JP3174704B2 - Numerical controller with position error correction function - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates also relates to a numerical control equipment for correcting the position of the feed axis of the machine tool, regarding the numerical controller for correcting the position error using the particular previously stored correction amount
You .

[0002]

2. Description of the Related Art A feed shaft of a machine tool, the operation of which is controlled by a numerical controller, has an error between a target position and a position actually reached. Factors that cause this error include a pitch error of the feed screw, a gear pitch error, and an error in straightness in the direction of the perpendicular axis.

Conventionally, in order to correct such an error caused by a mechanical system, a correction function such as pitch error correction, interpolation type pitch error correction, or straightness correction has been used in combination.

Further, the position error changes three-dimensionally. That is, even in the movement in the X-axis direction, the error at the target position changes depending on the values of the X-axis, Y-axis, and Z-axis. Therefore, conventionally, it has been considered to correct a three-dimensional error by combining pitch error corrections in three axial directions. As such an example, the present applicant has filed Japanese Patent Application Laid-Open No. 57-61906.

[0005]

However, in the conventional correction function, since a plurality of correction functions must be used, there is a problem that control for error correction becomes complicated.

Further, if correction is performed by a function combining pitch error correction in three axial directions in order to correct a three-dimensional position error, the number of data required in advance becomes enormous. Storing these data requires a large amount of memory,
There has been a problem that the cost of the numerical controller increases.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a numerical controller having a position error correction function capable of three-dimensionally correcting a position error with one correction function. .

[0008]

In order to solve the above-mentioned problems, the present invention provides a numerical controller for controlling the movement of a feed axis of a machine along each axis in a three-dimensional coordinate system. Grid point correction vector storage means for dividing into grid areas at predetermined intervals and storing grid point correction vectors measured in advance at the grid points of the grid area, and interpolating the movement amount of the feed axis according to a movement command. Predetermined interpolation cycle
And interpolating means for outputting in the period, the interpolation and the present position recognizing means for moving quantity by adding to recognize the current position of the axis, the current that is calculated based on the current location correction vector at the current position on the lattice point compensation vectors a position correction vector calculating means, the current position correction vector, the correction movement amount output means for comparing the starting position correction vector in the former current location before interpolation, and outputs the change amount as a correction moving amount, the
An adder for adding the correction movement amount to the interpolation movement amount at every predetermined interpolation cycle .

[0009]

The grid point correction vector storage means divides the coordinate system into grid-like areas at predetermined intervals in each axis direction, and stores grid-point correction vectors measured in advance at grid points set at the corners of the grid-like area. I do. The interpolation means outputs the interpolation movement amount of the feed axis in a predetermined interpolation cycle according to the position command. The current position recognizing means recognizes the current position specified by the position command based on the interpolation movement amount. The current position correction vector calculation means calculates a current position correction vector at the current position for each correction cycle based on the grid point correction vector. The corrected movement amount output means compares the current position correction vector with the start position correction vector at the old current position before interpolation, converts the change amount into a corrected movement amount, and outputs the corrected movement amount. Adder means interpolation
The correction movement amount is added to the interpolation movement amount for each cycle .

Thus, a position error at an arbitrary point in the three-dimensional space can be corrected by one interpolation function.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of the present invention. The lattice point correction vector storage means 6 stores lattice point correction vectors at lattice points. The lattice points are points set by dividing the three-dimensional coordinate system into lattice regions at regular intervals along each axis and intersecting the boundary lines of the lattice regions. The lattice point correction vector is a three-dimensional vector for correcting a position error measured in advance at each lattice point.

Various operations in the machine tool are commanded by the machining program 1. The preprocessing calculation means 2
It decodes the machining program 1 and outputs a movement command for each axis.
The interpolation means 3 converts the movement command of each axis into the interpolation movement amount (the actual movement amount).
In the embodiment, since the movement amount is output as a pulse,
Hereinafter, this interpolation movement amount is converted into an interpolation pulse) and output.

The current position recognizing means 4 fetches the interpolation pulse and recognizes the current position of each axis by dividing it into an X axis coordinate value 4a, a Y axis coordinate value 4b and a Z axis coordinate value 4c. The current position correction vector calculation means 5 calculates a current position correction vector at the current position based on the grid point correction vector. This is an interpolation-type error correction performed for each correction cycle. The correction pulse output means (correction movement amount output means) 7
Compares the starting point position-interpolation positive vector and the current location correction vector in the interpolation before the former current location, and converting a change amount of the correction pulse of each axis (correction movement amount), and outputs.

Adder means 8a, 8 provided for each axis
b and 8c are the interpolation pulses output from the interpolation means 3,
Add the correction pulse. Servo amplifiers 9a, 9b, 9c
Controls the rotation of the servo motors 10a, 10b, 10c according to the amounts of the input interpolation pulse and correction pulse.

FIG. 2 is a block diagram showing the configuration of an interactive numerical control device according to one embodiment of the present invention. The processor 11 controls the whole interactive numerical controller according to a system program stored in the ROM 12. As the ROM 12, an EPROM or an EEPROM is used. RA
An SRAM or the like is used for M13, and various data or input / output signals are stored. The non-volatile memory 14 stores the C backed up by a battery (not shown).
A MOS is used, and a parameter to be held even after the power is turned off, a tool correction amount, a grid point correction vector, and the like are stored.

The axis control circuit 21 receives an axis movement command from the processor 11 and outputs an axis command to the servo amplifier 22. The servo amplifier 22 drives the servo motor of the machine tool 50 in response to the movement command. The PMC (programmable machine controller) 23 receives a T function signal (tool selection command) and the like when executing the NC program. Then, these signals are processed by a sequence program to output signals as operation commands.
Control 0. Further, it receives a state signal from the machine tool 50, performs a sequence process, and transfers a necessary input signal to the processor 11.

All of the above components are connected to each other by a bus 19, and this bus 19
A processor 31 for dialogue is connected by a bus 29 separately from the processor 11 which is a CPU for communication.

The processor 31 has a bus 39 and a ROM
32, a RAM 33, a nonvolatile memory 34, a VRAM (video RAM) 35, and a graphic control circuit 36 are connected. The processor 31 executes an interactive processing program stored in the ROM 32, and displays settable operations or data in a menu format on a display device 43 described later on an interactive data input screen. In addition, a machining program is created from the data input in this way, and the overall motion trajectory of the tool is displayed as a background animation. The ROM 32 stores an input screen for the interactive data in addition to the interactive processing program. An SRAM or the like is used as the RAM 33, and various data for conversation and the like are stored. The non-volatile memory 34 uses a CMOS that is backed up by a battery (not shown), and stores program data and machining programs to be retained even after the power is turned off.
The VRAM 35 is a high-speed accessible RAM,
When performing a cutting simulation of the machine tool 50 based on the machining program stored as an NC statement in the non-volatile memory 34, graphic data for displaying an animation is stored. The graphic control circuit 36 is VR
The graphic data stored in the AM 35 is converted into a signal for display and output.

Also, a CRT / MDI (Cathod Ray Tube / Manual) which has a human interface with an operator.
Data Input) Panel 40 is connected to bus 19, and is provided with graphic control circuit 41, switch 42, display device 43, keyboard 44, and software keys 45.

The graphic control circuit 41 includes the processor 1
The digital signal output from 1 or the like is converted into a signal for display and output. The switch 42 is a graphic control circuit 3
6 or the display signal output from the graphic control circuit 41 is switched and applied to the display device 43. Display device 4
3 is a CRT or a liquid crystal display device. The keyboard 44 is composed of symbolic keys, numeric keys, etc.
Necessary graphic data and NC data are input using these keys. The software key 45 is a command key whose function changes according to a system program or the like, and its function name or the like is displayed at a predetermined screen position on the display device 43.

For displaying a background animation, the program data is transferred to the interactive processor 31.
Is converted into display data, converted into a display signal by the graphic control circuit 36, and displayed on the display device 43. Similarly, when a machining program is created in an interactive manner, the shape, machining conditions, and the like are controlled by the graphic control circuit 4.
The signal is converted into a signal for display by 1 and displayed on the display device 43.

Further, an input / output interface 46 is connected to the bus 19 to input / output data including NC data to / from an external device such as an FDD (floppy disk device), a printer, or a PTR (paper tape reader). Control.

FIG. 3 is a diagram showing a three-dimensional coordinate system divided into a lattice area. The three-dimensional coordinate system is divided into a grid at regular intervals. Point of intersection of the boundary line that divides the grid pattern is a lattice point P ₀ to P _26. It should be noted that what is shown in this figure is a part of the coordinate system, and in fact, the entire area in which the machine can be moved is divided into such a grid.

Then, the position error due to the mechanical system at each grid point is measured in advance. This error is represented by a three-dimensional vector. Therefore, a vector having the same absolute value as the position error and in the opposite direction is defined as a lattice point correction vector (in the figure, an arrow is shown on each lattice point). The grid point correction vector thus obtained is stored in the nonvolatile memory 14 (shown in FIG. 2) or the like. Note that this lattice point correction vector is an absolute value.

When the data is divided into lattices, if the division is made too fine, the data amount of the lattice point correction vector increases, and the required storage capacity of the memory increases. Therefore, the number of grid points that can calculate the correct amount of correction as much as possible and can reduce the data amount is about 10 per axis.

Here, a method of calculating a correction vector at an arbitrary machine position will be described. This is a function performed by the current position correction vector calculation means 5 shown in FIG. FIG. 4 is a diagram showing a grid-like area including a machine position at which a correction vector is to be calculated. In this example, the machine position at which the correction vector is to be calculated is a point P (P _x , P _y , P _z ).
_0, in _{P 1, P 3, P 4} , P 9, P 10, P 12, a region surrounded by the P _13. The grid spacing on the X axis is L _X , the grid spacing on the Y axis is L _Y , and the grid spacing on the Z axis is L _Z. The grid points P ₀ , P ₁ , P ₄ , P ₃ , P ₉ , P ₁₀ , P ₁₃ , and P ₁₂ have a grid point correction vector V1 (V1 _X , V1
_{1 Y, V1 Z) ~V8 (} V8 X, V8 Y, V8 Z) is set. Hereinafter, this area is assumed to be a vector field having linearity in which grid point vectors corresponding to the respective grid points are provided at the grid point positions.

When an area including the point P is obtained, the grid point P
_{0 (P 0X, P 0Y,} P 0Z) defined as a reference point. next,
To find the correction vector at point P, first the position in the grid is normalized to [0,1]. Let the lattice spacing on the X axis be Lx,
The coordinate value (x, y, z) of the normalized point P when the Y-axis grid interval is Ly and the Z-axis grid interval is Lz is determined by the following equation.

[0028]

X = (P _x −P _0x ) / L _X (1)

[0029]

Y = (P _y −P _0y ) / L _y (2)

[0030]

Z = (P _z −P _0z ) / L _z (3) Based on the coordinate values (x, y, z), the correction vector V (V _X , _VY , _VZ ) is calculated by the following equation.

[0031]

V _i = V1 _i · (1-x) (1-y) (1-z) + V2 _i · x (1-y) (1-z) + V3 _i · xy (1-z) + V4 _i · (1-x) y ( 1-z) + V5 i · (1-x) (1-y) z + V6 i · x (1-y) z + V7 i · xyz + V8 i · (1-x) yz ( i = x, y, z) (4) In this manner, a correction vector at an arbitrary point on the coordinate system can be calculated. That is, the current position correction vector calculation means 5 (shown in FIG. 1) can calculate the current position correction vector at the current position for each interpolation cycle using the above formula.

Incidentally, the calculated current position correction vector is an absolute value similarly to the lattice point correction vector. However, it must be converted to an incremental value to add to the interpolation pulse. Therefore, the amount of change of the current position correction vector from the start point correction vector at the old current position before interpolation is calculated. The starting point correction vector is
At the time of the previous interpolation, the current position correction vector has already been obtained. Assuming that the starting point correction vector at the position before the interpolation is U (U _X , U _Y , U _Z ), the correction vector C (C _X , C _Y , C _Z ) to be added is

[0033]

C = V−U (5) The value of the correction vector C is converted into a correction pulse and added to the interpolation pulse. The servo amplifier controls the movement of each axis according to the interpolation pulse to which the correction pulse has been added. As a result, the error caused by the mechanical system at the current position is corrected.

FIG. 5 is a flowchart showing the procedure for calculating the correction pulse. This flowchart is started when the current position is determined. [S1] An area including the current position is obtained, and one of the grid points is set as a reference point. [S2] The current position correction vector V is calculated based on the lattice point position correction vector. [S3] Start point correction vector U of current position correction vector V
Is calculated, and a correction vector C is calculated. [S4] The correction vector C is converted into a correction pulse and output.

As described above, every time an interpolation pulse is output, a three-dimensional correction vector at the current position is obtained and added as a correction pulse. Therefore, a position error in a three-dimensional space caused by a mechanical system is calculated by one interpolation pulse. It can be corrected by the shape error correction function. In addition, since the error is measured in advance at several points in the three-dimensional space and the error at an arbitrary position is calculated from the measurement result, it is necessary to take into account what causes the position error. Absent. In addition, since the number of grid point correction vectors to be measured in advance is small, the memory capacity can be small, and the numerical controller can be inexpensive.

[0036]

As described above, in the present invention, each time an interpolation pulse is output, a correction vector is calculated based on all three-dimensional coordinate values at the current position, and added as a correction pulse. The three-dimensional position error caused by various factors of the mechanical system can be corrected by one error correction function.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of the present invention.

FIG. 2 is a block diagram showing a configuration of an interactive numerical control device according to one embodiment of the present invention.

FIG. 3 is a diagram showing a three-dimensional coordinate system divided into a grid area.

FIG. 4 is a diagram illustrating a grid-like area including a machine position at which a correction vector is to be calculated.

FIG. 5 is a flowchart illustrating a procedure for calculating a correction pulse.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Machining program 2 Preprocessing calculation means 3 Interpolation means 4 Current position recognition means 5 Current position correction vector calculation means 6 Lattice point correction vector storage means 7 Correction pulse output means 8a, 8b, 8c Addition means 9a, 9b, 9c Servo amplifier 10a , 10b, 10c Servo motor

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-224003 (JP, A) JP-A-3-161247 (JP, A) JP-A-2-243247 (JP, A) JP-A-60-1985 205713 (JP, A) (58) Field surveyed (Int. Cl. ⁷ , DB name) G05B 19/404 G05B 19/4103

Claims (3)

(57) [Claims] 1. A numerical control device for controlling the movement of a feed axis of a machine along each axis in a three-dimensional coordinate system, wherein the coordinate system is divided into grid areas at predetermined intervals in each coordinate axis direction. A grid point correction vector storage means for storing grid point correction vectors measured in advance at grid points; and a predetermined interpolation of an interpolation movement amount of the feed axis according to a movement command.
Interpolating means for outputting at a periodic interval ; current position recognizing means for recognizing the current position of the axis by adding the interpolated movement amount; and current for calculating a current position correction vector at the current position based on the grid point correction vector. Position correction vector calculating means; comparing the current position correction vector with a start point position correction vector at the old current position before interpolation, and outputting a change amount as a correction movement amount; and the predetermined interpolation cycle And a adding means for adding the correction movement amount to the interpolation movement amount every time . 2. A current position correction vector calculating means, wherein the grid point correction vector is provided at the grid point, a vector field having linearity is assumed, and a vector amount of the current position in the vector field is corrected by a current position correction. The numerical control device with a position error correction function according to claim 1, wherein the numerical control is calculated as a vector. 3. The numerical control device with a position error correction function according to claim 1, wherein said lattice point correction vector storage means is a nonvolatile memory backed up by a battery.