JPH0659545B2 - Automatic welding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention relates to an automatic welding method, and is applied to all welding optical products (middle and thick plates).

[Conventional technology]

Conventionally, the method shown in FIG. 8 is known as an automatic welding method. In the figure, 1a and 1b are base materials forming the groove 2. A rail 3 is laid on the base material 1, and a welding carriage 4 travels on the rail 3 in the welding direction (arrow X). A controller 6 having a sensor 5 arranged in front of the carriage 4 and a welding torch drive system 8 having a welding torch 7.
Is placed. That is, the welding according to the prior art was obtained by oscillating the sensor 5 in the direction Y) perpendicular to the welding line to detect the groove cross-sectional shape, superimposing several cross-sections, and averaging. Welding is performed by obtaining the groove center position and groove cross-sectional area based on the information and performing welding line tracing and welding condition control.

[Problems to be solved by the invention]

However, according to the prior art, since the torch position can only be controlled with the groove center, it has only been used for one-pass finish welding of a relatively thin plate V-shaped groove. Therefore, it is unsuitable as control information for multi-layer welding (becomes an overlapping portion of the intersection of the groove wall at the torch target position and the bead). That is, in the prior art, an information processing method for satisfying the high-level construction technique required for multi-layer distribution welding has not been established. Therefore, the multi-layer distribution welding is
There is a problem that the operator's experience is unavoidable, and there is a demand for higher quality and higher efficiency by automation due to lack of skilled workers.

The present invention has been made in view of the above circumstances, and provides an automatic welding method capable of performing multi-layer welding by detecting groove information of a medium / thick plate structure with a sensor to obtain groove information with high accuracy. The purpose is to do.

[Means for solving problems]

According to the present invention, a step of averaging output signals of an optical distance sensor to form a shape signal, a step of subjecting the shape signal to multiple averaging processing, and a step of differentiating the signal subjected to the multiple averaging processing to obtain characteristic points. And a step of integrating the signals subjected to the multiple averaging process to obtain a groove cross-sectional area, and a step of controlling the torch position and welding conditions based on the information of the characteristic points and the groove cross-sectional area. It enables accurate multi-layer welding by obtaining accurate groove information.

[Action]

ADVANTAGE OF THE INVENTION According to this invention, the highly accurate groove information required for setting the torch aiming position and welding conditions from the 1st pass of a multilayer welding to the last pass can be provided, thereby enabling automatic control of a multilayer welding, and a hoist. It can be applied to all medium and thick plate welded structures such as products and chemical machines.

〔Example〕

 An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram schematically showing the control contents of a multi-layer distribution welding automation system incorporating the automatic welding method according to the present invention. Reference numerals 11a and 11b in the figure are base materials forming the groove 12. A laser light sensor (hereinafter referred to as a laser light sensor) 13, which is one example of the optical distance sensor, and a welding torch 15 connected to a motor 14 are arranged on the base materials 11a and 11b. In the system shown in FIG. 1, the distance signal obtained from the laser light sensor 13 and the sensor position signal at the time of rocking are combined, and the obtained two-dimensional information is subjected to the processing of the present invention. Weld shape A is obtained. Then, based on this information, the in-groove feature point and the unwelded cross-sectional area are obtained, and the appropriate torch position B is calculated by calculation with a preset arithmetic expression or by comparison with a reference value.
By controlling the welding conditions, multi-layer welding can be automated.

FIG. 2 is a diagram showing a torch position automatic control test result of multi-layer distribution welding using the first system. That is, first, after detecting the shape after the two-pass welding as shown in FIG. 2 (a), this groove information is processed to obtain 3 as shown in FIG. 2 (b).
Find the target position for the pass. After 3 passes of welding,
From the detection of the cross-sectional shape shown in FIG. 6C, the target position of the fourth pass shown in FIG. In this way, multi-layer welding is performed by sequentially finding the target position of the pass from the detection of the cross-sectional shape.

FIG. 3 is one of the groove information processing methods according to the present invention, and shows a method for averaging sensor signals and tackling them. In the figure, the data on the left is the groove shape when the sensor data is taken in without averaging (only half is shown).
Therefore, the effects of variations in the reflectance of the measurement surface are directly visible. On the other hand, the data on the right side of the figure has a large number of averaging processes, and has a stepwise shape due to the acquisition of the same data. Since such data is different from the actual groove, it is not possible to accurately detect the feature point in the groove, which also indicates that proper averaging processing is necessary. In addition, the data in the center of the figure is groove data that has been properly processed, and the shape as the actual groove was obtained. In addition, in FIG. 3, the sensor data was taken every 0.1 mm. The average number of times of processing is the data output frequency (10 KHz) divided by the sampling frequency.

FIG. 4 shows raw data before averaging the sensor signal (distance signal) according to the present invention and data (bar graph) obtained by differentiating the raw data. It is apparent from FIG. 4 that the shape of the measurement surface (groove) is varied due to scratches, spatter, etc., and the characteristic points cannot be detected by the differential data. On the other hand, FIG. 5 shows the multiple averaging process data and the differential process data after the averaging process. From the figure, it is clear that the differential processing data removes the above-mentioned noise, and the detection of the characteristic point P becomes easy and reliable. Here, the characteristic point P is determined by, for example, the height of the bar graph being increased or decreased by a certain amount.

FIG. 6 is a groove cross-sectional view (unwelded cross-sectional area) by a microcomputer.
It is a figure which shows the detection result of. The unwelded cross-sectional area serves as a reference for controlling the application of welding conditions, and as shown in FIG. 7, groove characteristic points P ₁ and P ₂ obtained by performing the processing shown in FIGS.
Is connected by a straight line, and the area S of the portion surrounded by the characteristic points P ₁ and P ₂ and the groove 12 is calculated from the sensor data. That is,
From the unit area Δa = 1 × d = 1 × 0.1, Becomes From FIG. 6, the microcomputer calculated value and the value obtained by actually cutting the welded portion and calculating the area were in good agreement, and the reliability as control data was also confirmed.

FIG. 9 is a diagram showing a groove information processing main flow according to the present invention.

First, when the control is started, a "standby" state is set. Here, when the "sensor swing command from the main CPU" is input, the "sensor swing" occurs. At this time, it is judged whether or not it has moved within a preset range by "whether it has moved by a predetermined amount".
It becomes a state. On the contrary, in the case of YES, "X / Y data acquisition" is performed and this is stored as groove information. Next, in order to check the reliability of this groove information, it is judged whether "in the sensor measurement range" or "is the received light amount appropriate?"
In the case of O, the data is invalid and the state is "standby". YE
In the case of S, information processing (calculation) for control is continued. The first process is "calculation / detection of feature points", and the content of the process is the differential process described in FIGS. 3 to 5 (however, one or two It is also possible in combination). Note that the calculation / processing of the characteristic points is based on the change points of the differentially processed data described in the above figures. Assuming that the sensor is tilted and used, "whether angle correction is necessary" is added, and if YES, "angle correction of feature point data" (using trigonometric function) is performed and the process continues. In the case of NO, as the second processing, “calculation / detection of cross-sectional area” is performed. The calculation and detection of the cross-sectional area is based on the integration method described in the figure. The content is the integration processing shown in FIG. Finally, the groove information (feature points within the groove, cross-sectional area) obtained in the first and second processings is summarized as "Main CP
Data transfer to U ”is performed and used as data for welding torch position control and welding condition control of multi-layer welding by the main CPU.

As is apparent from the above procedure, according to the present invention, the groove shape is captured by the groove sensing (sensor) by receiving only one simple command, and the information necessary for control is obtained from the enormous data obtained. It is possible to automatically obtain multi-layer welding by applying the present invention.

The present invention is also applicable to the case where the end shape of a steel plate or the like is detected and a groove suitable for the end shape is processed by gas cutting or the like.

〔The invention's effect〕

As described in detail above, according to the present invention, it is possible to provide an automatic welding method capable of performing multi-layer welding by detecting groove information of a medium / thick plate structure with a sensor and obtaining groove information with high accuracy.

[Brief description of drawings]

FIG. 1 is an explanatory view of an automatic system for multi-layer distribution welding of thick plates to which the present invention is applied, FIG. 2 is a view showing a torch position automatic control test result of multi-layer distribution welding using this system, and FIG. 3 is the present invention. FIG. 4 is a diagram showing the relationship between the number of averaging processes and output data according to the present invention, FIG. 4 is a diagram showing raw data and differential process data before the multiple averaging process according to the present invention, and FIG. 5 is a multiple average after the multiple averaging process. FIG. 6 is a view showing the grooved cross sectional area detected by the microcomputer, FIG. 7 is an explanatory view for obtaining the groove cross sectional area, and FIG. 8 is a conventional automatic welding method. And FIG. 9 is a main flowchart of the groove information processing according to the present invention. 11a, 11b ... Base metal, 12 ... Groove, 13 ... Laser light sensor, 14 ... Motor, 15 ... Welding torch.

Claims (1)

[Claims] 1. An automatic welding method in which an optical distance sensor is swung in a direction substantially perpendicular to a welding line to detect the position and shape of a groove to perform welding, and the output signals of the optical distance sensor are averaged. To form a shape signal, multiple averaging the shape signal, differentiating the multiple averaged signal to obtain a feature point, and integrating the multiple averaged signal to open it. An automatic welding method comprising: a step of obtaining a tip cross-sectional area; and a step of controlling a torch position and welding conditions based on the information of the characteristic points and the groove cross-sectional area.