JP2008164482A - Method for evaluating welding quality - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an evaluating method of welding quality capable of correctly evaluating the welding quality of a welding section when deposit such as a spatter exists in the welding section. <P>SOLUTION: The irregularity information on the surface of the welding section acquired with a light radiation type irregularity measuring instrument is low-pass-filter-processed in the direction parallel to a weld bead, and noise by the deposit such as the spatter existing in the welding section is removed (step 2), tangents of base materials positioned right and left in the plane perpendicular to the weld bead are determined, respectively, using the filter-processed irregularity data (step 3), the bead end of the weld bead is calculated over the weld bead overall length based on the irregularity information of the direction perpendicular to the weld bead and the tangent of each of right and left base materials (step 4), and the welding quality of the welding section is evaluated based on the feature amount of the appearance shape of the welding section determined using at least the bead end, of the calculated bead end, tangents of the base materials, and irregularity information. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method for evaluating welding quality, and more particularly, to the welding quality of a welded portion in which a base material made of a planar member is joined in parallel.

Conventionally, as a method for evaluating the weld quality of a welded part, light is irradiated on the weld bead constituting the welded part and the surrounding base material part to image a light cutting line, or laser scanning type two-dimensional displacement Welding quality that obtains unevenness information on the surface of the weld with a sensor, obtains the width, height, undercut, etc. of the weld bead from this unevenness information, and evaluates these external shape values according to preset evaluation criteria Are known (see, for example, Patent Documents 1 and 2 and Non-Patent Document 1).
JP-A-5-87539 JP-A-2005-14026 Kagoshima Prefectural Industrial Technology Center research report, Kagoshima Prefectural Industrial Technology Center, October 1, 2002, No. 15, p. 43-46

  However, in the evaluation methods described in each of the above-mentioned documents, the spatters generated during welding and adhering to the welded part such as spatters and scales adhering to the surface of the base material are directly taken into the unevenness information of the surface shape of the welded part as convex part data. For example, the weld bead width is calculated to be larger than the actual width, which causes errors in the external shape value, and the weld quality cannot be evaluated correctly. Neither the amount nor the amount thereof was adopted as an evaluation item of the welding quality, and the correct evaluation of the welding quality was not performed on the deposits such as spatter.

  The present invention has been made in view of the above points, and it is an object of the present invention to provide a welding quality evaluation method capable of correctly evaluating the welding quality of a welded portion when deposits such as spatter are present in the welded portion. .

  In order to achieve the above object, the welding quality evaluation method according to claim 1 is a welding quality evaluation method for a welded portion in which a base material made of a planar member is joined in parallel, and constitutes the welded portion. For the predetermined region of the base material located on the left and right sides of the weld bead and the weld bead, the surface unevenness information of the welded portion is acquired by a light irradiation type unevenness measuring instrument, and the acquired unevenness information is Low-pass filter processing is performed in a direction parallel to the bead to remove noise due to spatter and other deposits existing in the welded portion, and the left and right sides in a plane perpendicular to the weld bead are used using the filtered unevenness data. Tangent lines of the base metal positioned at the welding bead, and the welding bead over the entire length of the welding bead from the unevenness information perpendicular to the welding bead and the tangent lines of the left and right base metals. The weld quality of the welded portion is calculated based on at least the feature value of the appearance shape of the welded portion obtained using the bead end among the calculated bead end, tangent of the base material, and unevenness information. It is characterized by evaluating.

  According to the first aspect of the present invention, the appearance shape of the welded portion is determined based on the unevenness data obtained by removing noise caused by deposits such as spatter obtained by low-pass filtering the unevenness information on the surface of the welded portion. Since the feature quantity is obtained and the weld quality of the welded part is evaluated based on this feature quantity, the feature quantity of the external shape can be accurately calculated without being affected by the deposits such as spatter, and the weld part can be sputtered. Therefore, the weld quality of the welded portion can be correctly evaluated when the deposits are present.

  Further, the weld quality evaluation method according to claim 2 is the weld quality evaluation method according to claim 1, wherein the feature amount of the appearance shape of the welded portion is the width of the weld bead, the undercut, and the weld bead. It is characterized in that it is at least one of the height of the weld, the straightness of the weld bead, and the periodicity of the wave of the weld bead.

  According to the second aspect of the present invention, it is possible to accurately calculate each feature amount of the appearance shape of the weld enumerated in the second aspect without being affected by the deposits such as spatters. Therefore, the weld quality of the welded portion can be correctly evaluated when there are deposits such as.

  According to a third aspect of the present invention, there is provided a welding quality evaluation method for evaluating a welding quality of a welded portion in which a base material made of a planar member is joined in parallel. The weld bead constituting the welded portion and the weld bead With respect to a predetermined region of the base material that is positioned on the left and right sides of the base material, surface unevenness information of a welded portion is acquired by a light irradiation type unevenness measuring device, and the acquired unevenness information is low-pass filtered in a direction parallel to the weld bead. The noise due to the deposits such as spatter existing in the weld is removed, and the filtered concavo-convex data is used to tangent the base material located on the left and right in the plane perpendicular to the weld bead. Are calculated, the angular deformation of the base metal is calculated from the tangent lines of the left and right base metals, and the weld quality of the weld is evaluated based on the calculated angular deformation of the base material.

  According to the third aspect of the present invention, the corner deformation of the base material is performed based on the unevenness data obtained by removing noise due to deposits such as spatter obtained by low-pass filtering the unevenness information on the surface of the weld. Calculate and evaluate the weld quality of the welded part based on this angular deformation, so that the angular deformation of the base metal can be accurately calculated without being affected by the spatter and other deposits. Therefore, it is possible to correctly evaluate the weld quality of the welded portion.

  According to a fourth aspect of the present invention, there is provided a weld quality evaluation method for evaluating a weld quality of a welded portion in which a base material made of a planar member is joined in parallel. The weld bead constituting the welded portion and the weld bead For a predetermined region of the base material that is located on the left and right sides, surface unevenness information is acquired by a light irradiation type unevenness measuring device, and the acquired unevenness information is subjected to high-pass filter processing in a direction parallel to the weld bead. Obtaining the convexity data of the deposits such as spatter existing in the welded portion, and evaluating the deposits on the welded portion based on the convexity image obtained from the convexity data over the entire length of the weld bead. Features.

  According to the fourth aspect of the present invention, a convexity image is obtained based on the convexity data obtained by performing high-pass filtering on the irregularity information on the surface of the welded portion, and the deposit is based on the convexity image. Therefore, with the convex part image detected and calculated with high accuracy, it is possible to evaluate the welding quality as an object of quality evaluation, and when there are deposits such as spatter in the weld, The weld quality of the weld can be correctly evaluated.

  As described above, according to the present invention, it is possible to correctly evaluate the weld quality of a welded part when there is a deposit such as spatter in the welded part.

  Hereinafter, an example of an embodiment of the present invention will be described with reference to FIGS. First, FIG. 1 is a flowchart showing an overall welding quality evaluation method. As shown in FIG. 2, the unevenness information acquisition in step 1 is performed on a welded product W obtained by butt welding of flat base materials 11 and 12. For the welded portion, that is, the weld bead 13 and the predetermined regions of the base materials 11 and 12 located on the left and right sides, the unevenness information on the surface of the welded portion is acquired by the light irradiation type unevenness measuring instrument 14. In the drawing, as coordinate axes, an X axis is set in a direction perpendicular to the weld bead 13, a Y axis is set in a parallel direction, and an H axis is set in a direction perpendicular to the XY plane. Reference numeral 18 denotes deposits such as spatter.

In this example, a laser scan type two-dimensional displacement sensor is used as the unevenness measuring device 14 to measure the amount of unevenness in the H direction for a predetermined range in the X direction, and the output signal of the unevenness measuring device 14 is sent to the computer 16 via the amplifier 15. Introduced and stored in a memory provided in the computer 16, the computer 16 scans the unevenness measuring device 14 in the Y direction via the scanning controller 17 and a drive mechanism (not shown), and the unevenness measuring device 14 uses the predetermined region. The unevenness is measured over the entire area. As a result, as shown in FIG. 3, unevenness information consisting of m × n pieces of height data h ₁₁ to h _mn is acquired.

In step 2 (FIG. 1), the unevenness information thus obtained is subjected to low-pass filter processing in a direction parallel to the weld bead (Y direction). FIG. 4 shows the detailed processing procedure of this step 2. From the unevenness information acquired in step 1, the data sequence in the Y direction is processed, and the data sequence of the X coordinate j = 1, that is, h _{11 at the} left end in FIG. ˜h _1n data string is low-pass filtered and stored in memory, and this is performed for each of m data strings (steps 22 to 25). These data processing is performed by the CPU in the computer 16, and the other steps are the same in the following.

  By this low pass filter processing of step 2, noise due to the deposit 18 such as spatter and scale on the surface, which is a component (high frequency component) in which the unevenness changes rapidly, is removed from the unevenness information acquired in step 1. An error inducing factor in each data processing described later is removed. Hereinafter, all the information on the unevenness of the weld bead is referred to as “unevenness data”, and particularly the data after filtering is referred to as “filtered unevenness data”.

  Next, in step 3 (FIG. 1), the tangent lines of the base materials 11 and 12 in the plane (XH plane) perpendicular to the weld bead are obtained from the unevenness data after the low-pass filter processing in step 2.

  FIG. 5 and FIG. 6 show the detailed processing procedure of this step 3. As shown in FIG. 6, the left tangent detection region extending from X = a to b for the base material 11 and X = c for the base material 12 are shown. To the right side tangent region over ~ d and input to the computer 16 (step 31). First, an approximate straight line, that is, the left side base material tangent LL is calculated from each data string in the X direction in the range of X = a to b. The process of storing this approximate straight line data string in the memory is repeated over Y coordinates k = 1 to n (steps 33 to 36), and then for each data string in the X direction in the range of X = c to d. Similar processing, that is, calculation of the right base material tangent LR and storage of the data string in the memory are performed (steps 38 to 41).

  The above approximate straight line is calculated by the method of least squares, and a data string of values in the H direction corresponding to each X value is stored as a straight data string based on the obtained linear equation. An example of the obtained left base material tangent LL and right base material tangent LR is shown in FIG. Since each of the base material tangents is obtained from the low-pass filter processed unevenness data obtained in step 2 as described above, it is calculated as an accurate tangent with almost no influence by the deposits such as sputtering. . Note that a predetermined region (range of X = a to b and c to d) for obtaining the base material tangent is set in advance.

Next, in step 4 (FIG. 1), the bead end of the weld bead is calculated from the unevenness data in the direction perpendicular to the weld bead and the base material tangent obtained in step 3. FIG. 7 and FIG. 8 show the detailed processing procedure of Step 4, and in the range to the left of the reference point (X = e) provisionally set on the bead, the X-direction data string (the low-pass filter according to Step 2). Detection of the intersection of the data string of the unevenness data string before processing) and the data string of the left base metal approximate straight line (tangent) LL, and storage of the maximum X coordinate XL _{k of the} intersection and the H direction value HL _{k at} the coordinate position in the memory The process is repeated over Y coordinates k = 1 to n (steps 52 to 55). Next, in the range on the right side of the reference point, detection of the intersection of the same data string in the X direction as described above and the right base material approximate straight line (tangent) LR, the minimum X coordinate of the intersection, and the H direction value at the coordinate position Is stored in the memory in the same manner (steps 57 to 60). The intersection here means that the difference between the two data strings is within a predetermined threshold range. That is, most of the data corresponding to the base material portion of the concavo-convex data is on the base material tangent obtained by the above-described processing, and thus is an intersection described here. On the other hand, of the unevenness data, the data corresponding to the welded portion does not ride on the base material tangent. Therefore, for the left part from the weld bead, the intersection with the maximum X coordinate among the intersection points in the range of X = 1 to e is the left bead end. Note that even if there is convex part data such as spatter in the unevenness data, the maximum value of a plurality of intersections is detected, so that an accurate bead end can be calculated. The same applies to the right side portion from the weld bead, and the one with the smallest X coordinate is the right bead end. The reference point (X = e) is provisionally set within the weld bead width (substantially the center).

  The intersection coordinate values of the left bead end and the right bead end in each data string of y = 1 to n obtained in this way are based on the accurate base material approximate straight line (tangent) data obtained in step 3. Even if the unevenness data before the low-pass filter processing is used, the bead end is calculated as an accurate coordinate value, and it is possible to accurately calculate the appearance shape feature amount of the weld in each subsequent step. It is what.

  Next, in step 5 (FIG. 1), the bead width, undercut, weld bead height, and straightness of the weld bead are calculated using the bead end, unevenness data, and base material tangent calculated in each step. Calculate and evaluate the data. In addition, since the bead end is obtained from the tangent line and unevenness data of the base material for each row (line), a more accurate appearance feature amount can be obtained compared with the case of obtaining the bead end from the equation of the entire surface of the base material. Can be calculated.

FIG. 9 and FIG. 10 show the detailed processing procedure regarding the bead width, and the process of calculating the bead width BW _k as the difference between the X coordinates of the right and left bead ends obtained in step 4 and storing it in the memory. Repeat for Y coordinate k = 1 to n (step 62-65), the mean and variance of the total bead width data BW _k obtained to evaluate the bead width (step 66).

This evaluation is performed according to a predetermined evaluation criterion. For example, when the average value of the bead width BW _k is within a predetermined range, the evaluation is good, and when the variance value is larger than a predetermined threshold, the variation is large and poor. Are evaluated using a determination routine (not shown). Note that the evaluation based on the average and the variance may be performed for only one, and the same applies to the other steps.

FIG. 11 and FIG. 12 show the detailed processing procedure regarding the undercut in step 5, and calculate the difference between the data string in the X direction and the data string of the left base material approximate straight line (tangent) LL, and below the left bead end. The process of storing the maximum value UCL _k in the range (X = 1 to XL _k ) in the memory as the left undercut is repeated over Y coordinates k = 1 to n (steps 72 to 75), and then the data in the X direction The difference between the column and the data column of the right base material approximate straight line (tangent) LR is calculated, and the maximum value UCR _k in the range beyond the right bead end (X = XR _k -m) is stored in the memory as the right undercut. The process is repeated over Y coordinates k = 1 to n (steps 77 to 80). Specifically, a process of subtracting the value of the corresponding data string in the X direction from the value of the linear data strings LL and LR is performed. The maximum value of the subtracted value is determined as undercut. For example, for the weld bead portion, the subtracted value is “minus” and is not determined to be undercut. Then, undercut evaluation is performed on the obtained left undercut UCL _k and right undercut UCR _{k in the same} manner as in step 66 by averaging and variance (step 81).

FIG. 13 and FIG. 14 show a detailed processing procedure related to the bead height in step 5, and detects the maximum value HM _k of the unevenness data within the bead width, and this maximum value HM _k and the height HL of the left bead end. _The process of calculating the bead height BH _k from _k and the height HR _k of the right bead end and storing it in the memory is repeated over Y coordinates k = 1 to n (steps 92 to 96), and the total bead height obtained. Based on the average and variance of BH _k , the bead height is evaluated in the same manner as in step 66 (step 97).

FIG. 15 and FIG. 16 show the detailed processing procedure regarding the straightness of the bead in step 5, and calculate the bead center (bead center) BS _k as the average value of the X coordinates of the right and left bead ends, and store it in the memory. The storing process is repeated over Y coordinates k = 1 to n (steps 102 to 105), and when the obtained dispersion value of all bead center values BS _k is larger than a predetermined threshold value, the meander of the bead center greatly advances straight. Assessed to be poor.

  Next, in step 6 (FIG. 1), the unevenness data of the weld bead is Fourier-transformed, and the uniformity of the weld bead wave is quantified and evaluated from the power spectrum information.

FIG. 17 and FIG. 18 show the detailed processing procedure of this step 6. The bead height at the bead center BS _k obtained in the above steps 103 and 104 is obtained from the unevenness data, and the Y coordinates k = 1 to n are obtained. The unevenness data string is subjected to discrete Fourier transform (steps 111 and 112), and the presence / absence of the periodicity of the wave is evaluated based on the power spectrum (step 113).

  FIG. 18 shows an example of a power spectrum obtained by the Fourier transform described above. When there is a single peak exceeding the threshold Th as shown in FIG. 18 (a), it is evaluated that there is a periodicity of beads, and FIG. In the case where there is no single peak exceeding the threshold Th as in (), it is evaluated that there is no periodicity of beads. For example, if the frequency A (Hz) corresponds to a period for each B (mm), a peak appears at the frequency A (Hz) because welding is performed at a constant period for each B (mm). Represents that By setting such a predetermined cycle, the welding quality can be evaluated.

Next, in Step 7 (FIG. 1) branched from Step 3, the angular deformation of the base material is calculated from the base material tangent obtained in Step 3 and evaluated. FIG. 19 and FIG. 20 show the detailed processing procedure of step 7, and the process of calculating the slope AL of the data string of the left base material approximate straight line (tangent) LL obtained in steps 34 and 35 and storing it in the memory, Iterate over Y coordinates k = 1 to n (steps 122 to 125), and an average slope AL _av of all obtained slope data is calculated (step 126). Similarly, the process of calculating the slope AR of the data string of the right base material approximate straight line (tangent) LR obtained in steps 39 and 40 and storing it in the memory is repeated over Y coordinates k = 1 to n (steps 128 to n). 131) The average inclination AR _av of the obtained all inclination data is calculated (step 132). Then evaluate the angular distortion of the base material in comparison with the difference between the thresholds of these average gradient _{AL av} and _{AR av} (step 133).

  As described above, the weld bead width, undercut, weld bead height, weld bead straightness, weld bead wave periodicity, base material angular deformation, etc. The calculation and evaluation of the feature value of the appearance shape of the welded part is performed based on the base material tangent obtained from the filtered unevenness data obtained by removing noise caused by deposits such as spatter by the low-pass filter processing in Step 2. Therefore, it is possible to accurately calculate the feature value of the appearance shape and to accurately evaluate the welding quality without being substantially affected by the deposits such as spatter.

  On the other hand, in step 8 in FIG. 1, the unevenness information acquired in step 1 is subjected to high-pass filter processing in a direction parallel to the weld bead (Y direction), and deposits such as spatter are detected and evaluated.

  FIG. 21 and FIG. 22 show the detailed processing procedure of this step 8. From the unevenness information obtained in step 1, the data string in the Y direction is set as the processing target, and as in step 2, the X coordinate j = 1 to m. The high-pass filter process is performed on each data string of (1) and the data string after the high-pass filter process is stored in the memory (steps 142 to 145).

  In step 146, the convex data of the deposit such as sputtering is obtained by binarizing (detecting data exceeding a certain threshold value) the filtered data string obtained in each of the above steps. Welding caused by deposits 18 such as spatter is determined by comparing the number of lumps of each convex image 19 shown in FIG. 22 obtained from the data with the evaluation criteria such as the total area threshold value. Evaluate quality.

  As a result, it is possible to correctly evaluate the weld quality of the welded portion, with the adhering matter such as spatter as an evaluation object.

  The above describes the case where the weld quality of the welded portion is evaluated on the front surface side of the welded product W, but the weld quality is also evaluated on the back surface side of the welded product W with respect to the quality of penetration of the butt weld. FIG. 23 is a flowchart showing the entire evaluation method in this case.

  When evaluating the welding quality for the back side, as shown in FIG. 23, information processing is performed by the same steps 1 to 4 as the flowchart shown in FIG. 1 of the above example, and only the bead height is determined in step 5. Seek and evaluate. If this bead height is detected as a negative value, it is evaluated as a poor penetration.

  In addition to the evaluation of the weld quality of the welded part obtained by butt welding, the present invention is also applicable to the evaluation of the weld quality of the welded part (lap joint) obtained by lap welding. It can be done. FIG. 24 shows an example thereof. In the figure, the same or corresponding parts as those in the above-mentioned drawings are denoted by the same reference numerals. As shown in FIG. 24 (a), the unevenness information obtained by the apparatus similar to FIG. 2 is low-passed in the same manner as described above for the welded product W in which the base metals 11 and 12 are overlap-welded to form the weld beads 13. After the filtering process, from the base material tangent and the unevenness information (unevenness data), as shown in FIG. 24B, for example, the outer shape of each of the left and right bead ends, bead width, bead height, undercut, etc. Each feature quantity can be calculated with high accuracy in the same manner as in the above example, and the weld quality of the welded portion can be correctly evaluated based on these feature quantities.

  The present invention is not limited to the above-described example. For example, as the unevenness measuring device, an optical cutting type unevenness measuring device that obtains unevenness information by imaging an optical cutting line may be used.

It is a flowchart which shows the whole welding quality evaluation method which shows an example of embodiment of this invention. It is a perspective view of the measuring apparatus used for the method of FIG. It is explanatory drawing of the unevenness | corrugation information obtained with the unevenness measuring device in the apparatus of FIG. It is a flowchart which shows the process sequence of step 2 of FIG. It is a flowchart which shows the process sequence of step 3 of FIG. It is a perspective view which shows the detection area | region in step 31 of FIG. It is a flowchart which shows the process sequence of step 4 of FIG. It is a bead uneven | corrugated diagram of the X direction which shows the calculated value by the process sequence of FIG. It is a flowchart which shows the process sequence regarding the bead width in step 5 of FIG. It is a bead uneven | corrugated diagram of the X direction which shows the calculated value by the process sequence of FIG. It is a flowchart which shows the process sequence regarding the undercut in step 5 of FIG. It is a bead uneven | corrugated diagram of the X direction which shows the calculated value by the process sequence of FIG. It is a flowchart which shows the process sequence regarding the bead height in step 5 of FIG. It is a bead uneven | corrugated diagram of the X direction which shows the calculated value by the process sequence of FIG. It is a flowchart which shows the process sequence regarding the straight advanceability of the bead in step 5 of FIG. It is a bead uneven | corrugated diagram of the X direction which shows the calculated value by the process sequence of FIG. It is a flowchart which shows the process sequence of step 6 of FIG. It is a diagram which shows the power spectrum by the process sequence of FIG. It is a flowchart which shows the process sequence of step 7 of FIG. It is sectional drawing which shows the inclination | tilt angle of the base material computed by the process sequence of FIG. It is a flowchart which shows the process sequence of step 8 of FIG. It is explanatory drawing of the convex part image obtained by the process sequence of FIG. FIG. 3 is a view corresponding to FIG. 1 and showing another example of the embodiment of the present invention. It is the perspective view (a) of the welded product which shows the further another example of embodiment of this invention, and the bead uneven | corrugated diagram (b) of the X direction of a welding part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 ... Base material, 12 ... Base material, 13 ... Weld bead, 14 ... Concavity and convexity measuring device, 16 ... Computer, 18 ... Deposit, 19 ... Convex part image, W ... Welding product.

Claims (4)

A method for evaluating the welding quality of a welded portion in which a base material made of a planar member is joined in parallel,
For the predetermined region of the base material located on the left and right across the weld bead and the weld bead constituting the weld, obtain the uneven information on the surface of the weld with a light irradiation type unevenness measuring device,
The acquired uneven information is subjected to low-pass filter processing in a direction parallel to the weld bead to remove noise due to deposits such as spatter present in the weld,
Using this filtered uneven data, each tangent of the base material located on the left and right in a plane perpendicular to the weld bead,
From the unevenness information in the direction perpendicular to the weld bead and the tangent lines of the left and right base materials, calculate the bead end of the weld bead over the entire length of the weld bead,
Of the calculated bead end, base material tangent, and unevenness information, the weld quality of the weld is evaluated based on at least the feature value of the appearance of the weld determined using the bead end. To evaluate welding quality. The welding quality evaluation method according to claim 1,
The feature quantity of the outer shape of the weld is at least one of the width of the weld bead, the undercut, the height of the weld bead, the straightness of the weld bead, and the periodicity of the wave of the weld bead. Evaluation method. A method for evaluating the welding quality of a welded portion in which a base material made of a planar member is joined in parallel,
For the predetermined region of the base material located on the left and right across the weld bead and the weld bead constituting the weld, obtain the uneven information on the surface of the weld with a light irradiation type unevenness measuring device,
The acquired uneven information is subjected to low-pass filter processing in a direction parallel to the weld bead to remove noise due to deposits such as spatter present in the weld,
Using this filtered uneven data, each tangent of the base material located on the left and right in a plane perpendicular to the weld bead,
Calculate the angular deformation of the base material from the tangent of the left and right base materials,
A welding quality evaluation method, wherein the welding quality of the welded portion is evaluated based on the calculated angular deformation of the base material. A method for evaluating the welding quality of a welded portion in which a base material made of a planar member is joined in parallel,
With respect to a predetermined region of the base material positioned on the left and right sides of the weld bead constituting the weld and the weld bead, surface unevenness information is obtained by a light irradiation type unevenness measuring device,
The acquired unevenness information is subjected to high-pass filter processing in a direction parallel to the weld bead to acquire the convexity data of the deposits such as spatter existing in the weld,
An evaluation method for welding quality, characterized in that the deposit on the welded portion is evaluated based on a convex portion image obtained over the entire length of the weld bead from the convex portion data.

Images