DE19509325A1 - Automatic drift compensation method for position-controlled axles - Google Patents

Abstract

A known fixed portion of the drift is compensated by a machine datum stored in a memory element in the form of an offset. A speed value (Voffset) which is proportional to the fixed portion of drift is set to be unequal to zero on the output of the position controller (LR). For compensation of an additional variable, e.g. temperature dependent drift a resultant speed desired value (VVAR) at the output of the position controller is integrated for desired rest, until the sum of the fixed offset and the integrator value form a speed value (VDrift) which when switched to the output of the position controller compensates the true drift value.

Description

The drift in analog electronic components, e.g. B. existing in the input circuitry of an analog converter which can be leads to numerical controls, e.g. B. from Machine tools and robots, usually in Speed control loops to achieve a standstill of each because of a speed control loop with a speed setpoint un must be controlled equal to zero. A P position controller can only generate such a speed setpoint if on a slight following error even at standstill is available. However, this means that the Axis deviates from its target position. The deviation will as long as larger until the one on the position controller speed lag caused the speed setpoint to be so large has become that it corresponds to the drift. Only then reached the axis concerned the standstill.

Especially when using a drift from analog elec tronic components affected axis positioning movement to be carried out, one of axis position deviating from the target position as disadvantageous, because the precision required today, for example with the Manufacturing with numerically controlled machine tools is not can be guaranteed more or only insufficiently.

It is known that conventional methods for automatic Drift compensation to avoid such a drift error an additional speed setpoint on the output of the position switch on controller. In the conventional solution, numerical Controls has a position-controlled axis, e.g. B. a Feed, via a semi-automatic drift adjustment. Because of that ter one understands the necessity of an operating action of the  User for the purpose of determining a suitable drift Compensation value. This is usually done from finally at a standstill of the axis, with a speed setpoint value at the output of the position controller at the push of a button in a comm compensation branch of the numerical control is adopted. This speed setpoint remains until the next operator's hand effect unchanged.

In the case of a position-controlled spindle, e.g. B. Positionie workpiece slide is the problem of getting out wander of the spindle from its target position due to a Drift in analog electronic components still out more marked. For this reason you use conventional certainly a fully automatic drift adjustment. This fully automatic drift adjustment takes place with spindles, however only and only after performing a positioning instruction M19 instead, which is in the control language according to DIN 66025 ent a spindle stop in a defined end position speaks. Here, too, the compensation value remains up to next positioning instruction unchanged.

The above-mentioned standard procedures for Er Subsequent drift compensation therefore have the after part that they are not uniform for both position-controlled Axes can also be used for position-controlled spindles can. In addition, it is disadvantageous that in the case of automatic drift compensation for position-controlled spindles a drift adjustment only after executing one special positioning instructions can take place. This is However, it does not guarantee that the Spindle caused by analog electronic components gentle drift and the resulting deviation from actual posi tion and target position can be compensated. Come in addition, that a drift adjustment again and again over the whole th deviation range must be corrected, although in the Practice one caused by analog electronic components  gentle drift fluctuates only in a partial area while a fixed proportion of the drift usually occurs constantly. There is a drift adjustment across the entire deviation range however, very time consuming, so that drift compensation occurs this way is also very time consuming.

The invention is therefore based on the object of a method as well as the associated device for automatic drift To design compensation so that it is both for positional Applied axes as well as for position-controlled spindles uniform can be used. In addition, the drift compensation tion not only following the positioning instructions conditions, but each time the axis or spindle comes to a standstill, see above that drift compensation is independent at each standstill of any positioning instructions can. In addition, the adjustment of the drift should be possible done quickly.

According to the invention, this object is achieved by a method for automatic drift compensation for position-controlled axes and spindles with the following features:

• 1.1 a known fixed portion of the drift is over a in machine data stored in the form of a storage element of an offset compensated by a fixed portion of the Defined proportional speed value not equal Zero is switched to the output of the position controller,
• 1.2 to compensate for an additional variable, for example a wise temperature-dependent drift is a result ter speed setpoint at the position controller output at ge desired standstill until integrated until the total a speed value from a fixed offset and integrator value form which switched to the exit of the location regulates the true drift value.

An advantageous alternative embodiment, which the Solves the problem underlying the invention just as advantageously  and thus the possible uses of the Ver extended, contains the following features:

• 2.1 a known fixed part of the drift is over a in it stores the machine data in the form of a memory element of an offset compensated by a fixed portion of the Defined proportional following error on the Input of the position controller is switched,
• 2.2 to compensate for an additional variable, for example a wise temperature-dependent drift is a result ter following error at the input of the position controller at ge desired standstill integrated until the total a following error from fixed offset and integrator value form which intruded on the entrance of the location regulates the true drift value.

One with regard to the requirement, at every standstill automatic drift compensation for an axis or spindle make particularly advantageous embodiment it that the computational effort to determine the compensation limited and the automatic process for drift Compensation is accelerated. Such a way An advantageous embodiment has the following features:

• 3.1 a comparison of the integral part of the compensation values are due to possible changes in the Drifts automatically every time it stops,
• 3.2 as soon as the standstill is exited, the drift comm compensation switched off by no further values be integrated and the last current integral An part of the compensation value is saved.

Another advantageous embodiment of the present Er is characterized in that it has a desired Clear standstill of the corresponding axis or spindle and particularly reliably recognizes. At the same time, it ensures that changes in drift even during a standstill  be continuously compensated. Such an arrangement has the following features:

• 4.1 to detect a desired standstill is as Criteria used to ensure that setpoints do not change have more and / or a sentence end has been reached,
• 4.2 the automatic drift compensation remains active as long as fourth, how that criteria or criteria are met.

An advantageous device for performing the method According to the present invention, that they are effective with simple means and limited effort is realized. In addition, it can be particularly easily in integrate existing numerical controls. These The device has the following features:

• 5.1 in a memory element is a machine data as an offset stored, which is a known portion of the drift represented in the form of a speed setpoint,
• 5.2 an integrator is with a possibly at the output of the Applied to the position controller occurring speed setpoint,
• 5.3 the output of the integrator and the value of the machine end tums are switched to an adder,
• 5.4 Another adder is connected to the output of the first Adder and the output of the position controller,
• 5.5 the output of the second adder is with one on the Position controller following control level, especially one turn number control loop, connected.

An advantageous embodiment of the device is distinguished characterized in that they have a few additional components constant control enables whether a standstill actually is provided. This will make the system secure significantly improved. The components used are These are standard components, making them inexpensive Realization is made possible. This device has fol key features:

• 6.1 an activation input of the integrator is with a Signal connected, which signals whether a breastfeeding stand is desired
• 6.2 several signals to signal whether a standstill should take place on the input side to a logic build stone switched and its output with the activation input of the integrator connected.

Another advantageous embodiment of the present Sensation prevents settling and thus increases the Stability of drift compensation. It has the following note times:

• 7.1 the integrator has one compared to the camp large time constant.

The advantages achieved with the invention are in particular the other is that both position-controlled axes and position regulated spindles with a uniform procedure for auto Matic drift compensation can be operated. In addition comes that not only according to special positioning instructions, but every time an axis or spindle comes to a standstill Drift adjustment is carried out fully automatically. About that In addition, the drift adjustment is particularly quick because a Corresponding compensation value not at every standstill must be determined again completely. This is the required computing power low and the cost of one Realization accordingly low.

An embodiment of the invention is in the drawing shown and is described in more detail below. Here shows:

Fig. 1 is a block diagram of a device for automatic drift compensation using the example of a position-controlled axis.

The drawing of the figure shows a block diagram of a device for automatic drift compensation using the example of a position-controlled axis. Based on control data SD, these are processed in a pilot control circuit VSK and setpoints SW _{x are} formed from them. These setpoints reach a position controller LR at the output speed setpoints V _should be available. Due to the drift in used analog electronic components, e.g. B. in the input circuitry of an analog converter, a dar to the following speed controller DR with a small speed setpoint not equal to zero must be activated if the standstill of the underlying axis is to be reached. The position controller LR has such a speed setpoint V _soll at a standstill, since a small following error Δs is present at its input due to the drift even at a standstill. This is formed by the difference between the setpoint SW _x and the actual value S _{ist which} is fed back in the position control loop by means of a subtractor S. A resulting speed target value V _should not equal ZERO at the output of the position controller LR at a standstill is automatically compensated for by the present invention. For this purpose, an already known portion of the drift, which occurs regardless of the ambient conditions whenever the axis comes to a standstill, is stored using an _offset value V _{offset to be} entered in a memory element via machine data MD.

To determine a compensation value for a variable, z. B. temperature-dependent portion of the drift, the speed setpoint V Soll _{is integrated} at the output of the position controller LR at a standstill. For this purpose, the output of the position controller is led to the input IN of an integrator I. The integrator I advantageously has a time constant that is large in comparison to that of the position control loop in order to avoid transients. The speed setpoint V _{should be integrated} at the output of the position controller LR at a standstill until the sum of the machine data value MD, i.e. the offset V _offset and the integration value V _VAR , which corresponds to the variable component of the drift, corresponds to the actual drift value V _Drift . For this purpose, offset V _offset and the variable compensation value V _VAR at the output OUT of the integrator I are added in an adder AD1 and the sum V _drift of both values, which corresponds to the actual drift value, is fed to a further adder AD2. In the second adder AD2, the speed setpoint V Soll caused by the drift _{is added to} the output of the position controller LR and the total compensation value V _Drift present at the output of the first adder AD1. At the output of the second adder AD2 thus results a corrected rotation speed command value v _'to. This corrected speed setpoint V ' _{should be} fed to the speed controller DR, the output of which drives an axle drive A.

While a desired standstill of the axis is the pending at the beginning at the output of the position controller LR speed reference _set value V, which is supplied to the integrator I, integrated up there until the variable portion of the Kompensa tion value V _VAR, which at the output OUT of the integrator I to Is available, together with the fixed offset V _offset he gives a total compensation value V _Drift , which is proportional to the drift voltage that is applied to the input of the speed controller DR influenced by drift effects at the output of a speed value that results in a standstill of the Drive. This relationship can be seen in abbreviated form in the following calculation rule:

V _offset + V _VAR = V _drift (1)

At the moment when V _drift brings drive A to a standstill and the setpoint SW _x and the feedback actual value S _ist match, V _soll at the output of the position controller LR becomes ZERO and the compensation voltage is supplied exclusively by V _Drift . This compensates for the drift error at standstill and the system in a stable condition.

To prevent the drift compensation from being active even then remains when the machine is moving, according to the invention Verification criteria used to ensure that machine downtime is actually desirable. To do this, all processed criteria are based on a logic structure stone LB switched. Such a logic module LB is coming preferably a logical AND gate, the output of which is one Activation input E of integrator I controls. This Output of the logic block LB is only active if all criteria used are met.

Possible criteria for this are, on the one hand, that setpoints SW _{x have} small changes. In order to check whether a change in the setpoints SW _x occurs, the setpoints are switched directly to a comparator KP. In addition, the same setpoints SW _x are fed into a first-in-first-out memory FIFO, which they pass through. Due to the organization of such a memory FIFO, the values given on the input are only available with a time delay at its output. The register length n of the FIFO memory determines how far a comparison with previous setpoints SW _x is to be made. The output of the first-in-first-out memory FIFO is switched to the second input of the comparator KP. Depending on the register length n of the memory FIFO, the current setpoints SW _x and the values SW _xn corresponding to the register length n are present at the comparator KP. The comparator KP is wired in such a way that only when direct and delayed values match, setpoints SW _x thus experience no change, the output of the comparator KP is active. The comparator KP1 preferably has a hysteresis.

Another criterion for checking whether the machine has come to a standstill Machine is actually desired is to check whether a block end for the control data SD has been reached. The These two criteria are again in the following overview summarized:

K1: Setpoint shows no change
K2: Block end has been reached.

These two criteria K1 and K2 are switched together on the logic module LB. Only if the criteria K1 and K2 are met together is the output of the logic module LB active and thus the integrator I also activated. This ensures that the drift compensation is switched off as soon as drive A is desired according to the criteria K1, K2. As soon as the integrator I is deactivated, that is, no further integration of the speed setpoints V _{should occur} at the output of the position controller LR, the value last reached at the output GUT, that is to say the variable portion of the compensation value V _VAR , is stored. This has the result that the next time the integrator I is activated for automatic drift-off, this value can be used immediately and no complete redetermination of the variable portion of the compensation value V _{VAR is} necessary.

The elements shown in the block diagram of FIG. 1 can also be easily implemented using the standard means of a data processing system.

Claims (7)

1. Method for automatic drift compensation for position-controlled axes and spindles, in particular for use in machine tools and robots, with the following features:

• 1.1 a known fixed portion of the drift is compensated for by a machine data (MD) stored in a memory element in the form of an offset by switching a non-zero defined speed value (V _offset ) proportional to the fixed portion of the drift to the output of the position controller (LR) becomes,
• 1.2 to compensate for an additional variable, for example temperature-dependent drift, a resulting speed setpoint (V _VAR ) is integrated at the output of the position controller (LR) at the desired standstill until the sum of the fixed offset (V _offset ) and integrator value (V _VAR ) form a speed value (V _drift ) which, when connected to the output of the position controller (LR), compensates for the true drift value.

2. Method for automatic drift compensation for position-controlled axes and spindles, in particular for use in machine tools and robots, with the following features:

• 2.1 a known fixed part of the drift is compensated for by a machine data (MD) stored in a memory element in the form of an offset by switching a defined following error proportional to the fixed part of the drift to the input of the position controller (LR),
• 2.2 to compensate for an additional variable, for example temperature-dependent drift, a resulting following error at the input of the position controller (LR) is integrated at a desired standstill until the sum of the fixed offset and integrator value form a following error, which is applied to the input of the position regulator (LR) compensates for the true drift value.

3. The method according to claim 1 or 2 with the following features:

• 3.1 an adjustment of the integral part of the compensation value (V _VAR ) takes place automatically with every standstill due to possible changes in the drift,
• 3.2 As soon as the standstill is exited, the drift compensation (I) is switched off by no further values (V _soll ) being integrated and the last current integral part of the compensation value (V _VAR ) being stored.

4. The method according to any one of the preceding claims with the following features:

• 4.1 To detect a desired standstill, the criterion used is that target values have no more changes (K1) and / or a block end has been reached (K2),
• 4.2 the automatic drift compensation (I) remains activated as long as the criteria or criteria (K1, K2) are met.

5. Device for automatic drift compensation for position-controlled axes and spindles, in particular for use in machine tools and robots, with the following features:

• 5.1 a machine data (MD) is stored as an offset in a memory element, which represents a known portion of the drift in the form of a speed setpoint (V _offset ),
• 5.2 an integrator (I) is supplied with a speed setpoint (V _soll ) that may occur at the output of the position controller (LR),
• 5.3 the output (V _VAR ) of the integrator (I) and the value (V _offset ) of the machine data (MD) are connected to an adder (AD1),
• 5.4 another adder (AD2) is connected to the output (V _drift ) of the first adder (AD1) and the output (V _soll ) of the position controller (LR),
• 5.5 the output of the second adder (AD2) is connected to a control stage following the position controller, in particular a speed control loop (DR).

6. The device according to claim 5 with the following features:

• 6.1 an activation input (E) of the integrator (I) is connected to a signal which signals whether a standstill is desired,
• 6.2 several signals for signaling whether a standstill should take place are connected on the input side to a logic module (LB) and its output is connected to the activation input (E) of the integrator (I).

7. The device according to claim 5 or 6 with the following feature:

• 7.1 the integrator (I) has a large time constant compared to the position control loop.