JPH06187029A - Nc controller - Google Patents

Abstract

PURPOSE:To obtain a smooth working locus close to a normal curve by provid ing with a curve approximating circuit which approximates the working locus to the curve, and with a curve dividing circuit which divides the approximate curve in each sampling, and prepares new data. CONSTITUTION:When the operation of a curve approximation is discriminated by a curve approximation discriminating circuit 14, the calculation of the curve approximation is operated by a curve approximating circuit 15 based on the data of an NC program analyzing block 2. Afterwards, a parameter for deciding the number of division being the number of points on the curve is decided by a curve dividing circuit 16. Thus, the working locus close to the curve can be obtained, so that cutting can be attained by the smooth locus, and a grinded face can be finely finished without being given a mechanical shock.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an NC control device for machining program processing applied to an NC device of a machine tool.

[0002]

2. Description of the Related Art FIG. 7 shows processing blocks of an NC device from a conventional NC program to a speed command output and output data for the processing. That is, in the conventional NC device shown in FIG. 7, first, the NC program is read in the NC program reading block 1 and the NC program is analyzed in the NC program analysis block 2 to increase the movement amount of each axis for each program block. Calculate across blocks. The processing speed calculation block 3 calculates the target speed of each block, and the acceleration / deceleration circuit 4 performs acceleration / deceleration toward the target speed to determine the speed at each sampling time. The straight line division circuit 5 determines the combined movement amount for each sampling time along the command locus. In each axis distribution circuit 6, the amount of movement of each axis for each sampling time is calculated by distributing to each axis. The command position integration circuit 7 determines the command position by integrating the movement amount of each axis. By performing feedback control in the feedback control circuit 8, the speed output in feedback control is determined.

Further, the feedforward circuit 9 carries out feedforward correction of the inverse transfer function system by using the amount of movement of each axis for each sampling time, whereby accurate control is performed. Then, the speed command output circuit 10
Then, the output of the feedback control circuit 8 and the output of the feedforward circuit 9 are added to obtain the speed output to the servo amplifier.

[0004]

In the conventional block shown in FIG. 7, the inverse transfer function type feedforward correction is represented by the following equation (Equation 1).

[Equation 1]

The feedforward correction amount calculated by this equation is as shown by the shaded portion in FIG. 3, and it is a constant correction amount during one block of the command program, and it protrudes in a pulse shape at the joint of the blocks. It is the correction amount. Therefore, a smooth locus is not formed at the seam of the blocks, which gives an impact to the machine. This is because, as shown in the trajectory of the NC program shown in FIG. 2, the correction amount in the above (Equation 1) is only the term of (P _n −P _n−1 ) / Δt, since it moves linearly within one block, whereas in a state where acceleration and deceleration are not applied as a constant value, in the relay Ki eye blocks the second term _{(P n -P n-1 +} P n-2) × K /
Since (Δt) ² suddenly works, it is a pulse-like protruding correction amount.

An object of the present invention is to provide an NC control device which does not give a shock to a machine without causing a correction amount to project in a pulse shape.

[0007]

The present invention that achieves the above-mentioned object is (1) NC control for converting a machining locus for each block by an NC program into each axis movement amount for each sampling time and performing feedforward correction. In the device,
And a curve dividing circuit that divides the approximated curve for each sampling to generate new data.
(2) In the NC controller that converts the machining locus for each block by the NC program into each axis movement amount for each sampling time and performs feedforward correction, insert a curve approximation circuit that approximates the machining locus to a curve. The present invention is characterized in that the curve approximation circuit is provided with a determination output indicating whether or not the curve path is determined by the curve approximation determination circuit and an output for determining the number of curve divisions by the curve division circuit based on parameters.

[0008]

In order to prevent the correction amount shown in the shaded portion of FIG. 3 having a pulse-like protrusion, the protrusion correction amount is dispersed, and in addition, it is dispersed finely at each sampling time of the speed output command to the servo amplifier. The command locus is approximated to a curve to create a more regular locus, the correction amount is calculated by dividing the locus for each sampling time, and the correction amount is distributed for each sampling time. That is,
Aforementioned _{(P n -2P n-1 +} P n-2) × K / (Δt) the correction amount of the ^second term of each sampling time (P _'n -2P'
It is calculated as the correction amount of the term of ( _n-1 + P _n-2 ) × K / (Δt) ² . The formula of the correction amount according to the present invention is based on the following (Equation 2).

[Equation 2]

In another invention, after determining the curve approximation, the curve approximation circuit calculates an approximate trajectory, and the number of curve divisions is set according to the parameter setting value. The conventional tool path point group data is interpolated. As a result, detailed point cloud data that has been curve-approximated is obtained.

[0010]

Embodiments Two embodiments will now be described with reference to the drawings. The same parts as those in FIG. 7 are designated by the same reference numerals and the description thereof will be omitted. In FIG. 1, the block division number calculation unit 12 is provided at the next stage of the acceleration / deceleration circuit 4 in which the speed for each sampling is determined.
Is placed. The calculation unit 12 can calculate the combined movement value based on the speed for each sampling time, and knows the movement amount of one block, and thus the number of divisions for each block is calculated.

The curve approximation circuit 11 analyzes the NC program.
Obtain a large amount of movement per block by block 2
Are processed based on the values calculated over the period. This
The curve approximation of is performed, for example, in the curve approximation example of FIG.
It That is, vector A of the read data, vector
In B, the coordinates of the start point and the end point are P_i, P_{i + 1}, P
_{i + 2}And An example of a Bezier curve is the point P on the curve.
Coordinate is P = P _i(1-t)³+ Q_i・ 3 (1-t)
²t + Q_{i + 1}・ 3 (1-t)²t²+ P_{i + 1}t³And
To be done. At this time, Q_i, Q_{i + 1}Is point P_i, P_{i + 1}From
Vector T _i, T '_{i + 1}The coordinates are far apart. This place
Vector T_i, T '_{i + 1}Is obtained by the following equation (Equation 3)
It

[Equation 3]

By changing this t between 0 and 1, the coordinates of the point P on the curve are determined. That is, this t
Thus, the approximate curve can be divided at each sampling time. In the curve division circuit 13, as shown in FIG.
The above-mentioned t is determined so that the curve division is performed with the same division number as the block division number of the block division number calculation unit 12, and the combined movement amount for each sampling time is calculated. By performing feedback control and feedforward correction after such curve division, a correction amount such as the white pulse in FIG. 3 can be obtained and the feedforward correction amount can be dispersed without any impact on the machine.

FIG. 5 shows a circuit block of another embodiment. In FIG. 5, after the NC program analysis block 2,
The curve approximation determination circuit 14 determines whether or not the data of the movement amount of each axis by the block 2 approximates a curve.
Next, a curve approximation circuit 15 that receives the determination of the curve approximation determination circuit 14 and performs a curve approximation calculation based on the data of this block 2 is provided at the subsequent stage of the NC program analysis block 2. This curve approximation circuit is shown in FIG.
The point P on the curve is determined by the Bézier curve method shown in FIG. Next, the curve approximation circuit 15 is followed by a parameter-based curve division circuit 16 to determine a parameter for determining the number of divisions, which is the number of points P on the curve. That is, as shown in FIG. 6, a parameter, which is the number of divisions, for setting a new locus shown by a solid line with respect to the broken line of the curved approximation locus is set.

Since a machining locus close to a curve is obtained in this way, a smooth locus is used for cutting, so that no mechanical shock is applied and the cutting surface is finished cleanly.

[0015]

As described above, according to the present invention, it is possible to obtain a smooth machining locus close to a normal curve, and to disperse the feedforward correction amount without giving a shock to the machine. High precision processing is possible.

[Brief description of drawings]

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

FIG. 2 is an explanatory view of each locus and division of a command and approximation.

FIG. 3 is an explanatory diagram of a feedforward correction amount.

FIG. 4 is an explanatory diagram for curve approximation.

FIG. 5 is a block diagram of another embodiment of the present invention.

FIG. 6 is a comparison diagram of approximation, command, and new trajectory.

FIG. 7 is a block diagram of a conventional example.

[Explanation of symbols]

 11 curve approximation circuit 12 block division number calculation unit 13 curve division circuit 14 curve approximation determination circuit 15 curve approximation circuit 16 curve division circuit by parameters

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] April 15, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0011

[Correction method] Change

[Correction content]

The curve approximation circuit 11 analyzes the NC program.
Obtain a large amount of movement per block by block 2
Are processed based on the values calculated over the period. This
The curve approximation of is performed, for example, in the curve approximation example of FIG.
It That is, vector A of the read data, vector
In B, the coordinates of the start point and the end point are P_i, P_{i + 1}, P
_{i + 2}And An example of a Bezier curve is the point P on the curve.
Coordinate is P = P _i(1-t)³+ Q_i・ 3 (1-t)
²t + Q_{i + 1}・ 3 (1-t) t²+ P_{i + 1}t³Represented
Be done.Here, to explain a little about the Bezier curve,
4, the cubic function curve P = A from the point P _{i to the} point P _{i + 1}
Consider interpolation with t ³ + Bt ² + Ct + D. Figure
In 4, P is a point on the approximate curve and P _i Q _i is a point P _i .
Tangent line in P_{i + 1}Q _{i + 1}Is P_{i + 1}Tangent at point, expression
Middle t is from t = 0 to move from P _i point to P _{i + 1} point
It is a parameter up to t = 1. Therefore, the third order above
The function curve must further satisfy the following condition: t = 0
, P = P _i , when t = 1, P = P _{i + 1} , and t = 0
The slope P ′ (0) of the tangent line is K (Q _i −P _i ), where t = 1
The tangent slope P ′ (1) is then K ′ (Q _{i + 1} −P _{i + 1} ).
Become. If the above cubic curve is solved under this condition, P = P _i
(1-t)³+ Q _i・ 3 (1-t)²t + Q_{i + 1}・ 3
A ^{(1-t) t 2 +} P i + 1 t 3. However, K = 3
With K ′ = − 3, the size of P _i Q _i , P _{i + 1} Q _{i + 1} is empirical.
It was set to 1/3 of P _i P _{i + 1} which is considered to be good. Well, the above formula
The coordinate P can be obtained from the
Q of_i, Q _{i + 1}To identify_iPoint, P_{i + 1}
Consider the vectors T _i , T ′ _{i + 1} from the points. Where two
When paying attention to two connected vectors A and B,
A general expression for the tangent line between the cutouts A and B (for example, the vector T in FIG. 4)
_{i + 1} ) is obtained by the following equation (Equation 3) (1). This formula
By using (1), the vectors T _{i + 1} and T ′ _{i + 1}
Is obtained by the following equations (Equation 3) (2) (3).

[Equation 3]

Claims (2)

[Claims] 1. A machining locus for each block by an NC program is converted into each axis movement amount for each sampling time,
An NC control device for performing feedforward correction is provided with: a curve approximation circuit for approximating the machining locus by a curve; and a curve division circuit for dividing the approximation curve at each sampling time to generate new data. NC controller. 2. The machining locus for each block by the NC program is converted into each axis movement amount for each sampling time,
In an NC control device that performs feedforward correction, a curve approximation circuit that approximates the machining locus by a curve is inserted, and a determination output as to whether or not a curve route is made by a curve approximation determination circuit and a curve by a curve division circuit based on parameters An NC control device in which the determination output of the number of divisions is added.