JP5565904B2 - Method for identifying surface shape of ultrasonic testing specimen, identification program, aperture synthesis processing program, and phased array testing program - Google Patents

Description

  The present invention relates to a method for identifying a surface shape of a specimen to which an ultrasonic flaw detection method is applied, an identification program, an aperture synthesis processing program, and a phased array flaw detection program. More specifically, the present invention relates to a method for identifying a surface shape of a test specimen suitable for ultrasonic flaw detection using a phased array, an identification program, an aperture synthesis processing program, and a phased array flaw detection program.

  In the flaw detection of the pipe welded portion, since the mechanical scanning of the ultrasonic probe is limited by the welding surplus, flaw detection in the surface area of the shape change portion becomes difficult. When the access of the probe is poor as described above, a one-time reflection method is effective in which an ultrasonic beam is reflected on the inner back surface of the pipe for flaw detection. However, in this single reflection method, the flaw detection accuracy is lower than that in the direct irradiation method due to the influence of ultrasonic diffusion and attenuation, ultrasonic scattering due to the back surface shape, and change in propagation direction.

  On the other hand, high-chromium steel is used in high-temperature and high-pressure piping in thermal power generation facilities, and it has been confirmed that type IV damage due to creep occurs in the heat-affected zone fine-grained area of weld joints of the same material. This type IV damage is an important issue in ensuring the structural integrity of thermal power generation facilities. Type IV damage is a form of damage caused by creep that occurs in the heat-affected zone fine-grained area from inside the wall, occurs along the heat-affected zone fine-grained area, and proceeds in the surface layer area below the weld toe There is. For this reason, as shown in FIG. 15, in the flaw detection in the surface layer area below the weld toe, the mechanical scanning of the probe is limited due to the approach limit, and the incident ultrasonic wave is almost all depending on the inclination of the type IV damage. It may not return. Further, at the stage where creep voids before the occurrence of type IV cracks begin to be dense, detection is difficult because the signal intensity obtained by ultrasonic flaw detection is weak.

  For this reason, there is a demand for the development of a phased array method and an aperture synthesis method that can detect flaws by a direct method from a shape change portion such as a surplus. In particular, at the stage where creep voids before Type IV cracks start to gather, detection is difficult because the signal intensity obtained by ultrasonic flaw detection is weak, but signal amplification, azimuth resolution, and SN ratio are improved. Improvement of detection performance can be expected by applying a possible aperture synthesis method.

  Conventionally, as a technique for performing ultrasonic flaw detection by direct irradiation method from above the shape change part, a local water that uses a medium that can change its form along the surface shape of the specimen, for example, a medium such as water or gel, instead of a wedge. The immersion method is common. However, the phased array method and the aperture synthesis method are premised on being applied to a specimen having a flat surface shape (Patent Documents 1 to 3, Non-Patent Documents 1 to 3). Therefore, even if flaw detection by the phased array method is simply applied to this local water immersion method, etc., in the delay control assuming a flat specimen surface shape, the ultrasonic beam does not converge due to the extra shape, or the set direction May not propagate to The same applies to the case where the aperture synthesis method is simply applied. Therefore, prior to the ultrasonic test, the surface shape of the test specimen was grasped in advance by mechanical measurement using a general mold as a surface shape measurement technique, laser displacement measurement, etc. It is necessary to calculate the propagation path of the ultrasonic wave, control the delay time of each transducer, and match the phase of the received waveform.

  In addition, recently, there has been proposed one in which the transducer groups of the phased array probe are supported one by one so as to be vertically movable and are always urged toward the surface of the specimen (non-patent document). 4). According to this phased array probe, each of the transducers is moved along the shape changing portion, and the excitation of each transducer is controlled with delay while mechanically measuring the surface shape with this movable amount. Thus, the ultrasonic beam can be focused from above the complicated shape portion.

JP-A-8-201569 JP-A-10-142201 JP 2008-122209 A

H. Karasawa, M. Izumi, T. Suzuki, S. Nagai, M. Tamura and S. Fujimori, “Development of under sodium 3D visual inspection technique by using matrix-arrayed ultrasonic transducer”, Journal of Nuclear Science and Technology, 37 (2000), pp. 769-779. Motohisa Abe and Hirokazu Kurosawa “Portable Type 3D Ultrasound Inspection System Matrixey” Toshiba Review Vol.60 No.4, 48-51, 2005 Karasawa Hirokazu, Isobe Hideo, Hamashima Takayuki "3D Aperture Synthesis (3D-SAFT) Arrays and 9 Applications" Non-destructive inspection Vol.56 No.10, 520-524, 2007 S. Chatillon, G. Cattiaux, M. Serre and O. Roy, “Ultrasonic non-destructive testing of pieces of complex geometry with flexible phased array transducer”, Ultrasonics, Vol. 38, pp. 131-134, 2007.

  However, prior to the ultrasonic test, in order to grasp the surface shape in advance, when performing mechanical measurement using a mold, laser displacement measurement, etc., inspect after installing other measurement equipment of the flaw detector Therefore, there is a problem that a lot of inspection time is required and the cost is increased because other measuring instruments for the flaw detector are required.

  Further, in the method proposed in Non-Patent Document 4, since the vibrator itself is brought into contact by being individually displaced along the surface shape of the test body, the transverse wave cannot be incident by Snell's law, There is a problem that it is limited to longitudinal wave oblique angle flaw detection. Since shear waves are generally superior in resolution compared to longitudinal waves, it is possible to detect flaws with weak signal strength obtained by ultrasonic flaw detection, for example, the stage where creep voids before the occurrence of type IV cracks begin to crowd. It is desirable to be able to perform shear wave oblique angle flaw detection.

  Therefore, the present invention provides a method for identifying the surface shape of an ultrasonic flaw detection test body and an identification program that make it possible to grasp the surface shape of the test body using reflected waves from the surface when ultrasonic waves are incident on the test body. Another object is to provide an aperture synthesis processing program and a phased array flaw detection program.

  As a result of various studies conducted by the present inventors in order to achieve such an object, it is possible to reproduce the surface of the specimen by specifying the coordinates at which reflection occurs on the surface of the specimen due to the spread of the beam. As a result, it has been found that the surface shape can be identified by using the reflected wave from the surface when the sample is incident on the specimen. The present invention has been made on the basis of such knowledge, and the surface shape is measured using a reflected wave from the surface when ultrasonic waves are incident on the test body.

That is, the method for identifying the surface shape of a test body according to the first aspect of the present invention is a medium that can change the shape along the surface shape of the test body on the surface of the test body having a shape changing portion whose surface shape has changed. A phased array is arranged via each of the phased arrays, ultrasonic waves are emitted toward the test body for each transducer of the phased array, and a surface echo is received from the surface of the test body acquired for each transducer. Adjacent transducers are assumed by detecting a reflected wave and calculating a beam path from each transducer to the surface of the test body, and assuming a circle with a radius of the beam path determined for each transducer centered on each transducer. A common tangent line of the circle centered at the center, and there are sections where the common tangent line is obtained and sections where the common tangent line is not found. Identify this as the surface of the specimen In the section where the common circumscribing line is not found, the boundary between these vibrators is determined as the boundary of shape change, and the curve obtained by quadratic interpolation in each section is extrapolated and identified as the surface shape at the boundary, and the test The body surface shape is obtained.

According to a second aspect of the present invention, there is provided a program for identifying a surface shape of a test specimen, wherein the medium can change a shape along the surface shape of the test specimen on the surface of the test specimen having a shape changing portion whose surface shape has changed. A step of receiving a surface echo by emitting an ultrasonic wave toward the test body for each transducer of the phased array disposed via the antenna, and a reflected wave from the surface of the test body acquired for each transducer Detecting a beam path from each transducer to the surface of the specimen, and assuming a circle having a radius of the beam path centered on each transducer and having a radius determined for each transducer, determining a common outer tangent of the circle centered, the common outer tangent not obtained and determined period interval is present, the a common outer tangent is obtained interval a point on previous SL common outer tangent test what was secondary interpolation In the step of identifying the surface and the interval where the common tangent is not found, the shape between the transducers is determined as the boundary of shape change, and the points on the common tangent obtained in each interval are subjected to quadratic interpolation. The step of identifying the curve created by the insertion as the surface shape at the boundary is executed by a computer to obtain the surface shape of the specimen.

According to a third aspect of the present invention, there is provided an aperture composition processing program, on a surface of a test body having a shape changing portion whose surface shape is changed, via a medium whose form can be changed along the surface shape of the test body. A step of receiving a surface echo by emitting an ultrasonic wave toward the test body for each transducer of the arranged phased array, and detecting a reflected wave from the surface of the test body obtained for each transducer. Assuming a step of obtaining a beam path from each transducer to the surface of the specimen, and a circle centered on each transducer and having a radius of the beam path obtained for each transducer, and centering on adjacent transducers determining a common outer tangent of the circle, and identifying with the common outer tangent at is obtained interval before Symbol common outer point on the tangential secondary interpolated ones specimen surface, wherein the common outer tangent determined In the unending section Identifying the curve between the transducers as a shape change boundary, and identifying a curve created by extrapolating the points on the common tangent line obtained in each section as a surface shape at the boundary; Calculating the propagation distance of ultrasonic waves from any section of the flaw detection range inside the test body to the vibrator using the surface shape of the test specimen, and the ultrasonic wave of each vibrator Aperture synthesis processing for shifting the phase of the waveform so as to correct the delay of the A scope waveform signal received by each receiving element based on the propagation path, and superimposing the waves in the same phase, and after the phase operation And causing the computer to execute a step of drawing the flaw detection image of the image on the display.

Furthermore, the phased array flaw detection program according to the invention of claim 4 is provided on a surface of a test body having a shape changing portion whose surface shape is changed via a medium whose form can be changed along the surface shape of the test body. A step of receiving a surface echo by emitting an ultrasonic wave toward the test body for each transducer of the arranged phased array, and detecting a reflected wave from the surface of the test body obtained for each transducer. Assuming a step of obtaining a beam path from each transducer to the surface of the specimen, and a circle centered on each transducer and having a radius of the beam path obtained for each transducer, and centering on adjacent transducers determining a common outer tangent of the circle, and identifying with the common outer tangent at is obtained interval before Symbol common outer point on the tangential secondary interpolated ones specimen surface, wherein the common outer tangent determined Don't stop Determining the boundary between the vibrators as a shape change boundary, and identifying a curve created by extrapolating a quadratic interpolation of points on the common circumscribing line obtained in each section as a surface shape at the boundary; and Calculating the ultrasonic propagation distance from any section of the flaw detection range inside the specimen to all transducers using the identified surface shape of the specimen, and the ultrasonic propagation path for each transducer The computer executes a step of calculating a delay time of each transducer based on the phase, a step of performing phase array flaw detection by controlling the transducer based on the delay time, and a step of drawing a flaw detection image on a display I try to let them.

  According to the surface shape identification method and identification program for an ultrasonic flaw detector of the present invention, the surface shape is calculated by calculation by a central processing unit using an ultrasonic beam transmitted and received for each transducer of the flaw detector. Therefore, since the time required for the inspection hardly changes and the measurement can be performed using the phased array, no separate measuring device or instrument is required, and the equipment cost is hardly required.

  Further, according to the aperture synthesis processing program of the present invention, the surface shape is calculated by the calculation by the central processing unit using the ultrasonic beam transmitted and received for each transducer of the flaw detector, and the surface shape data is used. An aperture that shifts the phase of the waveform so as to correct the delay of the A scope waveform signal received by each receiving element based on the propagation path of the ultrasonic wave for each transducer to be received, and superimposes the wave phases in an aligned state. Since the synthesizing process can be executed, a weak echo can be detected and a weak echo can be detected while maintaining a high azimuth resolution by performing weak signal intensity amplification, azimuth resolution and SN ratio improvement over the entire flaw detection area inside the specimen. Therefore, it is possible to detect flaws in the surface area of the shape change portion such as welding surplus. In addition, an ultrasonic beam having a large spread that is not focused is made incident, so that the lower side of the weld toe as shown in FIG. Even when the type IV damage in the surface layer of the surface layer, that is, the propagation direction of ultrasonic waves and the type IV damage are parallel to each other, it is possible to detect defects at the stage where creep voids before the occurrence of type IV cracks begin to condense. I can expect.

  Further, according to the phased array flaw detection program of the present invention, the surface shape is calculated by the calculation by the central processing unit using the ultrasonic beam transmitted and received for each transducer of the flaw detector, and the surface shape data is used. The delay time of each transducer can be calculated by calculating the propagation distance of the ultrasonic waves from any section of the flaw detection range inside the specimen to all transducers. Can be controlled to perform phased array flaw detection. Therefore, since the transmission and reception are repeated so that the ultrasonic beam is successively focused in accordance with the surface shape obtained in advance in the entire flaw detection area, flaw detection can be performed with high accuracy. This phased array flaw detection also enables flaw detection equivalent to the above-described aperture synthesis process.

It is a principle figure of the identification method of the surface shape of the ultrasonic testing specimen concerning this invention. It is a principle figure at the time of carrying out an aperture synthetic process using the surface shape data of the test body identified by the identification method of the surface shape of an ultrasonic flaw detection test body. It is explanatory drawing which shows the state of the ultrasonic beam reflected from the test body surface, (a) shows a normal flat surface, (b) shows the reflection by an inclined surface. It is explanatory drawing which shows the determination method of the test body surface shape by a common tangent, (a) is a figure which shows the relationship between a test body and a vibrator | oscillator, (b) is a figure which shows the common external tangent between vibrator | oscillators. It is explanatory drawing which shows the approximation of the surface shape by a quadratic function. It is explanatory drawing which shows the determination method of the path | route between two points. It is explanatory drawing which shows the determination method of the path | route between a probe position and a converging point. It is explanatory drawing which shows the calculation method of an aperture synthetic method. It is the top view and side view of the test body which has a horizontal hole in the complicated shape part used for verification of the effectiveness of the surface shape identification method, (a) The test body which simulated the shape of butt welding, (b) Shape discontinuity The test body which simulated the shape of the part welding part is shown. It is a surface shape measurement model, (a) shows the vertical flaw detection in the butt welding model, (b) shows the oblique flaw detection in the butt welding model, and (c) shows the oblique flaw detection in the shape discontinuity welding model. It is a graph which shows the measurement result of the test body surface shape shown to (a)-(c) of FIG. 10, (a) is the vertical flaw detection in a butt welding model, (b) is an oblique flaw detection in a butt welding model, (c ) Shows the results of oblique flaw detection in the shape discontinuity welding model. It is a B scope image in a butt welding model, (a) shows a phased array method, (b) shows what is based on the aperture synthesis method of the present invention. It is a B scope image in a shape discontinuity part welding model, (a) echo from a horizontal hole by a transverse wave component and a longitudinal wave component by a phased array method, (b) is a horizontal hole by a transverse wave component by the aperture synthesis method of the present invention. (C) shows the echo from a horizontal hole by a longitudinal wave component by the aperture synthesis method of the present invention. It is explanatory drawing which shows refraction | bending of the ultrasonic beam in a shape change part. It is explanatory drawing which shows the flaw detection of a welding surplus shape change part surface layer area. It is a flowchart which shows the whole procedure of opening synthetic | combination processing. It is a flowchart which shows the procedure of surface shape identification. It is a flowchart which shows an opening synthetic | combination process procedure.

  Hereinafter, the configuration of the present invention will be described in detail based on embodiments.

  A part of the ultrasonic beam 4 transmitted from the probe 1 propagates inside the test body 2, and a part is reflected by the surface of the test body 2. Usually, the ultrasonic probe is designed so that echoes from the surface of the specimen are absorbed in the wedge and not detected by the probe, and only echoes from inside the specimen are detected. . The method for identifying the surface shape of a specimen having a shape changing portion according to the present invention identifies the surface shape by actively using echoes from the surface of the specimen.

  Here, the ultrasonic beam 4 is detected by the probe 1 as a reflected wave perpendicular to the surface of the specimen 2. Therefore, in the case of a spreading ultrasonic wave, as shown in FIG. 1, when the surface of the test body 2 and the surface of the vibrator 1 are parallel to each other, it is directly below the vibrator perpendicular to both the test body surface and the vibrator surface. An echo from the surface of the test body is detected, and an echo perpendicular to the surface of the test body is detected when the surface of the test body and the transducer surface are not parallel. However, it cannot be determined from the detection signal alone whether the ultrasonic wave is reflected on the surface of the test body immediately below the transducer or is reflected at a point away from directly below the transducer. For this reason, it is presumed that what is reflected by the inclined surface is reflected by the point A 'in FIG. 1, and an error occurs between the point A and the point A on the surface of the test body 2 directly under the vibrator. For this reason, even if the shape is specified by simply connecting the beam path derived from the surface echo of the specimen, an error occurs in the surface shape.

  That is, as shown in FIG. 3A, when the propagation direction of the ultrasonic beam and the normal direction of the specimen surface are parallel, the beam up to the reception echo of the surface reflected wave can be calculated by calculating the beam path as the actual distance. The path length is the distance from the transducer surface to the specimen surface. However, when the beam propagation direction and the normal direction of the specimen surface are tilted as shown in Fig. 3 (b), the reflected wave is received by the wave incident perpendicularly to the surface (indicated by the solid arrow in the figure). If the distance between the transducer surface and the specimen surface is calculated using this, an error is generated by Δy in the figure. When the surface of the test specimen is flat, it is easy to correct Δy because the incident angle is uniform, but it is difficult to correct the test specimen whose surface shape changes midway. Therefore, it was considered to reconstruct the surface of the specimen by specifying the coordinates at which surface reflection occurs due to beam expansion.

  As shown in FIG. 4, if it is assumed that the surface of the test specimen from which the reflected wave is obtained is a straight line between adjacent transducers, a circle is formed with the radius of the beam path r where the echo from the surface appears, centering on the transducer. Considering, the tangent to the circle drawn by the adjacent vibrator is the surface of the specimen. In the present invention, the contact point between the tangent line and the circle is considered as the coordinate of the surface of the specimen. In this method, a reflected wave can be obtained from the surface of the specimen having a large inclination by using a beam having a large spread. However, in the adjacent transducers A and B shown in FIG. If the beam path difference Δr where the echo appears is large, a common tangent cannot be created. This occurs when there is a significant change in the shape of the surface to be measured between the vibrators. For example, it occurs when the shape of the molten metal rises from the base metal, such as a surplus in butt welding. Therefore, in the present invention, it is considered that the gradient of the shape change becomes discontinuous at the point where such a common tangent cannot be created. And it assumed that the surplus shape of butt-welding can be approximated by a quadratic function, was divided at the boundary, and the surface coordinates in each section were subjected to quadratic interpolation.

Hereinafter, the procedure of the surface shape identification method will be described more specifically based on FIG. 16 and FIG.
First, the phased array 1 is arranged on the surface of the test body 2 having the shape changing portion whose surface shape has changed via the medium 3 whose form can be changed along the surface shape of the test body. At this time, the array probe 1 is fixed by the jig 5 or the like, and the distance from the surface of the specimen is maintained. Here, the medium 3 interposed between the phased array 1 of the flaw detector and the test body 2 is preferably a liquid such as water or a semi-solid body such as gel. When water is used as a medium, a phased array is used in a flaw detection method generally called a local water immersion method.

  Next, the ultrasonic flaw detector detects the amplitude resulting from the surface reflection of the specimen at each probe position by linear electronic scanning using one element (S1). That is, surface echoes are acquired by emitting ultrasonic waves toward the test body for each transducer of the phased array. For example, when a 64-element phased array is used, transmission / reception of one element is repeated in order from 1 element to 64 elements, and a surface echo for each transducer is acquired.

Next, the beam path length until the amplitude due to the surface reflection acquired for each transducer appears, that is, the beam path length from each transducer to the surface of the test body where the reflection occurs is obtained ( _S3-1 ).

Then, assuming a circle whose radius is the beam path obtained for each transducer centering on each transducer, a common outer tangent of the circles centering on adjacent transducers is obtained by calculation (S _3-2 ). The common tangent of two circles can be calculated if the radius (beam path) of the two circles and the center distance (transducer spacing) of the two circles are determined, and the common tangent is already calculated. Since the equation is derived and known, the description is omitted.

Next, it is determined whether or not a common tangent line is obtained in the section between the two vibrators ( _S3-3 ). When the common tangent line is obtained, a process for obtaining a common tangent line for all vibrators is further performed. It is determined whether or not (S _3-5 ). That is, a boundary where a common tangent cannot be created is searched while sequentially creating a common outer tangent. When the common circumscribing line cannot be obtained, it is determined that the transducer section is a boundary and a flag j is set or 1 is added to the flag j to identify the transducer section as a boundary ( _S3-4 ). Jump to step _S3-5 . When the process for obtaining the common tangent line is not added for all the vibrators, 1 is added to the vibrator flag i (S _3-6 ), the process jumps to step S _3-2 , and the next Steps S _{3-2 to} S _3-5 are repeated for the vibrator. And when the process which calculates | requires a common circumscribed line about all the vibrators is completed, it moves to step _S3-7 .

Next, in step _S3-7 , the initial value is k = 1, the group of points on the common tangent line in the sections k and k + 1 determined to be the boundary is subjected to quadratic interpolation, and these quadratic interpolated ones are tested. Identify as surface. Then, for all the sections it is determined as a boundary to determine whether the secondary interpolation (S3 _-8), adds 1 to the flag k in step S3 _-9 If not completed, step S3 _- Jump to ₇ , and quadratic interpolation is performed on the group of points on the common tangent line in the corresponding section.

Next, a surface shape is identified by extrapolating the quadratic curves of each group to the intersection of each other. Corresponding quadratic curves between the intersections become surface coordinates between the intersections (x coordinates) ( _S3-10 ). As for the surface shape data of the test body, for example, considering the xy plane with the position of the first transducer as the origin, the direction in which the transducers are arranged is the x-axis direction, the beam path direction is the y-axis direction, and secondary the curve is stored in memory as ^{y = ax 2 + bx + c} (S3 -11). Specifically, the coefficients a, b, and c of two quadratic curves in a group of sections in which two common circumscribed lines with the boundary as a boundary are obtained are stored in the memory. If necessary, an image is displayed on the display of the PC, or used for aperture synthesis processing or phased array flaw detection. Of course, the surface shape data of the specimen may be temporarily stored in a work memory or the like and used for the aperture synthesis processing in the next process or the arithmetic processing for the phased array flaw detection.

  The above calculation and identification of the surface shape and the like are executed by a personal computer by a program stored in a storage medium such as a ROM of a personal computer (PC). Here, the ultrasonic flaw detector and the PC process the surface echo data acquired by the flaw detector into the PC and then process it offline in the PC and display the image on the PC display. It is not particularly limited.

If the surface shape of the specimen is identified by the above procedure, the path until the ultrasonic beam radiated from the vibrator reaches the point in the specimen can be calculated using the surface shape data.
For example, a propagation path can be calculated according to Fermat's principle that a wave travels a path that takes the minimum time to propagate between two points. Now, as shown in FIG. 6, an orthogonal coordinate system xy is defined, and the media 1 and 2 are divided by a boundary line described by y = ax ² + bx + c. At this time, considering the propagation path between the point P (x _t , 0) of the medium 1 and the point Q (x _t + x ₁ + x ₂ , y ₁ + y ₂ ) of the medium 2, the following geometrically: The formula holds.
y ₁ tan θ ₁ + (y−y ₁ ) tan θ ₂ = x−x _t (3-1)
C ₂ sin (θ ₁ + θ ₀ ) = C ₁ sin (θ ₂ + θ ₀ ) (3-2)
tanθ ₀ = 2a (x _t + x ₁ ) + b (3-3)
y ₁ = a (x _t + x _i ) ² + b (x _t + x ₁ ) + c (3-4)
Here, C ₁ and C ₂ are sound speeds of the media 1 and 2. Since the above equation is non-linear, it is necessary to determine the angle by successive approximation or the like. Differentiating and organizing the formula gives the following formula.
(3-5)
By applying successive approximation to the above equation so that δx = 0, θ ₀ , θ ₁ , θ ₂ , y ₁ can be calculated. In this method, the surface of the specimen is approximated by a plurality of quadratic curves, and there are a plurality of paths. Therefore, it is necessary to determine a curve that gives an initial value when performing sequential calculation.

In the following, a method for determining an ultrasonic beam propagation path including how to assign an initial value will be described. The xy orthogonal coordinate system shown in FIG. 7 is defined.
1) Divide the surface of the test body into n pieces, and let the division point be S _i (i = 1, n).
2) starting from the arbitrary point Q of the test body to the point reaching through the points S _i on the specimen surface according to Fermat's principle flaw detection surface and _{X i (i = 1, n} ).
3) The points X _i and X _{i + 1} sandwiching the transducer P are searched, and the distances in the x-axis direction between P and the points X _i and X _{i + 1} are set to Δx _i and Δx _{i + 1} , respectively.
4) Considering the point R on the surface of the specimen, the distances in the x-axis direction between R and the points S _i and S _{i + 1} are respectively Δ
When s _i and Δs _{i + 1} are set, an x coordinate satisfying Δs _i : Δs _{i + 1} = Δx _i : Δx _{i + 1} is set as an x coordinate of the point R.
5) Using the point R as the initial value, determine the coordinates that satisfy Fermat's principle from the sequential calculation of Equation (3-5).
6) Check if the line segments PR and QR intersect the surface approximation curve other than the surface where R exists, and determine the propagation path.

Then, if the above procedure is executed by a computer and the propagation path of the ultrasonic wave incident on the shape changing portion is determined, the amplitude of an arbitrary point can be calculated by the following method. That is, the aperture synthesis process can be performed on a test body having a shape changing portion whose surface shape has changed.
As shown in FIG. 8, the amplitude value E (x _i , y _i ) of the pixel Q (x _i , y _i ) of interest in the B scope image was obtained at the transducer position P (x _k , 0). Using the A scope waveform S (x _k , t), the following equation is used.
(3-6)
here,
c ₁ and c ₂ : sound speeds of the media 1 and 2 N: total number of elements r ₁ (x _i , y _i , x _k ): ultrasonic waves from the transducer position P (x _k , 0) to the pixel Q (x _i , The distance r ₂ (x _i , y _i ,) from P (x _k , 0) to a point on the boundary between the medium 1 and the medium 2 on the path to reach y _i ) so as to satisfy Fermat's principle x _k ): a distance from a point on the boundary to the pixel Q (x _i , y _i ) in the path.

Specifically, the aperture synthesis processing program causes the following procedure shown in FIG. 16 to be executed in the central processing unit of each of the ultrasonic transmitter and the PC connected thereto.
1) A linear scan using one element for transmission / reception is executed by an ultrasonic flaw detector to acquire surface reflection data for each transducer (S1).
2) The coordinates of the surface shape of the specimen are determined from the received surface reflected wave by computer processing (S3). Specifically, the steps (S1, S3) of the above 1) and 2) are performed based on the surface echo data (S1) of the specimen taken from the ultrasonic flaw detector as shown in FIG. by executing the identification processing _(S3 -1 _{~S3 -12)} computer to calculate the surface shape data of the specimen represented by quadratic curve of ^{y = ax 2 + bx + c} .
3) Further, the inside of the test body is detected with an ultrasonic flaw detector, and a received waveform is acquired (S2). For example, a linear scan in which a plurality of vibration elements are simultaneously transmitted is performed to acquire internal echo data of a flaw detection area for flaw detection inside the specimen.
4) Then, from the surface shape coordinates determined in step (S3) of the above 2), a propagation path of the ultrasonic wave passing through the surface R from the element P and reaching the point Q in the test body is determined (S4). That is, the flaw detection range inside the specimen is divided into meshes, and the propagation distance of ultrasonic waves from each mesh to each vibrator to be received is calculated based on the identified specimen surface shape.
5) The amplitude of the received waveform corresponding to the path determined in 4) above is added to the amplitude of the Q point (S5 _-1 ).
6) Move the Q point of the test body while repeating the above 4 and 5 _(S5 -2). In other words, the phase of the waveform is shifted so as to correct the delay of the A scope waveform signal received by each receiving element based on the ultrasonic wave propagation path for each transducer for all sections of the flaw detection range. Aperture synthesis processing is performed in which the phases are aligned (S5).
In this embodiment, the surface echo data and the internal echo data are continuously taken into the PC, and then the surface shape is calculated using the surface echo for each transducer in the off-line PC (personal computer). Then, using this identified surface shape data, find the propagation path of the ultrasonic wave to any point inside the specimen for each transducer that receives the internal echo, and according to the propagation path The phase of the waveform is shifted so as to correct the delay of the A scope waveform signal received for each transducer, and aperture synthesis processing is performed in which the phases of the waves are aligned. Then, the flaw detection image generated by the aperture synthesis process is displayed on the display of the PC (S6).

  In the present embodiment, both the surface echo and the internal echo are acquired in steps 1 and 2, but the surface echo is used in step 1 and the echo inside the specimen is used in step 2. Therefore, if the echo from the internal crack can be received in the echo inside the specimen by transmission by one element in step 1, the surface shape of the specimen is identified by the echo data acquired in one scan, Therefore, the data acquisition in step 2 is not necessary. That is, in this embodiment, the linear scan using one vibration element for identifying the surface shape of the test specimen and the linear scan for simultaneously transmitting a plurality of vibration elements for flaw detection inside the test specimen are separate steps. However, the present invention is not particularly limited to this, and may be processed in one step as necessary. For example, when using a large vibrator, by transmitting from a single vibrator, it is possible to receive an echo from the surface of the specimen and an echo large enough for flaw detection from within the specimen, Both echoes can be acquired simultaneously by only a linear scan using one vibration element. However, since ultrasonic waves with only one element with a small transducer size are weak in energy and may not receive echoes large enough for flaw detection, the number of simultaneously transmitted elements must be set to collect echoes inside the specimen. It is preferable to increase the energy of ultrasonic waves. In some cases, a linear scan using one vibration element for identifying the surface shape of the test object may be executed after the linear scan for simultaneously transmitting a plurality of vibration elements for flaw detection inside the test object. good. In any case, when the surface echo acquired by the ultrasonic flaw detector and the echo data inside the specimen are once stored in the memory of the PC and then processed by the central processing unit of the offline PC. The data acquisition context is irrelevant.

The effectiveness of the identification method and identification program of the surface shape of the ultrasonic testing specimen of the present invention was verified.
A test specimen simulating the shape of the butt weld and the discontinuous part weld shown in Fig. 9 was produced, and a number of 1 mm diameter horizontal holes that assumed type IV damage were introduced at positions that assumed a fine grain region of the weld heat affected zone. did. Here, the material of the test body is SUS304. Then, water immersion testing was performed using a phased array probe having a frequency of 5 MHz, 64 elements, an opening area of each element of 0.6 mm × 10 mm, and a pitch of 0.7 mm. At this time, if the flaw detection surface is parallel, a jig capable of maintaining an angle at which a transverse wave is incident at a refraction angle of 45 degrees is attached to the probe. The jig is one channel of the array probe and has a height of 8 mm. For data acquisition, a three-dimensional aperture synthetic array inspection device (Matrixeye EX (registered trademark) manufactured by Toshiba Electric Power Systems Co., Ltd.) was used, and electronic scanning was linear scanning using one element for transmission and reception. Matrixeye performs aperture synthesis processing at high speed from the A scope waveform received by each element using the parallel arithmetic circuit and four A / D converters that incorporate the amplitude values of the pixels in the B scope image, and generates the B scope image in real time. Can be displayed.

  First, the specimen surface shape was measured by the method for identifying the surface shape of the specimen according to the present invention. Regarding the surface shape measurement, as shown in FIG. 10, vertical flaw detection was also performed using a jig capable of holding the flaw detection surface and a height of 20 mm. In FIG. 11, Method 1 (Method 1) is used to calculate the surface coordinates of the test specimen based on the beam path of the reflected echo from the surface, and Method 2 (Method 2) is used as the surface coordinates reconstructed by the method of the present invention. The surface coordinates of are also shown as a reference. 11 (a) and 11 (b), there is an error between the reference and method 2 in the portion surrounded by the dotted line. In this part, since the refraction angle of the ultrasonic main beam exceeds the critical angle and does not enter the specimen, it is considered that it does not contribute much to the result after the aperture synthesis method. Further, the portions other than the portion surrounded by the dotted line in FIG. 11 are in good agreement with the actual surface shape, and it has been found that the surface shape can be reproduced by the method of the present invention.

  Next, flaw detection results are shown. In electronic scanning, seven elements were used for simultaneous transmission, and 15 element groups having the same center as the transmitting element group were set for reception. As a comparison, an ultrasonic phased array flaw detector (Omniscan) was used to perform a transverse wave oblique water immersion flaw detection that does not take into account changes in the ultrasonic propagation path due to surface shape changes. Electronic scanning was linear scanning, the number of simultaneous transmitting and receiving elements was 16, and delay control was set so that the beam converged near the side hole when a flat plate was assumed. B scope images obtained by flaw detection of the butt welding simulated test specimen and the shape discontinuity weld simulated specimen are shown in FIGS. 12 and 13, respectively. FIGS. 12A and 13A are the results obtained by the phased array method without considering the shape change, and FIG. 12B is the flaw detection result to which the aperture synthesis method according to the present invention is applied. The vertical axis indicates the beam path length. Here, the beam path length is calculated based on the sound speed of water. In FIG. 12, T indicates an echo from a lateral hole caused by a transverse wave. Further, in the result of FIG. 12B, correction is performed by post-processing so that the actual coordinates are displayed in the test body so as to facilitate the analysis. The focusing of the echo by the aperture synthesis method is similar to the flaw detection result of the phased array method focused by using 16 elements, but the echo positional relationship is clearer in the result of the aperture synthesis method.

  In FIG. 13, L indicates an echo from a horizontal hole due to a longitudinal wave component. As shown in FIG. 14, in the ultrasonic beam radiated from the vibrator, the transverse wave that has undergone mode conversion after being incident on the specimen surface A reaches the horizontal hole, and after incident on the surface B, the longitudinal wave reaches the horizontal hole. In FIG. 13A showing the flaw detection result of the phased array method, T and L are displayed together. In FIG. 13B in which the aperture synthesis method processing is performed based on the transverse wave, it is surrounded by a broken line. L becomes thinner and only T is focused. Further, in FIG. 13C in which the aperture synthesis method processing is performed on the basis of the longitudinal wave, T disappears and only L is focused. In addition, for L, instructions from three horizontal holes are obtained, whereas for T, only two are obtained. This is because, as shown in FIG. 14, the transverse wave reaching the lateral hole near the surface has a smaller amplitude than the transverse wave reaching the other lateral hole, so that the obtained echo is also smaller than the other lateral holes and is hardly visible. Conceivable. As described above, by applying the aperture synthesis method in consideration of the propagation path of the ultrasonic beam due to the shape change, it is possible to remove the mixture of the longitudinal wave component and the transverse wave component of the echo and facilitate the flaw detection.

  As a result, it was verified that the surface shape of the test specimen can be identified using only the received waveform of the surface reflected wave. In addition, in the conventional phased array method, echoes due to both longitudinal and transverse modes were mixed due to the influence of the shape change part and analysis after flaw detection was difficult, but based on the identified specimen surface shape, According to the aperture synthesis method of the present invention that calculates the propagation path of the ultrasonic beam and controls the phase of the received waveform so as to be focused on all points in the test body, it is possible to amplify the signal intensity, improve the azimuth resolution, and the SN ratio. In addition, the longitudinal wave mode and the transverse wave mode can be selectively displayed, so that the identification accuracy of the reflection source position is improved.

  The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited thereto, and various modifications can be made without departing from the scope of the present invention. For example, the present invention has mainly been described with respect to aperture synthesis processing using a phased array. However, the present invention is not particularly limited to this. Identification of the surface shape of a specimen and determination of the propagation path of ultrasonic waves taking the surface shape into consideration. The steps up to here are common in setting the delay time for each transducer in the phased array flaw detection, and therefore can be easily applied to the phased array flaw detection.

  That is, when the surface shape of the test body having a shape change portion such as a weld surplus is specified by the above-described surface shape identification method of the test body, the flaw detection range inside the test body is meshed from the surface shape of the test body. Since the ultrasonic propagation distance to all transducers can be calculated for each mesh, the delay time of each transducer can be obtained based on the ultrasonic propagation path for each transducer. it can. Therefore, phase array flaw detection can be performed by inputting the delay time for each vibrator calculated by the computer to the flaw detector and controlling the transmission of each vibrator. As a result, the ultrasonic waves are focused at an arbitrary position in the flaw detection area inside the specimen by transmitting the ultrasonic waves with a unique delay time for every transducer. Since the ultrasonic waves transmitted from all the transducers gather at the focal point, the flaw detection accuracy increases. Similarly, since the transmission and reception are repeated so that the ultrasonic beam is sequentially focused in accordance with the surface shape obtained in advance in the entire flaw detection area, flaw detection can be performed with high accuracy. Then, a flaw detection image of the entire flaw detection area is depicted on the display.

Specifically, the phased array flaw detection program causes the following procedure to be executed in the central processing unit of each of the ultrasonic transmitter and the PC connected online thereto.
1) A linear scan using one element for transmission / reception is executed by an ultrasonic flaw detector to acquire surface reflection data for each transducer (S1).
2) Further, the inside of the specimen is detected by linear electronic scanning using one or several elements with an ultrasonic flaw detector, and a received waveform is acquired (S2). The surface echo data and the internal echo data acquired in steps 1 and 2 are taken into the PC, respectively.
3) Based on the acquired surface echo data, the coordinates of the surface shape of the specimen are determined by calculation processing on the PC (S3).
4) Then, from the surface shape coordinates determined in step (S3), the propagation path of the ultrasonic wave passing through the surface R from the element P and reaching the point Q in the test body is determined (S4).
5) Next, based on the propagation path of the ultrasonic wave calculated in step 4, the delay time during which the ultrasonic waves of all the transducers are focused at an arbitrary position in the flaw detection area inside the test body is calculated (S5 ′). ). Here, for example, UltraVision 3 (registered trademark) manufactured by Zetec Corporation in the United States is known as PC software that implements a function of calculating delay control of a phased array in accordance with the shape change portion. Therefore, if the surface shape of the specimen can be identified, the delay control of the phased array can be easily calculated with existing software.
6) The delay time for each transducer calculated by the central processing unit of the PC is input to the ultrasonic flaw detector to perform phased array flaw detection (S6 ′).
7) The flaw detection image is displayed on the display of the PC (S7 ′).

Claims (4)

A phased array is arranged on the surface of a test body having a shape change portion whose surface shape has changed via a medium whose shape can be changed along the surface shape of the test body, and each phased array has a super A beam path from each transducer to the surface of the specimen is detected by emitting a sound wave toward the specimen and receiving a surface echo and detecting a reflected wave from the surface of the specimen obtained for each transducer. Assuming a circle centered on each transducer and having the radius of the beam path obtained for each transducer, a common tangent of the circle centered on adjacent transducers is obtained, and the common tangent is determined. In the section where there is a section and an unobtained section and the common circumscribed line is obtained, a section obtained by quadratic interpolation of points on the common circumscribed line is identified as the specimen surface, and the common outer tangent is not obtained Then, the shape change between these vibrators The surface of the specimen having a shape change portion characterized by determining a boundary and extrapolating a curve obtained by quadratic interpolation in each section to identify the surface shape at the boundary and obtaining the specimen surface shape Shape identification method. Ultrasonic wave is applied to each of the transducers of the phased array arranged on the surface of the test body having the shape changing portion whose surface shape has been changed via a medium whose shape can be changed along the surface shape of the test body. And receiving a surface echo and detecting a reflected wave from the surface of the specimen obtained for each transducer to determine a beam path from each transducer to the surface of the specimen. Assuming a circle centered on each transducer and having a radius that is the beam path obtained for each transducer, obtaining a common tangent of the circle centered on adjacent transducers, and the common tangent the steps and interval not obtained and determined interval exists, to identify with the common outer tangent at is obtained interval before Symbol common outer point on the tangential secondary interpolated ones specimen surface, the common outer tangent In the section where is not found Determining a boundary between the transducers as a shape change boundary, and extrapolating a quadratic interpolation of points on the common circumscribing line obtained in each section to identify the surface shape at the boundary as a computer. A surface shape identification program for a test body having a shape changing portion, wherein the surface shape of the test body is obtained. Ultrasonic wave is applied to each of the transducers of the phased array arranged on the surface of the test body having the shape changing portion whose surface shape has been changed via a medium whose shape can be changed along the surface shape of the test body. And receiving a surface echo and detecting a reflected wave from the surface of the specimen obtained for each transducer to obtain a beam path length from each transducer to the surface of the specimen. Assuming a circle centered on each transducer and having a radius that is the beam path obtained for each transducer, obtaining a common tangent of the circle centered on adjacent transducers, and the common tangent and step calculated interval to identify the specimen surface obtained by secondary interpolation points on the leading SL common outer tangent, said common outer determine between them vibrators and the boundary of the shape change in the tangent line Motomara not interval And common obtained in each section A step of identifying a curve created by extrapolating a quadratic interpolation of points on a tangent line as a surface shape at the boundary, and an arbitrary section of a flaw detection range inside the specimen using the identified specimen surface shape Calculating the propagation distance of ultrasonic waves from each transducer to each transducer, and correcting the delay of the A scope waveform signal received by each receiving element based on the ultrasonic propagation path for each transducer An aperture composition processing program for causing a computer to execute an aperture composition process for shifting the phase of the waveform so that the phases of the waves are aligned and for causing the display to draw a flaw detection image after the phase operation. Ultrasonic wave is applied to each of the transducers of the phased array arranged on the surface of the test body having the shape changing portion whose surface shape has been changed via a medium whose shape can be changed along the surface shape of the test body. And receiving a surface echo and detecting a reflected wave from the surface of the specimen obtained for each transducer to obtain a beam path length from each transducer to the surface of the specimen. Assuming a circle centered on each transducer and having a radius that is the beam path obtained for each transducer, obtaining a common tangent of the circle centered on adjacent transducers, and the common tangent and step calculated interval to identify the specimen surface obtained by secondary interpolation points on the leading SL common outer tangent, said common outer determine between them vibrators and the boundary of the shape change in the tangent line Motomara not interval And common obtained in each section A step of identifying a curve created by extrapolating a quadratic interpolation of points on a tangent line as a surface shape at the boundary, and an arbitrary section of a flaw detection range inside the specimen using the identified specimen surface shape Calculating a propagation distance of ultrasonic waves from all the vibrators to each transducer, calculating a delay time of each transducer based on an ultrasonic propagation path for each transducer, and based on the delay time, A phased array flaw detection program for causing a computer to execute a phase array flaw detection by controlling a vibrator and a step of drawing a flaw detection image on a display.