OPERATION OF A MULTI-AXIS SYSTEM

Abstract

In order to provide an optimized multi-axis system having mechanically coupled axes, a feedforward control identification process is provided, during which actual identification variables occurring in each case at the motor are each provided to identification units associated with the feedforward controllers, wherein feedforward control parameters are identified using the actual identification variables, and closed-loop controllers are parameterized using the feedforward control parameters.

Claims

1. A method for operating a multi-axis system comprising a plurality of basic axes and at least one drive axis which is mechanically coupled to the basic axes at coupling points, the positions of the coupling points on the basic axes being changeable by an associated motor in order to move the drive axis in relation to the basic axes, closed-loop controller being provided which are associated with the motors and determine control input variables from specified setpoint variables, preferably setpoint positions, and from associated corresponding actual variables, preferably the positions, occurring at the motor and provide said control input variables to the motors in order to control the corresponding actual variables in accordance with the specified setpoint variables, a feedforward controller associated with the motors being provided in each case, which feedforward controller in each case determines a feedforward control value based on the associated specified setpoint variable and superimposes said feedforward control value on the associated control input variable, wherein a feedforward control identification process is provided, during which actual identification variables occurring on each motor are each provided to identification units associated with the feedforward controllers feedforward control parameters being identified using the actual identification variables and in that the feedforward controllers are parameterized using the feedforward control parameters.

2. The method according to claim 1, wherein, during the feedforward control identification process, mutually synchronized setpoint variables are provided to the identification units and are used by the identification units to identify the feedforward control parameters.

3. The method according to claim 1, wherein further feedforward control parameters are interpolated from the determined feedforward control parameters.

4. The method according to claim 2, wherein, during the feedforward control identification process, profiles of mutually synchronized setpoint variables are provided to the identification units and are used by the identification units to identify the feedforward control parameters.

5. The method according to claim 1, wherein the mutually synchronized setpoint variables are specified by a central setpoint generator or by setpoint generators which are each synchronized with one another.

6. The method according to claim 1, wherein, during the feedforward control identification process, profiles of the actual identification variables are provided to the identification units and are used by the identification units to identify the feedforward control parameters.

7. The method according to claim 1, wherein an actual current or an actual torque is in each case used as the actual identification variable for the feedforward control identification process.

8. The method according to claim 1, wherein the position is in each case used as the actual identification variable for the feedforward control identification process.

9. The method according to claim 1, wherein the feedforward control identification process takes place before normal operation of the multi-axis system.

10. The method according to claim 1, wherein the feedforward control identification process takes place during normal operation of the multi-axis system.

11. The method according to claim 1, wherein the basic axes are arranged in parallel with one another.

12. The method according to claim 1, wherein at least some of the motors, preferably each motor, are rotary motors.

13. The method according to claim 1, wherein at least some of the motors, preferably each motor, are linear motors.

14. The method according to claim 1, wherein, during the feedforward control identification process, acceleration-proportional and/or speed-proportional and/or direction-dependent and/or constant components of the feedforward control parameters are determined.

15. A multi-axis system comprising a plurality of basic axes and a drive axis, the drive axis being mechanically coupled to the basic axes at coupling points and each of the positions of the coupling points on the basic axes being changeable by an associated motor in order to move the drive axis in relation to the basic axes, closed-loop controllers being provided which are each associated with the motors and are designed to determine control input variables from specified setpoint variables, preferably setpoint positions, and from associated corresponding actual variables occurring at the motor and to provide said control input variables to the motors in order to control these corresponding actual variables in accordance with the setpoint variables, preferably the positions, feedforward controllers being provided which are each associated with the basic axes and are designed to determine feedforward control values from the setpoint variables and to superimpose said values on the control input variables, wherein identification units are provided which are each associated with the basic axes and are each designed to identify feedforward control parameters using actual identification variables occurring at the motor and to parameterize the feedforward controllers using the feedforward control parameters.

16. The multi-axis system according to claim 15, wherein the identification units are each designed to determine the feedforward control parameters using profiles of the actual identification variables.

17. The multi-axis system according to claim 15, wherein at least one setpoint generator is provided that is designed to specify mutually synchronized setpoint variables, and in that the identification units are each designed to determine the feedforward control parameters using the mutually synchronized setpoint variables.

18. The multi-axis system according to claim 17, wherein the at least one setpoint generator is designed to specify profiles of mutually synchronized setpoint variables, and in that the identification units are each designed to determine the feedforward control parameters using the profiles of the mutually synchronized setpoint variables.

19. The multi-axis system according to claim 17, wherein a central setpoint generator is provided which is designed to specify mutually synchronized setpoint variables in the feedforward control identification process.

20. The multi-axis system according to claim 17, wherein mutually synchronized setpoint generators are provided which are designed to specify mutually synchronized setpoint variables in the feedforward control identification process.

Description

[0024] The present invention is described in more detail in the following with reference to FIGS. 1 and 2, which schematically show, in a non-limiting manner, advantageous embodiments of the invention by way of example. In the drawings:

[0025] FIG. 1 shows an exemplary multi-axis system having basic axes which are mechanically coupled via a drive axis,

[0026] FIG. 2 shows an identification of feedforward control parameters in the multi-axis system.

[0027] FIG. 1 shows a multi-axis system 1 comprising a plurality of basic axes X1, X2 and a drive axis Y, the drive axis Y being mechanically coupled to the basic axes X1, X2 at coupling points K1, K2. The coupling points K1, K2 can also be regarded as coupling regions. The positions p1, p2 of the coupling points K1, K2 on the basic axes X1, X2 are each changeable by means of an associated motor M1, M2, as a result of which the drive axis Y can be moved with respect to the basic axes X1, X2. The motors M1, M2 can each be arranged on the drive axis Y or on the basic axis X1, X2. Furthermore, an object can be provided on the drive Y, the object position of which is preferably movable along the drive axis Y (not shown). A two-dimensional Cartesian coordinate system is provided in FIG. 1, the axis of abscissas y and the ordinate axis x spanning the xy plane. The basic axes X1, X2 are in this case, for example, parallel to the ordinate axis x, the drive axis Y being parallel to the axis of abscissas y.

[0028] Multi-axis systems 1 according to the invention can be used, for example in injection molding machines, laser cutters, glass cutters, woodworking machines, etc., to carry out machining processes or manufacturing processes; an object can in this case be arranged on the drive axis Y, which object is positioned with a high level of precision in a working region by movement of the drive axis Y. For example, a tool and/or a camera can be arranged as an object on the drive axis Y, which can thus be positioned with a high level of precision in a target region (e.g. in a working region) by movement of the drive axis Y, for example in machining processes or manufacturing processes. The object position of the object can preferably be moved along the drive axis Y (i.e. along the axis of abscissas y in FIG. 1), which can be carried out by a further motor. It is also conceivable that a fine positioning system is also provided, which, after the movement/positioning of the drive axis Y and/or the object, carries out an additional, even more precise adjustment of the positions p1, p2 and/or the object along the drive axis Y.

[0029] The system 1 shown in FIG. 1 corresponds to a gantry system. One or more further basic axes can also be provided, each of which is connected to the drive axis Y via further coupling points. The positions of the other coupling points on the further basic axes can be analogously controlled by a further motor.

[0030] An exemplary identification of feedforward control parameters is shown in FIG. 2. The motors M1, M2 are each controlled by a closed-loop control unit R1, R2 by means of a control input variable u1, u2, a plurality of actual variables occurring at the motors M1, M2, e.g. positions p1, p2, speeds v1, v2, accelerations a1, a2, torques T1, T2, currents i1, i2 etc. When controlling the motors M1, M2, servo amplifiers and/or electronic 1:1 transmissions can also be provided in each case.

[0031] A setpoint variable w1, w2 is provided to each of the closed-loop control units R1, R2. Furthermore, a corresponding actual variable x1, x2 from the number of actual variables, which corresponding actual variable is associated with the setpoint variables w1, w2, is fed back to the closed-loop control units R1, R2 in order to determine the control input variable u1, u2. As mentioned, the control input variable u1, u2 is provided to the associated motor M1, M2 in order to control the corresponding actual variable x1, x2 to the associated setpoint variable w1, w2 in each case. This means that, when specifying a setpoint position p1soll, p2soll as the setpoint variable w1, w2, the position p1, p2 is used as a corresponding actual variable x1, x2 in order to determine the control input variables u1, u2.

[0032] The corresponding actual variables x1, x2 of the respective axes X1, X2, i.e., of the coupling points K1, K2, can be changed fundamentally independently of one another via the motors M1, M2. However, since the axes X1, X2 or the coupling points K1, K2 are mechanically connected to one another via the drive axis Y, the closed-loop control units R1, R2 are coupled to a common setpoint generator 3 during normal operation of the multi-axis system 1, which common setpoint generator provides the setpoint variables w1, w2 to the closed-loop control units R1, R2 in a synchronized manner. Therefore, during operation, the motors M1, M2 are moved only as a group by the provision of the setpoint variables w1, w2 by the setpoint generator 3, in order to prevent mechanical tension.

[0033] In order to improve the dosed-loop control behavior, in particular the following error behavior, of the closed-loop control units R1, R2, one feedforward control unit V1, V2 is provided per axis X1, X2 in each case. The feedforward control units V1, V2 each receive the associated setpoint variables w1, w2 and each determine a feedforward control value v1, v2 therefrom for applying to the control input variable u1, u2. The feedforward control unit V1, V2 can in each case be an integral part of the associated closed-loop control unit R1, R2 or it can also be designed independently.

[0034] The feedforward control units V1, V2, however, have to be parameterized using suitable feedforward control parameters P1, P2. For this purpose, an identification unit l1, l2 is provided in each case according to the invention. The identification units l1, l2 each receive an actual identification variable x1, x2 from the plurality of actual variables of the associated axis X1, X2 and use this to determine the feedforward control parameters P1, P2. The actual identification variables x1, x2 preferably correspond to the corresponding actual variables x1, x2.

[0035] Setpoint variables w1, w2, preferably profiles of setpoint variables w1, w2, corresponding to mutually synchronized identification profiles are preferably provided by a centrally arranged setpoint generator 3 or mutually synchronized setpoint generators 3. These setpoint variables w1, w2 are provided to the associated closed-loop control units R1, R2 and the feedforward control units V1, V2 and preferably also to the identification units l1, l2.

[0036] The identification units l1, l2 can thus not only receive actual identification variables x1, x2 from the plurality of actual variables of the associated axis X1, X2, but also setpoint variables w1, w2, in order to identify the feedforward control parameters P1, P2 therefrom. The corresponding actual variables x1, x2 correspond to the setpoint variables w1, w2. The actual identification variables x1, x2 can correspond to the corresponding actual variables x1, x2 or be of a different type.

[0037] For example, the closed-loop control units R1, R2 may each be designed as a closed-loop position controller, as a result of which they receive setpoint positions pisoil, p2soll as setpoint variables w1, w2 and correspondingly control positions p1, p2 in as corresponding actual variables x1, x2. Actual identification variables x1, x2 (from the number of actual variables) are provided to the identification units l1, l2, in order to identify the feedforward control parameters P1, P2. The positions p1, p2, for example, can be used as actual identification variables x1, x2, as a result of which the actual identification variables x1, x2 correspond to the corresponding actual variables x1, x2. However, currents l1, l2 and/or torques T1, T2, speeds v1, v2, accelerations a1, a2, etc., can also be used as actual identification variables x1, x2. In addition, the setpoint variables w1, w2 (preferably setpoint positions p1soll, p2soll) can be provided to the identification units l1, l2 in order to determine the feedforward control parameters P1, P2. The setpoint variables w1, w2 (in the case mentioned, the setpoint positions p1soll, p2soll) are consistently specified for each axis X1, X2 by the setpoint generator 3, preferably as an identification profile, which, however, takes place in a mutually synchronized manner.

[0038] An identification of the feedforward control parameters P1, P2 on the basis of the associated identification profile (i.e. the profile of the setpoint w1, w2) and of the profile of the actual identification value x1, x2 takes place independently of one another in the individual identification units l1, l2. The identification units l1, l2 also parameterize the feedforward control units V1, V2 in accordance with the determined feedforward control parameters P1, P2.

[0039] The feedforward control parameters P1, P2 advantageously contain an acceleration-proportional component (e.g. a moment of inertia), a speed-proportional component (e.g. viscous friction), a direction-dependent component (e.g. static friction in the positive/negative direction) and/or a constant component (e.g. gravity).

[0040] If similar basic axes X1, X2 are mechanically coupled to the drive axis Y via coupling points K1, K2 and the coupling points K1, K2 are positioned centrally on the basic axes X1, X2, the respective feedforward control parameters P1, P2 of the feedforward control units V1, V2 are identical. However, if the coupling points K1, K2 are not arranged centrally on the basic axes X1, X2, then the feedforward control parameters P1, P2 of the feedforward control units V1, V2 vary, in particular the inertia as an acceleration-dependent component of the feedforward control parameters P1, P2.

[0041] As mentioned, an object can be arranged so as to be movable along the drive axis Y. It is advantageous if the object is arranged at one or more extreme positions along the drive axis Y (for example on the heads of the drive axis Y) and the feedforward control parameters P1, P2 are then identified according to the invention. In this way, the feedforward control parameters P1, P2 can be identified in the event that the object is located on a head (i.e. on a coupling point K1, K2 in FIG. 1) of the Y-axis (first extreme position). A further identification of the feedforward control parameters P1, P2 can be carried out in the event that the object is located on a different head (i.e. at the other coupling point K1, K2 in FIG. 1) (second extreme position). The feedforward control parameters P1, P2 for any object position along the drive axis Y can be determined from the identified feedforward control parameters P1, P2 for the respective extreme positions of the object by means of, for example, linear interpolation. This means that the feedforward control parameters P1, P2 are known as a function of the object position.