JP7743009B2 - Machine tool and machine tool control device - Google Patents

Description

The present invention relates to a machine tool and a machine tool control device.

In the above technical field, Patent Document 1 discloses technology for estimating frictional force to control the movement of an object.

Patent No. 5560068

However, with the technology described in the above document, friction compensation can cause an inverse response, resulting in low accuracy in moving the object.

The object of the present invention is to provide technology that solves the above-mentioned problems.

In order to achieve the above object, the machine tool according to the present invention comprises:
a feed drive unit for moving a moving body in the machine tool;
a friction compensation unit that adds, in advance, a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to the feed drive unit in order to compensate for rolling friction occurring in the feed drive unit;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by adding the friction compensation value;
It is a machine tool equipped with
In order to achieve the above object, another machine tool according to the present invention comprises:
a feed drive unit for moving a moving body in the machine tool;
a friction compensation unit that adds, in advance, a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to the feed drive unit in order to compensate for rolling friction occurring in the feed drive unit;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by adding the friction compensation value;
Equipped with
the step input unit includes a calculation unit that calculates timing and magnitude of the step input value from a tracking error when the friction compensation value before adding the step input value is used,
The machine tool starts adding the step input value between the timing when the absolute value of the tracking error becomes maximum before the sign of the tracking error is first reversed when the friction compensation value before adding the step input value is used , and the timing when the sign of the tracking error is first reversed.
In order to achieve the above object, the present invention provides a machine tool control device comprising:
a friction compensation unit that adds in advance a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to a feed drive unit for compensating for rolling friction that occurs in the feed drive unit for moving a moving body in a machine tool;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by the compensation for the rolling friction;
Equipped with
The step input unit is a machine tool control device that includes a calculation unit that calculates the timing and magnitude of the step input value from a tracking error when the friction compensation value before adding the step input value is used.

This invention enables highly accurate movement control of a moving object.

1 is a block diagram showing a configuration of a machine tool according to a first embodiment. FIG. FIG. 10 is a block diagram showing the hardware configuration of a machine tool according to a second embodiment. FIG. 10 is a block diagram showing the functional configuration of a machine tool according to a second embodiment. 10 is a flowchart showing a processing flow of a machine tool according to a second embodiment. FIG. 11 is a diagram showing a table of friction characteristics of the machine tool according to the second embodiment. FIG. 10 is a block diagram showing a control system of a machine tool according to a second embodiment. FIG. 11 is a diagram showing changes in various values of the machine tool according to the second embodiment. 10A and 10B are diagrams illustrating suppression of an inverse response of a machine tool according to a second embodiment. 10 is a flowchart showing a process for suppressing an inverse response of a machine tool according to a second embodiment.

Embodiments of the present invention will be described in detail below with reference to the drawings. However, the components described in the following embodiments are merely examples and are not intended to limit the technical scope of the present invention to these components alone.

[First embodiment]
An object moving device 100 according to a first embodiment of the present invention will be described with reference to Fig. 1. The object moving device 100 is incorporated into a machine tool and is a device for moving a workpiece placed on a stage.

As shown in Figure 1, the machine tool 100 includes a feed drive unit 101, a friction compensation unit 102, and a step input unit 103.

The drive unit 101 moves the moving body 120 in the machine tool 100.

To compensate for rolling friction that occurs in the feed drive unit 101, the friction compensation unit 102 adds in advance to the current value 121 input to the feed drive unit 101 a friction compensation value 122 obtained from rolling friction measurement data or a friction compensation value 122 calculated using a rolling friction model.

The step input unit 103 adds a predetermined step input value 123 to the friction compensation value 122 to reduce the inverse response caused by friction compensation.

This embodiment allows the moving object to be moved with greater precision.

Second Embodiment
Next, a machine tool 200 according to a second embodiment of the present invention will be described with reference to Figure 2 and subsequent figures. Figure 2 is a diagram for explaining the configuration of the machine tool 200 according to this embodiment. The machine tool 200 is equipped with a ball screw 202 as a feed drive unit that moves a stage 201 as a moving body. Such ball screw-driven stages have high energy conversion efficiency, little wear, and a long life, and are therefore often used as feed devices for industrial machinery such as machine tools.

The ball screw 202 is connected to a motor 204 via a coupling 203 and is supported by a bearing 205. As the ball screw 202 rotates, the nut 206 moves left and right in the figure, causing the stage 201 to move in the X-axis direction while being guided by a linear guide 207.

The machine tool 200 is equipped with a tool 208 and processes a workpiece 209, which is a workpiece fixed to the stage 201.

Figure 3 is a diagram explaining the configuration for drive control of the stage 201 in the machine tool 200. The stage 201 moves back and forth within the movement area 330. In such a ball screw-driven stage 201, rolling friction reduces tracking performance. Rolling friction is generated by the balls between the ball screw 202 and the nut 206, and by the balls within the linear guide 207.

Graph 350 shows the relationship between the stage position and the rolling friction of the ball screw. Rolling friction depends on the displacement from the velocity reversal point. For example, suppose the stage 201 begins moving from start position 351, then reverses its velocity at velocity reversal point 352, and the stage 201 returns to the original start position 351. In region 353, from start position 351 until the stage 201 moves a predetermined distance, the rolling friction exhibits nonlinear spring characteristics. On the other hand, in region 354, from when the stage 201 moves a predetermined distance from start position 351 until velocity reversal point 352, the rolling friction remains a constant value.

Furthermore, in region 355, from when the velocity reverses at velocity reversal position 352 until the stage 201 moves a predetermined distance, the rolling friction exhibits nonlinear spring characteristics, and then in region 356 up to the starting position 351, the rolling friction becomes a constant value. Therefore, the machine tool 200 controls the movement of the stage 201 taking these rolling friction characteristics into account.

Due to rolling friction, a large spike-like tracking error occurs near the velocity reversal point. This tracking error is sometimes called a quadrant protrusion. In order to perform high-precision tracking control of the stage 201, it is necessary to compensate for the rolling friction and suppress the quadrant protrusion.

As a method of compensating for rolling friction, model-based and learning-based feedforward approaches are more effective than feedback approaches such as disturbance observers.

To date, many rolling friction models have been proposed and evaluated, including the LuGre model, generalized Maxwell-slip model, rheology-based model, database friction model, and elastic-plasticity-based model. In model-based friction compensation approaches, rolling friction is precisely measured and a model is generated by curve fitting. Rolling friction is then compensated for using a control input calculated based on the obtained model. In contrast, learning-based approaches, such as iterative learning control, do not use a rolling friction model, but rather suppress quadrant projections by gradually shaping the friction compensation input over multiple motion control iterations.

To compensate for rolling friction in this manner, the machine tool 200 has a data acquisition unit 302, a derivation unit 303, a friction compensation unit 304, a storage unit 305, a control unit 306, an update unit 307, and a determination unit 308.

The data acquisition unit 302 acquires current data 322 (=f ^np _j+1 ) indicating the current value for driving the motor 204 and command data 321 (displacement data r _j ) indicating the displacement of the stage 201 .

The data acquisition unit 302 detects velocity reversal positions 352, 351 based on the velocity derived from the displacement data _rj , and determines whether the stage 201 is in a linear region 354, 356 or a nonlinear region 353, 355 based on the amount of displacement from the velocity reversal positions 351, 352.

The derivation unit 303 estimates a linear relational expression 333 (=(f ^p _j+1 = Ψ(r _j ) θ j+ ₁ )) between the basis function Ψ(r j ) based on the command data 321 and the current data 322 (=f ^np _j +1 ) from the current data 322 and command data 321 in _the linear regions 354 and 356. θ _j+1 in the linear relational expression 333 corresponds to the physical characteristics of the stage 201.

The calculation unit 304 calculates current data 341 using the inertia J and viscous friction coefficient D of the parameter θ _j+1 estimated by the derivation unit 303 in the nonlinear regions 353 and 355, and the acceleration and velocity of the basis function Ψ(r _j ).

Correction data 391 is derived based on current data 322 (=f ^np _j+1 ) acquired by the data acquisition unit 302 and current data 341 calculated by the calculation unit 304 in the nonlinear regions 353 and 355. The accumulation unit 309 then accumulates the derived correction data 391 in association with the displacement of the stage 201 from the velocity reversal position 352. Note that if the start position 351 or the velocity reversal position 352 changes along with a change in the command position, there may be a case where corresponding correction data 391 does not exist, and therefore the following measures are taken.

(1) Use the correction data 391 of the closest position. (2) Linearly interpolate from the correction data 391 of the previous and next positions. (3) Average the data of each position and store it in one table.
In (1), for example, if the rolling friction is −3 Nm at 1 μm from the speed reversal position 240 and −2 Nm at 2 μm from the speed reversal position 240, when 1.2 μm is input, −3 Nm is output.
In (3), for example, if there are multiple speed reversal positions 240, a table is created by averaging multiple pieces of rolling friction data.

That is, in the nonlinear regions 353 and 355, the accumulation unit 309 derives correction data 391 representing rolling friction from the current data 341 calculated by the calculation unit 304 and the current data 322 acquired by the data acquisition unit 302.

The control unit 306 calculates current data 363 for driving the motor 204 in the linear regions 354, 356 using a basis function Ψ(r _j+1 ) based on the command value trajectory data r _j+1 for the next operation and the linear relational expression 333 (=(f ^p _j+1 Ψ(r _j )θ _j+1 )) estimated by the derivation unit 303. The control unit 306 also derives current data 364 for driving the motor 204 in the nonlinear regions 353, 355 using the current data 322 and correction data 391. At this time, the step input unit 362 adds a predetermined step input value to the friction compensation value calculated using the linear relational expression 333 and correction data 391 to reduce the inverse response (control in the opposite direction to the command). The control unit 306 then drives the motor 204 using the calculated current data 363 and the derived current data 364.

The update unit 307 sequentially updates at least one of the linear relational expression 333 and the correction data 391. The update unit 307 also performs update processing based on the latest current data 322 and command data 321 acquired by the data acquisition unit 302 at predetermined timings. Here, the predetermined timings include, for example, timing when the tool of the machine tool 200 is away from the workpiece to be machined, timing during tool replacement, and timing when the internal temperature of the machine tool 200 reaches or exceeds a predetermined value. They may also be timing when the machine tool 200 has been in use for a certain period of time, timing when the movement distance of the stage 201 caused by the motor 204 exceeds a certain value, or timing when the cumulative driving time of the motor 204 has exceeded a certain period of time. Furthermore, the timings include timing when the internal temperature of the machine tool 200 falls below a certain value, and timing when the amount of change in the internal temperature of the machine tool 200 since the last update processing was performed exceeds a certain amount of change. The update unit 307 performs update processing at at least one of the predetermined timings described above.

FIG. 4 is a diagram illustrating the flow of friction compensation processing. In STEP 1, current data 322 (= f ^np _j+1 = Q(f _j + Le _j )) is obtained using a normal ILC (Standard-ILC). Here, f denotes the FF (Feedforward) input (current data 322), e denotes the tracking error, L denotes the Learning Filter, and Q denotes the Robust Filter. The superscript np on the right of f ^np indicates the FF input obtained by Standard-ILC. That is, the motor 204 is operated with a current (initial value at the start of learning is zero) based on the FF input f _j in response to command data 321 (= r _j ) that commands the desired position to which the stage 201 is to be moved. Displacement data of the stage 201 moved by the operation of the motor 204 is obtained.

The following error e _j = command data 321 - displacement data is calculated. Current data 322 (f ^np _j+1 = Q(fj + Lej)) is calculated from the FF input f _j and the following error e _j (FF input by normal ILC). Note that the linear relational expression 333 and friction model of the motor 204 are reflected in the current data 322 through the following error. Note that the command data 321 represents the command value trajectory (target trajectory) of the stage 201, that is, the input of the position at which the stage 201 is desired to move.

In STEP 2, the parameter θ _j+1 is estimated. Ψ(r _j ) is a basis function. The parameter θ _j+1 is found by the linear least squares method so that the FF input equation Ψ(r _j )θ _j +1, which is based on the linear relational equation 333 of the motor 204, fits f ^np _j+1 . Here, the accuracy of the parameter θ _j+1 is improved by excluding nonlinear regions 353 and 355 using the weighting matrix W j .

In STEP 3, rolling friction T _rf is calculated from parameter θ _j+1 , and table 501 ( FIG. 5 ) is created using r _j and T _rf . That is, current data 341 is derived from J and D of parameter θ _j+1 and acceleration and velocity of basis function Ψ(r _j ). Then, correction data 391 for nonlinear regions 353 and 355 is calculated and stored by subtracting current data 341 from current data 322 (= f ^np _j+1 ). Table 501 is then created from command data 321 and correction data 391. Here, J represents inertia, D represents the viscosity coefficient, R represents the ball screw lead (the distance the nut advances axially with one rotation of the screw), and K _T represents the torque constant. Note that current data 341 corresponds to T _rf = (f ^np _j+1 - J/RK _T <<r _j >> - D/RK _T <r _j >)K _T. <r _j > denotes the differential of r _j , and <<r _j >> denotes the differential of <r _j >.

In STEP 4, the FF input is recalculated from the parameter θ _j+1 estimated in STEP 2 and table 501 created in STEP 3. That is, in linear regions 354 and 356, the FF input is calculated from Ψ(r _{j+ 1} )θ _j+ 1 (J/(R×K t )×acceleration+D/(R×K _{T )} ×velocity+Tc/K _T ×b_rf) using Ψ(r _j+1 ) based on the next command r j+1 and the previously calculated θ _j +1. Next, in nonlinear regions 353 and 355, the next FF input f _j+1 is calculated using J/(R×K _T )×acceleration+D/(R×K _t )×velocity+(rolling friction stored in table 501)/K _T . Furthermore, a predetermined step input value is added to these FF inputs to reduce inverse response.

The parameter θ _j+1 corresponds to the mechanical characteristics of the motor 204 and is a value independent of the command. Therefore, if Ψ(r _j+1 ) is calculated according to the command, the current data becomes the appropriate FF input for any command. The FF input for the linear regions 354 and 356 is calculated using "J/(R×K _T )×acceleration + D/(R×K _T )×velocity + Tc/K _T ×b_rf." Note that b_rf is 1 in the linear regions 354 and 356 and is between -1 and 1 in the nonlinear regions 353 and 355 (depending on the distance from the velocity reversal position 240 and the friction model). Repeating steps 1 to 4 reduces the tracking error e (learning control). By saving the learned θ and correction data 391, appropriate FF correction can be performed even when a different command is used.
Although the processing flow for estimating parameters and performing friction compensation has been described above using FIG. 4, the present invention is not limited to this, and the parameters may be known.

Figure 5 is a diagram illustrating table 501 showing the relationship between displacement and rolling friction in nonlinear regions 353 and 355. Table 501 is stored in storage unit 309. In nonlinear regions 353 and 355, current data 364 is derived using table 501, which associates the nonlinear characteristics of rolling friction with the displacement (position) of stage 201 and correction data 391 (rolling friction), without using a formula. Table 501 stores rolling friction in association with the displacement from velocity reversal position 240.

The friction compensation approach using linear relational equation 333 and table 501 can effectively suppress quadrantal protrusions, but it can sometimes cause an inverse response, which is a tracking error in the opposite direction to the quadrantal protrusion. In machining, this inverse response can cause over-cutting and roughening of the workpiece surface.

In this embodiment, the step input unit 362 analyzes and suppresses the inverse response. Specifically, the step input unit 362 checks the inverse response through a simulation without a step input, analyzes the inverse response based on the results, and inputs a step with a timing and magnitude that suppresses the inverse response.

Figure 6 is a diagram showing the above processing in this embodiment using a control system. The purpose of this control system is to reduce the tracking error e = r - x between the position reference r and the output x. Because the tracking error (quadrant projection) increases when the velocity is reversed, feedforward friction compensation is considered effective. Therefore, a feedforward controller 601 is introduced as a stable inverse system of the nominal model P of the controlled object 607. The feedforward controller 601 achieves perfect tracking control if there are no modeling errors or disturbances. In addition, a proportional-integral-derivative (PID) controller is provided as a feedback controller 602 to suppress steady-state errors.

To correct rolling friction using feedforward control, a friction model ^Trf as shown in graph 350 is used. Furthermore, a friction compensation unit 603 is provided to compensate for rolling friction Trf using the rolling friction model ^Trf. To correct rolling friction using feedforward control, a friction model ^Trf as shown in graph 350 is used. The friction compensation value is calculated based on the friction model and position reference.

As described above, feedforward friction compensation can effectively reduce tracking errors, but friction compensation can sometimes result in an inverse response, which is a tracking error in the opposite direction to the quadrant projection. Therefore, in this embodiment, the inverse response is analyzed through simulation, and the inverse response is suppressed based on the analysis results. Specifically, a step input unit 362 is provided that adds a step input value and cancels the inverse response.

Figure 7 shows the simulation results. For a typical movement command such as Figure 7(a), friction compensation suppresses tracking errors (causing quadrant projections) as shown in Figure 7(b), but an inverse response 701 may occur, as shown in Figure 7(c). Figure 7(d) shows the rolling friction and its compensation input near the time of velocity reversal. Figures 7(e) and 7(f) show the equivalent input disturbance, defined as the difference between the applied rolling friction and the friction compensation value. The equivalent input disturbance is small at the sample points, but becomes relatively large between sample points because the control input is discretized by the zero-order hold block 606. Figures 7(e) and 7(f) show that the characteristics of the equivalent input disturbance are determined by friction compensation. The zero-order hold block 606 causes the friction compensation value to take a step-like value, resulting in a spike-like input disturbance. Without friction compensation, the input end disturbance would be the step disturbance shown in Figure 7(e) w/o comp.

It is generally known that in a closed-loop system in which the control target P and the feedback controller CFB are linear and time-invariant, the input terminal disturbance and tracking error to the control target P satisfy the following equations. Let the number of integrals held by the feedback controller 602 be i. In this case, the error ei when an impulse disturbance enters the input terminal of the control target P satisfies the following equation (1).




When a step disturbance is applied to the input terminal, the error es=rx without friction compensation is expressed by the following equation (2).





The feedback controller 602 has one integrator, and the equivalent disturbance is considered to be impulse-like. As a result, when friction compensation (without step input) is performed, an inverse response occurs according to equation (1). Even if there is no modeling error, the inverse response 701 is caused by the integrator of the feedback controller 602 and the control input discretized by the zero-order hold block 606. When friction compensation is not performed, the input-end disturbance is step-like, so no inverse response occurs according to equation (2).

To reduce this inverse response 701, as shown in FIG. 8, the step input value Gs calculated based on the parameters calculated by the parameter calculation unit 605 is added to the friction compensation value from timing Ts (604). This makes it possible to suppress the inverse response 701 (803) without increasing the tracking error 801 due to friction.

The parameter calculation unit 605 calculates the timing and magnitude of the step input value from the tracking error when using the friction compensation value before adding the step input value 604.

Figure 9 is a flowchart showing the flow of the parameter calculation process in the parameter calculation unit 605.

T represents time, and T=0 is the timing when the stage velocity reverses. First, in step S901, the tracking error Ec(T) due to friction compensation (no step input) is obtained.

Next, in step S902, the occurrence timing Tq and magnitude Eq of the quadrant projection, as well as the zero-cross timing Ti of the tracking error, are obtained.

Furthermore, in step S903, Tq is input as the initial value for Ts, and the process proceeds to step S904, where Eus(T-Ts) is calculated. Here, Ts represents the timing when the step input is added, and Eus(T) represents the tracking error due to the step input.

The feedback controller 602 is expressed by the following equation (3), where kp is the proportional gain, ki is the integral gain, kd is the differential gain, and Tf is the time constant of the pseudo differential.



The feedback controller 602 is a PID controller in which the closed-loop nominal model has a quadrupole at −wc. Therefore, due to the step input added at the timing when T=Ts, the tracking error Eus(Ts) is expressed as follows, where Sn is the nominal sensitivity function:




From the above formula, Eus can be calculated as follows:


Although the use of a PID controller has been described here, the present invention is not limited to this and can be applied to any feedback controller with a single integrator. Furthermore, even for PID controllers, a different design method (where equations 4 and 5 are changed) may be used.
Next, in step S905, the magnitude Gs of the step input to be added is calculated based on the following formula.



Furthermore, in step S906, the predicted tracking error Ep(T) with friction compensation (with step input) is calculated based on the following equation:
Ep(T)=Ec(T)+GsEus(T-Ts)

Next, in step S907, it is determined whether the following formula is true:
−Eq≦minimum value of Ep(T)≦0
If this formula is true, the process ends.

If the above inequality does not hold, proceed to step S908, where Ts is incremented, i.e., one sampling period is added to Ts. Then, in step S909, while it is determined that Ts≦Ti, the processes of steps S904 to S908 are repeated to determine Ts and Gs that satisfy the relationship -Eq≦Ep(T)≦0. In other words, the step input unit 362 (parameter calculation unit 605) adds the step input value Gs while changing the timing between the timing Tq at which the absolute value of the tracking error when using the friction compensation value before adding the step input value Gs is maximized and the timing Ti at which that absolute value becomes 0, and determines the step input value Gs so that the absolute value of the negative error does not exceed the maximum value Eq.

On the other hand, if Ts and Gs cannot be determined that satisfy the relationship -Eq≦Ep(T)≦0 by the timing up to Ti, a slight inverse response and a slight increase in the quadrant projection are allowed.

In other words, the parameter calculation unit 605 starts adding the step input value Gs between the timing Tq when the absolute value of the tracking error reaches its maximum and the timing Ti when the sign of the tracking error first reverses, before the sign of the tracking error first reverses when using the friction compensation value before adding the step input value. The parameter calculation unit 605 also determines the magnitude of the step input value Gs using the value of the tracking error when using the friction compensation value before adding the step input value.

It is also possible to repeat the process for all Ts in the range of 0≦Ts≦Ti, rather than terminating the process when the condition −Eq≦minimum value of Ep(T)≦0 is met for the first time, and then adopt Ts and Gs for which Ep(T)≦0 and the minimum value of Ep(T) is the largest.
Alternatively, Ts may be changed in a decreasing direction from Ti to Tq. That is, Ts may be set to Ti in S903, one sampling period may be subtracted from Ts in S908, and the condition determination Ts≦Ti may be changed to Ts≧0.
In each of the above cases, Ts may be changed by adding or subtracting multiple sampling periods.
By performing the above process, it is possible to determine an appropriate step input value and its timing, thereby suppressing the inverse response without increasing the tracking error that can cause quadrantal spikes.

In this embodiment, a ball screw has been described as an example of a feed drive unit, but the present invention is not limited to this. Also, in the above embodiment, a friction compensation value calculated using a rolling friction model is added to the current value, but the present invention is not limited to this, and a friction compensation value obtained from rolling friction measurement data may also be added to the current value.

Furthermore, the scope of the present invention also includes a machine tool control device equipped with a friction compensation unit and a step input unit to realize the above-mentioned control in a machine tool.

Other Embodiments
Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above embodiments. Various modifications can be made to the configuration and details of the present invention that are understandable to those skilled in the art within the technical scope of the present invention. Furthermore, systems or devices that combine the separate features included in each embodiment in any manner are also included in the technical scope of the present invention.

The present invention may also be applied to a system consisting of multiple devices, or to a single device. Furthermore, the present invention is also applicable when an information processing program that realizes the functions of the embodiments is supplied to a system or device and executed by an integrated processor. The technical scope of the present invention also includes programs installed on a computer to realize the functions of the present invention on the computer, media storing such programs, servers from which such programs are downloaded, and processors that execute such programs. In particular, the technical scope of the present invention includes at least non-transitory computer-readable media storing programs that cause a computer to execute the processing steps included in the above-described embodiments.

Claims (6)

a feed drive unit for moving a moving body in the machine tool;
a friction compensation unit that adds, in advance, a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to the feed drive unit in order to compensate for rolling friction occurring in the feed drive unit;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by adding the friction compensation value;
A machine tool equipped with The machine tool according to claim 1, wherein the step input unit includes a calculation unit that calculates the timing and magnitude of the step input value from the tracking error when the friction compensation value before adding the step input value is used. a feed drive unit for moving a moving body in the machine tool;
a friction compensation unit that adds, in advance, a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to the feed drive unit in order to compensate for rolling friction occurring in the feed drive unit;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by adding the friction compensation value;
Equipped with
the step input unit includes a calculation unit that calculates timing and magnitude of the step input value from a tracking error when the friction compensation value before adding the step input value is used,
a machine tool that starts adding a step input value between the timing when the absolute value of the following error becomes maximum before the sign of the following error is first reversed when the friction compensation value before adding the step input value is used, and the timing when the sign of the following error is first reversed . A machine tool as described in claim 3, wherein the step input unit determines the magnitude of the step input value using the value of the tracking error when the friction compensation value before adding the step input value is used. The machine tool described in claim 4, wherein the step input unit adds the step input value while changing the timing between when the absolute value of the tracking error when using the friction compensation value before adding the step input value is maximized and when the absolute value becomes zero, and determines the step input value such that the absolute value of the negative error does not become greater than the maximum absolute value of the tracking error. a friction compensation unit that adds in advance a friction compensation value obtained from measurement data of rolling friction or a friction compensation value calculated using a rolling friction model to a current value input to a feed drive unit for compensating for rolling friction that occurs in the feed drive unit for moving a moving body in a machine tool;
a step input unit that adds a predetermined step input value to the friction compensation value in order to reduce an inverse response caused by the compensation for the rolling friction;
Equipped with
The step input unit includes a calculation unit that calculates the timing and magnitude of the step input value from a tracking error when the friction compensation value before adding the step input value is used.