JP2002263742A - Press and pressing method for deep drawing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a press for deep drawing capable of improving the productivity of an excellent formed body. SOLUTION: The drawing speed of a punch is feedback-controlled based on a means for storing the basic data for fuzzy control as the database including an ideal working curve 200 which is obtained by performing the analysis to indicate the relationship between the retraction r of a blank 20 and the punch stroke S, and an evaluation function for relating the relationship between the actually obtained retraction of the blank and the punch stroke to the difference ϕbetween the actual working curve and the ideal working curve. An evaluation function which acquires the relationship between the actually measured retraction of the blank during the actual working and the punch stroke, identifies the basic data for fuzzy control adaptable to the blank from the database, obtains the actual time evaluation data related to the difference between the ideal working curve and the actual measurement data, and is included in the basic data identified to be the obtained actual time evaluation data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a deep drawing press machine and method for deep drawing a workpiece fixed by a die and a blank holder.

[0002]

2. Description of the Related Art Conventionally, as a deep drawing method using a press machine, a blank (for example, a plate-like aluminum material) to be processed is processed by using a die of the press machine and a blank holder. A method is known in which a blank is formed into, for example, a cylindrical shape by lowering and raising the punch while applying a wrinkle pressing force to the blank holder.

[0003] When performing such deep drawing, it is necessary to increase the punch speed of a press machine in order to increase the productivity. However, when the punch speed is increased, breakage of the workpiece may occur. This raises the problem that the processability increases and processing becomes difficult.

[0004] Therefore, in the conventional deep drawing, press working is performed within a speed range in which no breakage is guaranteed, and thus there is a limit in increasing productivity.

[0005] In addition, a technique for deep drawing by varying the punch speed has been studied. However, this kind of conventional technique avoids a condition in which the risk of breakage is theoretically high. However, since the punch speed is variably controlled, the calculation for the speed control becomes complicated, and the productivity of the molded product cannot be improved.

The present invention has been made in view of such problems, and an object of the present invention is to provide a press machine and a method for deep drawing capable of increasing the productivity of a good molded product. is there.

[0007]

Means for Solving the Problems (1) In order to achieve the above object, the present invention relates to a deep drawing method for deep drawing a workpiece fixed by a die and a blank holder by changing a punch speed. In the press machine, the relationship between the data of the ideal processing curve obtained by performing the analysis so as to represent the relationship between the drawing amount of the processing object and the punch stroke, and the drawing amount of the processing object and the punch stroke obtained by the actual measurement And the evaluation function data obtained from the evaluation function associated with the deviation between the actually measured machining curve and the ideal machining curve, and the fuzzy control basic data including at least the type of the processing target and the processing conditions associated with the type. Means for storing a database for each one of the items, and the amount of pull-in of the workpiece to be measured and the Along with acquiring the relationship with the first stroke, specifying the basic data for fuzzy control that matches the item of the processing target from the database, the data of the ideal processing curve included in the specified basic data and the actual measurement data are used. Control means for obtaining the real-time evaluation data associated with the deviation, based on the obtained real-time evaluation data and the evaluation function data included in the specified basic data, and performing feedback control on the drawing speed of the punch. Features.

Further, the deep drawing method of the present invention is directed to a deep drawing method using a press machine for performing a deep drawing process by varying a punch speed on an object fixed by a die and a blank holder. The ideal machining curve data obtained by performing an analysis so as to represent the association between the drawing-in amount of the object and the punch stroke, and the actually-measured processing curve representing the association between the drawing-in amount of the workpiece and the punch stroke obtained by the actual measurement, and Evaluation function data obtained from the evaluation function associated with the deviation from the ideal processing curve, and the basic data for fuzzy control including at least one of the type of the workpiece and the processing condition associated with the type for each item It is stored as a database, and the relationship between the retracted amount of the workpiece and the punch stroke measured during actual machining is And obtaining the fuzzy control basic data matching the item of the processing object from the database, and relating the actual processing data to the deviation between the ideal processing curve data included in the specified basic data and the actual measurement data. It is characterized in that time evaluation data is obtained, and the drawing speed of the punch is feedback-controlled based on the obtained real-time evaluation data and the evaluation function data included in the specified basic data.

Further, the program for deep drawing to which the present invention is applied is based on data of an ideal machining curve obtained by performing analysis so as to represent the association between the amount of drawing in the workpiece and the punch stroke, and actual measurement. Fundamental data for fuzzy control, including an evaluation function data obtained from an evaluation function associated with a deviation between the actually measured processing curve and the ideal processing curve, which represents the association between the amount of retraction of the object to be processed and the punch stroke, The punching speed of the workpiece fixed by the die and the blank holder is changed based on the data of the database constructed for at least one of the type of the workpiece and the processing condition associated with the type. Computer controlled press machine to perform deep drawing, readable by computer The program acquires the relationship between the drawing-in amount of the processing target and the punch stroke measured during the actual processing, and specifies the basic data for fuzzy control that matches the item of the processing target from the database, and specifies the program. Determine the real-time evaluation data associated with the deviation between the ideal processing curve data and the measured data included in the basic data, based on the obtained real-time evaluation data and the evaluation function data included in the specified basic data, Control means for feedback-controlling the aperture speed of the punch.

Preferably, the program is stored on a computer-readable storage medium.

Here, the type of the object to be processed is specified by, for example, a material and other type conditions.

The processing conditions include, for example, the material thickness, size, processing temperature, lubricant, wrinkle holding force,
There are tool shapes and other conditions.

According to the present invention, the basic data for fuzzy control is obtained in advance for each type of at least one of the type of the object to be processed and the processing conditions associated with the type, and this is stored as a database.

The basic data for fuzzy control includes data of an ideal machining curve and evaluation function data.

Here, the ideal machining curve is obtained by, for example, finite element analysis based on predetermined setting conditions (for example, constant thickness, constant volume, etc.). Represents the association.

The actually measured machining curve is obtained by actual measurement, and represents an association between the amount of drawing in the object to be processed and the punch stroke. For example, using the punch speed as a parameter and calculating a plurality of actually measured processing curves while gradually increasing the parameter speed, when the speed reaches a certain upper limit, the processing target Breaks. In the present invention, the actually measured processing curve obtained using the speed at this time as a parameter may be used as the actually measured processing curve.

The evaluation function for fuzzy control is used as an evaluation function for breakage.
It is obtained as a function associated with the deviation between the actually measured processing curve and the ideal processing curve. The evaluation function data is obtained, for example, as a minimum value and a maximum value of the evaluation function.

In the present invention, when actually processing the object to be processed, the relationship between the retracted amount of the object to be processed and the punch stroke is measured in real time using a sensor or the like, and the measurement data is obtained. .

Then, fuzzy control basic data matching the item of the workpiece is specified from the database, and the punch speed is feedback-controlled using the specified fuzzy control basic data and the measured data.

More specifically, real-time evaluation data associated with a deviation between the ideal processing curve data included in the specified basic data and the actually measured data is obtained in real time.

Then, based on the obtained real-time evaluation data and the evaluation function included in the basic data for fuzzy control, feedback control of the drawing speed of the punch is performed.

By adopting the above configuration, according to the present invention, when deep-drawing an object to be processed, the punch speed is adjusted to an optimum speed so as to increase the productivity without breaking the object. Feedback control can be performed, and it is possible to realize deep drawing with improved productivity of a good molded product.

In particular, according to the present invention, the punch drawing speed can be feedback-controlled to an optimum speed in accordance with the type of the object to be processed and the processing conditions excluding the above-described punching speed associated with the type. It is possible to realize deep drawing using a press machine capable of increasing the productivity of a good molded product without being affected by the skill of an operator.

(2) In the present invention, the basic data for fuzzy control is compiled into a database for each type of processing object and processing conditions, and the control means reads the type of processing object and processing conditions from the database. The basic data for fuzzy control conforming to the above is specified, and feedback control is performed on the drawing speed of the punch.

As described above, the basic data for fuzzy control is stored in a database for each type of processing target and processing conditions (excluding the drawing speed of the punch), so that even if the type of processing target is different, Even when the processing conditions are different, it is possible to select the optimal fuzzy control basic data and perform feedback control of the drawing speed of the punch to the optimum speed according to the type and the processing conditions.

In particular, according to the present invention, it is possible to efficiently produce these products without deteriorating their quality when performing high-mix, low-volume production.

(3) In the present invention, the feedback control is performed by fuzzy inference in accordance with a fuzzy inference rule created based on the evaluation function data.

As a method of the feedback control performed by such fuzzy inference, various methods can be adopted as needed.

For example, as one of such methods, the evaluation function data includes first evaluation function data obtained from a first evaluation function representing a deviation between the actually measured processing curve and the ideal processing curve; And second evaluation function data obtained from a second evaluation function obtained by differentiating the first evaluation function, wherein the feedback control is created based on the first and second evaluation function data. A method characterized by being performed by fuzzy inference according to the fuzzy inference rules described above may be employed. Here, as the first and second evaluation function data, for example, the minimum and maximum values of the first and second evaluation functions may be used.

Also, a method may be adopted in which the feedback control is performed by fuzzy inference in accordance with fuzzy inference rules created based on the first and second evaluation function data and a membership function using the data. Good.

[0031]

Next, embodiments of the present invention will be described in detail with reference to the drawings.

(1) Outline of Press Machine FIG. 1 shows an outline of a press machine for deep drawing according to the present embodiment. The press machine of the present embodiment forms a container with a bottom using a blank 20 which is a plate-shaped object to be processed, and specifically, a die 10 having a predetermined die hole 12. And punch 14
It is comprised including.

The blank 20 is held and fixed on the die 10 by the blank holder 30 with a desired wrinkle pressing force H.

In this state, FIGS. 1 (A), (B), (C)
As shown in (1), the blank 20 is deep drawn into a cup shape along the shape of the die hole 12 by lowering the punch 14 at a given drawing speed V, thereby producing a cup-shaped molded product.

At this time, if the drawing speed V of the punch 14 is low, the production efficiency of the molded product is low. Therefore, it is preferable to set the drawing speed V of the punch 14 as high as possible within a range in which the blank 20 does not break during the processing.

The feature of the present embodiment is that, at the time of actual processing of the blank 20, the retraction amount r and the punch stroke S of the blank 20 are measured in real time, and based on the measured values, a fuzzy inference technique is used to form the punch 14. The purpose is to feedback control the aperture speed V.

In particular, in this embodiment, the blank 2
For each type of 0 and processing conditions, basic data for fuzzy control is created, stored in a database, and optimal basic data for fuzzy control is set according to the type of blank 20 to be actually processed and processing conditions. A configuration is employed in which the punching speed V of the punch is feedback-controlled so that the deep drawing process can be efficiently performed without causing any breakage in the blank 20 by using this readout.

FIG. 3 shows a comparative example of the speed control characteristic of the press machine according to the present embodiment and the speed control characteristic of the press machine using the conventional method.

In FIG. 4, the horizontal axis represents the degree of progress of the drawing process of the blank 20 as a function of the amount of pull-in r. Specifically, as shown in FIG. The ratio of the amount r of the blank 20 drawn from the start to the radius Rp of the punch 14 is defined as ΔDR ^* = r / Rp, and the value is described.

The vertical axis represents the drawing speed V of the punch 14 corresponding to ΔDR ^* .

In the drawing, an area surrounded by a dashed line indicated by 100 is a blank 2 when the blank 20 is drawn.
This indicates an area where there is a high risk of breakage with respect to 0, and is referred to as a break limit area in analysis.

In the conventional technique of keeping the punch drawing speed V constant, the punch speed V is set as shown by 110a in the figure so as to avoid the breaking limit area 100 in this analysis. That is, the punch speed V is set to an upper limit value that avoids the limit area 100, and consideration is given to increasing the production efficiency of the workpiece.

The punch speed V is adjusted to the upper limit 110a.
If the above value is set to, for example, a value indicated by 110b in the drawing, the blank 20 being processed will be broken at a location indicated by a cross in the drawing during drawing.

On the other hand, when the punch speed feedback control is performed by using the fuzzy inference method of the present embodiment, the punch drawing speed V in the figure is changed under the same processing conditions for the same blank 20. As shown by reference numeral 120, it can be variably controlled, and even when the punch speed V takes a value within the analytical breaking limit area 100, the drawing process can be performed without actually breaking the blank 20. It was confirmed that it was possible.

As described above, according to the method of the present embodiment,
Conventionally, an analytical breaking limit area 1 which had to be avoided in order to avoid breaking during drawing of the blank 20
Even within 00, it is confirmed that the blank 20 can be efficiently and satisfactorily produced without causing breakage by performing feedback variable control of the punch speed V using a fuzzy inference method described later. According to the present embodiment, it is possible to increase the production efficiency of a pressed product while maintaining the quality of the product.

In particular, according to the method of the present embodiment, even in the case of performing a large variety of small-scale production of a pressed product, the basic data for fuzzy control suitable for the type of blank 20 to be used and the processing conditions is used. It is possible to realize low-volume production with high production efficiency while maintaining quality.

(2) Principle for Fuzzy Control of this Embodiment The principle for fuzzy control of this embodiment will be described below.

First, a method for obtaining basic data for fuzzy control for each type and processing condition of the object to be processed while changing the type and processing conditions of the blank 20, which is the object to be processed, and converting the basic data into a database will be described. A method of actually processing and forming a processing target object using this database will be described.

(Construction of Database) For each type of the blank 20 to be processed, the blank 20 is subjected to drawing processing control under the processing conditions, and the obtained data and a finite element analysis technique are used for fuzzy control. Generate basic data. The generation of such fuzzy control basic data is repeatedly performed by changing the type of blank 20 and the processing conditions, and is made into a database.

First Step First, it is assumed that the blank 20 of the target type has a constant thickness and volume. Then, using the method of finite element analysis (FEN) with the parameter ΔDR ^* and the punch stroke S for the inflow amount r from the flange end of the blank 20 during deep drawing as a variable, an ideal processing curve shown in FIG. Ask for 200.

[0051] Here,? DR ^*, as shown in FIG. 4 (A), inflow from the flange end of the blank 20 (pulled amount) r, is a parameter dimensionless radius _{R P} of the punch 14 .

Second Step Next, as shown in FIG. 4B, the blank 20 is actually press-drawn at a constant punch speed V, and an actual measurement representing the association between ΔDR ^* and the punch stroke S at this time. A processing curve 210 is obtained. This actually measured processing curve 210
Is measured data representing the relationship between ΔDR ^* and punch stroke S when deep drawing is performed with a constant punch speed V. This actually measured processing curve 210 has a speed V
Are obtained as a plurality of parameters.

The higher the punch speed V, the more
The measured processing curves are 210a, 210b, 210c, 210
As shown by d, it moves away from the ideal machining curve 200.

In FIG. 4B, reference numeral 270 denotes an area where the risk of breakage of the blank 20 is high when the blank 20 is drawn. Accordingly, when the actually measured processing curve 210 is obtained by gradually increasing the speed V, which is a parameter, using the punch speed V as a parameter, when the drawing speed V reaches a certain upper limit, before the forming is completed, A break occurs in the blank 20. The actually measured machining curve 210 obtained using the speed V at this time as a parameter is
As shown in FIG. 5A, it is used as an actual measurement processing curve 230 for obtaining a first evaluation function φ.

Third Step As shown in FIG. 5A, the ideal processing curve 200 obtained in the first step and the actually measured processing curve 230 obtained in the second step.
Is determined as a first evaluation function φ indicating the risk of breakage.

Then, the minimum value of the evaluation function φ is determined as φa, and when the punch stroke S is increased in accordance with the actually measured processing curve 230, the evaluation function φ at the position of the break generation point 272 where the blank 20 breaks occurs. Is obtained as φb.

In the present embodiment, φa and φb obtained from the first evaluation function φ are defined as first evaluation function data.

Then, based on the first evaluation function data φa and φb obtained as described above, FIGS.
An input-side membership function μφ for evaluating the risk of fracture shown in (A) is generated, and the membership function μ is generated.
Evaluation of fracture described later is performed using φ.

Fourth Step Further, in order to improve the reliability of the evaluation, the differential value φ ′ of the first evaluation function φ obtained in the third step is obtained as a second evaluation function, and the differential value is further obtained. Minimum value of φ 'φ
'A and the maximum value φ'b are obtained, and these are defined as second evaluation function data.

Then, based on the second evaluation function data φ′a and φ′b obtained in this way, FIGS.
An input-side membership function μφ ′ for evaluating the degree of risk of break shown in (B) is generated, and the break is evaluated using this.

This risk increases as the evaluation function φ increases, and the risk is maximized when it coincides with φb. On the other hand, when it coincides with φa, the degree of risk is minimized.

The differential component φ ′ of φ indicates whether the punch speed during processing changes in the direction of high fracture risk (φb direction) or in the low direction (φa direction). When φ ′ matches φb ′, the risk is maximum, and when φ ′ matches φa ′, the risk is minimum.

The membership function shown in FIG. 6A (or FIG. 5C) and the membership function shown in FIG.
By performing the evaluation in combination with the membership function shown in (C)), in the present embodiment, the risk of breakage of the blank 20 during actual processing can be more accurately evaluated.

In order to facilitate understanding, FIG.
(C), FIG. 6, FIG. 7 (A), and FIG. 8 show the membership functions, respectively, and their contents are basically the same.

Fifth Step As described above, the ideal machining curve 2 obtained in the first to fourth steps
00, the first evaluation function φ, its minimum value φa and its maximum value φ
b, a first evaluation function data, a second evaluation function φ ′,
The second evaluation function data representing the minimum value φ′a and the maximum value φ′b, and data for generating membership functions μφ and μφ ′ from the first and second evaluation function data, respectively, are fuzzy. As control basic data, a database is created in association with the type of blank 20 and processing conditions, and stored in the storage means 40 shown in FIG.

As described above, the processing of the first to fifth stages is repeatedly performed to obtain basic data for fuzzy control by changing the processing conditions for each blank 20 of various materials. The database constructed on the basis of the obtained data is stored in the storage means 40 in advance.

(Fuzzy Inference Rule) Next, using the basic data for fuzzy control associated with each type of blank 20 and each processing condition obtained as described above, the drawing speed V of the punch 14 is determined during actual processing. A fuzzy inference rule for feedback will be described.

First stage The basic data for fuzzy control associated with the type of blank 20 to be actually processed and the processing conditions is read out from the database stored in the storage means 40, and FIG.
A membership function μφ based on a first evaluation function and an input membership function μφ ′ based on a second evaluation function shown in FIGS. 6A and 6B are prepared.

Second Stage When the blank 20 is actually drawn according to the processing conditions as shown in FIG. 1, the punch stroke S and the amount r of drawing in the blank 20 shown in FIG. 4A are measured in real time.

[0070] Then, the pull-in amount r, calculates the ΔDR ^* = r / _{R P} and a radius _{R P} of the punch 14, the punch stroke S in the calculated value? DR ^* to the associated ideal working curve 200 on, A deviation φ from the punch stroke S obtained by actual measurement is obtained as first real-time evaluation data.

Further, a differential value φ ′ of the first real-time evaluation data φ is obtained as second real-time evaluation data.

Third Step FIG. 7B shows an If-then rule for estimating the amount of change ΔV in punch drawing speed V from the real-time evaluation data φ and φ ′ thus obtained. I have.

In FIG. 7B, if (φ, φ
') Represents the input condition of this fuzzy inference rule, the right (then) represents its output, and its value is specifically the output of the output control membership function of FIG. 7C. Is reflected in

For example, the first and second values obtained by actual measurement
Is assumed to be φ, φ ′ as shown in FIG. In this case, if-t shown in FIG.
Find the area of the input membership function of the hen rule. Specifically, as shown in FIG. 8, the input-side membership functions μφ, μφ ′ specified by the real-time evaluation data φ, φ ′ are 300-1, 300-2, 300-3,
The area of the triangle of 300-4 is _represented by A _φL , A _φS ,
_Determined as A _φ′L and A _{φ ′S.}

[0075] The fourth step Next, the aforementioned each part fuzzy set is the area of each triangle determined as _{(A φ L, A φ'L,} A φS,
The area of A _φ ′ _S ) is calculated and substituted into a membership function for determining an output value according to the if-then rule, as shown in FIG.

That is, the area of the input membership function shown in FIG. 8 is substituted for the output membership function shown in FIG. In FIG. 9, the substitution areas of the membership functions ΔV _LL , ΔV _LS , ΔV _SL , and ΔV _SS of the output value determination are 400-1, 400-2, 400-3, and 40.
This is the area indicated by 0-4.

Each of these areas 400-1, 400-2, 4
The area of 00-3 and 400-4 is A _LL, A_LS, A
_SL, A_SSIt is.

Specifically, it is represented by the following equation. _{A LL = A φL + A φ'L} A LS = A φL + A φ'S A SL = A φS + A φ'L A SS = A φS + A φ'S

Next, the partial fuzzy set (areas 400-1 to 400-4) assigned to the membership function for output value determination
From the area (00-4), the control value ΔV of the punch speed is obtained by the centroid method.

In this embodiment, each area 400-1 to 400- of the membership function for output value determination shown in FIG.
The area a and the center of gravity g of the area 4 are obtained, and this area 4 is calculated based on the following equation.
The position of the center of gravity of 00 is obtained by calculation as the variable amount ΔV of the speed V under feedback control.

[0081] _{ΔV = Σa n g n / Σa} n (Σa n = A LL ∪A LS ∪A SL ∪A SS)

[0082] Incidentally, a _n is the area of the region obtained by arbitrarily dividing the partial fuzzy set membership function for the output value determination,
g _n denotes the centroid of the region.

Fifth Step ΔV thus obtained represents the amount of change for obtaining the optimum punch drawing speed V without causing breakage when the blank 20 is actually deep drawn.

Accordingly, by performing feedback control of the drawing speed of the punch 14 by the amount of change ΔV inferred in real time in this way, as shown by 120 in FIG. Deep drawing can be performed efficiently at high speed.

In particular, according to the present embodiment, the deep drawing of the blank 20 can be performed without causing breakage even in the breakage limit area 100 on the analysis, which cannot be used in the conventional punch speed constant speed control. It is possible to do.

As shown in FIGS. 6A and 6B, not only φ as real-time evaluation data indicating the risk of fracture during actual machining but also this differential value φ ′ This is due to the fact that a method of determining the feedback amount ΔV of the punch speed by estimating the trend of the risk of fracture by generating the input-side membership function using the same is used. This makes it possible to achieve a punch speed in the analytical breaking limit area 100 that could not be achieved when the punch speed V is set to a fixed value, and it is possible to increase the productivity of drawing as compared with the conventional case.

(3) Configuration Block Diagram of Main Parts of Press Machine FIG. 2 is a functional block diagram of a configuration for performing feedback control of the drawing speed V of the punch using the above-described principle.

The press machine according to the present embodiment has a group of sensors 60 for measuring various parts, a group of actuators 70 for driving various parts, a control means 50, and a storage means 4.
0.

The sensor group 60 includes a punch stroke sensor 60a, a punch speed sensor 60b, a wrinkle pressing force sensor 60c, a pull-in amount sensor 60d, and other sensors.

The punch stroke sensor 60a is shown in FIG.
As shown in (A), a stroke S, which is an amount of movement of the punch 14 from its initial position, is detected, and a punch speed sensor 60
b is configured to measure the lowering and raising speeds of the punch 14, particularly the drawing speed V of the punch at the time of actual machining.

The wrinkle holding force sensor 60c measures the wrinkle holding force H for holding the blank 20 between the blank holder 30 and the die 10, and the pull-in amount sensor 60d
In a series of drawing processes shown in FIGS. 1A to 1C, how much distance the blank 20 is drawn by drawing of the punch 14 from the initial position before processing shown in FIG. The amount of pull-in r is measured. Specifically, as shown in FIG. 4A, the amount of pull-in r from the state before processing is measured in real time.

The actuator group 70 includes a plurality of actuators for driving the respective parts of the press machine. More specifically, the blank holder 30 generates the wrinkle pressing force H, raises and lowers the punch 14, and moves the punch 14 up and down. It is configured to perform speed control and other various driving.

As described above, the storage means 40 stores the fuzzy control basic data created for each type and processing condition of the blank 20 as a processing object in a database.

The control means 50 reads fuzzy control basic data suitable for the type and processing conditions of the blank 20 from the database stored in the storage means 40 when the blank 20 is actually drawn, and reads the read fuzzy control. The drawing speed V of the punch 14 is feedback-controlled in accordance with the above-described fuzzy inference method based on the basic data for use, the punch stroke S measured by the sensors 60a and 60d in real time, and the drawing amount r of the blank 20.

For this reason, the control means 50 of the present embodiment
Are configured to function as basic data specifying means 52, real-time evaluation data calculating means 54, and feedback control means 56.

FIG. 10 is a flowchart showing the operation of the drawing apparatus according to this embodiment.

Using the press machine of the present embodiment, FIG.
As shown in the figure, when drawing the blank 20,
First, the operator specifies the type (for example, material) of the blank 20 to be processed prior to processing (ST10), and specifies processing conditions for the blank 20 (ST11).

The input of the type and the processing condition may be displayed on a display as a selection screen, for example, and the operator may select the selection on the screen as appropriate.

When the type of blank 20 and the processing conditions are input, control means 50 functions as basic data specifying means 52, and reads out fuzzy control basic data suitable for these types and processing conditions from the database of storage means 40. (ST12).

When a series of such processes are completed, the control means 50 starts the movement of the punch 14 for drawing (ST13), and performs a fuzzy inference process (ST15).
A series of processing of feedback control of the drawing speed V of the punch 14 (ST16) is repeatedly performed until it is determined in ST14 that the processing is completed.

If it is determined that the processing has been completed, the punch 14 is stopped (ST17), and then the punch is lifted and retracted (ST18), and when returning to the predetermined reference position, a series of drawing operations is completed. .

Here, the above ST13, ST16,
In a series of processes such as ST17, the control unit 50 functions as the feedback control unit 56 and the actuator group 7
This is performed by driving 0.

FIG. 11 shows details of the fuzzy inference in ST15 described above.

This series of processing is performed by the control means 50 functioning as the real-time evaluation data calculation means 54 and the feedback control means 56.

First, real-time measured values of the punch stroke S and the drawing amount r of the blank 20 are obtained from the sensors 60a and 60d (ST30).
The first and second real-time evaluation data φ and φ ′ are calculated from the deviation between the data and the measured data (ST31). Then, the real-time evaluation data φ, φ ′ are converted into membership functions μφ,
used as an input value of Myufai', input membership functions of the partial fuzzy sets of the area _{A φL, A φS, A φ'L} ,
A _{φ ′S} is obtained (ST32).

Then, if-then shown in FIG.
Performs the determination process based on the rule (ST33), the area of each part fuzzy set of input membership functions described above according to the determination rule _{(A φL, A φ'L, A} φS,
A _{φ ′S} ) is substituted into the output-side membership function as shown in FIG. 9 (ST34).

Then, the center of gravity G of the substitution area 400 of the membership function for determining the output value shown in FIG. 9 is specified, and the deviation ΔV between the center of gravity G and the reference position is obtained as the punch speed change ΔV (ST35). ).

In ST16 shown in FIG. 10, feedback control is performed so that the drawing speed V of the punch is changed by the change amount ΔV thus obtained. (Verification of this embodiment)

Next, a verification experiment was performed to verify the results when deep drawing was performed by applying this embodiment.
The details will be described below.

FIG. 12A shows an outline of the processing apparatus used in this experiment, and FIG. 12B shows dimensional data of the apparatus used in the experiment.

Here, the blank 20 is 0.7 m
A cold-rolled steel plate having a thickness of m was used.

FIG. 13 shows the mechanical properties of this steel sheet.

The drawing ratio representing the size of the blank 20 is as follows.
It was set to 2.58. As the lubricant, a spray-type dry fluorine lubricant which is hardly affected by the speed was applied.

FIG. 3 shows actual measurement data when the punch drawing speed V is feedback-controlled using the fuzzy control method of the present embodiment under such experimental conditions. In the figure, reference numeral 120 denotes measured data at this time.

The punch speed V under feedback control is such that the speed at the start of machining is low, the speed rapidly increases to 130 mm / min in the initial stage of machining, and the condition of breaking at a constant speed in the middle to late stages of machining is 150 mm / min.
min, increasing to approximately 330 mm / min,
The drawing process was successful.

FIG. 14 shows the distortion of the minimum thickness in the punch portion of the deep drawing container formed by the experiment. As can be seen from the experimental results, the punch drawing speed is controlled by feedback control, as in the present embodiment, as compared with the measured value when deep drawing is performed with the punch drawing speed set to a constant value of 150 mm / min and 125 mm / min. It was confirmed that when the variable speed control was performed, the distortion was improved so as to be significantly reduced.

FIG. 15 shows that the punch drawing speed V is 12
A comparison is made between the processing time when deep drawing is performed at a constant speed of 5 mm / min and the processing time when performing variable speed control by feedback control using the fuzzy inference method as in the present embodiment. I have. From the figure, it was confirmed that the variable speed control according to the present embodiment was significantly reduced in processing time as compared with the case where deep drawing was performed at a constant processing time. In this experiment, it was specifically confirmed that the machining time could be reduced by 22%.

As described above, it has been confirmed that by using the system of the present embodiment and performing fuzzy adaptive control of the punching speed during deep drawing, it is possible to improve product accuracy and shorten the processing time. Was.

Note that the present invention is not limited to the above-described embodiment, and various modifications can be made within the scope of the present invention.

For example, a wrinkle holding force H of the blank holder with respect to the blank is added to one of the processing conditions, and a plurality of basic data for fuzzy control in a case where the wrinkle holding force is changed are prepared in advance, and are prepared as a database. You may comprise so that it may memorize | store in a memory | storage means.

As a result, even when the blanking force is changed and the punching drawing speed is variablely controlled for the same blank while changing the wrinkle pressing force, the value can be feedback-controlled to a more optimal value.

In the above embodiment, the case where the control means is formed as hardware has been described as an example.
The present invention is not limited to this, and a general-purpose computer is used as the control means, and a program for causing this computer to function as the control means, specifically, basic data specifying means, real-time evaluation data calculation means, feedback control means A program for causing the computer to function as such means may be stored in a storage medium, and the computer may be configured to function as control means based on the program.

In this case, the database stored in the storage means may be formed to be stored in the same storage means as the program, or may be formed to be stored in different storage means.

[Brief description of the drawings]

FIGS. 1 (A) to 1 (C) are schematic operation explanatory views of main parts of a press working machine for performing a series of deep drawing.

FIG. 2 is a functional block diagram of a punch drawing speed control mechanism in the press machine according to the present embodiment.

FIG. 3 is an explanatory diagram showing a relationship between a flange end reduction rate and a punch speed.

FIG. 4A is an explanatory diagram of a drawing amount of a blank during deep drawing, and FIG. 4B is an explanatory diagram of a relationship between a flange end reduction rate and a punch stroke.

FIG. 5A is an explanatory diagram showing a relationship between an ideal curve and an actually measured processing curve, and FIG. 5B is an explanatory diagram of a maximum value and a minimum value of an evaluation function.

FIG. 6 is an explanatory diagram of an input-side membership function.

FIG. 7A is an explanatory diagram of an input-side membership function, and FIG. 7B is a diagram illustrating if-th of fuzzy inference.
FIG. 7C is an explanatory diagram of an en rule, and FIG. 7C is an explanatory diagram of an output-side membership function.

FIG. 8 is an explanatory diagram of an input-side membership function for evaluating the risk of fracture when actual evaluation data is input.

FIG. 9 is a graph in which an area of an input membership function is substituted;
It is explanatory drawing of the output side membership function which evaluates a fracture | rupture risk.

FIG. 10 is a flowchart illustrating an example of the operation of the press device according to the present embodiment.

FIG. 11 is an operation flowchart of the fuzzy inference of the present embodiment.

FIG. 12 is an explanatory diagram of a press machine used for an experiment and dimensions thereof.

FIG. 13 is an explanatory diagram of material property values of a blank used in an experiment.

FIG. 14 is an explanatory diagram of actually measured values of minimum thickness distortion during deep drawing.

FIG. 15 is a diagram illustrating a comparison between machining times when the punch speed is controlled at a variable speed and when the punch speed is controlled at a constant speed.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Die 14 Punch 20 Blank 30 Blank holder 40 Storage means 50 Control means 60 Sensor group 70 Actuator group 200 Ideal processing curve 230 Actual processing curve

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Hiroshi Koyama 4-32-3 Sekido, Tama-shi, Tokyo F-term (reference) 4E089 EA01 EB02 EC01 EE01 FA09 FB06 FC05

Claims (12)

[Claims] 1. A deep drawing press machine for performing a deep drawing process by varying a punch speed on a workpiece fixed by a die and a blank holder, wherein a relation between a drawing amount of the workpiece and a punch stroke is determined. The data associated with the ideal machining curve obtained by performing the analysis as described above, and the evaluation associated with the deviation between the actually measured machining curve representing the relationship between the drawing stroke of the workpiece and the punch stroke obtained by the actual measurement and the ideal machining curve. Means for storing basic data for fuzzy control, including evaluation function data obtained from a function, as a database for at least one of the type of the processing target and the processing conditions associated with the type, and actual measurement during actual processing. The relationship between the drawn-in amount of the processed object and the punch stroke is obtained, and the data Specify the basic data for fuzzy control that conforms to the above item of the processing object from the data, determine the real-time evaluation data associated with the deviation between the data of the ideal processing curve and the measured data included in the specified basic data, Control means for performing feedback control on the drawing speed of the punch based on the obtained real-time evaluation data and the evaluation function data included in the specified basic data, a press machine for deep drawing. 2. The fuzzy control basic data according to claim 1, wherein the basic data for fuzzy control is stored in a database for each type of processing target and processing conditions, and the control unit matches the type of processing target and processing conditions from the database. A press machine for deep drawing, wherein basic data for fuzzy control to be specified is specified, and feedback control is performed on the drawing speed of the punch. 3. The press machine for deep drawing according to claim 1, wherein the feedback control is performed by fuzzy inference according to a fuzzy inference rule created based on the evaluation function. . 4. The evaluation function data according to claim 1, wherein the data of the evaluation function is first evaluation function data obtained from a first evaluation function representing a deviation between the actually measured processing curve and the ideal processing curve. And second evaluation function data obtained from a second evaluation function obtained by differentiating the first evaluation function, wherein the feedback control is performed based on the first and second evaluation function data. A press machine for deep drawing, which is performed by fuzzy inference according to the created fuzzy inference rules. 5. The fuzzy inference according to claim 1, wherein the feedback control is performed in accordance with a fuzzy inference rule created based on the first and second evaluation function data and a membership function using them. A press machine for deep drawing, characterized by being performed by: 6. The press machine for deep drawing according to claim 1, wherein the ideal working curve is obtained by finite element analysis based on predetermined setting conditions. 7. A deep drawing method using a press machine for performing a deep drawing process by varying a punch speed on an object fixed by a die and a blank holder, wherein a drawing amount of the object and a punch stroke are determined. The data of the ideal machining curve obtained by performing analysis to represent the association, the actual machining curve representing the association between the drawing stroke of the workpiece to be obtained by the actual measurement and the punch stroke, and the deviation between the ideal machining curve and Fuzzy control basic data including the evaluation function data obtained from the evaluated evaluation function and the processing object type and / or the processing conditions associated with the type are stored as a database for at least one item. At the same time, the relationship between the actually measured pull-in amount of the workpiece and the punch stroke is obtained, From the database, specify basic data for fuzzy control that matches the item of the processing target, and obtain real-time evaluation data associated with a deviation between the ideal processing curve data included in the specified basic data and the actual measurement data. A deep drawing method using a press machine, wherein feedback control of the drawing speed of the punch is performed based on the real-time evaluation data and the evaluation function data included in the specified basic data. 8. The fuzzy control basic data according to claim 7, wherein the basic data for fuzzy control is compiled into a database for each type of processing target and processing conditions, and the database is adapted from the database to the type of processing target and processing conditions. A deep drawing method comprising specifying data and performing feedback control on the drawing speed of the punch. 9. The deep drawing method according to claim 7, wherein the feedback control is performed by fuzzy inference according to a fuzzy inference rule created based on the evaluation function. 10. The evaluation function according to claim 7, wherein the evaluation function includes: first evaluation function data obtained from a first evaluation function representing a deviation between the actually measured processing curve and the ideal processing curve; And second evaluation function data obtained from a second evaluation function obtained by differentiating the first evaluation function, wherein the feedback control is created based on the first and second evaluation function data. A deep drawing method characterized by being performed by fuzzy inference according to a set fuzzy inference rule. 11. The fuzzy inference according to claim 7, wherein the feedback control is performed according to a fuzzy inference rule created based on the first and second evaluation function data and a membership function using the first and second evaluation function data. A deep drawing method characterized by being performed by: 12. The deep drawing method according to claim 7, wherein the ideal processing curve is obtained by finite element analysis based on predetermined setting conditions.