Automatic welding equipment, measurement anomaly detection method, and program

Abstract

To suppress the occurrence of a welding failure due to an abnormality in cross-section measurement at a groove part.SOLUTION: A sensor part 5 of an automatic welding device 1 measures the shape of a cross section perpendicular to a welding direction of a base material 91, 92 on the front side in the welding direction with respect to a welding torch 2. A first detection part 802 detects the positions of a feature point on the base material 91, 92 at each position in the welding direction by the first detection method on the basis of the output from the sensor part 5, as the first detection position. A second detection part 803 detects the position of the feature point at each position in the welding direction by a second detection method different from the first detection method on the basis of the output from the sensor part 5 as a second detection position. A determination part 804 detects an abnormality in the measurement result by the sensor part 5 on the basis of the comparison result between the first detection position and the second detection position at each position in the welding direction. With this, the occurrence of a welding failure due to an abnormality in cross-section measurement at a groove part 93 can be suppressed.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an automatic welding apparatus, as well as a measurement abnormality detection method and a program for detecting a measurement abnormality by a sensor unit in an automatic welding method.

Background Art

[0002] Conventionally, as automatic welding apparatuses, those having a function (automatic tracing function) of measuring the shape of a groove portion of a base material using a CCD camera, a laser displacement meter or the like, and adjusting the aiming position of a welding torch based on the measurement result have been used (Patent Document 1 and Patent Document 2).

[0003] In the automatic welding apparatus of Patent Document 2, the cross-sectional shape of the groove portion is represented by a sequence of points composed of measurement data from the laser displacement meter at a plurality of measurement points on the base material. In measurement by the laser displacement meter, when the surface condition of the welded portion is poor, measurement data may not be acquired and be missing at some measurement points. In the automatic welding apparatus, when the number of measurement data obtained by the laser displacement meter is less than the number of measurement data that should originally be obtained by a predetermined ratio or more, an error signal is output.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] Incidentally, while the automatic welding apparatus described in Patent Document 2 can detect missing measurement data from the laser displacement meter, it has difficulty detecting abnormal measurement data caused by poor reflection of the laser light or the presence of foreign matter below the laser displacement meter. Therefore, there is a risk that welding may be performed based on an incorrect cross-sectional shape of the groove obtained from abnormal measurement data. In this case, welding defects may be discovered through non-destructive testing after welding, potentially resulting in significant losses in the process due to welding repairs.

[0006] This invention has been made in view of the above problems, and aims to suppress the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove. [Means for solving the problem]

[0007] One aspect of the present invention is an automatic welding apparatus comprising: a welding torch for welding a groove extending in a predetermined welding direction in a base material; a moving mechanism for moving the welding torch relative to the base material in the welding direction; a position adjustment unit for adjusting the position of the welding torch in a width direction perpendicular to the welding direction; a sensor unit for measuring the shape of a cross-section of the base material perpendicular to the welding direction, located in front of the welding torch in the welding direction; a first detection unit for detecting the position of a feature point on the base material as a first detection position at each position in the welding direction using a first detection method based on the output from the sensor unit; a second detection unit for detecting the position of the feature point as a second detection position at each position in the welding direction using a second detection method different from the first detection method based on the output from the sensor unit; and a determination unit for detecting an abnormality in the measurement result by the sensor unit at each position in the welding direction based on a comparison result between the first detection position and the second detection position.

[0008] Aspect 2 of the present invention is an automatic welding apparatus according to aspect 1, wherein the characteristic feature is that the cross-section is included in the groove portion.

[0009] A third aspect of the present invention is an automatic welding apparatus according to the second aspect, wherein the characteristic feature is that welding by the welding torch is planned to be performed in the cross-section.

[0010] A fourth aspect of the present invention is an automatic welding apparatus according to the second aspect, wherein in the first detection method, the first detection position is detected based on the inclination of the cross-section of the groove with respect to the width direction, as measured by the sensor unit. In the second detection method, the second detection position is detected by fitting the cross-section of the groove, as measured by the sensor unit, with a pre-prepared reference cross-section.

[0011] Aspect 5 of the present invention is an automatic welding apparatus according to any one of aspects 1 to 4, wherein when the determination unit detects an abnormality in the measurement result, a warning is issued if the difference between the first detection position and the second detection position is greater than or equal to a first threshold, and welding by the welding torch is stopped if the difference between the first detection position and the second detection position is greater than or equal to a second threshold which is greater than the first threshold.

[0012] Aspect 6 of the present invention is an automatic welding apparatus according to any one of aspects 1 to 4 (or any one of aspects 1 to 5), wherein the automatic welding apparatus further comprises a storage unit that stores the comparison result between the first detection position and the second detection position at each position in the welding direction.

[0013] Embodiment 7 of the present invention is an automatic welding apparatus according to any one of embodiments 1 to 4 (or any one of embodiments 1 to 6), which performs multilayer welding by repeatedly welding the groove portion multiple times.

[0014] Aspect 8 of the present invention relates to a measurement anomaly detection method for detecting a measurement anomaly by the sensor unit in an automatic welding apparatus comprising: a welding torch for welding a groove extending in a predetermined welding direction in a base material; a moving mechanism for moving the welding torch relative to the base material in the welding direction; a position adjustment unit for adjusting the position of the welding torch in a width direction perpendicular to the welding direction; and a sensor unit located in front of the welding torch in the welding direction for measuring the shape of a cross-section of the base material perpendicular to the welding direction. The measurement anomaly detection method comprises: a) a step of detecting the position of a feature point on the base material at each position in the welding direction as a first detection position based on the output from the sensor unit using the first detection method; b) a step of detecting the position of the feature point at each position in the welding direction as a second detection position based on the output from the sensor unit using a second detection method different from the first detection method; and c) a step of detecting an anomaly in the measurement result by the sensor unit at each position in the welding direction based on the comparison result between the first detection position and the second detection position.

[0015] Aspect 9 of the present invention relates to a program for detecting measurement abnormalities by the sensor unit in an automatic welding apparatus comprising: a welding torch for welding a groove extending in a predetermined welding direction in a base material; a moving mechanism for moving the welding torch relative to the base material in the welding direction; a position adjustment unit for adjusting the position of the welding torch in the width direction perpendicular to the welding direction; and a sensor unit located in front of the welding torch in the welding direction for measuring the shape of a cross-section of the base material perpendicular to the welding direction. The program is executed by a computer, thereby performing the following steps: a) detecting the position of a feature point on the base material as a first detection position at each position in the welding direction based on the output from the sensor unit using the first detection method; b) detecting the position of the feature point at each position in the welding direction as a second detection position at each position in the welding direction based on the output from the sensor unit using a second detection method different from the first detection method; and c) detecting an abnormality in the measurement result by the sensor unit at each position in the welding direction based on the comparison result between the first detection position and the second detection position.

Effects of the Invention

[0016] In the present invention, it is possible to suppress the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove portion.

Brief Description of the Drawings

[0017] [Figure 1] It is a front view showing the configuration of the automatic welding device according to one embodiment. [Figure 2] It is a left side view of the automatic welding device. [Figure 3] It is a diagram showing the configuration of a computer that implements a control unit. [Figure 4] It is a block diagram showing the functions of the control unit. [Figure 5A] It is a diagram showing the flow of welding by the automatic welding device. [Figure 5B] It is a diagram showing the flow of welding by the automatic welding device. [Figure 6] It is a diagram showing an example of a measurement result obtained by a sensor unit. [Figure 7] It is a diagram showing an example of a measurement result obtained by a sensor unit. [Figure 8] It is a diagram showing an example of a measurement result obtained by a sensor unit.

Modes for Carrying Out the Invention

[0018] FIG. 1 is a front view showing the configuration of an automatic welding device 1 according to one embodiment of the present invention. FIG. 2 is a left side view of the automatic welding device 1. The automatic welding device 1 is a welding device with an automatic tracing function that performs continuous welding on the groove portion 93 while moving the welding position along the groove portion 93 of the base materials 91 and 92. In FIG. 2, the base materials 91 and 92 are shown in cross section.

[0019] In FIG. 1 and FIG. 2, three mutually orthogonal directions are indicated by arrows as the X direction, Y direction, and Z direction. In the example shown in FIG. 1 and FIG. 2, the X direction and the Y direction are horizontal directions perpendicular to each other, and the Z direction is the vertical direction, but the present invention is not limited thereto.

[0020] In the example shown in Figure 1, the automatic welding apparatus 1 moves from left to right in Figure 1 (i.e., from the (-Y) side to the (+Y) side) and performs welding on the grooves 93 provided in the base materials 91 and 92. This welding is, for example, a multi-layer welding in which welding is repeated multiple times along substantially the entire length of the groove 93. However, this welding may also be a single-layer welding. In the example shown in Figure 2, welding is performed to join base materials 91 and 92, each having a substantially V-shaped groove 93. The groove 93 extends substantially in a substantially straight line substantially parallel to the Y direction, for example. However, the groove 93 does not necessarily have to be straight and may extend in a curved manner. In the following description, the direction in which the groove 93 extends will also be called the "welding direction". The shape of the groove 93 is not limited to a V shape.

[0021] The automatic welding apparatus 1 comprises a welding torch 2, a moving mechanism 3, a position adjustment unit 4, a sensor unit 5, and a control unit 8. The welding torch 2 is supported by an arm 11 above the groove 93 of the base materials 91 and 92 (i.e., on the (+Z) side). The welding torch 2 performs welding on the groove 93 of the base materials 91 and 92 that extends in a predetermined welding direction (i.e., the Y direction).

[0022] The moving mechanism 3 moves the welding torch 2 relative to the base materials 91 and 92 in the Y direction. In the example shown in Figures 1 and 2, the moving mechanism 3 comprises a trolley 31 and rails 32. The rails 32 extend substantially parallel to the Y direction. The trolley 31 is a self-propelled trolley equipped with an electric motor or the like, and moves in the Y direction along the rails 32. The end of the arm 11 is fixed to the trolley 31, and as the trolley 31 moves in the Y direction, the welding torch 2 also moves in the Y direction together with the trolley 31. Note that the structure of the moving mechanism 3 is not limited to the above example and can be modified in various ways. For example, the moving mechanism 3 may be equipped with a linear motor.

[0023] The position adjustment unit 4 adjusts the position of the welding torch 2 in the X direction. The X direction is substantially perpendicular to the welding direction and substantially parallel to the upper surfaces of the base materials 91 and 92 (i.e., the main surfaces facing the welding torch 2). In other words, the X direction corresponds to the width direction of the groove 93 extending in the welding direction, and in the following description, it will also be simply referred to as the "width direction". The position adjustment unit 4 also adjusts the position of the welding torch 2 in the Z direction (i.e., the vertical direction).

[0024] In the automatic welding apparatus 1, the position adjustment unit 4 adjusts the widthwise and vertical positions of the welding torch 2 while it is positioned at various locations in the welding direction. The position adjustment unit 4 may, for example, be equipped with a linear motor, or it may be equipped with a ball screw mechanism and an electric motor. The structure of the position adjustment unit 4 can be modified in various ways.

[0025] The sensor unit 5 is positioned on the (+Y) side (i.e., in front of the welding direction) of the welding torch 2. The front of the welding direction refers to the front of the direction of movement when the welding torch 2 is moved by the moving mechanism 3 during automatic welding of the groove 93. In the example shown in Figure 1, the sensor unit 5 is supported by the arm 11 above the groove 93 of the base materials 91 and 92. Therefore, the sensor unit 5 is fixed relative to the welding torch 2. The sensor unit 5 measures the shape of the cross section of the groove 93 perpendicular to the welding direction (i.e., the cross section in the XZ plane). The sensor unit 5 is moved in the Y direction together with the welding torch 2 by the moving mechanism 3, and at each position in the Y direction, it measures the shape of the cross section of the groove 93 and the area near the groove 93 perpendicular to the welding direction (hereinafter also simply referred to as "cross-sectional shape").

[0026] The sensor unit 5 includes, for example, a light sectioning sensor. The light sectioning sensor irradiates the surface of the base materials 91 and 92 with a band-shaped laser beam extending in the X direction, and receives the reflected light diffusely reflected from the surface with an image sensor to acquire the cross-sectional shape of the base materials 91 and 92, including the groove 93. In the sensor unit 5, the cross-sectional shape of the groove 93 and the area near the groove 93 is acquired as a set of multiple points (i.e., a sequence of points) corresponding to multiple measurement points arranged in the X direction. These multiple measurement points are arranged, for example, at approximately equal intervals with respect to the X direction.

[0027] Figure 3 shows the configuration of the computer 80 that implements the control unit 8. The computer 80 is a normal computer comprising a processor 81, memory 82, input / output unit 83, and bus 84. The bus 84 is a signal circuit that connects the processor 81, memory 82, and input / output unit 83. The memory 82 stores various types of information. For example, the memory 82 reads and stores a program 89 that is pre-stored in a storage medium 95. The storage medium 95 is, for example, a USB memory or a CD-ROM.

[0028] The processor 81 performs various processes (e.g., numerical calculations) using the memory 82, etc., according to the program 89 etc. stored in the memory 82. The input / output unit 83 includes a keyboard 85 and a mouse 86 for receiving input from the operator, a display 87 for displaying the output from the processor 81 etc., and a transmission unit etc. for transmitting the output from the processor 81 etc. The control unit 8 may be implemented by a programmable logic controller (PLC) or a circuit board etc. The control unit 8 may be implemented by any combination of a computer, PLC and circuit board etc.

[0029] Figure 4 is a block diagram showing the functions of the control unit 8, which are realized when the program 89 is executed by the computer 80 described above. The control unit 8 controls the various operations of each component of the automatic welding apparatus 1. Figure 4 also shows components other than the control unit 8.

[0030] The control unit 8 comprises a storage unit 801, a first detection unit 802, a second detection unit 803, a determination unit 804, and a welding control unit 805. The storage unit 801 is mainly implemented by a memory 82 (see Figure 3) and stores various information related to the automatic welding apparatus 1. The first detection unit 802, the second detection unit 803, and the determination unit 804 are mainly implemented by a processor 81 (see Figure 3) and perform various calculations related to measuring the cross-sectional shape of the groove 93. The welding control unit 805 is mainly implemented by the processor 81 and performs various calculations related to welding the groove 93 and controls the welding torch 2, the moving mechanism 3, and the position adjustment unit 4, etc.

[0031] Next, the welding process using the automatic welding device 1 will be explained with reference to Figures 5A and 5B. When welding the base materials 91 and 92 is performed by the automatic welding device 1, first, the welding torch 2 and sensor unit 5 are moved in the Y direction by the moving mechanism 3, and the sensor unit 5 is positioned above the (-Y) end (i.e., the starting end) of the groove 93. Subsequently, the welding torch 2 and sensor unit 5 begin to move in the (+Y) direction. As the sensor unit 5 moves in the (+Y) direction by the moving mechanism 3, it continuously acquires the cross-sectional shape of the groove 93 and the area near the groove 93 at each position in the Y direction (i.e., the welding direction). The welding torch 2 also moves in the (+Y) direction by the moving mechanism 3, and continuously performs welding on the groove 93 at each position in the Y direction.

[0032] In the following explanation, we will focus on a specific position in the Y direction of the groove 93 and describe the operation of the automatic welding apparatus 1 with respect to that position. First, as the sensor unit 5 passes above the position, the cross-sectional shape of the groove 93 and the area near the groove 93 at that position (i.e., the cross-sectional shape of the portion of the base material 91, 92 that includes the groove 93) is acquired by the sensor unit 5 (step S11). In step S11, as shown in Figure 6, the approximately V-shaped cross-sectional shape of the groove 93 is acquired as a set of points at a predetermined number of positions in the X direction (i.e., a sequence of points 90). The sequence of points 90 acquired by the sensor unit 5 is output from the sensor unit 5 to the storage unit 801 (see Figure 4) and stored in the storage unit 801.

[0033] The specific cross-sectional shape of the groove 93 is a roughly trapezoidal shape with the lower side being shorter than the upper side. The cross-section of the groove 93 includes a first bending point 931, a second bending point 932, a third bending point 933, and a fourth bending point 934. The first bending point 931, the second bending point 932, the third bending point 933, and the fourth bending point 934 are multiple points that define the cross-sectional shape of the groove 93.

[0034] The first bending point 931 is the point where the surface of the base material 91 bends diagonally downward to the right in Figure 6 (i.e., on the (+X) and (-Z) sides) from the upper surface of the base material 91 toward the interior of the groove 93. In other words, the first bending point 931 is the boundary between the upper surface of the base material 91 and the inner surface of the groove 93. The second bending point 932 is the point where the groove side surface of the base material 91, which extends diagonally downward to the right in Figure 6 from the first bending point 931, bends approximately to the right (i.e., approximately on the (+X) side).

[0035] The third bending point 933 is the point where the surface extending to the right from the second bending point 932 in Figure 6 (i.e., the groove bottom surface) bends diagonally upward to the right (i.e., to the (+X) and (+Z) sides) in the base material 92, beyond the boundary between the base material 91 and the base material 92. The position of the third bending point 933 in the Z direction is approximately the same as the position of the second bending point 932 in the Z direction. In the example shown in Figure 6, the boundary between the base material 91 and the base material 92 is located approximately in the X direction between the second bending point 932 and the third bending point 933. The fourth bending point 934 is the point where the surface extending diagonally upward to the right from the third bending point 933 in Figure 6 (i.e., the groove side surface) in the base material 92 bends approximately to the right. That is, the fourth bending point 934 is the boundary between the inner surface of the groove 93 and the upper surface of the base material 92. The position of the fourth inflection point 934 in the Z direction is approximately the same as the position of the first inflection point 931 in the Z direction.

[0036] Once the cross-sectional shape of the groove 93 is obtained, the first detection unit 802 determines the position of a feature point included in the cross-section of the groove 93 by the first detection method based on the output from the sensor unit 5 (step S12). The feature point is, for example, one of the first bending point 931, the second bending point 932, the third bending point 933, and the fourth bending point 934. Alternatively, the feature point may be a point indicating the boundary between the base material 91 and the base material 92 located between the second bending point 932 and the third bending point 933 (i.e., the target position of the welding torch 2, hereinafter also referred to as the "welding point 94"). The above-mentioned position of the feature point is, for example, the position in the X direction and the Z direction, respectively (i.e., the X coordinate and the Z coordinate). The position of the feature point may be the X coordinate only, or the Z coordinate only. In the following description, the second bending point 932 is described as the feature point, and the position of the feature point is described as the X coordinate and Z coordinate of the second bending point 932.

[0037] The first detection method described above can be selected from, for example, various known detection methods. In the first detection method, the position of the feature point is detected based on, for example, the inclination of the cross-section of the groove 93 with respect to the width direction, as measured by the sensor unit 5. In the following description, this detection method will also be referred to as the "differential detection method".

[0038] In the differential detection method, for example, for each point in the sequence of points 90 (see Figure 6) that shows the cross-section of the groove 93, the slope of the line with respect to the width direction (i.e., dZ / dX) from that point to the next point (for example, the point adjacent to the (+X) side) is determined. Then, from the sequence of points 90, dZ / dX or (dZ / dX) 2 Four points where the coefficients change abruptly (i.e., inflection points) are extracted as the first inflection point 931, second inflection point 932, third inflection point 933, and fourth inflection point 934, and the X and Z coordinates of the second inflection point from the (-X) side are obtained as the position of the feature point, the second inflection point 932.

[0039] Alternatively, in the first detection method, the position of the feature point may be detected by fitting the cross-section of the groove 93 measured by the sensor unit 5 with a pre-prepared cross-section of the design groove 93 (i.e., the cross-section of the reference shape, hereinafter also referred to as the "reference cross-section"). In the following description, this detection method will also be referred to as the "fitting method".

[0040] In the fitting method, for example, from the sequence of points 90 representing the cross-section of the groove 93 measured by the sensor unit 5, the point group aligned along the X direction in the approximate center is selected as the groove bottom surface, and the point groups aligned inclined with respect to the X and Z directions on the (-X) and (+X) sides of the groove bottom surface are selected as a pair of groove sides. Subsequently, as shown by the thin lines in Figure 7, an approximate straight line is obtained from the point group constituting the groove bottom surface using the least squares method, and an approximate straight line is obtained from the point group constituting each of the pair of groove sides using the least squares method. Next, the shape of the approximate cross-section formed by these approximate straight lines is compared with the shape of the reference cross-section described above, and the degree of overlap between the two cross-sections is obtained. After that, changes are made to the points included in the point group corresponding to the groove bottom surface and the pair of groove sides (for example, moving the lowest point of the pair of groove sides to the point group of the groove bottom surface), and the degree of overlap between the approximate cross-section and the reference cross-section described above is repeated. Then, the X and Z coordinates of the second inflection point 932 in the approximate cross-section with the highest degree of overlap are obtained as the position of the feature point.

[0041] In the first detection method, the position of the feature points may be detected from the cross-section of the groove 93 measured by the sensor unit 5 using a learning model created in advance by machine learning. In the following description, this detection method will also be referred to as the "machine learning detection method". In the machine learning detection method, for example, a training dataset is generated based on a pre-prepared sequence of points or image showing the cross-sectional shape of the groove 93, and a trained model showing the relationship between the cross-sectional shape and the position of the feature points is generated by machine learning such as deep learning using the training dataset. Then, the sequence of points 90 showing the cross-section of the groove 93 measured by the sensor unit 5 is compared with the trained model, and the X and Z coordinates of the second bending point 932 are obtained as the position of the feature point.

[0042] When step S12 is completed, the second detection unit 803 determines the position of the aforementioned feature point (i.e., the second inflection point 932) included in the cross-section of the groove 93 based on the output from the sensor unit 5 using a second detection method (step S13). The second detection method is a different detection method from the first detection method described above. The second detection method can be selected from, for example, various known detection methods. The second detection method may be, for example, one of the differential detection method, fitting method, and machine learning detection method described above.

[0043] In this embodiment, the first detection method is a differential detection method, and the second detection method is a fitting method. In this embodiment, in step S12, the positions of the first inflection point 931, the third inflection point 933, and the fourth inflection point 934 are also determined by the first detection method. In step S13, the positions of the first inflection point 931, the third inflection point 933, and the fourth inflection point 934 are also determined by the second detection method. Note that the first and second detection methods may be methods other than the differential detection method, fitting method, and machine learning detection method described above.

[0044] When steps S12 and S13 are completed, the positions of the feature points determined in steps S12 and S13 are compared by the determination unit 804. In the following description, the position of the feature point determined by the first detection method in step S12 is also referred to as the "first detection position." The position of the feature point determined by the second detection method in step S13 is also referred to as the "second detection position." Based on the comparison result between the first detection position and the second detection position, the determination unit 804 determines whether the measurement result from the sensor unit 5 is normal or abnormal. The comparison result is stored in the storage unit 801 and displayed on the display 87 according to the operator's instructions, etc.

[0045] For example, as shown in Figure 6, if the cross-sectional shape of the groove 93 measured by the sensor unit 5 (i.e., the shape of the point sequence 90) is approximately the same as or close to the actual cross-sectional shape of the groove 93 (see Figure 2), then the first detection position and the second detection position are approximately the same. Therefore, the difference between the first detection position and the second detection position is less than a predetermined first threshold (step S14), and the determination unit 804 determines that the measurement result by the sensor unit 5 is normal (step S15).

[0046] The difference between the first detection position and the second detection position described above is, for example, the distance between a feature point obtained by the first detection method and a feature point obtained by the second detection method. Alternatively, the difference between the first detection position and the second detection position may be the distance in the X direction or the distance in the Z direction between the feature point obtained by the first detection method and the feature point obtained by the second detection method. Furthermore, if the determination of whether the measurement result by the sensor unit 5 is normal or abnormal is made based on the comparison result between the first detection position and the second detection position, it is not necessarily required to be based on the magnitude relationship between the difference between the first detection position and the second detection position and the first threshold.

[0047] If the measurement result from the sensor unit 5 at a position in the Y direction is normal, the position of the welding point 94 (i.e., the X and Z coordinates) is determined based on that measurement result. For example, the arithmetic mean of the X coordinates of the second bending point 932 and the third bending point 933, determined by the first detection method in step S12, is determined as the X coordinate of the welding point 94. Similarly, the arithmetic mean of the Z coordinates of the second bending point 932 and the third bending point 933, determined by the first detection method, is determined as the Z coordinate of the welding point 94. Alternatively, the X and Z coordinates of the welding point 94 may be determined as the arithmetic mean of the X and Z coordinates of the second bending point 932 and the third bending point 933, respectively, determined by the second detection method in step S13.

[0048] Alternatively, the position of the welding point 94 may be determined based on the positions of the second bending point 932 and the third bending point 933 determined by the first detection method, and the positions of the second bending point 932 and the third bending point 933 determined by the second detection method. That is, the position of the welding point 94 is determined based on the cross-sectional shape of the groove 93 determined by the first detection method, and / or the cross-sectional shape of the groove 93 determined by the second detection method. Note that the position of the welding point 94 may be determined by various other methods.

[0049] Then, when the welding torch 2 moves in the (+Y) direction and reaches the position, the position adjustment unit 4 is controlled by the welding control unit 805, and the position of the welding torch 2 in the X and Z directions is adjusted based on the X and Z coordinates of the welding point 94 (step S16). Then, welding is performed on the welding point 94 by the welding torch 2 (step S17). In the automatic welding apparatus 1, steps S11 to S17 described above are performed continuously at each position in the Y direction (i.e., the welding direction) of the groove 93, and automatic tracing welding of the groove 93 progresses in the Y direction. Note that when welding is performed on a groove 93 where there is almost no change in the position of the welding point 94 in the Z direction, the position adjustment of the welding torch 2 in the Z direction is not performed, and only the position adjustment in the X direction is performed. In this case, in steps S12 and S13, only the X coordinate of the feature point is detected as the position of the feature point, and the Z coordinate does not need to be detected.

[0050] On the other hand, as shown in Figure 8, if the cross-sectional shape of the groove 93 measured by the sensor unit 5 (i.e., the shape of the point sequence 90) differs relatively significantly from the actual cross-sectional shape of the groove 93 (see Figure 2), then the first detection position 951, which is the position of the feature point according to the first detection method, and the second detection position 952, which is the position of the feature point according to the second detection method, also differ relatively significantly. In Figure 8, the first detection position 951 is shown enclosed in a circle larger than the points constituting the point sequence 90, and the second detection position 952 is shown enclosed in a triangle. In the determination unit 804, if the difference between the first detection position 951 and the second detection position 952 is greater than or equal to the first threshold (step S14), the measurement result by the sensor unit 5 is determined to be abnormal (step S18).

[0051] Anomalies in the measurement by the sensor unit 5 include a state in which the actual cross-sectional shape of the groove 93 differs significantly from the cross-sectional shape of the groove 93 measured by the sensor unit 5, and a state in which, although the cross-sectional shape of the groove 93 measured by the sensor unit 5 is close to the actual cross-sectional shape of the groove 93, the position of the feature points obtained from the output of the sensor unit 5 differs significantly from the position of the actual feature points. Anomalies in the measurement by the sensor unit 5 can occur due to various causes, such as the influence of secondary reflected light of the laser beam emitted from the sensor unit 5, unsuitability of the sensor conditions in the sensor unit 5, unintended sudden operation of the automatic welding machine 1, bugs in the control unit 8, foreign matter present between the sensor unit 5 and the base materials 91, 92, and foreign matter such as burrs present on the inner surface of the groove 93.

[0052] If the measurement result is determined to be abnormal in step S18, the control unit 8 controls a warning unit (not shown) and issues a warning to the operator via a warning sound, warning light, etc. (step S19). The operator visually checks the groove 93, sensor unit 5, etc., and takes appropriate action as necessary. For example, if the cause of the abnormal measurement result is within the acceptable range, the operator stops the warning sound and warning light, etc., and continues welding the groove 93 with the welding torch 2 (steps S16-S17) while observing whether any abnormalities occur. Note that the warning may also be communicated to the operator by various means other than warning sound and warning light.

[0053] Furthermore, if the difference between the first detection position 951 and the second detection position 952 is greater than or equal to the second threshold (step S20), a warning is issued in step S19, and the control unit 8 controls the welding torch 2, the moving mechanism 3, etc., to emergency stop welding to the groove 93 (step S21). This prevents welding defects from occurring due to welding being performed based on an incorrectly determined welding point 94 position. In addition, by preventing welding from continuing in a defective state, repairs (i.e., rework) of the weld in subsequent processes can be suppressed.

[0054] As described above, in the automatic welding apparatus 1, steps S11 to S14 are performed continuously at each position in the Y direction (i.e., welding direction) of the groove 93, and it is continuously determined whether the measurement result from the sensor unit 5 is normal or abnormal. At this time, due to noise in the sensor unit 5, even if the measurement result from the sensor unit 5 is normal, the difference between the first detection position and the second detection position may momentarily exceed the first threshold or the second threshold.

[0055] Therefore, the determination unit 804 may determine that the measurement result from the sensor unit 5 is abnormal only if the difference between the first detection position and the second detection position is greater than or equal to the first threshold for a first predetermined time or longer. If the difference between the first detection position and the second detection position is greater than or equal to the first threshold for the first predetermined time or longer, the determination unit 8 may determine that the measurement result from the sensor unit 5 is normal. Furthermore, the emergency stop of welding by the control unit 8 is performed only if the difference between the first detection position and the second detection position is greater than or equal to the second threshold for a second predetermined time or longer. If the difference between the first detection position and the second detection position is greater than or equal to the second threshold for the second predetermined time or longer, the emergency stop of welding may not be performed. Note that the first predetermined time and the second predetermined time may be the same or different.

[0056] In step S12, the position of the feature point detected by the first detection method (i.e., the first detection position) may be corrected by a predetermined correction method. One example of such correction method is to use the cross-sectional shape of the groove 93 that has already been detected on the (-Y) side of the current measurement position of the sensor unit 5 (i.e., the position in the Y direction where the cross-section of the groove 93 is acquired by the sensor unit 5). Specifically, for example, statistical values ​​(e.g., mean, maximum, minimum, etc.) of the coordinate groups of the first inflection point 931, second inflection point 932, third inflection point 933, and fourth inflection point 934 detected in a predetermined range on the (-Y) side of the measurement position of the sensor unit 5 are used to correct the coordinates of the first inflection point 931, second inflection point 932, third inflection point 933, and fourth inflection point 934 at the current measurement position so as not to deviate significantly from the statistical values.

[0057] In step S12, the position of the feature point may be corrected by another correction method instead of the above correction method, or in addition to the above correction method. Similarly in step S13, the position of the feature point detected by the second detection method (i.e., the second detection position) may be corrected. The method for correcting the position of the feature point used in steps S12 and S13 may be the same or different.

[0058] Furthermore, in the automatic welding apparatus 1, the above-mentioned characteristic points do not necessarily have to be included in the groove 93, and may be provided on the base material 91 or base material 92 outside the groove 93. For example, if a linear protrusion or linear recess extending substantially parallel to the groove 93 is provided near the groove 93 on the base material 91, the top of the linear protrusion, or the bottom or upper edge of the linear recess may be considered the characteristic point.

[0059] As described above, the automatic welding apparatus 1 comprises a welding torch 2, a moving mechanism 3, a position adjustment unit 4, a sensor unit 5, a first detection unit 802, a second detection unit 803, and a determination unit 804. The welding torch 2 performs welding on a groove 93 extending in a predetermined welding direction (Y direction in the above example) on the base materials 91, 92. The moving mechanism 3 moves the welding torch 2 relative to the base materials 91, 92 in the welding direction. The position adjustment unit 4 adjusts the position of the welding torch 2 in the width direction (X direction in the above example) perpendicular to the welding direction at each position in the welding direction. The sensor unit 5 measures the shape of the cross-section of the base materials 91, 92 perpendicular to the welding direction at a position in front of the welding torch 2 in the welding direction ((+Y) side in the above example). The first detection unit 802 detects the position of a feature point on the base materials 91, 92 as a first detection position based on the output from the sensor unit 5 using a first detection method at each position in the welding direction. The second detection unit 803 detects the position of the feature point at each position in the welding direction as a second detection position based on the output from the sensor unit 5 using a second detection method different from the first detection method. The determination unit 804 detects an abnormality in the measurement result by the sensor unit 5 at each position in the welding direction based on the comparison result between the first detection position and the second detection position.

[0060] As a result, abnormalities in the cross-sectional measurement of the sensor unit 5, which occur due to various causes as described above, can be easily detected without having to prepare separate detection methods for each cause. Consequently, in the automatic welding apparatus 1 with an automatic tracing function, the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove 93 can be suppressed.

[0061] In the automatic welding apparatus 1, it is preferable that the above-mentioned characteristic point is a point included in the groove 93 in the cross-section described above. By positioning the characteristic point, which forms the basis for determining whether the measurement by the sensor unit 5 is normal or abnormal, on the groove 93, abnormalities in the cross-sectional measurement of the groove 93 can be detected with high accuracy.

[0062] Furthermore, it is even more preferable that the above-mentioned characteristic point is a single point that defines the shape of the groove 93 (in the above example, one of the first bending point 931, the second bending point 932, the third bending point 933, and the fourth bending point 934). This makes it easy to determine the position of the characteristic point from the measurement results of the sensor unit 5. In addition, based on the determined position of the characteristic point, the position of the welding point 94 can be determined accurately and easily.

[0063] In the automatic welding apparatus 1, it is also preferable that the above-mentioned feature point is the point in the cross-section where welding by the welding torch 2 is planned to be performed (i.e., the welding point 94). In this case, by detecting the position of the feature point, the target position of the welding torch 2 can also be detected, thus simplifying the control of automatic tracing welding in the automatic welding apparatus 1.

[0064] As described above, it is preferable that the position of the welding point 94 is determined based on the cross-sectional shape of the groove 93 determined by the first detection method and / or the cross-sectional shape of the groove 93 determined by the second detection method. This simplifies the control of automatic tracing welding in the automatic welding apparatus 1.

[0065] In the automatic welding apparatus 1, when the determination unit 804 detects an abnormality in the measurement result, it is preferable that a warning is issued if the difference between the first detection position and the second detection position is greater than or equal to the first threshold. This allows the operator to easily recognize the occurrence of an abnormality in the cross-sectional measurement by the sensor unit 5. Furthermore, it is even preferable that welding by the welding torch 2 is stopped if the difference between the first detection position and the second detection position is greater than or equal to the second threshold, which is greater than the first threshold. This prevents welding from continuing in a defective state. As a result, repair of the weld in subsequent processes can be suppressed.

[0066] As described above, it is preferable that the automatic welding apparatus 1 further includes a storage unit 801 that stores the comparison results between the first detection position and the second detection position at each position in the welding direction. By outputting the comparison results to a display 87 or the like, the operator can easily understand what kind of abnormality occurred in the measurement of the cross-sectional shape of the groove 93 (i.e., the nature of the measurement abnormality). As a result, it is possible to easily identify the cause of the abnormality and correct the abnormality.

[0067] As described above, the automatic welding apparatus 1 performs multi-layer welding, in which welding is repeated multiple times on the groove 93. In multi-layer welding, if welding defects occur, the subsequent repair work on the welds becomes extensive. As described above, the automatic welding apparatus 1 can suppress the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove 93, thereby reducing the need for subsequent repair work on the welds. For this reason, the structure of the automatic welding apparatus 1 is particularly suitable for automatic welding apparatuses that perform multi-layer welding.

[0068] As described above, in the first detection method, the first detection position is preferably detected based on the inclination of the cross-section of the groove 93 with respect to the width direction as measured by the sensor unit 5, and in the second detection method, the second detection position is preferably detected by fitting the cross-section of the groove 93 measured by the sensor unit 5 with a pre-prepared reference cross-section. This makes it relatively easy to measure the cross-section with the sensor unit 5. Furthermore, it is possible to suitably detect abnormalities in the cross-sectional measurement.

[0069] The measurement abnormality detection method for detecting measurement abnormalities by the sensor unit 5 in the above-described automatic welding apparatus 1 comprises: a step (S12) of detecting the positions of characteristic points on the base materials 91 and 92 as first detection positions based on the output from the sensor unit 5 using a first detection method at each position in the welding direction; a step (S13) of detecting the positions of the characteristic points at each position in the welding direction as second detection positions based on the output from the sensor unit 5 using a second detection method different from the first detection method; and a step (S14) of detecting an abnormality in the measurement result by the sensor unit 5 based on the comparison result between the first detection position and the second detection position at each position in the welding direction. As a result, as described above, the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove portion 93 can be suppressed.

[0070] As described above, program 89 is a program that detects measurement abnormalities by the sensor unit 5 in the automatic welding apparatus 1. When program 89 is executed by the computer 80, the following steps are performed: step S12, in which the positions of feature points on the base materials 91 and 92 at each position in the welding direction are detected as first detection positions by a first detection method based on the output from the sensor unit 5; step S13, in which the positions of the feature points at each position in the welding direction are detected as second detection positions by a second detection method different from the first detection method based on the output from the sensor unit 5; and step S14, in which an abnormality in the measurement result by the sensor unit 5 is detected at each position in the welding direction based on the comparison result between the first detection position and the second detection position. As a result, as described above, the occurrence of welding defects caused by abnormalities in the cross-sectional measurement of the groove 93 can be suppressed.

[0071] Various modifications are possible to the automatic welding apparatus 1, measurement anomaly detection method, and program 89 described above.

[0072] For example, the moving mechanism 3 may be a boom that supports the welding torch 2. Alternatively, the moving mechanism 3 and the position adjustment unit 4 may be combined by a 6-axis robot arm. Or, the moving mechanism 3 and the position adjustment unit 4 may be combined by a 6-axis robot arm and a slider that moves the 6-axis robot arm in the Y direction.

[0073] The moving mechanism 3 does not necessarily need to move the welding torch 2; for example, it may be a slider or turning roller that moves the base materials 91 and 92 in the Y direction below the welding torch 2, which is in a fixed position. The base materials 91 and 92 may have various shapes, such as being roughly flat or roughly cylindrical.

[0074] The sensor unit 5 does not necessarily need to be fixed to the arm 11 together with the welding torch 2, nor does it need to be moved together with the welding torch 2 by the moving mechanism 3. For example, the sensor unit 5 may be supported by a support part different from the arm 11, and may be able to move in the Y direction independently of the welding torch 2 by a moving mechanism different from the moving mechanism 3. In this case, the movement of the sensor unit 5 in the (+Y) direction may be synchronized with the movement of the welding torch 2 in the (+Y) direction, or it may be done independently. For example, the sensor unit 5 may be moved along the groove 93 in the welding direction before welding of the groove 93 by the welding torch 2 is started, and the cross-sectional shape of the groove 93 may be acquired over substantially the entire length of the groove 93.

[0075] The sensor unit 5 does not necessarily need to be equipped with an optical sectioning sensor, and various structures can be adopted. For example, the sensor unit 5 may include a camera that images the surface of the base materials 91 and 92 located on the (-Z) side, and an image analysis unit that analyzes the image captured by the camera to obtain the shape of the groove 93. Alternatively, the sensor unit 5 may include a laser displacement meter that measures the distance in the Z direction to the surface of the base materials 91 and 92 at a plurality of measurement points arranged in the X direction.

[0076] In step S14, the feature points compared between the first detection position and the second detection position are not necessarily limited to one, but may be two or more.

[0077] In the automatic welding apparatus 1, if the measurement result from the sensor unit 5 is determined to be abnormal, it is not necessarily required that a warning be issued or that welding be stopped as an emergency measure be taken; it is sufficient if the appropriate processing is carried out as needed.

[0078] In the automatic welding apparatus 1, it is not necessarily required that the comparison results between the first detection position and the second detection position at each position in the welding direction be stored.

[0079] The position of the welding point 94 may be determined based on the cross-sectional shape of the groove 93 (for example, the positions of the second bending point 932 and the third bending point 933) determined by a third detection method different from the first and second detection methods.

[0080] The shape of the groove 93 of the base materials 91 and 92 is not limited to a V shape, but may be a U shape, X shape, J shape, or any other shape.

[0081] The configurations in the above embodiments and each modified example may be combined as appropriate, as long as they do not contradict each other. [Explanation of Symbols]

[0082] 1. Automatic welding equipment 2 Welding torch 3 Moving mechanism 4 Position adjustment part 5. Sensor section 80 Computers 89 Programs 91,92 Base material 93 Bevel section 94 welding points 801 Storage section 802 First detection unit 803 Second detection unit 804 Judgment section 931 1st bending point 932 Second bending point 933 3rd bending point 934 4th bending point 951 First detection position 952 Second detection position S11~S21 Step

Claims

1. Automatic welding equipment, A welding torch that performs welding on a groove extending in a predetermined welding direction in the base material, A moving mechanism that moves the welding torch relative to the base material in the welding direction, A position adjustment unit for adjusting the position of the welding torch in the width direction perpendicular to the welding direction, A sensor unit located in front of the welding torch in the welding direction measures the shape of the cross-section of the base material perpendicular to the welding direction, A first detection unit detects the position of a feature point on the base material at each position in the welding direction as a first detection position based on the output from the sensor unit using a first detection method, A second detection unit detects the position of the feature point at each position in the welding direction as a second detection position based on the output from the sensor unit using a second detection method different from the first detection method, A determination unit detects an abnormality in the measurement result by the sensor unit at each position in the welding direction based on the comparison result between the first detection position and the second detection position, An automatic welding apparatus characterized by being equipped with the following features.

2. An automatic welding apparatus according to claim 1, An automatic welding apparatus characterized in that the aforementioned feature point is a point included in the groove portion in the cross-section.

3. An automatic welding apparatus according to claim 2, The aforementioned feature point is a point in the cross-section where welding by the welding torch is to be performed, and is characterized by this automatic welding apparatus.

4. An automatic welding apparatus according to claim 2, In the first detection method, the first detection position is detected based on the inclination of the cross-section of the groove with respect to the width direction, as measured by the sensor unit. The automatic welding apparatus is characterized in that the second detection position is detected by fitting the cross-section of the groove measured by the sensor unit with a pre-prepared reference cross-section in the second detection method.

5. An automatic welding apparatus according to any one of claims 1 to 4, When the determination unit detects an abnormality in the measurement result, If the difference between the first detection position and the second detection position is greater than or equal to the first threshold, a warning is issued. An automatic welding apparatus characterized in that welding by the welding torch is stopped when the difference between the first detection position and the second detection position is greater than or equal to a second threshold, which is greater than the first threshold.

6. An automatic welding apparatus according to any one of claims 1 to 4, An automatic welding apparatus further comprising a storage unit that stores the comparison results of the first detection position and the second detection position at each position in the welding direction.

7. An automatic welding apparatus according to any one of claims 1 to 4, An automatic welding apparatus characterized by performing multi-layer welding, which involves repeating welding to the groove multiple times.

8. An automatic welding apparatus comprising: a welding torch for welding a groove extending in a predetermined welding direction in a base material; a movement mechanism for moving the welding torch relative to the base material in the welding direction; a position adjustment unit for adjusting the position of the welding torch in the width direction perpendicular to the welding direction; and a sensor unit located in front of the welding torch in the welding direction for measuring the shape of the cross-section of the base material perpendicular to the welding direction, wherein a measurement abnormality detection method for detecting a measurement abnormality by the sensor unit, a) A step of detecting the position of a feature point on the base material at each position in the welding direction as a first detection position based on the output from the sensor unit using the first detection method, b) A step of detecting the position of the feature point at each position in the welding direction as a second detection position based on the output from the sensor unit using a second detection method different from the first detection method, c) A step of detecting an abnormality in the measurement result by the sensor unit at each position in the welding direction based on the comparison result between the first detection position and the second detection position, A method for detecting measurement anomalies, characterized by comprising the following features.

9. An automatic welding apparatus comprising: a welding torch for welding a groove extending in a predetermined welding direction in a base material; a movement mechanism for moving the welding torch relative to the base material in the welding direction; a position adjustment unit for adjusting the position of the welding torch in the width direction perpendicular to the welding direction; and a sensor unit located in front of the welding torch in the welding direction for measuring the shape of the cross-section of the base material perpendicular to the welding direction, wherein the program for detecting an abnormality in measurement by the sensor unit is provided. When the program is executed by the computer, a) A step of detecting the position of a feature point on the base material at each position in the welding direction as a first detection position based on the output from the sensor unit using the first detection method, b) A step of detecting the position of the feature point at each position in the welding direction as a second detection position based on the output from the sensor unit using a second detection method different from the first detection method, c) A step of detecting an abnormality in the measurement result by the sensor unit at each position in the welding direction based on the comparison result between the first detection position and the second detection position, A program characterized by the following:

Images