section of the first test material P2. That is, a cross-sectional image of the first test material is generated at each relative movement position of the ultrasonic probe 1, and a plurality of cross-sectional images of the first test material are obtained by sequentially changing the relative movement position of the ultrasonic probe 1. ..

At this time, the first test material P2 detects the flaw under the same flaw detection conditions as when the flaw-detected material P1 is detected. Specifically, the distance between the ultrasonic probe 1 and the outer surface of the first test material P2 is the distance between the ultrasonic probe 1 and the outer surface of the flawed material P1 when detecting the flawed material P1. Set equal to the distance. Further, the refraction angle of the ultrasonic wave seen from the axial direction of the first test material P2 is the refraction angle of the ultrasonic wave seen from the axial direction of the flaw-detected material P1 when the flaw-detected material P1 is detected (FIG. 1A). Set the same as the angle α) shown in. Further, the control method of each oscillator 10 in the control / signal processing means 2 (control of the transmission / reception timing of ultrasonic waves of each oscillator 10, etc.) is made the same, and the ultrasonic probe 1 to the first test material P2 are used. Set the ultrasonic waves of the fan-shaped wavefront to be transmitted at one time. Further, the aperture synthesis method is used as a method for two-dimensionally imaging the flaw detection signal in the control / signal processing means 2 as in the case of flaw detection of the flaw detection material P1. As for the specific content of the aperture synthesis method, for example, a known method as described in Japanese Patent Application Laid-Open No. 2014-55885 can be applied, and therefore detailed description thereof will be omitted here.

However, the cross-sectional image of the first test material generated in the process of generating the flaw detection data of the first test material is synthesized and used for calculating the correction coefficient, as will be described later. Therefore, the ultrasonic probe 1 is moved in the axial direction and the circumference of the first test material P2 so that the position of the flaw pixel region corresponding to the omnidirectional artificial flaw in each first test material cross-sectional image changes at a small pitch. It is preferable that the speed of relative movement along the direction is slower than the speed at which the flaw-detected material P1 is detected. Alternatively, in the case of intermittently (repeating movement and stop) relative movement of the ultrasonic probe 1 along the axial direction and the circumferential direction of the first test material P2, the cross-sectional image of the first test material is displayed. It is preferable to reduce the pitch of the generated stop position.

FIG. 5 shows an example of a plurality of first test material cross-sectional images generated in the first test material flaw detection data generation step. The horizontal axis of each first test material cross-sectional image indicates the axial position of the first test material, and the vertical axis indicates the position of the first test material in the wall thickness direction. As shown in FIG. 5, each first test material cross-sectional image is generated by relatively moving the ultrasonic probe 1 along the axial direction and the circumferential direction of the first test material to detect the first test material. Therefore, the position of the flaw pixel region F corresponding to the omnidirectional artificial flaw in each first test material cross-sectional image also changes according to the relative movement position of the ultrasonic probe 1.

In the present embodiment, as a preferred method, in the first test material flaw detection data generation step, the notch flaw detection data obtained from the notch flaw detection signal is generated by detecting the first test material further provided with the notch. Specifically, as in the case of an omnidirectional artificial flaw, the ultrasonic probe 1 is placed facing the outer surface of the first test material and relatively moved along the axial direction and the circumferential direction of the first test material. By detecting the first test material, a plurality of notch cross-sectional images as notch flaw detection data obtained from the flaw detection signal of the notch provided in the first test material are generated. More specifically, the ultrasonic probe 1 is placed in the axial direction and circumference of the first test material so that a pixel region corresponding to a notch simulating an inclined flaw provided in the preparation step exists in any cross-sectional image. The first test material is detected by moving it relative to each other along the direction. As a result, the control / signal processing means 2 generates a notch cross-sectional image which is a plurality of two-dimensional images of the cross-section of the first test material. That is, a notch cross-sectional image is generated at each relative movement position of the ultrasonic probe 1, and a plurality of notch cross-sectional images are obtained by sequentially changing the relative movement position of the ultrasonic probe 1.

In the first test material flaw detection data generation step, when generating a notch cross-sectional image, the flaw detection sensitivity of the ultrasonic probe 1 is appropriately changed, and the notch is created by comparing the notch cross-section images generated by each flaw detection sensitivity. Identify suitable flaw detection sensitivities for detection. Alternatively, the flaw detection sensitivity of the ultrasonic probe 1 remains constant to identify a threshold suitable for detecting a notch in the generated notch cross-section image.

(Anidirectional artificial flaw data generation process)
In the omnidirectional artificial flaw data generation step, omnidirectional artificial flaw data related to the strength of the flaw detection signal corresponding to the omnidirectional artificial flaw is generated based on the generated first test material flaw detection data. Specifically, in the present embodiment, a plurality of first test material cross-sectional images as shown in FIG. 5 are synthesized by the control / signal processing means 2 (concentration of each corresponding pixel in each first test material cross-sectional image). Is added) to generate an omnidirectional artificial flaw image as omnidirectional artificial flaw data as shown in FIG. The horizontal axis of the omnidirectional artificial flaw image indicates the axial position of the first test material, and the vertical axis indicates the position of the first test material in the wall thickness direction. The region indicated by the reference numeral FT in FIG. 6 is a composite flaw pixel region corresponding to the omnidirectional artificial flaw in the omnidirectional artificial flaw image. Since the composite flaw pixel region FT is used for calculating the correction coefficient as described later, it is preferable that the composite flaw pixel region FT extends as much as possible in the horizontal axis direction and the vertical axis direction of the omnidirectional artificial flaw image. The spread of the synthetic flaw pixel region FT in the horizontal axis direction depends on the range of the relative movement distance along the axial direction of the first test material of the ultrasonic probe 1 when generating the cross-sectional image of each first test material. The spread of the synthetic flaw pixel region FT in the vertical axis direction is the range of the relative movement distance along the circumferential direction of the first test material of the ultrasonic probe 1 when generating the cross-sectional image of each first test material. It changes according to. Therefore, the composite flaw pixel region FT in the omnidirectional artificial flaw image is along the axial direction and the circumferential direction of the first test material so as to spread as much as possible in the horizontal axis direction and the vertical axis direction of the omnidirectional artificial flaw image. It is preferable to set the range of the relative movement distance of the ultrasonic probe 1.

(Correction coefficient calculation process)
In the correction coefficient calculation step, the correction coefficient is calculated based on the generated omnidirectional artificial scratch data (omnidirectional artificial scratch image). Specifically, when the density of each pixel in the composite flaw pixel region FT in the omnidirectional artificial flaw image is corrected by using the correction coefficient, the density of each pixel in the composite flaw pixel region FT is equal to each other. The correction coefficient 2 is calculated for each pixel in the composite flaw pixel region FT by the control / signal processing means 2. Hereinafter, a more specific description will be given.

FIG. 7 is an explanatory diagram illustrating the procedure of the correction coefficient calculation process. 7 (a) is an example of an omnidirectional artificial flaw image, FIG. 7 (b) is an example of extracting a synthetic flaw pixel region FT, and FIG. 7 (c) is a correction of each pixel in the synthetic flaw pixel region FT. An example of displaying the coefficient as an image is shown.
As shown in FIG. 7, in the correction coefficient calculation step of the present embodiment, among the pixels of the omnidirectional artificial flaw image shown in FIG. 7A, the pixels having a density exceeding a predetermined threshold value are combined. It is extracted as a pixel region FT, and the remaining pixels are used as noise components to set the density to 0 (FIG. 7 (b)). In addition, in FIG. 7 (b), pixels other than the composite flaw pixel region FT (pixels located above FIG. 7 (b)) are also extracted, but in this example, the vicinity of the inner surface of the tube is defined as the flaw detection range in advance. Since the pixels other than the extracted composite flaw pixel region FT are out of the flaw detection range, they are not used for calculating the correction coefficient.
Then, in the correction coefficient calculation step of the present embodiment, assuming that the maximum density of the pixels in the extracted composite flaw pixel region FT is A and the density of each pixel in the composite flaw pixel region FT is B, the following equation (1) ), The correction coefficient X is calculated for each pixel in the composite flaw pixel region FT.
X = 20 ・ log (A / B) ・ ・ ・ (1)

FIG. 7 (c) displays the correction coefficient X calculated for each pixel in the composite flaw pixel region FT based on the above formula (1) after converting it into the density of each pixel according to a predetermined conversion formula. For pixels within the flaw detection range (near the inner surface of the tube in this example) and outside the composite flaw pixel region FT, the correction coefficient X represented by the above equation (1) is not calculated and will be described later. In the correction step of, the correction is made so that the corrected densities are all 0.
The calculated correction coefficient X is stored in the control / signal processing means 2.

(Scratch-detected material flaw detection data generation process)
In the process of generating flaw detection data for the flawed material, the ultrasonic probe 1 is arranged facing the outer surface of the flawed material P1 and relatively moved along the axial direction and the circumferential direction of the flawed material P1 to detect the flawed material P1. By doing so, the control / signal processing means 2 generates the flaw detection data obtained from the flaw detection signal for the cross section of the flaw detection material P1. That is, in the present embodiment, a cross-sectional image of the scratched material, which is a two-dimensional image of the cross section of the scratched material P1, is generated.
When the flaw detection sensitivity suitable for detecting the notch is specified in the first test material flaw detection data generation step described above, this is specified as the flaw detection sensitivity of the ultrasonic probe 1 in the flaw detection material generation data generation step. Use flaw detection sensitivity. That is, the flaw-detected material P1 is flaw-detected by the ultrasonic probe 1 having the flaw-detection sensitivity set specified in the first test material flaw-detection data generation step.

(Correction process)
In the correction step, the corrected flaw detection data (cross-sectional image of the flaw-detected material) is generated by making corrections using the correction coefficient calculated for the generated flaw-detection data (cross-sectional image of the flaw-detected material). .. Specifically, the control / signal processing means 2 calculated the density of each pixel in the correction pixel region corresponding to the composite flaw pixel region FT in the scratched material cross-sectional image of the scratched material P1 in the correction coefficient calculation step. By performing correction using the correction coefficient X of each pixel, a corrected cross-sectional image of the damaged material is generated. Since the cross-sectional image of the scratched material P1 is sequentially generated along with the relative movement of the ultrasonic probe 1, the corrected cross-sectional image of the scratched material is also sequentially generated. Hereinafter, the correction process will be described more specifically.

FIG. 8 is an explanatory diagram illustrating the procedure of the correction step. FIG. 8A shows an example of a cross-sectional image of the damaged material P1 to be detected, and FIG. 8B shows an example of a corrected cross-sectional image of the material to be detected.
In the correction step of the present embodiment, the density of each pixel in the correction pixel region (pixel region located at the same coordinate as the synthetic flaw pixel region FT shown in FIG. 7) in the scratch-detected material cross-sectional image shown in FIG. 8A is adjusted. The control / signal processing means 2 makes corrections using the correction coefficients of each pixel calculated and stored in the correction coefficient calculation step, and generates a corrected cross-sectional image of the scratched material shown in FIG. 8 (b). In the correction step of the present embodiment, assuming that the density of each pixel in the correction pixel region is C, the density D of each corrected pixel is calculated based on the following equation (2).
D = C ・ 10 ^{(X / 20)}・ ・ ・ (2)
In reality, the control / signal processing means 2 corrects not only the density of each pixel in the correction pixel area but also the density of all the pixels in the flaw detection range (near the inner surface of the tube in this example). I do. However, as described above, the pixels within the flaw detection range and outside the composite flaw pixel region FT are corrected so that the corrected density is all 0.
Further, the corrected density D calculated based on the above equation (2) is the maximum value of the density that can be set by the control / signal processing means 2 (for example, when the density is represented by 8 bits, the maximum value is 255). If it exceeds the value, the value of the correction coefficient X calculated in the equation (1) may be uniformly set small (for example, set to X-6 or the like).

Comparing the flaw pixel region F corresponding to the flaw shown in FIG. 8 (a) with the flaw pixel region F shown in FIG. 8 (b), the density of the pixels in the flaw pixel region F is changed by the correction. You can see that.

(Inclination flaw detection process)
In the tilted flaw detection step, the tilted flaw of the flawed material P1 is detected by using the corrected flaw detection data of the flawed material. That is, the control / signal processing means 2 detects the inclined flaw of the scratched material P1 by using the corrected cross-sectional image of the scratched material as shown in FIG. 8B. Specifically, the corrected cross-sectional image of the scratched material is compared with a predetermined threshold value set in advance, and pixels having a density exceeding the predetermined threshold value are detected as tilted flaws.
When a threshold value suitable for detecting a notch is specified in the first test material flaw detection data generation step described above, this specified threshold value is used in the tilt flaw detection step.

According to the ultrasonic flaw detection method according to the first embodiment described above, in the preparatory step, a first test material formed by providing an omnidirectional artificial flaw in a tube having the same material and cross-sectional dimensions as the flaw-detected material P1. In the first test material flaw detection data generation step, the first test material is flawed with the ultrasonic probe 1 to generate a cross-sectional image of the first test material. Therefore, even if it is a single omnidirectional artificial flaw, a cross-sectional image equivalent to that in the case where a plurality of artificial flaws having different inclination angles are provided and flaw detection can be obtained. Further, in the cross-sectional image of each first test material, the ultrasonic probe 1 is relatively moved along the axial direction and the circumferential direction of the first test material in the same manner as in the case of detecting the flawed material P1, in the first test. It is generated by searching for scratches on the material. Therefore, the position of the flaw pixel region F in each first test material cross-sectional image also changes, and in the omnidirectional artificial flaw data generation step, an omnidirectional artificial flaw generated by synthesizing each first test material cross-sectional image. In the same way as when omnidirectional artificial flaws of the same dimensions are provided at multiple positions in the image for flaw detection, when the positional relationship between the ultrasonic probe 1 and the omnidirectional artificial flaw is changed and flaw detection is performed. The flaw pixel region (composite flaw pixel region FT) is included.

Next, according to the ultrasonic flaw detection method according to the present embodiment, in the correction coefficient calculation step, when the density of each pixel in the synthetic flaw pixel region FT in the omnidirectional artificial flaw image is corrected using the correction coefficient. A correction coefficient is calculated for each pixel in the composite flaw pixel region FT so that the densities of the pixels in the composite flaw pixel region FT are equal to each other. That is, if the density of each pixel in the synthetic flaw pixel region FT is corrected using this correction coefficient, even if the artificial flaw has various tilt angles (omnidirectional artificial flaw), ultrasonic detection can be performed. Regardless of the positional relationship between the child 1 and the omnidirectional artificial flaw, if they are the same omnidirectional artificial flaw, each pixel in the composite flaw pixel region FT will have the same density.

Then, according to the ultrasonic flaw detection method according to the present embodiment, in the flaw detection data generation step, the ultrasonic probe 1 is relatively moved along the axial direction and the circumferential direction of the flaw detectable material P1 to be flawed. By detecting the material P1, a cross-sectional image of the flawed material P1 is generated, and in the correction step, the composite flaw pixel region FT in the cross-section image of the flawed material P1 is supported. The density of each pixel in the correction pixel area to be corrected is corrected by using the correction coefficient of each pixel calculated in the correction coefficient calculation step. Therefore, when the pixel region F corresponding to the tilted flaw exists in the corrected pixel region in the corrected cross-sectional image of the flawed material P1 with respect to the cross section of the flawed material P1, the tilt angle of the tilted flaw does not matter. Regardless of the positional relationship between the ultrasonic probe 1 and the tilted flaw (regardless of the position of the flaw pixel region F in the corrected cross-sectional image of the scratched material), if the tilted flaws have the same dimensions, the same density is obtained. Can be expected to become.
Therefore, in the tilted flaw detection step, when detecting the tilted flaw of the flawed material P1 using the corrected cross-sectional image of the scratched material, the threshold value set in the corrected cross-sectional image of the flawed material is appropriately set. It is possible.

As described above, according to the ultrasonic flaw detection method according to the first embodiment, the cross-sectional image of the flaw detection signal obtained by the ultrasonic probe 1 (the cross-sectional image after correction in the correction step) is used for the shaft of the tube. It is possible to detect tilted flaws extending in a direction tilted with respect to the direction with high accuracy.
In the present embodiment, the case where the scratched material P1 and the first test material are pipes has been described as an example, but the present invention is not limited to this, and the scratched material P1 and the first test material are used. It can also be applied to a bar having a substantially circular cross section.

Further, in the present embodiment, an example is a case where an inclined flaw extending in a direction inclined with respect to the axial direction of the scratched material P1 having a substantially circular cross section is detected (when the reference direction is the axial direction of the scratched material P1). As described above, the present invention is not limited to this, and is a case where an inclined flaw extending in an inclined direction with respect to the radial direction (thickness direction) of the scratched material P1 having a substantially circular cross section is detected (the reference direction is). It can also be applied (when it is in the radial direction of the scratched material P1). However, in this case, the ultrasonic probe 1 is to be detected so that the arrangement direction of the plurality of vibrators 10 included in the ultrasonic probe 1 is along the circumferential direction of the to be detected material P1 and the first test material. It is necessary to detect flaws while facing the outer surface of the material P1 or the first test material. Further, in this case, the omnidirectional artificial flaw provided on the first test material is preferably a hole having a substantially circular cross section extending in the axial direction of the first test material.

Further, although the ultrasonic flaw detector 100 can detect an inclined flaw extending in a direction inclined with respect to the axial direction of the flawed material P1 having a substantially circular cross section, the ultrasonic flaw detector 100 can detect an inclined flaw. There is actually a range at the corner. Therefore, the arrangement direction of the ultrasonic probe 1 (the arrangement direction of the plurality of oscillators 10) is set based on the inclination angle of the inclination flaw to be detected. In the present embodiment, the inclination angle of the inclined flaw to be detected extending in the direction inclined with respect to the axial direction of the scratched material P1 having a substantially circular cross section is arranged along the axial direction of the scratched material P1 having a substantially circular cross section. Although the case where the ultrasonic flaw detector 100 including the ultrasonic probe 1 is within the detectable range has been described, the present invention is not limited to this. That is, the inclination angle of the inclined flaw to be detected extending in the direction inclined with respect to the axial direction of the flawed material P1 having a substantially circular cross section is within the range of the inclination angle of the inclined flaw that can be detected by the ultrasonic flaw detector 100. In some cases, the ultrasonic probe 1 is preferably arranged along the axial direction of the scratched material P1 having a substantially circular cross section. On the other hand, the inclination angle of the inclined flaw to be detected extending in the direction inclined with respect to the axial direction of the flawed material P1 having a substantially circular cross section exceeds the range of the inclination angle of the inclined flaw that can be detected by the ultrasonic flaw detector 100. If so, it is preferable to arrange the ultrasonic probe 1 along the circumferential direction. Specifically, the range of the tilt angle of the tilted flaws that can be detected by the ultrasonic flaw detector 100 is in the range of ± 45 °, and the tilted flaws to be detected are in the axial direction of the flawed material P1 having a substantially circular cross section. On the other hand, when tilted within a range of ± 45 °, the reference direction is the axial direction of the flawed material P1 and the first test material, and the ultrasonic probe 1 is along the axial direction of the flawed material P1 having a substantially circular cross section. Is placed. On the other hand, the range of the tilt angle of the tilted flaw that can be detected by the ultrasonic flaw detector 100 is within ± 45 °, and the tilted flaw to be detected is ± with respect to the axial direction of the flawed material P1 having a substantially circular cross section. When tilting beyond the range of 45 ° (when the tilted flaw tilts within the range of ± 45 ° with respect to the circumferential direction of the flawed material P1 having a substantially circular cross section), the reference direction is the flawed material P1 and the first test. The ultrasonic probe 1 is arranged in the circumferential direction of the material along the circumferential direction of the flawed material P1 having a substantially circular cross section.

Further, the present invention is not limited to the case where the scratched material P1 and the first test material have a substantially circular cross section, and can be applied to the case where the scratched material P1 and the first test material are rectangular plates in a plan view. Is. When the scratched material P1 and the first test material are rectangular plates in a plan view, the inclined flaw extends in a direction inclined with respect to the longitudinal direction of the scratched material P1 (the reference direction is the longitudinal direction of the scratched material P1). In the case), the ultrasonic probe 1 is placed on the flawed material P1 so that the arrangement direction of the plurality of vibrators 10 included in the ultrasonic probe 1 is along the longitudinal direction of the flawed material P1 and the first test material. Or, the flaw is detected in a state where it is arranged facing the outer surface of the first test material. Further, when the inclined flaw extends in the direction inclined with respect to the longitudinal direction of the scratched material P1, the omnidirectional artificial flaw provided in the first test material has a hole having a substantially circular cross section extending in the thickness direction of the first test material. It is preferable to do so.
On the other hand, when the flaw-detected material P1 and the first test material are plate materials having a rectangular shape in a plan view, when the inclined flaw extends in the direction inclined with respect to the thickness direction of the flaw-detected material P1 (the reference direction is the thickness direction of the flaw-detected material P1). In the case of The flaw is detected in a state of being arranged facing the outer surface of the flaw detection material P1 or the first test material. Further, when the inclined flaw extends in the direction inclined with respect to the thickness direction of the scratched material P1, the omnidirectional artificial flaw provided in the first test material has a hole having a substantially circular cross section extending in the longitudinal direction of the first test material. It is preferable to do so.

Hereinafter, the results of a test for confirming the effect of correction by the ultrasonic flaw detection method according to the first embodiment will be described using the flaw-detected material P1 and the plate material S having a rectangular plan view as the first test material.
FIG. 9 is an explanatory diagram illustrating the test conditions and methods. As shown in FIG. 9, a hole having a substantially circular cross section extending from the bottom surface side of the plate material S in the thickness direction was provided, and this was used as an omnidirectional artificial flaw (corresponding to the preparation step of the present embodiment). A plurality of first test material cross-sectional images are generated by detecting this omnidirectional artificial flaw using the ultrasonic probe 1 arranged opposite to the plate material S, and the omnidirectional artificial flaw is synthesized by synthesizing these. An image was generated (corresponding to the first test material flaw detection data generation step and the omnidirectional artificial flaw data generation step of the present embodiment). Next, the correction coefficient was calculated by the equation (1) for each pixel in the composite flaw pixel region in the generated omnidirectional artificial flaw image (corresponding to the correction coefficient calculation step of the present embodiment).

On the other hand, in another portion of the same plate material S, the inclination angle is in the range of 0 to 45 ° with respect to the oscillator arrangement direction (the X direction which is the longitudinal direction of the plate material S in FIG. 9) included in the ultrasonic probe 1. By providing a plurality of inclined flaws (artificial flaws) on the bottom surface side of the plate material S and detecting flaws using the ultrasonic probe 1, a cross-sectional image of the flawed material was generated for each inclined flaw (in the present embodiment). Corresponds to the flaw detection data generation process for the flawed material). Next, the correction coefficient represented by the formula (2) is used for the density of each pixel in the correction pixel region (region corresponding to the composite flaw pixel region) in each of the cross-sectional images of the scratched material, using the calculated correction coefficient of each pixel. A cross-sectional image of the scratched material after correction was generated (corresponding to the correction step of the present embodiment).

Then, the relationship between the maximum density of the pixels in the flaw pixel region corresponding to the tilted flaw and the tilted angle of the tilted flaw was evaluated for each of the cross-sectional images of the scratched material before and after the correction.

FIG. 10 is a diagram showing the results of the above test. The maximum density of the pixels in the flaw pixel region on the vertical axis of FIG. 10 is converted into dB units based on the maximum density when the tilt angle of the tilted flaw is 0 °.
As shown in FIG. 10, according to the ultrasonic flaw detection method according to the present embodiment, by applying the correction, if the tilt flaws have the same dimensions as each other, the tilt flaws in the cross-sectional image of the flawed material are shown regardless of the tilt angle. It was confirmed that the density change in the scratch pixel region corresponding to the above can be suppressed within ± 1 dB.

<Second Embodiment>
Next, the ultrasonic flaw detection method according to the second embodiment will be described focusing on the differences from the ultrasonic flaw detection method according to the first embodiment, and the same points will be basically omitted.
Similar to the ultrasonic flaw detection method according to the first embodiment, the ultrasonic flaw detection method according to the second embodiment also includes an ultrasonic flaw detector 1 for a one-dimensional array type and a control / signal processing means 2. Although it is executed using the device 100, in the ultrasonic flaw detection method according to the second embodiment, the flaw detection signal (signal waveform) itself is used without using the cross-sectional image obtained from the flaw detection signal as in the first embodiment. Detects tilted flaws.
Similar to the ultrasonic flaw detection method according to the first embodiment, the ultrasonic flaw detection method according to the second embodiment also has at least a preparation step, a first test material flaw detection data generation step, and an omnidirectional artificial flaw data generation step. It includes a correction coefficient calculation step, a flaw detection data generation step for the scratched material, a correction step, and a tilt flaw detection step. In the preparatory step, as in the first embodiment, it is preferable to provide a notch simulating an inclined flaw to be detected at another portion of the first test material provided with the omnidirectional artificial flaw. The ultrasonic flaw detection method according to the second embodiment is different from the first embodiment in the specific contents of each step after the preparation step. Specifically, it is as follows.

In the second embodiment, in the first test material flaw detection data generation step, the maximum strength of the flaw detection signal for the cross section of the first test material is generated as the first test material flaw detection data for each predetermined flaw detection condition. As this predetermined flaw detection condition, in the present embodiment, the deflection angle γ of the ultrasonic wavefront when the ultrasonic probe 1 is sector-scanned is used.
Further, also in the second embodiment, it is preferable to generate notch flaw detection data obtained from the flaw detection signal of the notch by detecting the first test material further provided with the notch in the first test material flaw detection data generation step. In the second embodiment, the ultrasonic probe 1 is arranged facing the outer surface of the first test material and relatively moved along the axial direction and the circumferential direction of the first test material to detect the first test material. As the notch flaw detection data obtained from the notch flaw detection signal provided in the first test material, the maximum strength of the flaw detection signal for the cross section of the first test material (maximum strength of the flaw detection signal for the notch) for each predetermined flaw detection condition. To generate. In the first test material flaw detection data generation step, when the maximum strength of the flaw detection signal for the cross section of the first test material (maximum strength of the flaw detection signal for the notch) is generated, the flaw detection sensitivity of the ultrasonic probe 1 is appropriately changed. Then, by comparing the maximum intensities of the flaw detection signals generated at each flaw detection sensitivity, the flaw detection sensitivity suitable for detecting the notch is specified. Alternatively, the flaw detection sensitivity of the ultrasonic probe 1 remains constant to specify a threshold suitable for detecting a notch in the maximum intensity of the generated flaw detector signal.

Further, in the second embodiment, in the omnidirectional artificial flaw data generation step, based on the maximum strength for each predetermined flaw detection condition generated, the omnidirectional artificial flaw data is set to each predetermined flaw detection condition (deflection angle γ). Generates a correspondence with each maximum intensity of the corresponding flaw detection signal for omnidirectional artificial flaws.
Further, in the second embodiment, in the correction coefficient calculation step, when the maximum intensities of the flaw detection signals corresponding to the omnidirectional artificial flaws in the generated correspondence relationship are corrected by using the correction coefficients, the maximum intensities are mutually exclusive. Equivalent correction coefficients are calculated for each flaw detection condition.

Further, in the second embodiment, in the flaw detection data generation step of the flaw detection material, the maximum strength of the flaw detection signal for the cross section of the flaw detection material P1 under a predetermined flaw detection condition (deflection angle γ) is generated as the flaw detection data. do. When the flaw detection sensitivity suitable for detecting the notch is specified in the first test material flaw detection data generation step described above, this is specified as the flaw detection sensitivity of the ultrasonic probe 1 in the flaw detection material generation data generation step. Use flaw detection sensitivity. That is, the flaw-detected material P1 is flaw-detected by the ultrasonic probe 1 having the flaw-detection sensitivity set specified in the first test material flaw-detection data generation step.
Further, in the second embodiment, in the correction step, the predetermined flaw detection condition (deflection angle γ) calculated to the maximum intensity of the flaw detection signal for the cross section of the flawed material P1 under the predetermined flaw detection condition (deflection angle γ). By making a correction using the correction coefficient of, the maximum intensity of the corrected flaw detection signal is generated.
Further, in the second embodiment, in the inclined flaw detection step, the inclined flaw of the flaw-detected material P1 is detected by using the maximum strength of the corrected flaw detection signal. When a threshold value suitable for detecting a notch is specified in the first test material flaw detection data generation step described above, this specified threshold value is used in the tilt flaw detection step.

Then, in the correction coefficient calculation step of the second embodiment, the maximum value of each maximum intensity of the flaw detection signal corresponding to the omnidirectional artificial flaw in the correspondence relationship is set to A, and the omnidirectional artificial flaw in the correspondence relationship is dealt with. Assuming that each maximum intensity of the flaw detection signal is B, the correction coefficient X is calculated for each flaw detection condition (deflection angle γ) based on the following equation (1).
Further, in the correction step of the second embodiment, assuming that the maximum intensity of the flaw detection signal for the cross section of the flawed material P1 under the predetermined flaw detection condition (deflection angle γ) is C, the predetermined flaw detection condition (deflection angle γ). ) Is used to calculate the maximum intensity D of the corrected flaw detection signal based on the following equation (2).
X = 20 ・ log (A / B) ・ ・ ・ (1)
D = C ・ 10 ^{(X / 20)}・ ・ ・ (2)

Hereinafter, the results of a test for confirming the effect of the correction by the ultrasonic flaw detection method according to the second embodiment will be described using the flaw-detected material P1 and the plate material S having a rectangular plan view as the first test material.
First, an example of the contents and results of the first test material flaw detection data generation step in the ultrasonic flaw detection method according to the second embodiment will be described.
FIG. 11 is a diagram showing an example of test conditions and methods for confirming the effect of correction by the ultrasonic flaw detection method according to the second embodiment, and the result of the first test material flaw detection data generation step. FIG. 11A shows the test conditions and methods for confirming the effect of the correction, and FIG. 11B shows an example of the result of the first test material flaw detection data generation step.
As shown in FIG. 11A, a hole having a substantially circular cross section extending from the bottom surface side of the plate material S in the thickness direction was provided, and this was used as an omnidirectional artificial flaw (corresponding to the preparation step of the present embodiment). This omnidirectional artificial flaw is deflected by sequentially detecting flaws by sector scan while changing the deflection angle γ in the range of 0 to 32 ° at a pitch of 2 ° using an ultrasonic probe 1 in which the omnidirectional artificial flaw is arranged facing the plate material S. By generating multiple strengths of the flaw detection signal corresponding to the omnidirectional artificial flaw at each measurement point for each angle γ and comparing these multiple strengths with each other, the maximum strength of the flaw detection signal for each deflection angle γ (first). (Corresponding to the test material flaw detection data) was generated (corresponding to the first test material flaw detection data generation step of the present embodiment). Then, based on these, a correspondence relationship between each deflection angle γ and each maximum intensity of the flaw detection signal corresponding to the omnidirectional artificial flaw was generated (corresponding to the omnidirectional artificial flaw data generation step of the present embodiment).
On the other hand, an inclination angle of 0 to 45 ° with respect to the oscillator arrangement direction (X direction which is the longitudinal direction of the plate material S in FIG. 11A) provided in the ultrasonic probe 1 at another portion of the same plate material S. A plurality of notches inclined at a pitch of 7.5 ° are provided on the bottom surface side of the plate material S in the range of By generating multiple strengths of the flaw detection signal corresponding to the notch at the measurement point and comparing these multiple strengths with each other, the maximum strength of the flaw detection signal (corresponding to the notch flaw detection data) for each deflection angle γ was generated (this book). Corresponds to the first test material flaw detection data generation step of the embodiment).

The data plotted by “◯” in FIG. 11B shows the maximum strength of the flaw detection signal corresponding to the omnidirectional artificial flaw when flaw detection is performed at each deflection angle γ in the first test material flaw detection data generation step. In other words, the data plotted with "○" shows the correspondence between each deflection angle γ generated by the omnidirectional artificial flaw data generation process and each maximum intensity of the flaw detection signal corresponding to the omnidirectional artificial flaw. ing. Further, the data plotted by "◇" indicates the maximum intensity of the flaw detection signal corresponding to each notch when flaw detection is performed at each deflection angle γ in the first test material flaw detection data generation step. The numerical value described near the maximum intensity of the flaw detection signal corresponding to each notch represents the inclination angle θ of each notch. The maximum intensity of the flaw detection signal, which is the vertical axis of FIG. 11B, is based on the maximum intensity of the flaw detection signal (0 dB) when a notch having an inclination angle θ = 0 ° is flawed at a deflection angle γ = 0 °. Converted to dB unit.
As shown in FIG. 11B, the maximum intensity of the flaw detection signal changes according to the deflection angle γ of the sector scan between the case of detecting a notch simulating a tilted flaw and the case of detecting an omnidirectional artificial flaw. However, if the flaw detection sensitivities of the ultrasonic probe 1 are the same, it was found that the maximum intensity of the flaw detection signal has a large difference of about 18 dB. Therefore, as described above, in the first test material flaw detection data generation step, it is necessary to specify the flaw detection sensitivity suitable for detecting the notch or the threshold value suitable for detecting the notch. From the results shown in FIG. 11B, the flaw detection sensitivity suitable for detecting a notch is about 18 dB smaller than the flaw detection sensitivity suitable for detecting an omnidirectional artificial flaw. Further, the threshold value suitable for detecting the notch is about 18 dB higher than the threshold value suitable for detecting the omnidirectional artificial flaw. Then, the flaw detection sensitivity specified in the first test material flaw detection data generation step is used as the flaw detection sensitivity of the ultrasonic probe 1 in the flaw detection data generation step, or the first test material flaw detection data is used in the tilt flaw detection step. It is necessary to use the threshold value specified in the generation process. In the example shown below, the flaw detection sensitivity specified in the first test material flaw detection data generation step is used as the flaw detection sensitivity of the ultrasonic probe 1 in the flaw detection data generation step.

In this test, the correction coefficient was calculated by the equation (1) for each deflection angle γ using the correspondence relationship generated as described above (data plotted by “◯” in FIG. 11 (b)) (the correction coefficient was calculated for each deflection angle γ). Corresponds to the correction coefficient calculation process of this embodiment).
Further, a plurality of notches provided in the first test material in the preparatory step and used in the first test material flaw detection data generation step are considered to be inclined flaws existing in the flaw detected material P1 (plate material S), and in FIG. 11 (b), " The data plotted in "◇" was also used as the maximum intensity of the flaw detection signal for each deflection angle γ for the flaw detectable material P1 (corresponding to the flaw detection data generation step of the present embodiment).
Next, the maximum intensity of the flaw detection signal at the predetermined deflection angle γ is corrected by the equation (2) using the calculated correction coefficient of the predetermined deflection angle γ, whereby the corrected flaw detection signal is applied. Generated the maximum intensity of (corresponding to the correction step of this embodiment).

FIG. 12 is a diagram showing the results of the above test. FIG. 12 shows the values before and after the correction of the maximum intensity of the flaw detection signal of each inclined flaw (notch) when flaw detection is performed at each deflection angle γ. As shown in FIG. 12, according to the ultrasonic flaw detection method according to the present embodiment, the correction is performed when there is a difference in maximum intensity of 4.4 dB at the maximum depending on the inclination angle θ before the correction. It was confirmed that the difference in maximum strength of 2.0 dB or less can be suppressed regardless of the inclination angle θ if the inclination flaws have the same dimensions.

In the ultrasonic flaw detection method according to the first and second embodiments, an example in which a one-dimensional array type ultrasonic probe 1 including a plurality of vibrators 10 arranged in a row is used as the ultrasonic probe. However, the present invention is not limited to this. For example, it is also possible to use a two-dimensional array type ultrasonic probe having a plurality of oscillators arranged in a matrix. It is also possible to use a plurality of ultrasonic probes having a single oscillator arranged one-dimensionally or two-dimensionally.

1 ... One-dimensional array type ultrasonic probe 2 ... Control / signal processing means 10 ... Oscillator 100 ... Ultrasonic flaw detector P1 ... flaw detector P2 ... First test material

Claims (3)

The ultrasonic probe is placed facing the outer surface of the flawed material having a substantially circular cross section, and the ultrasonic probe is relatively moved along the axial and circumferential directions of the flawed material to obtain the flawed material. It is an ultrasonic flaw detection method that detects tilted flaws extending in a direction tilted with respect to the axial direction.
A preparatory step for preparing a first test material provided with an omnidirectional artificial flaw composed of a hole having a substantially circular cross section extending in the radial direction in a material having the same material and cross-sectional dimensions as the scratched material.
The ultrasonic probe is placed facing the outer surface of the first test material, and the ultrasonic probe is relatively moved along the axial direction and the circumferential direction of the first test material to move the first test material. A first test material cross- sectional image generation step of generating a plurality of first test material cross- sectional images obtained from a flaw detection signal for the first test material by detecting a flaw,
An omnidirectional artificial flaw image generation that generates an omnidirectional artificial flaw image including a synthetic flaw pixel region which is a pixel region corresponding to the omnidirectional artificial flaw by synthesizing the plurality of first test material cross-sectional images. Process and
A correction coefficient calculation step of calculating a correction coefficient based on the following equation (1) for each pixel in the composite flaw pixel region, and a correction coefficient calculation step.
A process of generating a cross-sectional image of the material to be detected, which generates a cross- sectional image of the material to be detected obtained from a flaw detection signal of the material to be detected, by detecting the material to be detected.
A correction step of generating a corrected cross- sectional image of the scratched material by applying a correction based on the following equation (2) to the corrected pixel region located at the same coordinates as the synthetic flaw pixel region in the cross-sectional image of the scratched material. When,
An ultrasonic flaw detection method comprising a tilted flaw detecting step of detecting an inclined flaw of the flawed material using the corrected cross- sectional image of the flawed material.
X = 20 ・ log (A / B) ・ ・ ・ (1)
D = C ・ 10 ^{(X / 20)} ・ ・ ・ (2)
In the above equations (1) and (2), X is the correction coefficient. In the above formula (1), A is the maximum density of the pixels in the synthetic flaw pixel region, and B is the density of each pixel in the synthetic flaw pixel region. In the above equation (2), C is the density of each pixel in the correction pixel region before correction, and D is the density of each pixel in the correction pixel region after correction. In the preparatory step, the first test material is provided with a notch having an inclination angle within the range of the maximum inclination angle of the inclination flaw to be detected and simulating the inclination flaw.
In the first test material cross- sectional image generation step, by detecting the first test material provided with the notch , a notch cross- sectional image obtained from the flaw detection signal of the notch is generated.
Based on the notch cross- sectional image , a threshold for setting the flaw detection sensitivity of the ultrasonic probe in the flaw-detected material cross-section image generation step or detecting the tilted flaw in the tilted flaw detecting step. The ultrasonic flaw detection method according to claim 1 , wherein a value is set. In the preparatory step, a material having the same material and cross-sectional dimensions as the scratched material has an tilt angle within the range of the maximum tilt angle of the tilted flaw to be detected of the scratched material, and the tilted flaw is simulated. Prepare the second test material with the notch ,
The ultrasonic probe is placed facing the outer surface of the second test material, and the ultrasonic probe is relatively moved along the axial direction and the circumferential direction of the second test material to move the second test material. Further including a second test material flaw detection data generation step of generating a second test material flaw detection data obtained from the flaw detection signal for the second test material by flaw detection.
Based on the second test material flaw detection data, the flaw detection sensitivity of the ultrasonic probe in the flaw detected material cross- sectional image generation step is set, or the tilt flaw is detected in the tilt flaw detection step. The ultrasonic flaw detection method according to claim 1 , wherein a threshold value is set.

Images