JP2742232B2 - Arc welding machine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an arc welding machine, and more particularly to a welding power source.

[0002]

2. Description of the Related Art In a conventional technique of an arc welding machine, welding output control is performed by selecting an output value from a fixed range according to a predetermined rule such as differential amplification according to a fed-back arc state. I was going.

[0003] Further, in an electric discharge machine, there is one disclosed in Japanese Patent Application Laid-open No. Hei 2-303718 as one in which machining conditions are changed by fuzzy control based on a discharge state.

[0004]

However, when welding output control is performed by selecting an output value from a fixed range in accordance with a predetermined rule according to the arc state fed back as described above, feedback is given. The output signals selected for the signals have a one-to-one relationship, and an optimum output signal for stabilizing the arc at that time is not necessarily selected, which is a factor that causes the arc to become unstable.
For example, when the gain is low, it is not possible to respond quickly to a change in the arc state, and when the gain is set high, the arc state may be rather unstable due to overshoot.

[0005] Further, the one described in the above-mentioned Japanese Patent Application Laid-Open No. Hei 2-303718 is described in the upper right column, line 16 to line 1 of the second page.
As described in line 7, "In general, to increase the processing speed, it is considered that efficient processing is performed by flowing the maximum processing current within a range that does not shift to arcing." The control is performed from the viewpoint of not generating an arc, and the control is performed from a completely different viewpoint from the viewpoint of keeping the arc stable in an arc welding machine.

SUMMARY OF THE INVENTION In view of the above-mentioned conventional problems, the present invention performs an inference based on a plurality of feedback signals based on fuzzy logic and outputs an output signal necessary for realizing an optimum arc state. The purpose is to provide.

[0007]

According to the present invention, there is provided an arc welding machine comprising: an output control unit for outputting an arc according to an output control signal; and at least one of a signal corresponding to an output arc voltage value and a signal corresponding to an output arc current value. One signal and short circuit times
An output detection unit that outputs the number as a feedback signal,
A fuzzy inference control unit that forms a welding output waveform necessary for realizing an optimum arc state by using a feedback signal as an input by fuzzy inference and outputs the output control signal to the output control unit as an output control signal.

[0008]

According to the present invention, the basic output signal, the signal corresponding to the output arc voltage value, and at least one of the signals corresponding to the output arc current value are short-circuited by the above configuration.
Inference using fuzzy logic is performed on the basis of the number and the output waveform can be controlled so as to obtain an optimum output arc state.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS.

In this embodiment, the description will be made with the consumable electrode arc welding machine in mind, but it goes without saying that the present invention is not limited to the consumable electrode arc welding machine.

In the block diagram shown in FIG. 1, 1 is a fuzzy inference control unit, 2 is an output control unit, and 3 is an output detection unit. The FBS outputs the fuzzy inference control unit 1 from the output detection unit 3.
Is an output arc state feedback signal output toward the output control unit 2 and CS is an output control signal output from the fuzzy inference control unit 1 to the output control unit 2.

The output detection unit 3 observes and measures the state of the arc currently output from the output control unit 2 which is an output drive unit of the welding machine, and outputs the output arc state feedback signal F.
Output as BS. The fuzzy inference control unit 1 receives the output arc state feedback signal FBS signal as an input,
According to a preset membership function, fuzzy inference is performed to form an output waveform realizing an optimal arc state, and an output control signal CS is set and output. The output control unit 2 controls driving based on the output control signal CS.

2, the fuzzy inference control unit 1 includes a fuzzy inference unit 11 and a basic output setting unit 1.
2, and the output detection unit 3 includes a short-circuit frequency measurement unit 31 and a signal conversion unit 32.

The signal converter 32 takes in the arc current value and the arc voltage value currently output from the welding machine as the output arc current signal Ao and the output arc voltage signal Vo and converts them into signals, and outputs the output arc current value signal AS and the output. Output as the arc voltage value signal VS.

The number-of-short-circuits measuring section 31 monitors the arc voltage with the output arc voltage signal Vo as an input, determines that the voltage has fallen below a predetermined voltage as a wire short-circuit, and determines the number of short-circuits within a predetermined time. Is counted and output as a short circuit count signal SS.

The basic output setting section 12 selects a welding output current waveform and a voltage waveform, which are output basic waveforms for a welding output current and voltage value commanded from the outside, from a memory data bank and generates a basic output signal KS. Output.

The fuzzy inference unit 11 outputs the basic output signal KS
And a short circuit frequency signal SS, an output arc current value signal AS, and an output arc voltage value signal VS. The short circuit frequency signal SS, the output arc current value signal AS, and the output arc voltage value signal VS are substituted into membership functions. The basic output signal KS is finely adjusted based on the result, and is output as an output control signal CS for controlling an output value so as to realize an optimal arc state.

As a membership function for the short-circuit count signal SS, the output arc current value signal AS, and the output arc voltage value signal VS, for example, as shown in FIG.
B (Negative Big), NS (Negative Small), ZO
(Zero), PS (Positive Small), PB (Positive B)
ig). In this case, the membership function μ (x) of the fuzzy variable NB is expressed as follows: x ≦ b ₁ : μ (x) = 1 b ₁ ≦ x ≦ b ₂ : μ (x) = 1− (x−b ₁ ) / ( b ₂
−b ₁ ) b ₂ ≦ x: μ (x) = 0 Membership function μ of fuzzy variables NS, ZO, PS
(X) is given by μ (x) = max (0, 1− | x−b _n | / (b _{n + 1} −
b _n )) (n = 2, 3, 4) The membership variable μ (x) of the fuzzy variable PB is: x ≦ b ₄ : μ (x) = 0 b ₄ ≦ x ≦ b ₅ : μ (x) = (X−b ₄ ) / (b ₅ −b
₄₎ b ₅ ≦ x: given respectively μ (x) = 1.

The rules of the fuzzy inference are as follows:
The following three-input / one-output rule groups described in IF to (consequent part) THEN (consequent part) are set based on the rules of thumb of a skilled welding operator.

IF AS = NS and VS = ZO and SS = PS THEN C ₁ S = A1 IF AS = ZO and VS = ZO and SS = PS THEN C ₂ S = A2 IF AS = NB and VS = ZO and SS = PS THEN C ₃ S = A3 IF AS = NS and VS = NS and SS = ZO THEN C ₄ S = A4 IF AS = ZO and VS = NS and SS = ZO THEN C ₅ S = A5 IF AS = NB and VS = _{NS and SS = ZO tHEN C 6} S = A6 ····· ·· IF ····· tHEN C n S = An fuzzy inference adopted in this embodiment, well-known which is widely in domestic electrical appliances, etc. Is a simple fuzzy inference. That is, the consequent parts A1 to A6... Of the above rules are constant values, and regardless of the magnitude of the membership values obtained for the short circuit count signal SS, the output arc current value signal AS, and the output arc voltage value signal VS. not (of course, 0 is excluded), C _n S values of all the rules obtained membership values are involved, pick up A1~An its center of gravity, that is, the average value determined, the fuzzy inference that value Part 11
, The basic output signal KS is adjusted as the output value, and is output as the output control signal CS. Therefore, it is not necessary to calculate a membership value for each input signal by a membership function. Actually, a pickup table is provided for obtaining a C _n S value of a rule involved by the value of each input signal. The output control signal CS is calculated based on the output and output.

The fuzzy inference unit 11 and the basic output setting unit 12 can be easily realized by using a microprocessor or the like, and the short circuit frequency measuring unit 31 can be easily realized by using a comparator and a counter IC.

In the fuzzy inference unit 11 of the above embodiment, a simple fuzzy inference using a pickup table with the consequent part of the rule as a constant value is applied. It goes without saying that regular fuzzy inference as a function may be applied. In this embodiment, as one element of the fuzzy inference,
Although the number of short circuits is taken into account, the elements of fuzzy inference are not necessarily limited to those of the present embodiment, but the essential point is that any element capable of inferring an output waveform so as to obtain an optimal output arc state can be used. .

[0023]

As described above, according to the arc welding machine of the present invention, depending on the state of the output arc,
The optimum output waveform to be fed back can be inferred.
It is possible to realize a stable and optimal arc, and an operator can easily perform an optimal arc welding, thereby exerting a great industrial effect.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a block diagram showing a fuzzy inference control unit and an output detection unit in more detail.

FIG. 3 is an explanatory diagram of a membership function.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Fuzzy inference control part 2 Output control part 3 Output detection part 11 Fuzzy inference part 12 Basic output setting part 31 Short circuit frequency measurement part 32 Signal conversion part

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Junki Nishida 1006 Kazuma Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (56) References JP-A-2-303718 (JP, A) Kenji Oshima et al. "Application of Fuzzy Control to Pulsed MIG Welding," Proc. Of the National Meeting of the Japan Welding Society, Vol. 46 (1990), pp. 160-161

Claims (1)

(57) [Claims] An output control unit for performing arc output in accordance with an output control signal, and outputs, as a feedback signal, at least one of a signal corresponding to an output arc voltage value and a signal corresponding to an output arc current value and a number of short circuits. An output detection unit, and a fuzzy inference control unit that forms a welding output waveform necessary for realizing an optimum arc state by using a feedback signal as an input by fuzzy inference and outputs the output control signal to the output control unit as an output control signal. An arc welding machine characterized by the following.