WO2021111545A1 - Welding abnormality diagnosis device - Google Patents

Abstract

This welding abnormality diagnosis device comprises: a camera system for imaging the rear surface of a material to be welded; a welding part capturing camera image collection unit for collecting a light emission capture image captured by the camera system during the welding of the material to be welded; a welding control information collection unit for collecting a welding condition set value; and a welding state diagnosis unit that calculates, on the basis of the light emission capture image and the welding condition set value, a welding light feature amount for diagnosing the welding state of the material to be welded, and that determines whether the welding is good or not, on the basis of the welding light feature amount and the upper and lower limits of welding abnormality determination.

Description

Welding abnormality diagnostic device

This application relates to a welding system for welding between steel sheets supplied to a continuous cold processing line mainly for a continuous cold rolling line for steel sheets.

In the continuous cold rolling line for steel sheets, steel sheets are continuously supplied and cold rolled. The steel sheet supplied to the continuous cold rolling line is mainly a steel sheet rolled by the hot rolling line and having a thickness of about 1.0 to 10 mm and a length of 100 m to 1 km.

In the continuous cold treatment line, which is mainly a continuous cold rolling line, the tail end of the leading material and the tip of the trailing material are welded, and the steel plate is continuously supplied. This improves productivity.

If a defect occurs in welding here, the steel plate will break from the weld during cold rolling. The breakage causes misrolling, which leads to a decrease in productivity or damage to the cold rolling equipment. For this reason, welding between steel sheets is one of the important processes in a continuous cold processing line mainly a continuous cold rolling line.

Welding of the leading material and the trailing material is performed by an automatic welding device. Normally, in an automatic welding device, the tail end of the leading material and the tip of the trailing material are cut by a shear to make their welded portions parallel. After that, the gap between the leading material and the trailing material is abutted and welded. Depending on the material, after preheating, the gap between the leading material and the trailing material is abutted.

In order to execute the desired welding with the automatic welding device, for example, predetermined welding conditions (hereinafter referred to as preset information) are input. However, welding may not be successful with preset information alone. One example of the reason why welding does not work is the distortion of the steel sheet itself. Another example of the reason is non-uniform thermal expansion when preheated. Another reason is, for example, a change in the gap distance due to a temperature drop from the beginning to the end of welding. Therefore, welding may be executed by feedback control using the actual information from the welding equipment or the accompanying instrument. However, there are cases where useful information used for feedback control cannot be obtained.

Therefore, when automatic welding is performed using only preset information, for example, an optical sensor may be used to detect the shape of the weld bead after welding is completed to determine whether the welding is good or bad. In addition, for example, in Patent Document 1, the quality of welding is determined by measuring the temperature near the welded portion immediately after welding. Further, Patent Document 2 proposes to use an infrared sensor to judge the quality of welding from the shape of the welding pond that is visually recognized immediately after welding, and to perform welding control using the judgment result. ing.

Japanese Patent No. 5058707 Japanese Patent Application Laid-Open No. 2000-351071

Judgment of good or bad welding by detecting the shape of the welding bead has become mainstream in recent years. However, even if a good judgment is made based on the weld bead, fracture may occur in continuous cold rolling, and it cannot always be said that an accurate judgment can be made. In addition, since the final judgment is made by the manager of the automatic welding device or the operator of the welding system, it may be difficult to make a judgment.

Patent Document 1 can be applied only to a specific welding method in which the temperature distribution around the welded portion is relatively visible. Therefore, for example, it is difficult to apply it to laser welding having a small heat-affected zone. Patent Document 2 focuses on the surface of the material to be welded on the welding head 12 side, and monitors the quality of welding based on the shape characteristics of the welding bead on the surface. However, it is difficult to distinguish the shape of the weld bead when the welded part is in the proper position and the shape when the welded part is greatly deviated from the proper position.

As explained above, there was a fact that it was still difficult to make a high-precision quality judgment with the conventional technology. Therefore, as a result of diligent studies by the inventor of the present application, a technique for determining the quality of welding with high accuracy has been found based on a new technical idea different from the conventional one.

This application was made in view of the above, and an object of the present application is to provide an improved welding abnormality diagnostic device so as to judge the quality of welding with high accuracy.

The welding abnormality diagnostic device provided as one of the embodiments of the present application captures a camera system for photographing the back surface of the material to be welded and a luminescence photographed image taken by the camera system at the time of welding the material to be welded. Welding part imaging camera image collecting unit to collect, welding control information collecting unit to collect welding condition set values, and welding light feature amount for diagnosing the welding status of the material to be welded are the luminescence photographed image and the welding condition. It is provided with a welding status diagnosis unit that calculates based on the set value and determines the quality of welding based on the welding light feature amount and the upper and lower limit values for determining welding abnormality.

The welding light feature amount may be a numerical value or the like representing the contour shape of the welding light taken from the back surface side of the material to be welded. The welding light feature amount may be one feature amount selected from the group consisting of the spatial moment, the area, the center of gravity, the welding light, the spark roughness, and the roundness of the welding light of the luminescence photographed image.

The welding condition set value is the power supplied to the welding head in the welding system, the moving speed of the welding head, the height or feed speed of the welding rod, and the amount of gap between the leading material and the trailing material. It may be included. The welding condition set value may include a set value or an actual value for the height position or pressure of the guide roller. The welding condition setting values are the specifications of the preceding material (for example, steel type, plate thickness and plate width), the specifications of the trailing material (for example, steel type, plate thickness and plate width), the presence or absence of the preheating process, and the temperature at the time of preheating. May include settings and.

For convenience of data processing, a welding abnormality judgment upper and lower limit table and a welding status information database may be provided. The welding abnormality determination upper and lower limit value table stores the welding abnormality determination upper and lower limit values for use in determining the quality of welding. The welding status information database stores the output result of the welding status.

In order to improve the convenience of the welding system administrator, at least one of the welding status output unit and the welding abnormality alarm unit may be provided. The welding status output unit is an interface for the welding system administrator to visually recognize the output result of the welding status diagnosis unit. The welding abnormality alarm unit notifies the welding system administrator of the welding abnormality based on the determination of the welding abnormality in the welding condition diagnosis unit.

According to the above welding abnormality diagnosis device, the welding condition can be accurately evaluated based on the welding light feature amount calculated from the welding light on the back surface of the material to be welded. The welding light obtained by photographing the back surface of the material to be welded has a characteristic that differences are likely to appear depending on the quality of the welding condition. By treating this difference in the appearance of welding light as a welding light feature amount, it is possible to accurately determine whether the welding is good or bad.

It is a figure which showed the welding system which concerns on the 1st Embodiment. It is a block diagram which showed the structure of the welding abnormality diagnosis apparatus. It is a figure which showed the example of the image collected by the welding part photographing camera image collecting part. It is a figure which showed the welding abnormality judgment upper and lower limit table. It is a figure which showed the execution procedure in the welding condition diagnosis part. It is a flow chart which showed the process of the diagnosis function for every photographed image. It is the figure which showed the example of the table for storing the actual value of the information of the welding light and the welding control output from moment to moment. It is a figure which showed the other processing example of the diagnosis function for every photographed image. It is a figure which showed the other processing example of the diagnosis function for every photographed image. It is a figure which showed the characteristic of the welding light every moment, the welding abnormality judgment upper and lower limit values, and the welding abnormality warning upper and lower limit values. It is a flow chart which showed the process of the diagnostic function after welding completion. It is a flow chart which showed the process of the welding tendency diagnosis function. It is a figure which showed an example of the tendency of the statistic of the welding light feature amount. It is a flow chart which showed the process of the diagnosis function for every photographed image in 2nd Embodiment. It is a flow chart which showed the process of the diagnosis function for every photographed image in 2nd Embodiment. It is a figure which showed the acquisition example of the gradient of the characteristic of the welding light in the 2nd Embodiment. It is a flow chart which showed the process of the diagnosis function after the completion of welding in the 2nd Embodiment. It is a flow chart which showed the process of the diagnosis function after the completion of welding in the 2nd Embodiment. It is a figure which showed an example of the characteristic of the welding light and the upper control limit and the lower control limit in the post-weld diagnosis function of the second embodiment. It is a flow chart which showed the process of the welding tendency diagnosis function in 2nd Embodiment. It is a flow chart which showed the process of the welding tendency diagnosis function in 2nd Embodiment. It is a figure which showed the statistic of the welding light feature amount in the welding tendency diagnosis function of the 2nd Embodiment, and an example of the upper control limit and the lower control limit. It is a figure which showed the processing flow which concerns on 3rd Embodiment. It is a figure which showed the processing flow which concerns on 3rd Embodiment. It is a figure which shows an example of the welding system which welds between steel plates supplied to the continuous cold treatment line mainly which a continuous cold rolling line of a steel plate. It is a figure for demonstrating the difference of appearance of welding light on the back surface according to a welding situation. It is a figure for demonstrating the difference of appearance of welding light on the back surface according to a welding situation. It is a figure for demonstrating the difference of appearance of welding light on the back surface according to a welding situation. It is a figure for demonstrating the problem in the case where the cut surface of a material to be welded is non-parallel. It is a figure for demonstrating the variation of the welding light feature quantity.

In the following description, the configurations with the same reference numerals in the specification and the drawings are assumed to be the same or substantially the same as each other. For example, in the flowchart, the corresponding steps are designated by the same reference numerals.

The first embodiment.
FIG. 1 shows a configuration example of the welding system 10 according to the first embodiment. The welding head 12 is welded in the width direction of the material 9 to be welded along a guide rail, a guide roller, or the like equipped in the device. Specifically, the material to be welded 9 includes a leading material 9a and a trailing material 9b. The welding head 12 welds the leading material 9a and the trailing material 9b.

The welded portion front surface photographing camera 13 and the welded portion back surface photographing camera 14 move so as to follow the movement of the welding head 12, and always photograph the welding situation at the same position with respect to the welding head 12. The welding head 12 may be for laser welding or for arc welding.

The electric power supplied from the output device 16 is controlled by the control device 15 to obtain a desired welding output. At this time, the control amount that determines the welding conditions is set based on the information of the material 9 to be welded. The controlled amounts are, for example, the electric power supplied to the welding head 12, the moving speed of the welding head 12, the height and feed speed of the welding rod of the welding head 12, and the amount of gap between the leading material 9a and the trailing material 9b. (Hereinafter, also referred to as a gap) and the height position and pressure of the guide roller may be included. The information of the material to be welded 9 may include, for example, the steel type, plate thickness and plate width of the leading material 9a, and the steel type, plate thickness and plate width of the trailing material 9b.

These control amounts, information on the material to be welded 9, and preset information. In addition, the actual information at the time of actual welding is taken into the control device 15 and may be used for feedback control or the like. The welding abnormality diagnosis device 20 collects images taken moment by moment by the welding portion front surface photographing camera 13 and the welding portion back surface photographing camera 14, information on welding control output, and other information incidental to welding.

The welding abnormality diagnosis device 20 determines whether the welding is good or bad by executing the calculation of the welding light feature amount and the statistic thereof for diagnosing the welding condition. The welding system configuration described is an example, and the configuration may be added or omitted as appropriate. For example, the welded surface photographing camera 13 may be omitted.

FIG. 20 illustrates a rolling system 40 to which the welding system 10 is applied. The welding system 40 includes a facility for supplying a hot coil 41, a welding system 10 for welding these, a cold rolling mill 43 which is a continuous cold rolling line, and a facility for winding a thin plate after rolling as a cold rolled coil 44. And.

The welding abnormality diagnosis device 20 is composed of the block diagram shown in FIG. The welding abnormality diagnosis device 20 includes a welding portion photographing camera image collecting unit 21, a welding control information collecting unit 22, a welding status diagnosis unit 23, a welding abnormality determination upper and lower limit table 24, a welding abnormality alarm unit 25, and a welding status. It includes an output unit 26, a welding control information database 27, and a welding status information database 28.

Welding control information collecting unit 22 collects welding control output information and other welding-related information. Information on the welding control output includes, for example, the power supplied to the welding head 12, the moving speed of the welding head 12, the height and feed speed of the welding rod, and the amount of gap between the leading material 9a and the trailing material 9b (hereinafter, gap). ), The height position of the guide roller, the set value of the pressure, and the actual value.

Other information associated with welding includes, for example, the steel type, plate thickness, and plate width of the preceding material 9a, the steel type, plate thickness, plate width of the trailing material 9b, the presence or absence of a preheating process, and the temperature setting during preheating. , Information used to set welding conditions. The information collected by the welding control information collecting unit 22 is stored in, for example, the welding control information database 27.

The welded portion photographing camera image collecting unit 21 collects images taken every moment by the welded portion front surface photographing camera 13 and the welded portion back surface photographing camera 14.

FIG. 3 shows an example of an image collected by the welded portion photographing camera image collecting unit 21. The welding head 12 moves from the lower part to the upper part of FIG. The welded portion front surface photographing camera 13 or the welded portion back surface photographing camera 14 moves so as to follow the welding head 12. After the welding head 12 has passed, the welding pond B _d1 is formed.

The welded portion front surface photographing camera 13 or the welded portion back surface photographing camera 14 photographs the dotted line portion of FIG. 3, and obtains an image in which the _{welding light L w1 as shown in FIG. 3 is reflected as an example.} The image obtained by the welded portion back surface photographing camera 14 is also referred to as a welded portion back surface photographing camera image 19 for convenience.

_{It is desirable that the welding light L w1} can be appropriately visually recognized in the welded portion back surface photographing camera image 19 by adjusting the contrast, brightness, exposure, and the like. These adjustments may be processed by the welded portion photographing camera image collecting unit 21, the welded portion front surface photographing camera 13, or the welded portion back surface photographing camera 14.

21 to 23 are diagrams for explaining the difference in how the _{welding light L w1 appears on the back surface according to the welding situation.} FIG. 21 illustrates the case of laser welding. As shown in FIG. 21, when the welding laser beam is appropriately applied to the contact portion between the leading material 9a and the trailing material 9b, the welding light L _w1 appears on the front surface and the back surface of the material 9 to be welded. On the other hand, as shown in FIG. 22, when the welding laser beam deviates from the contact portion, the welding light L _w1 appears on the front surface, but the welding light L _w1 does not appear on the back surface. As shown in FIG. 23, for some reason, the intensity (energy) of the irradiated welding laser beam may be weaker than in the normal state. In this case, the size and shape of the _{welding light L w1} appearing on the back surface are smaller than in the appropriate case of FIG.

As shown in FIG. 24, the cross section of the trailing member 9b may be slanted with respect to the length direction. In this case, since an oblique gap ga1 is generated, the contact condition of the welding laser beam changes during welding. In this case, a clear change in the size and shape of the welding light Lw1 on the back surface tends to appear during welding.

In the above example, the case of laser welding has been illustrated, but in arc welding as well, a difference appears in _{the welding light L w1 on the back surface due to the same circumstances.}

_{As described above, the welding light L w1} photographed from the back surface of the material 9 to be welded has a feature that a difference is likely to appear depending on the quality of the welding condition. Therefore, by _{treating the difference in the appearance of the welding light L w1} as a feature amount, it is possible to accurately determine whether the welding is good or bad. Based on such a principle, in the embodiment, the welding condition can be accurately evaluated based on the welding condition on the back surface of the material 9 to be welded.

The welding status diagnosis unit 23 calculates the welding light feature amount and its statistic for diagnosing the welding status based on the information collected by the welding control information collecting unit 22 and the welding unit photographing camera image collecting unit 21. , Diagnose welding abnormalities. The calculation results of the welding light features and the statistics for diagnosing the welding status are stored in the welding status information database 28.

For the diagnosis of welding abnormality, for example, the welding light feature amount for diagnosing the welding condition and its statistic are compared with the welding abnormality diagnosis standard, and when the standard value is exceeded, it is determined as a welding abnormality. Here, for example, the welding light feature amount for diagnosing the welding condition and its statistic are compared with a predetermined value in the welding abnormality determination upper and lower limit table 24 in which the welding abnormality determination upper and lower limit values are predetermined, and welding is performed. An abnormality may be determined.

FIG. 4 is an example of the welding abnormality determination upper / lower limit table 24. As shown in FIG. 4, the upper limit value and the lower limit value of the welding abnormality determination upper and lower limit values may be set according to various categories. The various categories may include the steel grades of the leading material 9a and the trailing material 9b, the average plate thickness of the leading material 9a and the trailing material 9b, the gap setting value between the leading material 9a and the trailing material 9b, and the like.

Further, the upper limit value and the lower limit value (upper and lower limit values of the welding abnormality warning) for warning that a numerical value close to the welding abnormality judgment is detected may be set in a narrower range than the range where the welding abnormality judgment is judged. Alternatively, as described in the second embodiment, the welding abnormality is determined by both the welding light feature amount for diagnosing the welding condition and the welding abnormality determination upper and lower limit values calculated using the statistic thereof. May be good.

When it is determined that the welding is abnormal, the welding abnormality alarm unit 25 sends a warning to the administrator. The welding status output unit 26 outputs the welding status moment by moment, the results of a plurality of welding statuses, and the like. This output may be displayed on a display screen for the administrator to confirm welding.

The welding status diagnosis unit 23 has a diagnosis function block 23A for each photographed image, a diagnosis function block 23B after the completion of welding, and a welding tendency diagnosis function block 23C.

The diagnostic function block 23A for each photographed image calculates a statistic for diagnosing the welding situation for each image photographed every moment from the start to the end of the welding process, and determines whether the welding is good or bad.

Each time the welding of the material 9 to be welded is completed, the post-welding diagnostic function block 23B uses the statistic for diagnosing the welding condition calculated by the above-mentioned photographed image-by-photographed image-based diagnostic function block 23A in one welding. Calculate the statistics for diagnosing the welding situation and judge the quality of welding.

The welding tendency diagnosis function block 23C is a statistic for diagnosing the welding situation at an arbitrary number of welds M from the newest welded time after the number of welds exceeds a predetermined number of welds N after the welding system is started. Welding quality is judged from the tendency of quantity.

Both functions use the welding portion backside photographing camera image 19 collected by the welding portion photographing camera image collecting unit 21 and the information collected from the welding control information collecting unit 22 as input sources.

FIG. 5 shows a flow chart showing the execution procedure of each function. The logic for determining from the start to the end of welding (step S1) may be determined based on the actual value of the welding control output information indicating the welding sequence ON. Alternatively, it may be determined by confirming the existence _{of the welding light L w1 in} the welding portion front surface photographing camera image or the welding portion back surface photographing camera image 19 obtained from the welding portion photographing camera image collecting unit 21.

The diagnostic function block 23A for each photographed image operates according to the welding process ON. Information collected by the welding control information collecting unit 22 is given to the diagnostic function block 23A for each captured image via the welding control information database 27.

The diagnostic function block 23A for each captured image calculates the analysis result of the welding status for each captured image every moment. The analysis result includes the "welding light feature amount" described later. The analysis result for each image is stored in the welding status information database 28.

Step S2 includes a logic for determining the completion of welding. In step S2, it is also determined whether or not the database has been updated. After the welding is completed, the diagnostic function block 23B operates after the welding is completed. After the welding is completed, the analysis result (welding light feature amount) of each image taken every moment stored in the welding status information database 28 is used in the post-welding diagnostic function block 23B.

The post-welding diagnostic function block 23B outputs the analysis result of the welding status in one welding to the welding status information database 28. Specifically, the analysis result includes a statistic calculated from the welding light feature amount. The details of this statistic will be described later.

The logic for determining the completion of welding (step S2) may be determined based on the actual value of the information of the welding control output representing the welding sequence ON, or the welding portion obtained from the welding portion photographing camera image collecting unit 21. It may be judged by confirming the existence _{of the welding light L w1 in} the front surface photographing camera image or the welding portion back surface photographing camera image 19.

_{As a method of determining the presence of the weld light L w1} in the welded portion front surface photographed camera image obtained from the welded portion photographed camera image collecting unit 21 or the welded portion back surface photographed camera image 19 in steps S1 and S2, for example, an image. The following threshold processing in the processing may be used. The numerical array img of the color space of the image is represented by the following equations (1) and (2).

Here, w is the number of pixels in the width direction of the image, h is the number of pixels in the height direction of the image, and I is the pixel value at the width direction pixel position x and the height direction pixel position y. Further, as an example, the color space is described in RGB space, but this is not the case. At this time, for example, after the image is grayscaled, _{the presence of the welding light L w1} can be determined by performing threshold processing based on the average value of all the pixels.

Here, No Welding is a judgment from the start to the end of welding. I ^d (x, y) is a pixel value after grayscale at the width direction pixel position x and the height direction pixel position y. ^{The coefficients for calculating Id} (x, y) shown here are ITU-R BT.601 (Studio encoding parameters of digital television for standard), which is an international standard for converting analog and digital signals. 4: 3 and wide screen 16: 9 aspect ratios International Telecommunication Union). However, each coefficient may rely on other standards.

However, the method using the above average value is an example. Not limited to the average value, various known threshold treatments may be applied.

Further, since the welding beam L _w1 flickering necessarily because the welding beam L _w1 in all images during welding can not always be confirmed, determination using an image of an arbitrary number is desirable.

Next, it is determined whether or not the number of welds since the system is started is N or more (step S3). If it is not N or more, this routine ends.

When it is determined in step S3 that the number of welds is N or more, the welding tendency diagnosis function block 23C is activated. The welding tendency diagnosis function block 23C receives a statistic based on the welding light feature amount for diagnosing the welding status of the number of welds J (J ≧ N) from the welding status information database 28.

<Diagnosis function for each captured image>
The diagnostic function block 23A for each captured image will be described with reference to the flow chart of FIG. First, the back side photographed camera image 19 of the welded portion is acquired. The light source other than the _{welding light L w1} existing in the acquired backside photographing camera image 19 of the welded portion _{is removed, and the portion corresponding to the welding light L w1} and the other portion are binarized (step S101).

As an example of the method of extracting only the welding light L _{w1, the} case where the Gaussian filter and the binarization of Otsu are applied will be shown. The Gaussian filter smoothes the image by weighting the neighboring pixel values with the Gaussian distribution g.

Apply Otsu's binarization process to the image smoothed by the Gaussian filter. Otsu's binarization process calculates the threshold value at which the degree of separation is greatest within the range of the maximum and minimum pixel values in the image. The pixel value is binarized according to the calculated threshold value.

The above process is an example, and other modified examples are also possible. As a modification, average value filtering may be used, smoothing by median filtering may be used, or processing such as adaptive binarization can be applied.

From an image obtained by extracting the welding beam _{L w1,} it calculates the contour of the welding beam _{L w1} (step S102). The contour extraction method may be a method of applying a filtering process such as a first-order differential filter or a Laplacian filter. Alternatively, the contour extraction method may simply be a method of extracting the contour having the maximum area belonging to one of the numerical values obtained by the binarization process. In either method, a pixel position group for expressing the contour of the _{welding light L w1 is obtained.}

In the embodiment, by using the pixel position group representing the outline of the welding beam L _w1, welding light feature amount is calculated indicating a feature of the welding beam L _w1 (step S103). The welding light feature amount numerically represents the feature of the figure represented by the contour shape of the _{welding light L w1.}

For example, various welding light features such as the following can be used. Welding light feature amount may be a spatial moment of welding light L _w1 may be in the area of the welding beam L _w1 may be in the center of gravity of the welding beam L _w1 may be a circumference of the welding beam L _w1, spark welding light L _w1 The roughness may be used, or the roundness of the welding light L _{w1 may be used.}

The spatial moment _mijf of the welding light L _w1 may be calculated by the following equation (6). The area A _f of the welding light L _w1 may be calculated by the following formula (7). The center of gravity C _f of the welding light L _w1 may be calculated by the following equation (8). The peripheral length P _f of the welding light L _w1 may be calculated by the following equation (9). The spark roughness R _f of the welding light L _w1 may be calculated by the following equation (10). Roundness _{C IRCF} welding light _{L w1} may be calculated by the following equation (11).

Here, m _10f and m _01f are spatial primary moments in the width direction and the height direction of the image, respectively. The subscript f is the number of images (f = 1toF, F: welding end image) from the start to the end of welding. Welding light feature amount, at least, the area of the welding beam L _w1, the circumferential length of the welding beam L _w1, the center of gravity of the welding beam L _w1, it is desirable to include a roundness of the welding beam L _w1.

The amount is not limited to each of the above welding light feature amounts, and other amounts may be used. FIG. 25 is a diagram showing variations in welding light features. The welding light Lw1 of FIG. 25 includes a first portion and a second portion extending to the left and right of the paper surface with respect to the center of gravity G. The distance from the center of gravity G to the first part is r1, and the distance from the center of gravity G to the second part is r2. _{Let r br be} the absolute value of the difference between r1 and r2. This r _br may be used as a welding light feature amount.

The welding light feature amount is stored in the welding status information database 28 (step S104). At the time of storage, the welding light feature amount is associated with the information used for setting the welding conditions obtained from the welding control information collecting unit 22. In addition, the welding light feature amount is stored together with the actual value of the welding control output information obtained from the welding control information collecting unit 22. That is, as in the table shown in FIG. 7, the actual value of the welding light feature amount and the welding control output information is stored in a predetermined table linked to the information used for setting the welding conditions. Will be done.

From the obtained welding light feature amount, the quality of welding in the acquired welded portion back surface photographed camera image 19 is evaluated and diagnosed. In the embodiment, first, the welding abnormality determination criterion is acquired (step S105).

As an example, the upper and lower limit values for welding abnormality judgment, which is an example of welding abnormality judgment criteria, are used for evaluation / diagnosis. The welding abnormality determination upper and lower limit values are stored in advance in the welding abnormality determination upper and lower limit table 24. Among the welding condition setting information obtained from the welding control information collecting unit 22, the steel grades of the leading material 9a and the trailing material 9b, the average plate thickness of the leading material 9a and the trailing material 9b, and the leading material 9a and the trailing material 9b The set value of the gap may be used as a table reference parameter. The corresponding category of the welding abnormality determination upper and lower limit table 24 is referred to based on the table reference parameter, and the welding abnormality determination upper and lower limit values of the welding abnormality determination upper and lower limit values are acquired.

Here, if a welding abnormality warning upper / lower limit value is provided, this may be acquired at the same time as the welding abnormality determination upper / lower limit value.

When the welding abnormality warning upper and lower limit values are provided, each welding light feature amount is compared with the corresponding welding abnormality warning upper and lower limit values (step S106). Further, each welding light feature amount is compared with the corresponding welding abnormality determination upper and lower limit values (step S107).

When each welding light feature amount exceeds the welding abnormality warning upper and lower limit values and does not exceed the welding abnormality upper and lower limit values, a warning is notified that the state is close to the welding abnormality (step S108). When the characteristics of each welding light L _w1 exceed the upper and lower limits of the welding abnormality, it is determined that the welding abnormality is present and the welding abnormality is notified (step S109). At this time, the welding light feature amount used for determining the welding defect may be limited to any one, or a plurality of welding light features may be selected.

However, in the welding abnormality determination using the welding light feature amount obtained from the welded portion backside photographed camera image 19 acquired every moment, there are also extreme numerical values output due to various disturbances, so the manager is notified of the welding abnormality. As a means, it is not always effective.

Therefore, as a modification, the flow chart described in the combination of FIGS. 8A and 8B is provided. For convenience, FIGS. 8A and 8B may be collectively referred to as FIG. In the flow of FIG. 8A, by adding steps S200 to S203 after step S104, the welding light feature amount _{in an arbitrary number of images section (Fr) is acquired.} Next, smoothing is performed using the average value or the median value of the plurality of welding light features acquired in step S203 (step S204), and the welding abnormality determination criterion is acquired (step S205). Welding abnormalities may be determined after smoothing.

Figure 9 shows an example of the welding abnormality judging a welding light feature amount A momentary lower limit _{A err1, A err2} and welding abnormality Warning threshold _{A wrn1, A wrn2.} The welding light feature amount A is arbitrarily selected from various quantities such as the above-mentioned spatial moment and area. The horizontal axis is the number of images from the beginning to the end of welding.

_{Since the welding light feature amount A is below the welding abnormality warning lower limit value A wrn2} in the middle from the start to the end of welding, a warning is notified to the administrator. After that, since the welding light feature amount A is lower _{than the welding abnormality determination lower limit value Aerr2} , the welding abnormality is notified.

<Diagnosis function after welding is completed>
Next, the diagnostic function block 23B after the completion of welding will be described. The processing flow of the diagnostic function block 23B after the completion of welding is shown in FIG. Upon completion of welding, the welding light feature amount of the completed welding is acquired from the welding status information database 28 every moment (step S300). At this time, the welding light feature amount between any number of images may be taken out from the collected momentary welding light feature amount, or may be taken out by the moving average or the moving median value in an arbitrary number of image number sections.

Next, in the embodiment, the statistics for each of the collected momentary welding light features are calculated (step S301). By obtaining the statistic, the characteristic amount of welding light from moment to moment is reduced to the characteristic for each number of weldings.

The statistic described here may be an average value, a standard deviation, a variance, a maximum value, a minimum value, or a skewness. It may be sharp, it may be median.

The statistic may be calculated by the following generally known formulas (12) to (19).

The standard deviation σ may be calculated by the following equation (13). The variance s ² may be calculated by the following equation (14). The maximum value may be calculated by the following formula (15). The minimum value may be calculated by the following equation (16). The skewness β ₁ may be calculated by the following equation (17). The kurtosis β ₂ may be calculated by the following equation (18). The median value may be calculated by the following equation (19).

One of the features of the welding light L _w1 is represented by the following formula (20). An arbitrary type of welding light feature is selected from the various welding light features described above as exemplified by the formulas (6) to (11). The statistics of the selected weld light features are calculated.

Here, f is the number of images (f = 0toF, F: welding end image) from the start to the end of welding.

Next, the calculated welding light feature amount statistic is linked to the information used for setting the welding conditions obtained from the welding control information collecting unit 22, and is stored in the welding status information database 28 (step). S302).

Next, the quality of welding is evaluated and diagnosed from the beginning to the end of welding from the statistics of welding light features. Here, the welding abnormality determination criterion is acquired (step S105). For the evaluation / diagnosis of the quality of welding from the beginning to the end of welding, the welding abnormality judgment upper / lower limit value is acquired from the welding abnormality judgment upper / lower limit table 24, and the upper / lower limit value is used as the upper / lower limit value in the same manner as the processing performed by the diagnostic function block 23A for each captured image. Evaluation by comparison and determination of welding abnormality (steps S105 to S109).

In the welding abnormality determination upper and lower limit table 24 for welding abnormality determination at the beginning and end of welding using the statistic of the welding light feature amount, upper and lower limit values corresponding to each statistic are provided. Further, the welding light feature amount and the statistic thereof used for determining the welding defect may be limited to any one, or a plurality of them may be selected.

<Welding tendency diagnosis function>
Next, the welding tendency diagnosis function block 23C will be described. FIG. 11 is a processing flow of the welding tendency diagnosis function block 23C. In the welding tendency diagnosis function block 23C, first, the statistics of the welding light feature amount for an arbitrary number of welds M are acquired from the welding status information database 28 (step S400). It is desirable that the arbitrary number of welds M is relatively large.

FIG. 12 shows an example of a tendency of the statistical amount of the welding light feature amount for an arbitrary number of welds M. ^{FIG. 12 shows a statistic C d} of a certain welding light feature amount C. Figure 12, statistics ^C welding abnormality judging upper limit for ^{the d} value ^C _{d err1} welding abnormality determination lower limit ^C _{d err2} welding abnormality warning upper limit ^C _{d WRN1} welding abnormality warning limit value ^C _{d Wrn2} the regression line _{L C} ^d _{and stat} are exemplified.

Next, a regression line is obtained for each category defined in the welding status information database 28 for the acquired statistic of the welding light feature amount for the number of welds M (step S401). The regression line of the welding light feature statistic may be calculated by the following equation (21).

Here, feat is _{a feature of the welding light L w1} , stat is a statistic, a _{feat and stat} are slopes of the regression line, and b _{feat and stat} are intercepts of the regression line.

As a result, the slope of the regression line is obtained (step S402). From the slope of the obtained regression line, the long-term tendency of the characteristics of the _{welding light L w1 is evaluated and diagnosed.} Similar to the diagnostic function block 23A for each captured image and the diagnostic function block 23B after the completion of welding, the welding abnormality determination upper and lower limit values are set in advance from the welding abnormality determination upper and lower limit table 24, which corresponds to the features and statistics of _{the welding light L w1.} Acquire the upper and lower limit values of the welding abnormality determination upper and lower limit values for the welding tendency (step S403).

When the slope of the regression line exceeds the upper and lower limit values of the welding abnormality determination upper and lower limit values in the welding tendency abnormality, the first condition is satisfied (step S404). _{The second condition is satisfied when the characteristics of the welding light L w1 corresponding to} the number of welding lights L w1 going back from the new welding of the welding time exceed the upper and lower limit values of the welding abnormality determination upper and lower limit values (step S405). When both the first condition and the second condition are satisfied, it is diagnosed that there is a long-term change in the tendency regarding the characteristics of the _{welding light L w1.} The diagnosis result is notified to the outside (step S406).

For example, the dirt on the lens of the welded portion back surface photographing camera 14 due to long-term use may be obtained by capturing the tendency change in the roundness of the _{welding light L w1.} Further, the welding light feature amount includes the area or the peripheral length of _{the welding light L w1.} If the slope of the regression line in these changes in the welding light feature amount tends to decrease, it can be seen that there is a risk of abnormality in the welding output. In particular, if the welding system is laser welding, it can be suggested that the protective glass of the laser output source may be dirty.

The second embodiment.
Next, the second embodiment will be described. The description of the parts that overlap with the first embodiment will be omitted. In the first embodiment, various welding abnormality determination upper and lower limit values are acquired as preset numerical values and used for welding abnormality determination. On the other hand, in the second embodiment, various welding abnormality determination upper and lower limit values are calculated.

<Diagnosis function for each captured image>
The diagnostic function block 23A for each captured image in the second embodiment is shown in combination with FIGS. 13A and 13B. For convenience, FIGS. 13A and 13B may be collectively referred to as FIG.

In the diagnostic function block 23A for each captured image, the first-order differential component of the welding light feature amount is calculated from the welding light feature amount acquired in the arbitrary number of images section R (that is, the gradient of the welding light feature amount in the section R). ), The quality of welding is evaluated and diagnosed based on the change. Here, R may be a relatively long section between the beginning and end of one welding. Further, as described in the first embodiment, the welding light feature quantity obtained from the welded portion back surface photographed camera image 19 acquired every moment has an extreme numerical output due to various disturbances, and thus is different. use welding light feature amount smoothed in any number of images section (F d _^r), it may calculate the first derivative. However, in this case, it is necessary to make the ^{_R> F} _{d r.}

In FIG. 13, after the processes of steps S101 to S104 described above are executed, the identifier FrmCnt for counting the number of images is compared with the predetermined value R in step S500. Up to the point before step S503 in FIG. 13, the same processing as in the first embodiment is performed. In step S503, the welding light feature amount is acquired in R. Obtained welded light feature quantity may be smoothed by any image acquisition sections F ^{d r} (step S504).

The latest gradient Qk and the previous gradient Qk _-1 are compared. Here, the number of image section positions k is the number of number of image section positions at the beginning and end of one welding (k = 0toK (K = F / R, F: welding end image)).

FIG. 14 shows an example of acquiring the gradient of the welding light feature amount. Each of the image number section positions k, k-1, and k-2, the gradient Q _k at the position k, and the gradient Q _{k-1 at} the position k-1 are shown.

If the difference between the gradient exceeds an arbitrary threshold value (D _grad), significant change is determined to have occurred welding light feature quantity (step S506). In this case, the manager is notified of the welding abnormality (step S109).

<Diagnosis function after welding is completed>
In the post-welding diagnostic function block 23B in the second embodiment, a control chart, which is one of the quality control methods, is applied. In the control chart, generally, the upper control limit and the lower control limit are set to 3σ (σ: standard deviation), and when they are exceeded, it is determined to be abnormal.

In the post-welding diagnostic function block 23B in the second embodiment, the upper control limit and the lower control limit for the welding light feature amount are calculated for each time series using the welding light feature amount when welding is completed normally. To do. These control limits are used as the upper and lower limits for determining welding abnormalities to diagnose the quality of welding.

However, it is necessary to give a code indicating the quality of welding in advance for each welding. This code may be assigned by the administrator, as a result of the diagnostic method in the first embodiment, or by another welding quality determination facility (for example, a bead inspection device). You may.

15A and 15B are flow charts of the post-welding diagnostic function block 23B in the second embodiment. 15A and 15B may be collectively referred to as FIG. After the welding is completed, it is confirmed whether or not the number of welds when the welding is normal is N ^{d or more (step S600).}

If the number is less than N ^d , the same processing as that of the post-welding diagnostic function block 23B in the first embodiment is performed.

When the number is N ^{d or more, the welding light feature amount at an arbitrary number of welds M d} among the number of welds in which welding was normal is acquired (step S601).

From the collected welding light features, the upper control limit and the lower control limit are calculated for each number of images (step S602). Even if the upper control limit and the lower control limit are calculated according to the definition of the upper control limit and the lower control limit in the Shewhart control chart (JIS Z 9020-2: 2016 control chart-Part 2: Shewhart control chart), for example. Good. Specifically, the following formulas (22) to (25a) and (25b) may be used.

The upper control limit UCL _i may be calculated by the following equation (22). The lower control limit LCL _i may be calculated by the following equation (23). _{The average value of the features of the welding light L w1} in the normalized number of images i may be calculated by the following equation (24). _{The standard deviation of the characteristics of the welding light L w1} in the normalized number of images i may be calculated by the following formula (25a).

Here, i is a number image normalized (i = 0toI ^d).

sigma _i is the standard deviation of the welding number M ^d present in the number of images i normalized. Since the number of images may differ depending on the welding conditions, the welding light feature amount corresponding to the normalized number of images may be obtained from an approximate value or the like and complemented. In this case, for example, simple linear interpolation may be used, or interpolation by a spline function may be used.

Also, here, the upper control limit and the lower control limit are set to 3σ. However, as a modification, 2σ may be used instead of 3σ, and the upper and lower limit settings may be changed.

Further, the upper control limit and the lower control limit may be set for each category similar to the welding abnormality determination upper / lower limit table 24. The classification of the welding abnormality determination upper and lower limit table 24 is the steel type of the leading material 9a and the trailing material 9b obtained from the welding condition setting information of the welding control information collecting unit 22, the average plate thickness of the leading material 9a and the trailing material 9b, and the leading material. It is a set value of the gap between the material 9a and the trailing material 9b.

The obtained upper control limit and lower control limit are set as welding abnormality determination criteria, and are used for determining the quality of welding (step S603). That is, when there are many points exceeding the upper control limit and the lower control limit, it is determined that the welding abnormality is found, and the manager is notified of the welding abnormality.

FIG. 16 shows an example of the welding light feature amount and the upper control limit and the lower control limit corresponding to each time series. FIG. 16 shows an upper control limit D _m1 and a lower control limit D _{m2 of a certain welding light feature amount D.} The first welding example D _ex1 illustrates the case where the upper control limit D _{m1 is exceeded.} The second welding example D _ex2 is an example in which welding is completed normally. The normalized number of images i is I ^d .

<Welding tendency diagnosis function>
In the welding tendency diagnosis function block 23C in the second embodiment, the upper control limit and the lower control limit in the control chart are set to the upper and lower limit values for welding abnormality determination, as in the case of the post-welding diagnosis function block 23B in the second embodiment. Used for.

17A and 17B show a flow chart relating to the processing of the welding tendency diagnosis function block 23C in the second embodiment. 17A and 17B may be collectively referred to as FIG. In the second embodiment, the post-welding diagnostic function block 23B ^{determines whether or not the number of normally completed welds is N d} or more (step S600).

When the number of welds completed normally is N ^d or more, the statistic of the welding light feature amount at an ^{arbitrary number of welds M d} among the number of welds in which welding was normal is acquired (step). S601). Further, the upper control limit and the lower control limit are calculated from the acquired statistics (step S702). These control limits are set as welding abnormality determination criteria (step S603). After that, the processes of steps S400 to S406 are executed in the same manner as in the flow chart of FIG.

That is, as shown in FIG. 18, the upper control limit D ^d _{mx 1} and the lower control limit D ^d _{mx 2} are uniquely determined for a certain number of welds. The obtained upper control limit D ^d _mx1 and lower control limit D ^d _mx2 are used as the upper and lower limit values for determining welding abnormality to evaluate the quality of welding.

As one of the methods for determining the upper and lower limits of welding abnormality determination, the upper control limit and the lower control limit obtained from the control chart are shown, but they are not necessarily limited to this, and for example, pattern recognition is used. The method may be used.

In this case, the distance between the welding light features obtained for each welding is calculated for a certain number of welds M ^dd , and the boundary between the case where the welding is completed normally and the case where the welding becomes abnormal is obtained based on the distance. .. Here, the number of welds M ^dd is the number of welds completed regardless of whether the welds are good or bad. The distance between each welding light feature amount obtained for each welding may be obtained by, for example, a mean square error, or may be calculated by, for example, the following equation (26).

Here, d _{s and t} are the distances (s> t, s, t = 1 toM) between the welding light feature amount in the sth welding and the welding light feature amount in the tth welding within the number of welds M ^{dd. dd} ). The mean square error is an example, and the distance between features may be calculated by another method. Welding quality is judged based on the boundary between the case where welding is completed normally and the case where welding becomes abnormal.

Various machine learning may be used as a method for finding boundaries. For example, a support vector machine, a neural network, or the like may be used to determine the boundary between when welding is completed normally and when welding becomes abnormal.

Third embodiment.
In the third embodiment, the diagnostic function block 23A for each photographed image is used for welding from the welding light feature amount for diagnosing the welding condition calculated for each image photographed every moment from the start to the end of the welding process. Correct the control output information.

The control output includes the electric power supplied to the welding head 12, the moving speed of the welding head 12, and the gap. For example, when the welding system is arc welding, the feed rate of the welding torch may be included in the correction target. Various other control outputs may be corrected. These control outputs are "welding condition set values" in the welding system 10.

The third embodiment will be described with reference to FIGS. 19A and 19B. 19A and 19B may be collectively referred to as FIG. It should be noted that the parts that overlap with the first embodiment and the second embodiment are not mentioned.

In the flow chart of FIG. 19, steps S100 to S104 and steps S200 to S204 are executed in the same manner as described in FIG. In the diagnostic function block 23A for each photographed image in the third embodiment, the welding light feature amount obtained from the acquired welded portion back surface photographed camera image 19 and the welding light feature equivalent to the number of welds W in the number of matching images. From the difference in the average value of the amounts, the correction amount for the welding control output information is calculated, and the correction is added to the target control output (steps S800 to S802). At this time, the correction amount for the controlled object may be given by, for example, the calculation formula (27) below.

Here, α _S is a correction coefficient for the target control output. cur is the number of images from the start of welding in the welding. f _match is the number of images that matches cur.

For example, the target control output is the electric power supplied to the welding head 12. At this time, for example, an area is used as one of the welding light feature quantities. This area is, at a certain time cur _^d, it is assumed that is smaller than the average value of the welding number W present in the area of the welding beam _{L w1} in _{f match} ^d. In this case, the electric power supplied to the welding head 12 may be increased by a correction amount so that the area of the _{welding light L w1 is constant.}

For example, the target control output is the moving speed of the welding head 12. At this time, for example, the peripheral length is used as one of the welding light feature quantities. It is assumed that this circumference becomes smaller than the average value of the number of welds W of the circumference of the welding light L _w1 in f _match ^{dd at} a certain time cur ^dd. In this case, the moving speed of the welding head 12 may be slowed by the correction amount so that the peripheral length of _{the welding light L w1 is constant.}

Further, the correction coefficient to the control output is the steel type of the leading material 9a and the trailing material 9b, the leading material 9a and the trailing material obtained from the information used for setting the welding conditions obtained from the welding control information collecting unit 22. It is desirable to provide each of the same categories as the welding abnormality determination upper and lower limit table 24, which divides the average plate thickness of 9b, the set value of the gap between the leading member 9a and the trailing member 9b, and the like. Further, the target control output is not necessarily one, but may be a plurality.

A correction amount is added to the control output based on the welding light feature amount every moment, but when _{the feature of the welding light L w1} exceeds the welding abnormality warning upper and lower limit values, the same as in the first embodiment. , Notify the administrator with a warning. Further, when the welding abnormality judgment upper and lower limit values are exceeded, a welding abnormality is notified.

As described above, the welding abnormality diagnosis device 20 according to the embodiment utilizes the characteristics of light emission (that is, welding light) at the time of welding obtained from the image obtained by photographing the back surface of the material 9 to be welded to determine the welding status. Analyze and determine the quality of the weld during and after the weld. Welding conditions are automatically adjusted from the good or bad state of welding judged during welding, and the subsequent welding state is made good. Furthermore, after welding, the judgment criteria used in the analysis of good / bad judgment during welding and the welding conditions are used to predict the welding situation and recommend measures to the manager to avoid the situation caused by the welding failure. ..

In the embodiment, the welding status diagnosis based on the welding light is applied to the back surface of the material 9 to be welded, but even if the welding status diagnosis of the embodiment is applied to the front surface of the material 9 to be welded. Good.

The welding abnormality diagnosis method according to the embodiment may be provided by reading the processing step of each flowchart executed by the welding abnormality diagnosis device 20 according to the embodiment as a method step.

9 Welded material, 9a leading material, 9b trailing material, 10 welding system, 12 welding head, 13 welded part front surface photographing camera, 14 welded part back surface photographing camera, 15 control device, 16 output device, 19 welded part back surface photographing camera Image, 20 Welding abnormality diagnostic device, 21 Welding part imaging camera image collecting unit, 22 Welding control information collecting unit, 23 Welding status diagnosis unit, 23A Diagnostic function block for each photographed image, 23B Diagnostic function block after welding completion, 23C Welding tendency diagnosis Functional block, 24 Welding abnormality judgment upper and lower limit table, 25 Welding abnormality alarm unit, 26 Welding status output unit, 27 Welding control information database, 28 Welding status information database, A, C, D Welding light feature quantity, C ^d , D ^{d _statistics,} A _{err1, C d} err1 welding abnormality determination upper ^_limit, A _{err2, C d} err2 welding abnormality judging lower _^limit, A _{wrn1, C d wrn1} welding error warning _^limit, A _{wrn2, C d wrn2} welding abnormality warning lower limit , _{B d1} weld _{^{pool, D m1, D d mx1,}} UCL i upper control _{^{limit, D m2, D d mx2,}} LCL i lower control ^{limit, F} d r image acquisition sections, FRMCNT number of images _{identifiers, L} ^{C d} regression line, L _w1 welding light, Q _k , Q _k-1 gradient, x width direction pixel position, y direction pixel position

Claims (10)

1. A camera system that captures the back side of the material to be welded,
   A welded part photographing camera image collecting unit that collects luminescence photographed images taken by the camera system at the time of welding the material to be welded
   Welding control information collection unit that collects welding condition setting values,
   Welding that calculates the welding light feature amount for diagnosing the welding condition of the material to be welded based on the luminescence photographed image and the welding condition set value, and determines the quality of welding based on the welding light feature amount. Situation diagnosis department and
   Welding abnormality diagnostic device equipped with.
2. The welding status diagnosis unit includes a diagnostic means for each photographed image.
   The diagnostic means for each photographed image calculates the welding light feature amount from moment to moment from the light emission photographed image obtained from the welding part photographing camera image collecting unit in one welding process, and the welding. The welding abnormality diagnostic apparatus according to claim 1, wherein the quality of welding is determined based on the amount of optical features and the upper and lower limits of welding abnormality determination.
3. The diagnostic means for each photographed image calculates the welding light feature amount from moment to moment from the light emission photographed image obtained from the welding part photographing camera image collecting unit in one welding process, and determines in advance. The welding abnormality diagnostic apparatus according to claim 2, wherein the quality of welding is determined by comparing the gradients of the welding light feature amount within the period.
4. The diagnostic means for each photographed image corrects the welding condition setting value based on the momentary welding light feature amount obtained from the momentarily emitted light emission image obtained from the one welding portion photographing camera image collecting unit. The welding abnormality diagnostic apparatus according to claim 2.
5. The welding status diagnosis unit includes a post-welding diagnostic means.
   The welding according to claim 1, wherein the post-welding diagnostic means calculates a statistic from a plurality of the welding light feature quantities, and determines whether the welding is good or bad based on the statistic and the upper and lower limit values for determining welding abnormality. Abnormality diagnostic device.
6. The welding abnormality diagnostic apparatus according to claim 5, wherein the welding abnormality diagnosis means determines the upper and lower limits of the welding abnormality determination by using a statistical method based on the statistic.
7. The welding abnormality diagnostic apparatus according to claim 5, wherein the welding abnormality diagnosis means determines the upper and lower limits of the welding abnormality determination by using machine learning based on the statistic.
8. The welding status diagnosis unit includes a welding tendency diagnosis means.
   The welding tendency diagnosing means is described in claim 1, wherein a statistic is calculated from a plurality of the welding light feature quantities, and the quality of welding is determined based on the tendency of the statistic and the upper and lower limits of welding abnormality determination. Welding abnormality diagnostic device.
9. The welding abnormality diagnosing device according to claim 8, wherein the welding tendency diagnosing means determines the welding abnormality determination upper and lower limit values by using a statistical method based on the statistic.
10. The welding abnormality diagnosing device according to claim 8, wherein the welding tendency diagnosing means determines the welding abnormality determination upper and lower limit values by using machine learning based on the statistic.

Images