JP2020085526A - Method for evaluating quality of weld zone - Google Patents

Abstract

To provide a method for evaluating the quality of a weld zone with which it is possible to obtain the throat thickness of a weld zone with good accuracy.SOLUTION: A method for evaluating the quality of a weld zone pertaining to the present invention evaluates the quality of a weld zone 100 in lap fillet welding carried out on a first plate 10 and a second plate 20 in such a way that a bead 11 is formed on the first plate 10 side. This method comprises the steps of: acquiring the image of cross section of the weld zone 100 that intersects the longitudinal direction of the bead 11 at right angles; binarizing the acquired cross section image; removing a nose component from the binarized image; extracting an end point Pof the first plate 10 before welding and a contour Lof surface of the bead 11 from the image having had the noise component removed; and calculating a throat thickness Lof the weld zone 100 using the length of a perpendicular dropped from a pre-welding end point Pof the first plate 10 before welding toward the contour Lof surface of the bead 11.SELECTED DRAWING: Figure 3

Description

The present invention relates to a weld quality evaluation method, and more particularly to a weld quality evaluation method in overlap fillet welding.

Generally, the quality of the weld is evaluated by the penetration depth, which indicates how deeply the metal has melted during welding, the leg length, which indicates the width of the weld, and the throat thickness, which indicates the thickness of the bead formed. be able to. Specifically, it is possible to determine whether or not the welding has been successfully performed depending on whether or not the penetration depth, the leg length, and the throat thickness length satisfy predetermined criteria.

Patent Document 1 discloses a method of obtaining a penetration depth by oscillating ultrasonic waves toward a welded portion and creating a sectional distribution map based on the reflected wave. The quality of welding can be judged by the penetration depth obtained by the method.

JP, 2002-214207, A

The cross-sectional distribution map created based on the reflected waves of ultrasonic waves and the image obtained by capturing the actual cross-section may contain a lot of noise. Therefore, as described in Patent Document 1, when the shape of the welded portion was obtained based on the image of the cross section, the length such as the penetration depth could not be obtained accurately. Further, the method described in Patent Document 1 has a problem that the penetration depth cannot be obtained while the penetration depth can be obtained.

The present invention has been made in view of the above circumstances, and provides a method for evaluating the quality of a welded portion, which can accurately obtain the throat thickness of the welded portion.

The quality evaluation method of a welded part according to the present invention is a quality of a welded part in a lap fillet welding performed by forming a bead on the first plate material side of a first plate material and a second plate material. In the evaluation method, a step of acquiring an image of a cross section of the welded portion orthogonal to the longitudinal direction of the bead, a step of binarizing the acquired image of the cross section, and the binarization processing A step of removing a noise component from the image; a step of extracting an end point of the first plate material before welding from the image from which the noise component has been removed; and a surface of the bead from the image from which the noise component has been removed From the end point before the welding of the first plate material before the welding, using the length of the vertical line lowered toward the contour of the surface of the bead, the throat thickness of the welded portion. It is characterized in that it comprises the steps of obtaining.

In the above configuration, since the binarization process and the noise removal process are performed on the cross-sectional image of the welded portion, the evaluation error caused by the noise can be reduced. Further, since the end points of the plate material and the contour of the bead surface can be extracted, the throat thickness of the welded portion can be obtained based on these distances. Therefore, in the welded part quality evaluation method according to the present invention, the value of each evaluation item such as the throat thickness can be accurately obtained.

ADVANTAGE OF THE INVENTION According to this invention, the quality evaluation method of a welded part which can obtain|require the throat thickness of a welded part accurately can be provided.

It is an image which picturized a welding part. It is a schematic diagram of the quality evaluation system which performs the quality evaluation method of the welding part concerning a 1st embodiment. It is the whole flowchart of the quality evaluation method of the welding part concerning a 1st embodiment. It is an image (cross-sectional image) of a cross section of the welded portion shown in FIG. 1. 5 is an image (binarized image) obtained by binarizing the cross-sectional image of FIG. 4. It is a detailed flowchart of step S30 of FIG. 6 is a Fourier spectrum obtained by Fourier transforming the binarized image of FIG. 5. 8 is an image (masking image) obtained by masking the Fourier spectrum of FIG. 7. 9 is an image obtained by inverse Fourier transforming the masking image of FIG. 8 (Fourier inverse transform image). 10 is an image (morphology image) obtained by performing morphology processing on the inverse Fourier transform image of FIG. 9. It is a detailed flowchart of step S40 of FIG. It is a figure which shows a mode that the distribution of the pixel value of a row direction is extracted for every column from the morphology image of FIG. 13 is a graph of pixel value distribution in the row direction extracted for each column in FIG. 12. It is the image (basic information image) which extracted the basic information of the characteristic of a welding part from the morphology image of FIG. It is an image (evaluation information image) in which information used for quality evaluation is extracted from the basic information image of FIG. 14. It is a detailed flowchart of step S50 of FIG. It is a figure for demonstrating the method of calculating|requiring values, such as throat thickness, in step S50. It is the schematic of the quality evaluation system which performs the quality evaluation method of the welding part which concerns on 2nd Embodiment. 3 is an image of a cross section of the welded portion shown in FIG. 1 acquired using an ultrasonic imaging apparatus.

Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings. However, the present invention is not limited to the following embodiments. Further, the following description and drawings are simplified as appropriate for the sake of clarity. Further, the plurality of embodiments described below can be implemented independently or can be implemented in combination as appropriate. These embodiments have novel features that are different from each other. Therefore, these plurality of embodiments contribute to solving different purposes or problems, and contribute to achieving different effects.

[First Embodiment]
First, an outline of a quality evaluation method for a welded portion according to the first embodiment will be described with reference to FIGS. 1 to 3. FIG. 1 is an image of the welded portion 100. FIG. 2 is a schematic diagram of a quality evaluation system S1 that executes the quality evaluation method for a welded portion according to the first embodiment. FIG. 3 is an overall flowchart of the quality evaluation method for a welded portion according to the first embodiment.

The quality evaluation method for a welded portion according to the first embodiment uses a first plate material 10 and a second plate material 20 as shown in FIG. 1 so that a bead 11 is formed on the first plate material 10 side. It is a method for evaluating the quality of the welded portion 100 in the overlap fillet welding performed by the above method.
The first plate member 10 and the second plate member 20 are made of, for example, a metal or alloy such as iron, aluminum, or copper, and are formed so as to extend from the front side to the back side in FIG. 1. The first plate member 10 and the second plate member 20 are integrated by welding the end faces of the first plate member 10 by lap fillet welding in a state where their respective end portions are overlapped. At this time, the bead 11 is formed along the end surface of the first plate member 10. Therefore, the bead 11 is formed with the longitudinal direction extending from the front side to the back side in FIG. Further, at this time, a part of the first plate material 10 is melted into the second plate material 20 side, and the melted portion 12 is formed.

The quality evaluation method for a welded portion according to the first embodiment can be executed using a quality evaluation system S1 as shown in FIG. 2, for example. As shown in FIG. 2, the quality evaluation system S1 includes a microscope 1, a cable 2, a processing device 3, and a monitor 4.

The microscope 1 images the cross section of the welded portion 100 between the first plate member 10 and the second plate member 20. The microscope 1 is communicatively connected to the processing device 3 via a cable 2, and the imaged image of the cross section is transmitted to the processing device 3. The processing device 3 includes, for example, a CPU (Central Processing Unit), and executes a process of a quality evaluation method for a welded part, which will be described later, by a preset program to evaluate the quality of the welded part 100. The monitor 4 displays the calculation result of the processing device 3 and the image acquired or generated by the processing device 3.
In the example of FIG. 2, the microscope 1 and the processing device 3 are connected by the cable 2, but the microscope 1 and the processing device 3 may be wirelessly connected so as to be communicable. ..

FIG. 3 is an overall flowchart of the quality evaluation method for a welded portion according to the first embodiment. As shown in FIG. 3, the welding part quality evaluation method according to the first embodiment includes steps S10 to S60. The subsequent operations of steps S10 to S60 may all be performed by the processing device 3 (see FIG. 2), or may be performed by the user who uses the processing device 3 if possible.

First, in step S10, the processing device 3 acquires an image of a cross section of the welded portion 100 orthogonal to the longitudinal direction of the bead 11. Specifically, for example, as described above, the image of the cross section of the welded portion 100 captured by the microscope 1 is acquired via the cable 2.
Next, in step S20, the processing device 3 performs binarization processing on the acquired image of the cross section.
Next, in step S30, the processing device 3 performs a process of removing a noise component from the binarized image.
Next, in step S40, the processing device 3 extracts basic information on the characteristics of the welded portion 100 from the image from which the noise component has been removed. For example, the processing device 3 extracts the end point of the first plate material 10 before welding and the contour of the surface of the bead 11.
Next, in step S50, the processing device 3 uses the extracted information to determine the value of each evaluation item in the welded portion 100. For example, the processing device 3 obtains the throat thickness of the welded portion 100 by using the length of the perpendicular line drawn from the end point of the first plate material 10 before welding obtained in step S40 toward the contour of the surface of the bead 11. .
Finally, in step S60, the processing device 3 determines the quality of the welded portion 100 based on the magnitude of the value of each evaluation item obtained in step S50.
As described above, the quality of the welded part 100 can be evaluated.

Hereinafter, details of each step will be described with reference to FIGS. 4 to 17. FIG. 4 is an image (cross-sectional image 101) of the cross section of the welded portion 100 shown in FIG. 1 orthogonal to the longitudinal direction of the bead 11.
For convenience of explanation, the top and bottom of the cross-sectional image 101 in FIG. 4 and each image shown in FIG. 5 and subsequent figures are displayed by reversing the top and bottom of the image of the welded portion 100 shown in FIG.

First, in step S10, the microscope 1 captures the cross-sectional image 101 of the welded portion 100 and transmits it to the processing device 3. In this way, the processing device 3 acquires the cross-sectional image 101.
In addition, when the microscope 1 captures the cross-sectional image 101, it is preferable that the first plate material 10 and the second plate material 20 have different brightness values, for example, by irradiating light from an oblique direction. For example, in the cross-sectional image 101 in FIG. 4, the second plate member 20 is imaged so as to be brighter than the first plate member 10. By imaging in this manner, the first plate material 10 and the second plate material 20 can be easily discriminated, and the quality evaluation of the welded portion 100 can be performed more accurately.

In addition, it is preferable that the cross-sectional image 101 be captured so that the boundary between the first plate material 10 and the second plate material 20 (that is, the boundary of the melted portion 12) and the contour of the surface of the bead 11 are bright. For example, in the cross-sectional image 101 of FIG. 4, the boundary of the melted portion 12 and the contour of the surface of the bead 11 are imaged so as to be brighter than other regions. By picking up images in this way, the boundary shape of the welded portion 12 and the contour shape of the surface of the bead 11 can be easily determined, and the quality of the welded portion 100 can be evaluated with even higher accuracy.

When proceeding to step S20, the processing device 3 performs a binarization process on the cross-sectional image 101, which divides the cross-sectional image 101 into two colors using a predetermined pixel value as a threshold. The pixel value used as the threshold value at this time is a value for dividing the first plate member 10 and the second plate member 20. The pixel value used as the threshold value may be determined based on a calibration experiment performed in advance, or may be appropriately determined by the user's judgment.

FIG. 5 is a binarized image 102 obtained by dividing a high lightness region into white and a low lightness region into black in the cross-sectional image 101. In the binarized image 102, the area corresponding to the second plate member 20 is mainly displayed in white. This is because the second plate member 20 is imaged brighter than the first plate member 10 in the cross-sectional image 101 of FIG. In the binarized image 102, the boundary between the first plate material 10 and the second plate material 20 (the boundary of the melted portion 12) and the contour of the surface of the bead 11 are also displayed in white. This is because in the cross-sectional image 101 of FIG. 4, the boundary corresponding to the melted-in portion 12 and the portion corresponding to the surface of the bead 11 are imaged relatively brightly.

Next, details of step S30 will be described with reference to FIGS. FIG. 6 is a detailed flowchart of step S30. As shown in FIG. 6, step S30 includes steps S31 to S34.

First, in step S31, the processing device 3 Fourier transforms the binarized image 102 to generate a Fourier spectrum 103 as shown in FIG. At this time, the processing device 3 may perform a DFT (Discrete Fourier Transform) process or a FFT (Fast Fourier Transform) process as a Fourier transform process.

The Fourier spectrum 103 is a spectrum obtained by plotting the horizontal frequency and the vertical frequency of the binarized image 102. In the Fourier spectrum 103, the low frequency component of the binarized image 102 is displayed in the center.
In the Fourier spectrum 103, the two white lines intersecting at the center are reference lines shown for convenience of description, and are not included in the actual Fourier spectrum of the binarized image 102.

In step S32, the processing device 3 performs the masking process for removing the high frequency component in the Fourier spectrum 103 to generate the masking image 104 as shown in FIG.
After that, the process proceeds to step S33, and the processing device 3 performs the inverse Fourier transform on the masking image 104 to generate the inverse Fourier transform image 105 as shown in FIG.
When the processing proceeds to step S34, the processing device 3 performs a morphological process in which expansion and contraction of pixels of the inverse Fourier transform image 105 are combined several times to generate a morphology image 106 as shown in FIG.

The morphological image 106 obtained as described above is one in which the high frequency components of the binarized image 102 are removed in steps S31 to S33, and the discontinuous pixel components are removed in step S34. Therefore, it can be said that the morphology image 106 is an image in which the noise component of the binarized image 102 is removed.
The threshold value of the frequency to be removed in the masking process in step S32 and the number of expansions and contractions in the morphology process in step S34 can be appropriately adjusted according to the user's judgment.

Next, details of step S40 will be described with reference to FIGS. 11 to 15. FIG. 11 is a detailed flowchart of step S40. As shown in FIG. 11, step S40 includes steps S41 to S48.
Of these, in steps S41 to S44, the processing device 3 extracts the boundary portion between the white region and the black region in the morphology image 106 and extracts the basic information of the welded portion 100 (basic image shown in FIG. 14). The information image 107) is generated.
Further, in steps S45 to S48, the processing device 3 generates an image (evaluation information image 108 shown in FIG. 15) in which information used for quality evaluation is further extracted based on the basic information image 107. As described above, the processing device 3 extracts the basic information of the characteristics of the welded portion 100 from the morphological image 106.

First, in step S41, the processing device 3 scans the morphological image 106 in the column direction as shown in FIG. 12, and obtains the row-direction distribution of pixel values in each column. FIG. 13 is a graph showing the distribution in the row direction of pixel values of each column of the morphology image 106 obtained in FIG. 13A is a row-direction distribution of pixel values in the column AA of FIG. 12, FIG. 13B is a row-direction distribution of pixel values in the column BB of FIG. 12, and FIG. ) Indicates the distribution of pixel values in the row direction in the column C-C in FIG. 12, respectively.
In FIGS. 13A to 13C, the black portion has a pixel value of 0, and the white portion has a pixel value of 1.

Further, the processing unit 3 in step S41, as shown in FIG. 13 (A) ~ (C) , a point where the pixel value changes, extracted from the top point _P a, as _{P b.} Further, as illustrated in FIG. 13C, the processing device 3 extracts a point in a region where the pixel value is 1 as a point P _c .

Next, in step S42, the processing device 3 obtains an approximate straight line for the set of points P _a extracted in each row and extracts it as the end face a of the second plate member 20 (see FIG. 14). In addition, when the horizontal direction of the cross-sectional image 101 is parallel to the surface of the second plate member 20, the processing device 3 obtains the average coordinate in the vertical direction of the points P _a extracted in each row, and the horizontal direction at the average coordinate. The line of the direction may be the end surface a.
Further, the processing device 3 extracts a line that is translated from the end face a by the thickness T of the second plate member 20 as the other end face d of the second plate member 20 (see FIG. 14 ). In this way, the contour of the second plate member 20 is obtained.

When the processing proceeds to step S43, the processing device 3 extracts the set of points P _b extracted in each column as a boundary b where the pixel value changes in the vicinity of the merged portion 12 (see FIG. 14). Further, in step S44, the processing device 3 extracts the set of points P _c extracted in each column as the surface c of the bead 11 (see FIG. 14). In this way, the basic information image 107 shown in FIG. 14 can be obtained.

When proceeding to step S45, the processing device 3 determines the penetration center P ₁ which is the deepest position of the penetration part 12. For example, specifically, the processing device 3 approximates the shape of the boundary b in the basic information image 107 by an upwardly convex curve using the cross-correlation method. Then, the approximated peak position of the curve is determined as the penetration center P ₁ (see FIG. 15 ).

When the processing proceeds to step S46, the processing device 3 determines the contour L _b of the welded portion 12 from the shape of the boundary b. For example, specifically, the processing device 3 differentiates the shape of the boundary b to extract only the upwardly convex portion, and determines the upwardly convex portion as the contour L _b of the welded portion 12 (see FIG. 15 ). ).

When the processing proceeds to step S47, the processing device 3 obtains an approximate straight line with respect to the set of the surfaces c of the beads 11 and determines it as the contour L _c of the surfaces of the beads 11 (see FIG. 15). The contour L _c does not necessarily have to be approximated by a straight line, but may be determined by approximation by a function of a curved shape.

Proceeding to step S48, the processing device 3, the penetration portion 12 of the end points of the contour L _b, the end point distance of the long side of the outline L _c of the surface of the bead 11, before welding the first plate 10 It is determined as the end point P ₂ .
As described above, the evaluation information image 108 shown in FIG. 15 can be obtained.

Next, details of step S50 will be described with reference to FIGS. FIG. 16 is a detailed flowchart of step S50. As shown in FIG. 16, step S50 includes steps S51 to S53. FIG. 17 is a diagram for explaining a method of obtaining the value of each evaluation item such as throat thickness in step S50.

First, in step S51, the processing apparatus 3 obtains the distance from the end surface d of the second plate member 20 to the penetration center P ₁ and determines the distance as the penetration depth L ₁ (see FIG. 17).
Further, in step S52, the processing device 3 obtains the length of the perpendicular line drawn from the end point P ₂ of the first plate material 10 before welding to the contour L _c of the surface of the bead 11 and the length thereof. The thickness is determined as the throat thickness L ₂ (see FIG. 17).
Then, determination in step S53, the processing unit 3, of the outline L _b of the penetration portion 12, a width in the region larger than a predetermined value is the distance between the end face d of the second plate 20 as a leg length L ₃ (See FIG. 17).

After obtaining the value of throat thickness _{L 2,} etc. of the welded portion 100 by the above steps S51 to S53, processing unit 3 in step S60, the penetration depth _{L 1,} throat thickness _{L 2,} leg length _{L 3} is a predetermined value It is determined whether or not the above. When the penetration depth L ₁ , the throat thickness L ₂ , and the leg length L ₃ are equal to or more than the predetermined values, it is determined that the welded portion 100 is well formed. On the other hand, if any of the penetration depth L ₁ , the throat thickness L ₂ , and the leg length L ₃ is less than the predetermined value, it is determined that the welded portion 100 was not formed well.

As described above, in the quality evaluation method for the welded portion 100 according to the present embodiment, since the binarization process and the noise removal process are performed on the cross-sectional image 101 of the welded part 100, the evaluation error caused by noise is reduced. can do. Further, since the end point P ₂ of the first plate member 10 and the contour L _c of the surface of the bead 11 can be extracted, the throat thickness L ₂ of the welded portion 100 can be obtained based on these distances. Therefore, in the quality evaluation method for a welded portion according to the present invention, the throat thickness L _{2 and the} like can be accurately obtained.

[Second Embodiment]
Next, the outline of the quality evaluation method for the welded portion according to the second embodiment will be described with reference to FIGS. 18 and 19. FIG. 18 is a schematic diagram of a quality evaluation system S2 that executes the weld part quality evaluation method according to the second embodiment. The quality evaluation system S2 is different from the quality evaluation system S1 according to the first embodiment in that an ultrasonic imaging device 5 is provided instead of the microscope 1.

The ultrasonic imaging device 5 is a device that oscillates ultrasonic waves on an object and creates a cross-sectional distribution map based on the reflected waves. FIG. 19 is an image (cross-sectional image 201) of the cross section of the welded portion 100, which is imaged using the apparatus. In FIG. 19, the upper bright portion corresponds to the second plate member 20, and the lower dark portion corresponds to the first plate member 10. Further, of the darkly shown portions, the upwardly convex portion corresponds to the melt-in portion 12, and the brightly colored portion below the first plate member 10 is the surface portion of the bead 11. Corresponding to.

In the present embodiment, instead of the cross-sectional image 101 in the first embodiment, the cross-sectional image 201 described above is used, and the quality of the welded portion 100 is evaluated through steps S10 to S60 similar to those in the first embodiment. .. In such a configuration, the image of the cross section of the welded portion 100 can be acquired nondestructively, so that nondestructive inspection is possible.

It should be noted that the present invention is not limited to the above embodiment, and can be modified as appropriate without departing from the spirit of the present invention.
For example, although the case where each step of steps S31 to S34 is used has been described as an example of the method of removing the noise of the cross-sectional image 101 in step S30, the method of processing the noise of the binarized image 102 is not limited thereto. is not. Other known methods may be appropriately used or combined with the above-described method as long as the method can remove noise from the binarized image 102 and improve the accuracy of extraction in step S40.

Further, although the case where each step of steps S41 to S48 is used as an example of the method of extracting each information of the welded portion 100 from the morphological image 106 in step S40 has been described, the method of extracting each information is not limited to these. is not. For example, a known method such as labeling processing may be appropriately used or may be combined with the above method.
Furthermore, the execution procedure of steps S41 to S48 and steps S51 to S53 is not limited to the order described above. If the penetration depth L ₁ , the throat thickness L ₂ , and the leg length L ₃ can be determined, the respective steps may be appropriately interchanged, or may be performed simultaneously.

1 Microscope 2 Cable 3 Processor 4 Monitor 5 Ultrasonic Imaging Device 10 First Plate Material 11 Bead 12 Penetration Part 20 Second Plate Material 100 Weld Part 101 Cross Section Image 102 Binarized Image 103 Fourier Spectrum 104 Masking Image 105 Inverse Fourier Transform Image 106 Morphology image 107 Basic information image 108 Evaluation information image 201 Cross-sectional image L ₁ Penetration depth L ₂ Throat thickness L ₃ Leg length L _b Contour of welded portion L _c Contour of bead surface P ₁ Center of fusion P ₂ First plate material End point S1 and S2 quality evaluation system before welding

Claims (1)

A first plate material and a second plate material are a quality evaluation method of a welded portion in lap fillet welding performed so that a bead is formed on the first plate material side,
A step of acquiring an image of a cross section of the weld, which is orthogonal to the longitudinal direction of the bead;
Binarizing the acquired image of the cross section,
Removing noise components from the binarized image,
Extracting an end point of the first plate material before welding from the image from which the noise component has been removed;
Extracting the contour of the surface of the bead from the image from which the noise component has been removed,
From the end point before the welding of the first plate material before the welding, using the length of the perpendicular line lowered toward the contour of the surface of the bead, the step of determining the throat thickness of the welded portion,
Weld quality evaluation method.