KR19990070430A - Welding line tracking device and method of arc welding robot using fuzzy inference - Google Patents

Abstract

The present invention relates to a welding line tracking device and method of the arc welding robot using fuzzy inference,

In the present invention, the arc welding robot apparatus for welding a steel structure including a welding power source, a gas, a touch unit, and a robot body and a robot controller, the robot controller comprising a main CPU board and an input / output board. The welding line tracking board is installed to control the welding torch by controlling the position and trajectory of the welding torch. The welding line tracking board includes an amplifier for amplifying a welding current detection value and a low pass filter through which the amplified welding current value passes. And analog / digital converters that convert the analog values passed through the low pass filter to digital, the CPU processing the digitized information, and the rate of change of the current error and error, which are control variables, from the detected welding current values. And fuzzy, define fuzzy set and membership function, and build rule base by designer Basis by a seam tracking apparatus and method of arc welding robot using fuzzy inference to be made up of software with an algorithm to perform deeper ground cover indications (Defuzzification) to change it back to the physical quantity after performing a fuzzy inference.

Description

Welding line tracking device and method of arc welding robot using fuzzy inference

The present invention relates to a welding line tracking device and method for arc welding robots using fuzzy inference, in particular to the robot controller for controlling the welding robot is equipped with a welding line tracking board device for controlling the welding torch trajectory of the welding robot and operating the board Weld seam tracking method of welding torch using algorithm fuzzy inference.

In general, the characteristics of steel structures are large, heavy, and processing errors are large.

In order to automatically weld the steel structure with the arc welding robot, it is necessary to weld while tracking the welding line by detecting a machining error and a jig error of the base metal to be welded. Conventionally, a visual apparatus using a laser beam is used.

However, this method is not only expensive but also requires additional equipment for processing the laser beam scanning device, the CCD camera, and the image captured by the CCD camera. Image processing is difficult because the sparks such as and spatter are strong.

The present invention is to solve the above problems, by detecting the welding current during welding to infer the position correction value between the welding line and the welding torch by fuzzy inference and to control the position and trajectory of the arc welding robot An object of the present invention is to provide an apparatus and a method for tracking an arc welding robot that includes a machining error and a jig error.

In addition, another object of the present invention is a device capable of welding while welding the welding wire including the machining error and jig error of the base material by using an inexpensive and simple structure without using an expensive visual device and the arc welding robot; To provide a method.

1 is a schematic overall configuration diagram including an apparatus of the present invention

2 is a schematic diagram of the hardware of the present invention;

Figure 3 is a schematic diagram showing the process of the present invention

4 is a schematic diagram showing an arc welding process

Figure 5 is a schematic embodiment showing left / right and up / down correction

The construction, embodiments and functions of the present invention for achieving the object as described above and removing the conventional drawbacks will be described in detail with reference to the accompanying drawings.

1 is a schematic overall configuration diagram including the apparatus of the present invention,

2 is a schematic diagram of the hardware of the present invention;

3 is a schematic configuration diagram showing a process of the present invention;

Figure 4 is a schematic diagram showing an arc welding process,

5 is a schematic embodiment showing left / right and up / down corrections.

In the configuration of the present invention, the arc welding robot device for welding the steel structure provided with a welding power source and gas and a touch unit, and a robot body and a robot controller,

It is configured to control the position and trajectory of the welding torch by mounting a welding line tracking board to the robot controller consisting of a main CPU board and an input / output board,

The welding line tracking board includes an amplifier for amplifying a welding current detection value; Low pass filter through which the amplified welding current value passes, Analog / Digital converter for converting the analog value passed through the low pass filter into digital, CPU processing digitized information, Detected welding current Calculate and fuzzy change rate of control error, current error and error from values, define fuzzy set and membership function, perform fuzzy inference based on rule base constructed by designer, and then convert it back to physical quantity It consists of software with algorithms that perform defuzzification.

In the principle of the present invention, when arc welding while weaving to the left and right, the welding current during welding is detected because the length of wire protrusion increases as the distance between the base metal and the welding torch increases and the resistance heat (I ² R) increases slightly. By calculating the distance inversely proportional to the current, the center point of the weaving is controlled to be located at the center of the base materials A and B. Since the welding current detected during the actual welding is not uniform due to the influence of wave control on the welding power supply and the factors such as spatter, it should be stabilized by using a low pass filter, and the distance between the base material and the welding torch is calculated from the welding current. In order to achieve this, there are many nonlinear factors between current and distance, so the method by fuzzy reasoning is effective.

The fuzzy inference algorithm used in the welding line tracking board apparatus is as follows.

For the left and right correction, the welding current is detected and stored when the torch reaches the A position in FIG. 4, and the welding current is detected when the torch reaches the B position, and current errors and errors at the A and B positions are detected. Calculate the rate of change, add left / right control inputs to the robot controller, configure the control variables, multiply the scaling factors by each, and fuzzy variables to define fuzzy sets and membership functions, Fuzzy inference is performed based on the rule base established by the designer, and physical left / right position correction amount is calculated by defuzzifying the output value.

For the up / down correction, the welding current is detected and stored when the torch reaches the A position in FIG. 4, and the welding current is detected when the torch reaches the B position, and the current values at the A and B positions are averaged. Calculate the error with the reference current value and the rate of change of the error, add up / down control inputs to be supplied to the robot controller, configure the control variables, multiply the scaling factors by each, and make fuzzy parameters to multiply them. Define a membership function, perform fuzzy inference based on the rule base established by the designer, and calculate physical up / down position correction by defuzzifying the output value. do.

The operation of the present invention as described above is as follows.

As shown in FIG. 4, a welding line tracking device of an arc welding robot using fuzzy inference in an arc welding robot generating arcs on a base metal and a welding torch to weld steel structures by performing welding along a welding line, and As shown in FIG. 5, welding can be performed along the center of the base material A and the base material B in the process of performing left / right correction and up / down correction as shown in FIG. 5, thereby enabling arc welding while tracking the direction of the welding line. .

The present invention as described above is a welding line tracking device and method of the arc welding robot using fuzzy inference by detecting the current in the welding without adding a separate device such as a visual device to automatically weld the steel structure to the arc welding robot. In addition, since the welding line including the machining error and the jig error of the base material can be welded by the left / right correction and the up / down correction, the welding quality can be obtained irrespective of the machining error or the jig error.

Claims (2)

In the arc welding robot device for welding a steel structure having a welding power source, a gas and a touch unit, a robot body and a robot controller, It is configured to control the position and trajectory of the welding torch by mounting a welding line tracking board to the robot controller consisting of a main CPU board and an input / output board, The welding line tracking board includes an amplifier for amplifying a welding current detection value; A low pass filter through which the amplified welding current value is passed; An analog / digital converter for converting the analog value passed through the low pass filter into digital; A CPU for processing the digitized information; From the detected welding current values, the current error and the rate of change of the control variables are calculated and fuzzy, the fuzzy set and membership function are defined, and fuzzy inference is performed based on the rule base constructed by the designer. The welding line tracking device of the arc welding robot using fuzzy inference, characterized in that the software is configured with an algorithm for performing a defuzzification (Defuzzification). Algorithm used for welding line tracking board device, Detects the welding current at the left and right of the weaving point for left and right correction, calculates the current error and the change rate of the error, adds the left and right control inputs to the robot controller, configures the control variables, and calculates the scaling index ( multiply each scaling factor into fuzzy variables to define fuzzy sets and membership functions, perform fuzzy inference based on rule bases established by designers, and output values calculating a physical left / right position correction amount by defuzzification; Detects the welding current at the left and right of the weaving point for up / down correction, calculates the error and rate of change between the average value and the reference current value, and adds up / down control input to supply to the robot controller. And fuzzy variable by multiplying scaling factors to each one to define fuzzy set and membership function, and perform fuzzy inference based on rule base established by designer Welding line tracking method of arc welding robot using fuzzy inference, characterized in that the calculation process calculates physical up / down position correction amount by defuzzification.