KR101593642B1 - Inverter dc resistance spot welding system, control method for welding process thereof and design method for fuzzy controller thereof - Google Patents

Abstract

The present invention relates to an inverter DC resistance spot welding system, a method of controlling the welding process thereof, and a method of designing the fuzzy controller, and a fuzzy controller for constant current control is designed without using a system identification model and simulation, The conversion coefficient can be optimized.

Description

TECHNICAL FIELD [0001] The present invention relates to an inverter DC resistance spot welding system, a welding method control method thereof, and a method of designing a fuzzy controller of the inverter DC resistance spot welding system.

The present invention relates to an inverter DC resistance spot welding system, a method of controlling a welding process thereof, and a fuzzy controller design method thereof, and more particularly, to an inverter DC resistance spot welding system capable of designing a controller with minimal experimentation, A welding process control method and a fuzzy controller design method thereof.

In general, resistance spot welding systems combine mechanical equipments and electrical equipments to control the welding current, welding force, and welding time required for welding. Thereby joining metals.

Resistance spot welding is divided into AC (alternating current) welding and DC (direct current) welding according to the characteristics of the welding power source to be used. The AC power source is a single-phase power source of 50 to 60 Hz, and a 50 to 60 Hz welding power source, which is the same as the main power source, is derived from the secondary side through a transformer. The AC welder controls the welding current by controlling the current waveform by firing a thyristor. The welding current is amplified by passing through the transformer and is led to a large current which can be welded.

 Recently, the DC resistance spot welder receives the spotlight because the manufacturing cost for manufacturing the DC resistance spot welder is reduced due to the miniaturization of the power device required for the inverter configuration such as the IGBT and the diode and the price drop, Because. In addition, the DC resistance spot welder can be finely controlled as compared with the AC resistance spot welder. The control of the AC resistance spot welder widely used in the industrial field uses the phase control of the thyristor, so it is possible to control 120 times per second. The AC resistance spot welder can not cope with the change in dynamic resistance of a short welding process, and a large amount of spatter is generated because of the large instantaneous heat input. The generated spatter has a disadvantage of deteriorating the welding quality and causing contamination around the welded portion.

On the other hand, DC resistance spot welding is controlled by PWM (Pulse Width Modulation) at 1 kHz, so 2000 control is possible per second. Therefore, the DC resistance spot welder can overcome the disadvantage of the AC resistance spot welder through fine control. And DC welding machine can save energy by suppressing current loss, and it is getting attention as environmental regulation and green IT technology in the future.

On the other hand, in the case of a resistance spot welding system in which a mechanical device and an electric device are complexly connected, a mathematical model capable of replacing an actual system is mainly used in order to analyze the dynamic characteristics of the system with respect to external input . The reason for this is that it is more efficient to use the mathematical model than the actual system because the control system design has to go through various processes such as selecting the proper control algorithm through analyzing the characteristics of the system and evaluating the characteristics thereof. In particular, when using large currents such as resistance spot welding systems, it is necessary to use a mathematical model in order to eliminate the risk factors that occur when the output current response of various system inputs attempted at the control system development stage is excessive. Do.

As a modeling method for a mathematical model, we approximate each element of the system using a govern equation based on the physical principle and then use a method of expressing it as a differential equation. However, since there are nonlinear elements in the welding circuit such as melting of metal, mechanical reaction to welding gun, switching of electric device, etc., it is difficult to express the welding system mathematically using the physical law and linearize it to be. In fact, the complexity of a system means that many assumptions about system behavior are required because information about all situations is unknown.

Thus, there is a growing need for a more efficient method for modeling resistance spot welding systems.

In addition, there is a need to develop an inverter DC resistance spot welding system that can overcome the disadvantages of conventional AC single-phase AC resistance spot welding.

Embodiments of the present invention provide an inverter DC resistance spot welding system that can compensate for the disadvantages of AC single-phase AC resistance spot welding.

An embodiment of the present invention provides a method of controlling a welding process of an inverter DC resistance spot welding system using embedded software.

Embodiments of the present invention provide a method of designing a fuzzy controller of an inverter DC resistance spot welding system using a system identification modeling technique.

Embodiments of the present invention provide a fuzzy controller design method for an inverter DC resistance spot welding system capable of designing a fuzzy controller in an inverter DC resistance spot welding system and adjusting a conversion coefficient of the fuzzy controller.

Embodiments of the present invention provide a method for designing a fuzzy controller of an inverter DC resistance spot welding system capable of at least experiments without many prior experiments.

According to an aspect of the present invention, there is provided an inverter DC resistance spot welding system comprising: a power conversion unit for generating a welding current and controlling a current amount or a welding time of the welding current; A welding transformer for generating a large current to be delivered to the workpiece; A welding part having a welding gun for transmitting an electrode pressing force required to the welding part of the workpiece and a pneumatic device for generating the electrode pressing force; And a welder controller for controlling driving of the welded portion, wherein the power converter includes a rectifier diode for rectifying three-phase AC power; A capacitor for smoothing the rectified power supply to generate a DC waveform; And an inverter for converting the DC waveform generated in the capacitor into an AC having a predetermined pulse width.

By constituting as described above, it is possible to realize an inverter DC resistance spot welding system which can compensate for the disadvantages of the conventional AC resistance spot welding and can save energy.

The inverter is formed of a full bridge inverter using four IGBTs, and two IGBTs of the four IGBTs are combined to form two pairs of IGBT modules. A PWM pulse is applied to one of the two pairs of IGBT modules The output of the full bridge is positive and the output of the full bridge may be negative when a PWM pulse is applied to the other module.

In this manner, when the PWM pulses are alternately applied to the two pairs of IGBT modules, the output of the full bridge can be an AC waveform.

The AC waveform is converted into a DC waveform through the welding transformer, the rectifier diode, and the weld.

According to another aspect of the present invention, there is provided a method for controlling a PWM output, the method comprising: a first process for adjusting a PWM output to be a constant current, an electrostatic force, a constant voltage, or a heat quantity; A second process for monitoring voltage or current; A third process of controlling a welding press force applied to the workpiece; A fourth process for performing intelligent control; A fifth process of managing the welding process; A sixth process of performing display or crisis management; And a seventh process of communicating with a computer terminal (PC).

By controlling the welding process using the embedded software installed in the resistance spot welding system as described above, the welding process can be precisely controlled and the user can be prevented from being injured during the welding process, have.

The first process is a process of fuzing control of the constant current, the constant voltage, the constant electric power or the fixed amount by a command value selection of the computer terminal.

The second process is the process of storing the measured voltage or current from the ADC of the controller during welding to a specified location and processing the measurement data using a digital filter.

The third process is a process of applying the welding pressing force to the electrode of the welding gun.

The fourth process is a process of predicting the welding result using the result measured during welding and changing the current and the pressing force command value during welding.

The fifth process is a process of managing pressurization, stabilization time, energization time, or retention time of welding, and coping with the user's emergency situation.

The sixth process is a process of displaying the command value inputted by the user and displaying the state of the welding system.

The seventh process is a process of communicating data input by a user at the computer terminal and transmitting measured data during welding to the computer terminal.

The first process to the seventh process may be executed simultaneously or independently. Thus, in managing the operation of the welding system and the welding process, it is possible to prevent another process from being affected by any one of the processes.

According to an embodiment of the present invention, there is provided a method of manufacturing an inverter DC resistance spot welding system, the method comprising: modeling the inverter DC resistance spot welding system as a system identification model; Verifying the system identification model; Designing a fuzzy controller using the system identification model obtained as a result of the verification; Verifying performance of the fuzzy controller; Re-adjusting the optimum coefficient of the fuzzy controller obtained as a result of the verification; And optimizing the fuzzy controller to provide a fuzzy controller design method for an inverter DC resistance spot welding system.

By designing the fuzzy controller or the intelligent adaptive controller by the above-described method, it is possible to design the fuzzy controller or the constant current controller of the welding system by using simulation even if the experiment is not excessive.

Modeling the inverter DC resistance spot welding system comprises: designing an experiment of the inverter DC resistance spot welding system and measuring input and output for system identification; Extracting or filtering valid values from the measured inputs and outputs; Selecting a structure of the system identification model; Calculating an optimal system identification model that specifies a reference value and satisfies the reference value; And verifying the performance of the computed system identification model.

If the performance is not satisfied at the step of verifying the performance of the system identification model, the step 218 of calculating the optimal system identification model may be performed again.

Wherein designing the fuzzy controller comprises: selecting a structure of the fuzzy controller; Selecting a fuzzy inference method; Extracting a control rule of the fuzzy controller; Modifying a control parameter adjustment or control rule of the fuzzy controller; And evaluating performance of the fuzzy controller.

Wherein modeling the system identification model includes experimentally obtaining input data that reflects characteristics of the inverter DC resistance spot welding system and designing the fuzzy controller comprises adjusting the coefficient of the fuzzy controller appropriate to the system identification model A computer simulation can be used.

The step of re-adjusting the optimum coefficient of the fuzzy controller may use the reactive surface method to re-adjust the coefficients, and the step of optimizing the fuzzy controller may use a genetic algorithm.

The step of designing the fuzzy controller may design a rate-based fuzzy PI controller using the system identification model.

The step of selecting the fuzzy inference method uses a simple reasoning method, and if the performance is not satisfied in the step of evaluating the performance of the fuzzy controller, the conversion coefficient of the fuzzy controller can be adjusted.

INDUSTRIAL APPLICABILITY As described above, the inverter DC resistance spot welding system according to the embodiment of the present invention can overcome the disadvantage of AC single-phase AC resistance spot welding.

The inverter DC resistance spot welding system according to the embodiment of the present invention can reduce the energy consumed.

According to the welding process control method according to the embodiment of the present invention, since the welding process is controlled using the embedded software, various situations that may occur in the welding process can be considered and the user can stably perform the welding process.

Since the inverter DC resistance spot welding system according to the embodiment of the present invention graphically expresses the input of the user, various input patterns can be made into a waveform, and the user can see the measured data after welding, Can be transmitted quickly and accurately.

The fuzzy controller design method according to the embodiment of the present invention uses a system identification technique to design a fuzzy controller of an inverter DC resistance spot welding system so that many assumptions on system operation are not required and even if the system becomes complicated, Can be designed.

The fuzzy controller design method according to the embodiment of the present invention can improve the performance of the controller again when the performance of the controller is unsatisfactory.

The fuzzy controller design method according to the embodiment of the present invention can easily design the fuzzy controller with minimal experiment without prior experiment.

1 is a schematic view of an inverter DC resistance spot welding system according to an embodiment of the present invention.
Fig. 2 is a block diagram schematically showing the construction of the welding system according to Fig. 1;
Fig. 3 is a schematic circuit diagram of a welding system according to Fig. 1;
Fig. 4 is a view showing a change in the current waveform in the welding system according to Fig. 1;
Fig. 5 is a view showing a switching control operation of an inverter of the welding system according to Fig. 1;
Figure 6 is a diagram showing the inter-process relationship of the embedded software used in the welding system according to Figure 1;
Figure 7 is an illustration of an example of a graphical user interface (GUI) of a weld management program for use in the welding system according to Figure 1;
8 to 10 are flowcharts illustrating a method of designing a fuzzy controller of an inverter DC resistance spot welding system according to an embodiment of the present invention.
11 is a diagram showing a configuration of a fuzzy PI controller used in a fuzzy controller design of an inverter DC resistance spot welding system according to an embodiment of the present invention.
FIG. 12 is a diagram showing membership functions for inputs and outputs used in the fuzzy controller according to FIG. 11. FIG.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to or limited by the embodiments. Like reference symbols in the drawings denote like elements.

Fig. 1 schematically shows an inverter DC resistance spot welding system according to an embodiment of the present invention. Fig. 2 is a block diagram schematically showing a configuration of a welding system according to Fig. 1, Fig. Fig. 4 is a view showing a variation of the current waveform in the welding system according to Fig. 1, and Fig. 5 is a view showing a switching control operation of the inverter in the welding system according to Fig. 1 .

1 to 5, an inverter DC resistance spot welding system 100 according to an embodiment of the present invention generates a welding current and controls the amount of current or the energization time of the generated welding current A welding transformer 170 for generating a large current to be delivered to the workpiece, a welding gun for transmitting an electrode pressing force necessary for welding the welding object, and a pneumatic device for generating the electrode pressing force And a welding portion controller 110 for controlling the driving of the welding portion 190 and the welding portion 170. Here, the welding drive unit 150 may be configured as a welding robot, but the present invention is not limited thereto and may be implemented in various forms.

The power conversion unit 130 can control the magnitude of the welding current applied to the object to be welded (for example, a thin plate such as an automobile body) and the time for applying the welding current to the workpiece, that is, the energization time. The welding transformer 170 may include a transformer for generating a large current. The welding portion 190 is a welding gun for physically transferring an electrode force to a welding portion of a workpiece, a pneumatic device such as a pneumatic valve and a pneumatic cylinder for generating an electrode pressing force, a welding jig . ≪ / RTI > The welding gun is a pneumatic type, and when a servo gun is used as a welding gun, a pressing force can be generated by using a motor instead of a pneumatic device.

The power conversion unit 130 includes a bridge diode for rectifying three-phase AC power, a condenser for smoothing the rectified power to generate a DC waveform, and a capacitor for condensing the DC waveform generated from the capacitor to a predetermined pulse And may be an inverter for making an AC having a width. A more detailed description of the power conversion unit 130 will be described later.

Meanwhile, as shown in FIG. 2, the inverter DC resistance spot welding system 100 according to an embodiment of the present invention can largely include three modules. That is, it may include a power module, a control module, and a driving module. The power module is composed of IGBT and transformer as main elements, and the control module is composed of DSP and LCD which are high-speed processors. The driving module can be composed of a pneumatic valve and a pneumatic valve which generate an electrode pressing force on the welding gun. As described above, the three modules constituting the resistance spot welding system 100 can perform a desired welding process while mutually operating with each other. Here, the welding gun is electrically connected (E) to the power module and mechanically connected (M) to the driving module.

FIG. 3 shows a simplified circuit constituting the power conversion unit 130, the welding transformer unit 170, and the like of the inverter DC resistance spot welding system 100 according to an embodiment of the present invention. The current waveform changes in the inverter DC resistance spot welding system 100 according to an embodiment of the present invention. The principle of the welding current conversion will be described with reference to FIGS. 3 and 4. FIG.

First, the main power source, 50 ~ 60 Hz, 3 phase 440V AC power is rectified by using a bridge diode and smoothed by a condenser. A diode and a capacitor are combined to form a DC-link. Noises such as ripples are removed by passing the DC link capacitor, resulting in a DC waveform of a clean waveform. Since the output of the DC-link is a direct current, an IGBT (Insulated Gate Bipolar Transistor) can be used to generate an AC of a desired pulse width.

5, the inverter of the power conversion unit 130 of the resistance spot welding system 100 according to an embodiment of the present invention may be configured as a full-bridge inverter using four IGBTs . The inverter can be switched at a cycle of 1 kHz using a PWM (Pulse Width Modulation) control technique to generate a square wave AC of 1 kHz on the primary side of the welding transformer 170. Since the transformer of the welding transformer 170 has a winding ratio of 55: 1, the current can be amplified and the voltage can be reduced when passing through the welding transformer 170. The alternating current passing through the welding transformer 170 is finally rectified by the diode to generate a DC welding current. The rectified waveforms are pulsed and require smoothing. Since the welding gun of the welding part 190 is a structure made of copper, it has its own resistance and inductance component. Therefore, due to the inductance component existing in the welding gun, it can be smoothed to have a direct current form even without attaching a reactor. The finally converted large current is transferred to the workpiece through the welding gun, and the resistance point welding can be performed by generating heat by the contact resistance of the welding part of the workpiece.

As described above, the inverter of the power conversion unit 130 according to an embodiment of the present invention may include four IGBTs to generate square wave AC.

5, the inverter of the power conversion unit 130 is formed of a full bridge inverter using four IGBTs, and two IGBTs of the four IGBTs are combined to form two pairs of IGBT modules Can be formed. 5 (a), when the PWM pulse is applied to the IGBT 1 and the IGBT 4, the output of the full bridge circuit can output a positive waveform. Conversely, when a PWM pulse is applied to the IGBT 2 and the IGBT 3 as shown in FIG. 5 (b), the output of the full bridge circuit may have a negative waveform. Thus, when PWM pulses are alternately applied to the IGBT module including the IGBT 1 and the IGBT 4 and the IGBT module including the IGBT 2 and the IGBT 3, the output of the full bridge circuit can be an AC waveform. That is, when a PWM pulse is applied to any one of the two pairs of IGBT modules, the output of the full bridge becomes positive, and when a PWM pulse is applied to the other module, the output of the full bridge may become negative When the PWM pulses are alternately applied to the two pairs of IGBT modules, the output of the full bridge can be an AC waveform.

This AC waveform can be converted into a DC waveform through the welding transformer 170, the rectifying diode, and the welding gun of the welding portion 190. At this time, when a waveform having a large duty ratio (duty ratio) is applied, the current becomes large. When a waveform having a small duty ratio is input, the current becomes small.

Meanwhile, the inverter DC resistance spot welding system 100 according to an embodiment of the present invention may be equipped with embedded software for controlling a welding process.

That is, the inverter DC resistance spot welding system 100 according to an embodiment of the present invention includes a first process P1 for controlling the PWM output so as to be a constant current, an electrostatic force, a constant voltage or a heat quantity, A third process (P3) for controlling the welding press force applied to the workpiece, a fourth process (P4) for performing intelligent control, a fifth process (P5) for managing the welding process, A sixth process P6 for performing management, and a seventh process P7 for communicating with the computer terminal.

FIG. 6 is a diagram showing the inter-process relationship of the embedded software used in the welding system 100 according to FIG. Referring to FIG. 6, the first process P1 is a process of fuzing control of the constant current, the constant voltage, the constant power, or the fixed amount by the instruction value selection of the computer terminal PC. The first process P1 may be regarded as a process of managing the PWM output so as to be a constant current, a constant voltage, an electrostatic force, or an amount of heat. The second process P2 is to monitor the voltage or current measured from the ADC (Analog-to-Digital Converter) of the welder controller 110 during welding or to store the measured voltage or current at a designated position, . The third process P3 is a process of controlling the welding pressure or applying the welding pressure to the electrode of the welding gun. The fourth process (P4) is an intelligent control process, which is a process having a function of predicting a welding result using a result measured during welding, and changing welding current and pressing force command value during welding to improve weldability. The fifth process (P5) is a process for managing the welding process during welding, and is a process for managing the pressurization, stabilization time, energization time or maintenance time of the welding and coping with the emergency situation of the user. The sixth process P6 is a display and crisis management process, which is a process of displaying a command value inputted by a user on an LCD or the like and displaying the state of the welding portion 190. [ The seventh process (P7) is a process of communicating with a computer terminal (PC). The seventh process (P7) is a process of communicating data input by the user from a computer terminal (PC) through communication and transmitting data measured during welding to a computer terminal Process.

The first to seventh processes P1 to P7 may be executed simultaneously or independently so as to control the cooling process of the inverter DC resistance spot welding system so as to be stably welded. That is, the welding process control method of the inverter DC resistance spot welding system 100 according to an embodiment of the present invention can be performed by the first to seventh processes P1 to P7.

Referring to FIG. 6, it can be seen that the first to seventh processes P1 to P7 operate in association with each other. The HMI process (Human Machine Interface Process) shown in FIG. 6 allows a user to start the welding system using a teaching panel (not shown) or the like to operate the resistance spot welding system 100 And a communication between the teaching panel and the user's computer terminal.

In other words, the first to seventh processes (P1 to P7) mounted on the resistance spot welding system 100 described above are used to control the welding process of the resistance spot welding system 100, an operating process and a management process.

Figure 7 is an illustration of an example of a graphical user interface (GUI) of a weld management program for use in the welding system according to Figure 1; As shown in FIG. 7, the graphical user interface of the management program of the inverter DC resistance spot welding system 100 according to an exemplary embodiment of the present invention can graphically express a user's input, The input pattern can be made into a waveform, and various current waveforms can be graphically implemented. And can transmit data graphically generated by a user to a digital signal processor (DSP), which is a controller of the welding system. In addition, it may have a function of displaying the data measured after welding in the welding part 190 to the user, and may have a function of storing the welding result in a database. To this end, the graphical user interface of the management program shown in FIG. 7 includes a menu for selecting a control mode, a current mode, a pressure mode, a menu for transmitting welding setting information, A menu capable of storing welding measurement information, and the like.

Hereinafter, a method of designing a fuzzy controller of the inverter DC resistance spot welding system 100 according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 8 is a flowchart illustrating a method of designing a fuzzy controller of an inverter DC resistance spot welding system according to an embodiment of the present invention, FIG. 11 is a flowchart illustrating a method of designing an inverter DC resistance spot welding system according to an embodiment of the present invention. Fig. 12 is a diagram showing the configuration of a fuzzy PI controller used in the fuzzy controller design, and Fig. 12 is a diagram showing membership functions for input and output used in the fuzzy controller according to Fig.

The fuzzy controller of the inverter DC resistance spot welding system 100 according to an embodiment of the present invention can precisely control the intensity of the welding current or the energization time in the power conversion unit 130, Which allows designers to design in a trial and error manner by performing a number of actual experiments. However, in the intelligent adaptive controller design method according to an embodiment of the present invention, a controller is calculated using computer simulation, a minimum experiment is performed, and an optimal controller is designed by comparing the experimental result with a controller obtained through computer simulation . That is, the method of designing a fuzzy controller according to an embodiment of the present invention can provide a method of designing an optimum controller having a relatively low performance and a relatively high performance. Hereinafter, the present invention will be described in detail with reference to the drawings.

Referring to FIG. 8, a method of designing a fuzzy controller of an inverter DC resistance spot welding system 100 according to an embodiment of the present invention includes modeling an inverter DC resistance spot welding system 100 as a system identification model A step 220 of verifying the system identification model, a step 230 of designing a fuzzy controller using the system identification model obtained as a result of the verification, the step 230 of verifying the performance of the fuzzy controller, (240), re-adjusting (250) the optimal coefficient of the fuzzy controller obtained as a result of the verification, and optimizing (260) the fuzzy controller.

The fuzzy controller design method of the resistance spot welding system 100 as described above is a method for effectively designing a constant current controller and tuning a coefficient to the inverter DC resistance spot welding system 100. First, The system identification model can be verified by modeling the welding system as a system identification model, and a fuzzy controller can be designed using the obtained system identification model. Here, the fuzzy controller is a type of intelligent adaptation controller.

In addition, a computer simulation may be used to adjust the scaling factor of the fuzzy controller appropriate to the system identification model, and a genetic algorithm may be used to optimize the coefficient of the fuzzy controller. In addition, since the system identification model differs from the actual inverter DC resistance spot welding system 100, the response surface method is used to apply the optimum coefficient of the fuzzy controller obtained by computer simulation to the actual resistance spot welding system. Adjust the coefficients by using

Among the above design methods, the method includes modeling (210) the inverter DC resistance spot welding system (100) with a system identification model (220), verifying the system identification model (220) Wherein the step of designing a fuzzy controller using the fuzzy controller and the step of verifying the performance of the fuzzy controller using the computer simulation are performed using computer simulation and the optimal coefficient of the fuzzy controller obtained as a result of the verification is readjusted (250) and optimizing the fuzzy controller (260) may be performed using an actual resistance spot welding system (100) and comparing the experimental results with the results obtained in the computer simulation.

The design procedure of the controller of the inverter DC resistance spot welding system 100 according to an embodiment of the present invention and the method used are summarized in Table 1 below.

Process Algorithm

System identification model

System Modeling ARX model or ANN model Controller design Fuzzy PI controller Performance index
(Performance Index) IAE (integral of the absolute error) Coefficient control
(Parameter tuning) Scaling factor
(GE, GDE, GDU) Optimization algorithm Genetic algorithm
(Genetic Algorithm)
Actual system
(Resistance spot welding system) Performance index IAE (integral of the absolute error) Coefficient adjustment
(Parameter re-tuning) Scaling factor
(GE, GDE, GDU) Optimization algorithm Reaction surface method
(Response surface method)

9, modeling the inverter DC resistance spot welding system 100 into a system identification model 210 may include designing an experiment of the inverter DC resistance spot welding system 100 and inputting an input for system identification (214) for extracting or filtering valid values from the measured inputs and outputs, selecting a structure of the system identification model (216), designating a reference value and determining the reference value (Step 218) computing the optimal system identification model to satisfy the step 222, verifying 222 the performance of the calculated system identification model, and verifying 224 the optimal model. Here, if performance is not satisfied in step 222 of verifying the performance of the system identification model, step 218 of calculating an optimal system identification model is performed again. The system identification modeling step 210 may be repeatedly performed until a desired reference model is obtained through selection of a model structure, selection of an optimal model index, and analysis of a model.

As a modeling method for a mathematical model, we approximate each element of the system using a govern equation based on the physical principle and then use a method of expressing it as a differential equation. However, since the resistance spot welding system 100 has nonlinear elements existing in the welding circuit such as melting of the metal, mechanical reaction to the welding gun, electric element switching, etc., the welding system can be expressed mathematically using physical laws And linearization is a difficult task. In fact, the complexity of a system means that many assumptions about system behavior are required because information about all situations is unknown. Therefore, an efficient method is required for modeling of the welding system. For this purpose, the system identification technique is introduced in the present invention. System identification is a method of mathematically expressing the relationship between input and output without using physical equations in unknown systems. Mathematical models include linear models and nonlinear models. Linear models include ARX, ARXMAX, OE, BJ, and models. Nonlinear models include NARMA and Nonlinear ARX using artificial neural networks.

A method for designing an experiment of an inverter DC resistance spot welding system (100), comprising the steps of measuring inputs and outputs for system identification and extracting and filtering valid values in the measured inputs and outputs (214) Is defined as the PWM control value in the welder controller, and the welding current, which is the output of the welding system, can be defined as the value measured by the ADC of the welder controller, in the on / off pulse signal applied to the IGBT have. In addition, the step 212 of measuring the input and output may acquire input data reflecting the characteristics of the inverter DC resistance spot welding system 100 through an experiment.

Meanwhile, the design method according to an exemplary embodiment of the present invention defines a model structure as shown in Equation (1) in a step 216 of selecting a structure of a system identification model.

Where y is the output, u is the input, e is the white noise, and there may be several models depending on the presence of the polynomials A, B, C, D,

In one embodiment of the present invention, after determining the model structure based on Equation (1) described above, the performance of the system identification model is examined in an error required by the inverter DC resistance spot welding system 100 or in a satisfactory range, Performance is verified by comparing the performance of various model structures. In the system identification modeling method according to an embodiment of the present invention, the ARMAX (25, 18, 15) model is used as the system identification model most suitable for the inverter DC resistance spot welding system 100 according to an embodiment of the present invention. Respectively. Here, ARMAX is an autoregressive moving average model with exogenous input models, and 25, 18, and 15 are values corresponding to A, B, and C in Equation (1), respectively. Using the ARMAX (25, 18, 15) model according to the input of each verification data, the current value by the system identification model and the current value of the resistance spot welding system 100 measured by the actual experiment were compared. As a result, (25, 18, 15) model is similar to that of actual welding system for various input types.

Hereinafter, a step of designing the fuzzy controller will be described with reference to the drawings.

In the resistance spot welding system according to an embodiment of the present invention, since the resistivity varies depending on the material to be welded and the resistance changes during welding, a robust control characteristic is required, and therefore, it is preferable to use a fuzzy controller. In the present invention, a speed type fuzzy PI (Proportional Integral) controller is designed using the system identification model described above for the fuzzy controller design for constant current control. In the inverter DC resistance spot welding system 100 according to the embodiment of the present invention, since the IGBT is turned on / off at a switching frequency of 1 kHz, the microcontroller of the control board generates a control input every 0.5 msec . Also, since the total welding current flow time is usually several hundreds of milliseconds, it is necessary to select a high-speed microprocessor and a controller considering the operation speed for precise control during a short process. In the present invention, a speed type fuzzy PI controller using a simple reasoning method is used for continuous control rule extraction in a single loop control.

The method of designing the speed type fuzzy PI controller 230, that is, the step 230 of designing the fuzzy controller of FIG. 9, includes a step 232 of selecting the structure of the fuzzy controller, selecting a fuzzy inference method A step 236 of extracting a control rule of the fuzzy controller, a step 238 of adjusting a control parameter or a control rule of the fuzzy controller, and a step 239 of evaluating the performance of the fuzzy controller. . ≪ / RTI >

FIG. 11 shows a fuzzy PI controller used in an embodiment of the present invention. In FIG. 11, the plant uses the ARMAX (25,18,15) model described above, and Ref and Yk are the target values and actual outputs of the welding system, respectively. The input variables of the controller are the error (e) and the error variation (Δe), and the controller output is the difference of the output variation (Δu). The relationship between the error (e), the variation amount of error (? E), and the output (u) is as shown in the following equation (2).

Thus, step 230 of designing the fuzzy controller, i.e., selecting 232 the structure of the fuzzy controller, may utilize computer simulation to adjust the coefficients of the fuzzy controller appropriate to the system identification model. Here, the step 234 of selecting the fuzzy inference method may use a simple reasoning method. The fuzzy inference method can be divided into three types as fuzzy fuzzy, fuzzy inference engine, and defuzzyification. Fuzzification can be done using an isosceles triangle method, which converts a stochastic property of an input value into an appropriate fuzzy number when considered to be uncertain due to the effect of disturbance. In addition, the fuzzy inference method uses the min-max method, and the non-fuzzy method uses the center of gravity.

The control rule extracted in the step 236 of extracting the fuzzy control rule may be composed of seven stages of fuzzy sets for the inputs e and e and the output u of the equation 2, respectively. The membership functions for the inputs e and e and the output u are as shown in Fig. Fig. 12A is a membership function of the error e, Fig. 12B is a membership function of the error change amount e, and Fig. 12C is a membership function of the output u.

Meanwhile, if the performance of the fuzzy controller designed in the step 240 of verifying the performance of the fuzzy controller is unsatisfactory, the performance of the fuzzy controller can be improved again. The performance of the fuzzy controller is mainly determined by the scaling factor, It can be influenced by membership functions and control rules. In the present invention, a method of adjusting the conversion coefficient is used as a method for improving the performance of the fuzzy controller (refer to 250 in FIG. 8).

The step 260 of optimizing the fuzzy controller uses a genetic algorithm to adjust the conversion factor of the fuzzy controller. In the present invention, a scaling factor adjustment method is used to improve the control performance of the fuzzy controller, and the objective function is defined as a performance index (IAE) of an integral of the absolute error And the genetic algorithm is used as the optimization algorithm. The conversion coefficient of the simulation that optimizes the fuzzy controller of the system identification model is the genetic algorithm converged at 53 generation, GE = 0.4936, GD = 2.48683, GDU = 221.035.

On the other hand, the fuzzy controller designed by using the system identification model and the optimization algorithm and designed through simulation is very effective in time and cost than the control of the fuzzy controller for the actual plant, that is, the welding system from the beginning. However, since the system identification model basically includes the modeling error due to the complexity and nonlinearity of the actual system, the controller designed based on the system identification model may be applied to the actual welding system to be readjusted. Thus, the step 250 of re-adjusting the optimal coefficients of the fuzzy controller may re-adjust or optimize the fuzzy control conversion factor using the response surface method.

When the fuzzy control conversion coefficient of the computer simulation is applied to the actual resistance spot welding system, the fuzzy controller designed by the computer simulation works well because the system identification model represents the actual welding system well. In order to design an existing controller, many prior experiments have been performed. However, in the present invention, a controller can be sufficiently designed by only a system identification model and a simulation.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

100: Inverter DC Resistance Spot Welding System
110: welding part controller 130: power conversion part
150: welding robot 170: welding transformer
190:

Claims (21)

delete delete delete delete delete delete delete delete delete delete delete delete delete A power converter for generating a welding current and controlling an amount of current or an energizing time of the welding current; A welding transformer for generating a large current to be delivered to the workpiece; A welder having a welding gun for transmitting an electrode pressing force necessary for a welding part of the workpiece and a pneumatic device for generating the pressing force of the electrode; And a welder controller for controlling driving of the weld portion, wherein the power converter includes: a rectifier diode for rectifying three-phase AC power; A capacitor for smoothing the rectified power supply to generate a DC waveform; And an inverter for converting the DC waveform generated from the capacitor into an AC having a predetermined pulse width and using four IGBTs, the method comprising the steps of:
Modeling the inverter DC resistance spot welding system as a system identification model;
Verifying the system identification model;
Designing a fuzzy controller using the system identification model obtained as a result of the verification;
Verifying performance of the fuzzy controller;
Re-adjusting the optimum coefficient of the fuzzy controller obtained as a result of the verification; And
Optimizing the fuzzy controller;
Lt; / RTI >
Wherein modeling the inverter DC resistance spot welding system comprises:
Designing an experiment of the inverter DC resistance spot welding system and measuring input and output for system identification;
Extracting or filtering valid values from the measured inputs and outputs;
Selecting a structure of the system identification model;
Designating a reference value and calculating an optimal system identification model satisfying the reference value; And
Verifying performance of the computed system identification model;
/ RTI >
In the step of measuring input and output for system identification, an input defines an interval size corresponding to an on / off pulse signal applied to the IGBT as a PWM control value in the welder controller and an output is defined as an ADC Wherein the fuzzy controller is defined as a value measured by the fuzzy controller.
delete 15. The method of claim 14,
And if the performance is not satisfied at the step of verifying the performance of the system identification model, calculating the optimal system identification model again.
15. The method of claim 14,
Wherein designing the fuzzy controller comprises:
Selecting a structure of the fuzzy controller;
Selecting a fuzzy inference method;
Extracting a control rule of the fuzzy controller;
Modifying a control parameter adjustment or control rule of the fuzzy controller; And
Evaluating performance of the fuzzy controller;
A method for designing a fuzzy controller of an inverter DC resistance spot welding system.
18. The method of claim 17,
Modeling the system identification model may include experimentally obtaining input data that reflects the characteristics of the inverter DC resistance spot welding system,
Wherein designing the fuzzy controller utilizes computer simulation to adjust coefficients of the fuzzy controller appropriate to the system identification model.
15. The method of claim 14,
The step of re-adjusting the optimal coefficients of the fuzzy controller may include re-adjusting the coefficients using a reactive surface method,
Wherein the step of optimizing the fuzzy controller utilizes a genetic algorithm.
15. The method of claim 14,
Wherein the step of designing the fuzzy controller designes a velocity-type fuzzy PI controller using the system identification model.
18. The method of claim 17,
Wherein the step of selecting the fuzzy inference method uses a simple reasoning method,
Wherein the conversion coefficient of the fuzzy controller is adjusted if the performance is not satisfied in the step of evaluating the performance of the fuzzy controller.

Images