JPH08297B2 - Automatic arc welding method by real-time control - Google Patents

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention belongs to the field of arc welding, and relates to an automatic arc welding method by real-time control.

[0002]

2. Description of the Related Art The conventional automatic welding method based on the arc sensing method is applicable only when there is no deviation in the groove shape variation in the work, only when there is deviation in the welding line, and when the deviation in groove variation is uniform. However, it is difficult to control all of the welding line copying control, the torch height control, and the bead height constant control in real time.

[0003]

SUMMARY OF THE INVENTION In view of the above, the present invention provides an automatic arc welding method in which the welding reference value is corrected in real time according to the situation of the work and the bead height is constant.

[0004]

Means for Solving the Problem The present invention detects and calculates a positional deviation of a welding start point and a groove shape variation by a detection means, and weaves the weaving width and a welding speed with respect to a reference value based on the calculation result. Start welding operation in the arc welding method that corrects the input values of width and welding speed
Immediately after, the arc current of the current detected by the arc sensor
Real-time data to calculate the value, the upper and lower welding torch, calculating the lateral direction compensation reference value and the weaving width correction reference value, corresponding to these reference values and the reference values hereinafter them as initial setting value (reference value) by comparison with, Wibi
The welding bead height is constantly controlled by correcting the welding width and the welding speed.

[0005]

[Operation] The welding start point and welding end point of the welding robot are corrected by confirming the setting state of the workpiece, and the welding current data is sampled at the beginning of welding while performing the welding operation based on the setting program. By calculating the vertical and horizontal correction reference values of the welding torch and the weaving width correction reference values, and then comparing each current integral value of the real-time data regarding the vertical and horizontal of the welding torch and the weaving width, which are input moment by moment, with each of the above reference values. , Each correction value is calculated by a predetermined correction function, and the vertical and horizontal directions of the welding torch, the weaving width, and the welding speed are corrected in real time.

[0006]

EXAMPLES The present invention will be described by way of examples. As shown in FIG. 1, a welding torch 1 is an arc generated by a voltage applied by a welding power source 4 between a welding wire 2 and a base material 3 to be fed. The base material 3 is arc-welded by means of the arc welding. The arc current generated at this time is detected by the arc sensor unit 5, and the data of the arc current is sent to the microprocessor 6, and the microprocessor 6 is stored in the storage device 7. The storage data such as the correction reference value and the correction function are input, and the welding torch 1 is controlled by the driver unit 8 which operates according to the instruction of the microprocessor 6, such as the movement of the base material 3 along the welding line and the weaving operation. It

Hereinafter, in detail according to the control flowchart shown in FIG. 2, as shown in FIG. 3 in step 15,
The position deviation of the welding start point 9'on the actual base material 3 on the jig with respect to the welding start point 9 on the program is detected by means such as "touch sensor" or "image processing" (not shown), and the same is done at step 16. Dimensional variation of the groove is detected by the means, and in step 17, the welding start and end point positions and the initial welding speed which are taught in advance are corrected based on the detection data. Next, in step 18, the welding is started while detecting the welding current by the arc sensor unit 5, and in step 19, as shown in FIG. 4, between the path ABC and the EFG (IJK, Data sampling is performed by using the welding currents of the end point portion 10 and the root portion 11 during the MNO (repeating like below) as current integrated values as shown in FIG. Here, the reason why the sampling is performed only from the end point portion 10 to the route portion 11 is that the welding current waveform ripple is small and stable sampling data can be obtained, as is clear from FIG.

Next, in step 20, as shown in FIG.
The current value between the weaving path ABC on the left side and the path IJK on the next cycle is integrated and the average current integrated value S _OL is used as the left torch lateral correction reference value, and even between the weaving path EFG on the right side and the path MNO on the next cycle. The same calculation is performed, and the average current integrated value S _OR is used as the right side torch horizontal direction correction reference value.

In step 21, as shown in FIG. 7, the weaving end point AA ′ and the root portion CC ′ on the left side and the weaving end point II ′ and the root portion K′K of the next periodic path are integrated, and their respective average current integrated values are obtained. Respectively S _IL ,
S _2L , the weaving end point EE ′ and the root portion G′G on the right side, and the weaving end point MM ′ of the next periodic path and the root portion O′O are similarly integrated and their respective average current integral values are respectively set to S _{1R. ,} and S _2R, in which the above average current <br/> sum _{S oB (S oB = S 1L} + S 2L + S 1R + S 2R) the vertical correction reference value of the torch of the integral value.

In step 22, as shown in FIG. 8, the respective average current integral values between the weaving end point AA ′ and the root portion C′C on the left side and between the weaving end point II ′ and the root portion K′K of the next periodic path are respectively calculated. Let S _1L and S _2L be the right side weaving end point EE ′ and the root part G′G, and the mean current integral values between the weaving end point MM ′ of the next periodic path and the root part O′O be S _{1R and} S _{2R ,} respectively. ,
Average of sum of left and right integrated deviations, that is, average current integrated value S
_{OW (S OW = {| S} 1L -S 2L | + | S 1R -S 2R |} / 2)
Is the weaving width correction reference value.

Next, in step 23, step 2
The left side torch left / right correction reference value S _OL calculated and stored at 0 is compared with the current integrated value S _nL corresponding to the above S _OL between the left side paths abc in real time, and in step 24, S _OL = S _nL
If there is no correction, if S _OL > S _nL , to the left, S
_{When OL} <S _nL, the correction is performed rightward according to the correction amount calculated by the correction function shown in FIG.

In step 25, the calculation in step 20,
Wherein comparing the current integral value S _nR corresponding to S _OR between the stored right torch lateral direction correction reference value S _OR and real-time right path efg, in the case of S _OR = S _nR no correction, S _OR
> S _{nR to} the right, S _OR <S _{nR to} the left, and in step 26, the torch vertical correction reference value S _OR calculated and stored in step 21 and the real-time path end point aa. ', ee', ii ', mm' and the root section C'c, compared g'g, k'k, the average current integral S _nB corresponding to the S _OB between o'o, S _OB = for S _nB no correction, in the case of S _OB> S _nB downward direction, S _OB <
Upward direction in the case of S _nB, according to the correction amount calculated by the respective correction function at each step 27, the left-right direction in the vertical direction correction simultaneously (base-vector synthesized with) performed.

Next, in step 28, the weaving width correction reference value S _OW calculated and stored in step 22, the real-time integrated deviation Δac between the weaving end point aa ′ and the root portion c′c, and similarly the right weaving end ee. ′ And the root part g′g, the average current integral value S of the integral deviation Δeg
Compare with _nW, and if it exceeds “threshold ew”,
At step 29, cancel the vertical and horizontal corrections, and _click S _OW >
In the case of S _nW , the weaving width is small, in the case of S _OW <S _nW , the weaving width is large. The weaving width is corrected according to the correction amount calculated by the correction function, and the weaving width is corrected until it becomes equal to or less than the threshold value ew. The welding speed is also corrected from the data.

After repeating the correction completion a specified number of times in step 30, the real-time correction is interrupted and step 3
At 1, the vertical and horizontal corrections are restarted again, and the correction is continued with a constant correction amount until the weaving width and the welding speed exceed the threshold value in the final correction data of the weaving width.

If it is determined in step 32 that the ending condition is satisfied, the process ends in step 33. If it is determined that the end condition is not satisfied, the detection and correction by the arc sensor are repeated until the end condition is satisfied.

In the correction function used here, the correction amount increases in accordance with the amount of difference between the correction reference value and the real-time value,
In addition, it is considered that the correction is promptly performed while the shift amount is small. 1 / n 1 / n For example, ΔC = G × {(S / S _O ) −1}: S> S _O 1 / n 1 / n ΔC = G × {(S _O / S) −1 }: In the case of S <S _O Here, ΔC: correction amount S: real-time welding current integration value S _O : welding current reference integration value n: parameter setting value (n = 2 to 4) G: gain The correction function to be satisfied is the optimum, and FIG. 9 shows the correction function when n = 1 and 2 with respect to G = 1.

[0017] Real-time weaving width and welding speed relationship is more determined constant quantity correction by the real-time correction and final correction data. For example, W _n = W _o + ΣΔC V _n = V _o × {W _o / (W _o + W _n )} where W _n : real-time weaving width by correction W _o : initial setting weaving width V _n : by correction Real-time welding speed V _o : Initial setting welding speed The weaving width is corrected and the corresponding welding speed is corrected by the formula.

[0018]

According to the present invention, the base point of the work on the jig is automatically confirmed, and the deviation correction at the left and right ends of the welding torch, the deviation correction in the vertical direction, and the regular operation are started while welding is started. Since the weaving width deviation can be corrected in real time with respect to the groove dimension variation or the irregular groove dimension variation, and the welding speed compensation corresponding to the variation can be performed in real time, a weld bead shape with a uniform weld height dimension can be guaranteed. effective.

[Brief description of drawings]

FIG. 1 is a block diagram of an arc sensor control according to an embodiment of the present invention.

FIG. 2 shows a control flowchart.

FIG. 3 shows a welding pattern at the position where the reference position taught by the welding robot is corrected to the actual position of the work on the jig.

FIG. 4 is a current waveform diagram during one reciprocating weaving from A to 1 on the welding correction pattern.

5 shows current integration values at left and right data sampling positions (end point portion and root portion) in the weaving path of FIG.

FIG. 6 shows respective average current integral values S _OL and S _{OR which are} reference values for lateral correction of the torch at the left and right ends of the weaving path.
Show.

FIG. 7 shows average integration values S _1L , S _2L , S _1R , and S _2R for calculating correction reference values in the vertical direction of the torch at the left and right end points and the root portion in the weaving path.

FIG. 8 shows average integration values S _1L , S _2L , S _1R , and S _2R for calculating a weaving width correction reference value at each of the left and right end points and the root portion in the weaving path.

FIG. 9 shows an example of a correction function.

[Explanation of symbols]

 1 Welding torch 2 Welding wire 3 Base metal 4 Welding power source 5 Arc sensor unit 6 Microprocessor 7 Memory device 8 Driver unit 9 Welding start point 10 End point part 11 Root part

Claims (1)

[Claims] 1. A detecting a positional deviation and groove shape variation of the welding start point by the detecting means, and calculating, by the calculation result, U
In the arc welding method, in which the input values of the weaving width and welding speed are corrected with respect to the standard values of the weaving width and welding speed, the current value of the arc current detected by the arc sensor is calculated immediately after starting the welding operation , and the welding is performed. upper and lower torch, calculating the lateral direction compensation reference value and the weaving width correction reference value, they (as the reference value.) initial setting as subsequent latest welding corresponding to these reference values and the reference value
An automatic arc welding method by real-time control, characterized in that the weaving width and welding speed can be corrected and the weld bead height can be controlled to a constant value by comparing with the current value (hereinafter referred to as real-time data ).