Numerical control device and numerical control method

Abstract

Provided are a numerical control device and a numerical control method with which it is possible to stabilize a control system by operating a predetermined control stabilizing means when an unstable operation is detected. When an unstable operation of a machine tool 1 is detected after a predetermined time T or a predetermined period C has elapsed from the start of the operation of the machine tool 1, the present invention switches and operates, by means of a switching unit 26, at least one of: a speed gain changing means 21 for changing a speed gain; a learning condition changing means 22 for storing, in a learning memory 12, correction data 16 obtained by attenuating the high frequency components of a positional deviation through a bandwidth limiting filter 11, and performing learning control of the positional deviation by implementing phase advancement on the correction data 16 using a dynamic characteristic compensation element 13; a learning control disabling means 23 for disabling the learning control; and a machining condition changing means 24 for changing the machining conditions, which has a periodic operation pattern.

Description

Numerical control device and numerical control method

[0001] The present disclosure relates to a numerical control device and a numerical control method.

[0002] Learning control is known as a method for improving machining accuracy when machining or the like is performed using repeated commands of the same pattern. This learning control uses the time of a pattern operation, such as one rotation of the workpiece, as a learning cycle, rotates the workpiece multiple times, determines the position deviation for each predetermined control cycle, stores correction data based on this position deviation in memory, and attempts to converge the position deviation to zero by adding the correction data for the corresponding control cycle in the pattern cycle immediately preceding the one stored in memory to the position deviation for each control cycle of the pattern cycle. Techniques related to this type of learning are described, for example, in Patent Document 1 and Patent Document 2.

[0003] JP 2011-123616 A JP 2020-163487 A

[0004] Here, repetitive control utilizes the repeatability of position deviations and commands, but if the position deviations and commands contain asynchronous components that are not repeatable, the control system may become unstable. As a countermeasure, conventional methods aim to stabilize the control system by passing the position deviations and repetitive commands through a low-pass filter to remove asynchronous disturbances. However, even when passing through a low-pass filter, depending on the machining conditions and learning conditions, mechanical resonance may be induced, causing the control system to become unstable.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and aims to provide a numerical control device and a numerical control method that can stabilize a control system by activating a predetermined control stabilization means when unstable operation is detected.

[0006] One aspect of the present disclosure is a numerical control device that controls a machine tool that operates according to a periodic operation pattern, the numerical control device comprising: position information acquisition means that acquires position information of the machine tool; position deviation calculation means that calculates a position deviation, which is the deviation between a movement command issued by the numerical control device and the position information; a learning controller that stores correction data in a learning memory in which high-frequency components of the position deviation have been attenuated by a band-limiting filter, and further performs learning control of the position deviation by applying a phase advance to the correction data using a dynamic characteristic compensation element; and unstable operation detection means that detects when the position deviation exceeds a threshold value as the occurrence of unstable operation of the machine tool, and when unstable operation is detected by the unstable operation detection means after a predetermined time or a predetermined period has elapsed since the start of operation of the machine tool, the numerical control device switches between and operates at least one of speed gain change means that changes a speed gain, learning condition change means that changes the learning conditions of the learning controller, learning control disable means that disables the learning control, and machining condition change means that changes the machining conditions of the periodic operation pattern using a switching unit.

[0007] One aspect of the present disclosure is a numerical control method for controlling a machine tool that operates according to a periodic operation pattern, the method comprising: position information acquisition means for acquiring position information of the machine tool; position deviation calculation means for calculating a position deviation, which is the deviation between a movement command issued by the numerical control device and the position information; a learning controller that stores correction data in a learning memory in which high-frequency components of the position deviation have been attenuated by a band-limiting filter, and further performs learning control of the position deviation by applying a phase advance to the correction data using a dynamic characteristics compensation element; and unstable operation detection means that detects when the position deviation has exceeded a threshold value as an occurrence of unstable operation of the machine tool, and when unstable operation is detected by the unstable operation detection means after a predetermined time or a predetermined period has elapsed since the start of operation of the machine tool, the numerical control method switches between and operates at least one of speed gain change means for changing a speed gain, learning condition change means for changing learning conditions of the learning controller, learning control disable means for disabling the learning control, and machining condition change means for changing the machining conditions of the periodic operation pattern using a switching unit.

[0008] Fig. 1 is a schematic diagram showing the configuration of a machine tool equipped with a numerical control device according to the present disclosure; Fig. 2 is a functional block diagram showing the configuration of a numerical control device; Fig. 3 is a block diagram showing the configuration of a learning controller; Fig. 4 is a control system diagram of a numerical control device including a speed gain changing means; Fig. 5 is a graph showing the operation start timing of an unstable operation detecting means; Fig. 6 is a control system diagram of a numerical control device including a learning condition changing means; Fig. 7 is a control system diagram of a numerical control device including a learning control disabling means; Fig. 8 is a control system diagram of a numerical control device including a machining condition changing means; Fig. 9 is a control system diagram of a numerical control device including a switching unit; Fig. 10 is a control system diagram of a numerical control device including a swing condition changing means; Fig. 11 is a flowchart showing the procedure of stabilization control according to the present embodiment.

[0009] An example of an embodiment of the present disclosure will be described below. Fig. 1 is a schematic diagram showing the configuration of a machine tool 1 equipped with a numerical control device 2 according to the present disclosure. The machine tool 1 is an NC lathe that cuts a workpiece W, which is an object to be machined, using a cutting tool B. The machine tool 1 has three axes: a main spindle S1 that rotates the workpiece W, a feed axis S2 that feeds and moves the cutting tool B relative to the workpiece W in a direction parallel to the main spindle S1, and a cutting axis S3 that moves the cutting tool T in a radial direction of the main spindle S1 relative to the workpiece W. The main spindle S1, the feed axis S2, and the cutting axis S3 are driven by a main spindle motor M1, a feed axis motor M2, and a cutting axis motor M3, respectively.

[0010] The numerical control device 2 is composed of a computer device equipped with a CPU, memory, etc. that executes a machining program, and controls the machine tool 1 by causing the main spindle S1, feed axis S2, and cutting axis S3 to operate in coordination. The machining program can be written in, for example, G-code or the like.

[0011] 2 is a functional block diagram showing the configuration of the numerical control device 2. The numerical control device 2 has position information acquisition means 3 for acquiring position information of the machine tool 1, i.e., motor position information, position deviation calculation means 4 for calculating a position deviation which is the deviation between a movement command issued by the numerical control device 2 and the position information, position deviation storage means 5 for storing the position deviation which is the deviation between the position information and the movement command, steady-state position deviation deduction means 6 for deducting a steady-state position deviation determined by the movement command and a position gain, swing command calculation means 7 for calculating a swing command from the movement command issued by the numerical control device 2 and the swing condition, unstable operation detection means 8 for detecting that the position deviation has exceeded a threshold value ±L (see FIG. 6 ) as the occurrence of unstable operation of the machine tool 1, threshold calculation means 9 for calculating a threshold value based on the position deviation stored in the position deviation storage means, and a learning controller 10 for performing learning control of the position deviation.

[0012] The swing command calculation means 7 calculates a swing command for periodically reciprocating the feed shaft S2 based on the swing frequency and swing amplitude. The learning controller 10 stores correction data in a learning memory, in which high-frequency components of the position deviation have been attenuated by passing the data through a band-limiting filter, and further performs learning control of the position deviation by applying a phase advance to the correction data using a dynamic characteristic compensation element.

[0013] The numerical control device 2 also has a control stabilization means 20. The control stabilization means 20 includes at least one of a speed gain change means 21 that changes the speed gain, a learning condition change means 22 that changes the learning conditions of the learning controller 10, a learning control disabling means 23 that disables learning control, a machining condition change means 24 that changes the machining conditions of the periodic operation pattern, and an oscillation condition change means 25 that changes the oscillation conditions. A motor drive means 27 drives and controls the spindle motor M1, the feed axis motor M2, and the cutting axis motor M3 based on output signals from the above-mentioned means, etc. The numerical control device 2 may be configured to have either one of the machining condition change means 24 and the oscillation condition change means 25.

[0014] When unstable operation of the machine tool 1 is detected by the unstable operation detection means 8, the numerical control device 2 switches between and operates at least one of the means included in the control stabilization means 20 using the switching unit 26, thereby making it possible to converge the unstable operation of the machine tool 1. Note that the above-mentioned components are distinguished functionally, and do not necessarily have to be clearly distinguishable in terms of physical configuration and program configuration.

[0015] 3 is a block diagram showing the configuration of the learning controller 10. The learning controller 10 has an adder 15 to which a position error, which is the difference between the motor position information and the repeat command, a band-limiting filter 11 as a frequency filter that passes only a predetermined frequency included in the position error, a learning memory 12 that stores correction data 16 that has passed through the band-limiting filter 11, and a dynamic characteristic compensation element 13 that applies a phase advance to the correction data 16.

[0016] For example, a finite impulse response (FIR) filter serving as a digital filter can be applied to the band-limiting filter 11. By changing the value of a weighting parameter of the FIR filter, the filter characteristics can be changed to, for example, a low-pass filter or a band-pass filter. Note that the band-limiting filter 11 is not limited to an FIR filter, and an infinite impulse response (IIR) filter or the like may also be applied.

[0017] As described above, even when the signal passes through the band limiting filter 11, it may not be possible to completely remove disturbances that are asynchronous with the repetitive commands, or the position deviation may be compensated for at an inappropriate timing, depending on the machining conditions, learning conditions, and settings of the band limiting filter 11 and dynamic characteristic compensation element 13, thereby inducing mechanical resonance and making the control system unstable. When unstable operation of the machine tool 1 is detected, the numerical control device 2 according to this embodiment activates at least one means included in the control stabilization means 20, thereby making it possible to quickly converge this unstable operation.

[0018] FIG. 4 is a control system diagram of the numerical control device 2. The same reference numerals as those used above indicate the same or equivalent parts. The position error calculation means 4 calculates the position error, which is the deviation between the movement command 30 issued by the numerical control device 2 and the position information of the machine tool 1. The position information of the machine tool 1 is the position information of the motors M1, M2, and M3. In this diagram, the operation switch 32 of the learning controller 10 is on, and the feedforward control operation switch 33 is off. When the feedforward control operation switch 33 is on, a feedforward term obtained by multiplying the differential term of the movement command 30 by a feedforward coefficient 31 is applied to the speed command. The speed command is calculated from the position error compensated for by the learning controller 10 and the position gain 17. The current controller 36, which controls the motors M1, M2, and M3, is controlled by a torque command obtained by superimposing a proportional gain 35 of the speed command and an integral gain 34 based on the integral term of the speed command. The position information of the motors M1, M2, and M3 is input to a subtractor for feedback control. The feedforward control activation switch 33 may be either off or on.

[0019] 5 is a control system diagram of the numerical control device 2 including the speed gain changing means 21. The same reference numerals as those used above indicate the same or equivalent parts. In a control system with repetitive commands, if the speed gain is high, the stability margin becomes small and the system may become unstable. In particular, in learning control, if the gain is high and the stability margin is small, it may be difficult to adjust the dynamic characteristics compensation element 13. When unstable operation of the machine tool 1 is detected, the speed gain changing means 21 reduces the speed gain to increase the stability margin, thereby making it possible to stabilize the control.

[0020] When unstable operation detection means 8 detects unstable operation of machine tool 1, numerical control device 2 activates speed gain change means 21. Speed ​​gain change means 21 changes at least one of integral gain 34 and proportional gain 35 to reduce the speed gain.

[0021] FIG. 6 is a graph showing the timing at which the unstable operation detection means 8 starts operating. The numerical control device 2 activates the control stabilization means 20 when the unstable operation detection means 8 detects unstable operation of the machine tool 1. It takes several cycles from the start of learning for the learning control to converge to a sufficiently small position error. In the present invention, unstable operation is defined as a situation in which the position error gradually increases and diverges due to insufficient adjustment of the learning conditions or machining conditions after the position error has converged to a sufficiently small value after the start of learning. Therefore, the system is configured to wait until the position error has converged to a sufficiently small value before starting detection of unstable operation by the unstable operation detection means 8. In this embodiment, the system is configured to start detecting unstable operation at time t1, which is a predetermined time T or a predetermined period C after the start of operation of the machine tool 1 at time t=0. This prevents the unstable operation detection means 8 from operating before the position error has converged immediately after the start of operation of the machine tool 1, resulting in a false detection of unstable operation. The numerical control device 2 activates the control stabilization means 20 when the position error exceeds the threshold value ±L after time t1 has elapsed.

[0022] FIG. 7 is a control system diagram of the numerical control device 2 including the learning condition changing means 22. The same reference numerals as those described above indicate the same or equivalent parts. When the unstable operation detection means 8 detects unstable operation of the machine tool 1, the numerical control device 2 activates the learning condition changing means 22. The learning condition changing means 22 changes at least one of the cutoff frequency of the band-limiting filter 11 and the order or coefficient of the dynamic characteristic compensation element 13. Specifically, if the cutoff frequency of the band-limiting filter 11 is set near a disturbance asynchronous with the repeat command, the disturbance asynchronous with the repeat command cannot be completely removed, resulting in unstable learning control. To address this, lowering the cutoff frequency of the band-limiting filter 11 can stabilize control. Furthermore, if the dynamic characteristic compensation element 13 is not properly adjusted, position deviations are compensated for at an inappropriate timing (phase), resulting in unstable learning control. To address this, changing the order or coefficient of the dynamic characteristic compensation element 13 to an optimal value can stabilize control.

[0023] FIG. 8 is a control system diagram of the numerical control device 2, including the learning control disabling means 23. The same reference numerals as those described above indicate the same or equivalent parts. When the unstable operation detection means 8 detects unstable operation of the machine tool 1, the numerical control device 2 activates the learning control disabling means 23. The learning control disabling means 23 disables learning control by turning off the operation switch 32 of the learning controller 10, and then turns on the feedforward control operation switch 33 to compensate the speed command with a feedforward term 37, which is the product of the differential value of the periodic operation pattern multiplied by the feedforward coefficient 31. While learning control can improve the tracking ability to repetitive commands and improve machining accuracy, it can also cause unstable operation depending on the settings of the band-limiting filter and dynamic characteristic compensation element. In this case, switching from learning control to feedforward control can stabilize control, although tracking ability is slightly reduced. Note that the higher the operating frequency of the repetitive commands, the greater the difference between the tracking ability of learning control and that of feedforward control.

[0024] 9 is a control system diagram of the numerical control device 2 including the machining condition changing means 24. The same reference numerals as those used above indicate the same or equivalent parts. When the unstable operation detection means 8 detects unstable operation of the machine tool 1, the numerical control device 2 activates the machining condition changing means 24. When the machining condition changing means 24 is activated, the activation switch 32 of the learning controller 10 is turned off and the activation switch 33 of the feedforward control is turned on.

[0025] The machining condition changing means 24 changes at least one of the feed speed of the cutting tool B, the rotation speed of the spindle S1, and the frequency of the periodic operation pattern. Here, if the machining conditions are inappropriate, such as if the operating frequency of the repeat command is close to a resonance or anti-resonance frequency, the operation of the machine tool 1 may become unstable. Furthermore, if the operating frequency is set close to a resonance or anti-resonance frequency, the operation is likely to become unstable. In this case, changing (including stopping) the operating frequency as a machining condition can stabilize the control.

[0026] 10 is a control system diagram of the numerical control device 2 including the switching unit 26. The same reference numerals as those described above indicate the same or equivalent parts. When unstable operation is detected by the unstable operation detection means 8 after a predetermined time T or a predetermined cycle C has elapsed since the start of operation of the machine tool 1, the numerical control device 2 switches between and operates at least one of the speed gain change means 21, learning condition change means 22, learning control invalidation means 23, and machining condition change means 24 using the switching unit 26.

[0027] The threshold calculation means 9 divides the position deviation 70 stored in the position deviation storage means 5 into predetermined cycles, calculates the maximum value of the position deviation for each interval, and calculates the threshold ±L by multiplying the minimum or average value of the maximum values ​​of the position deviation by a predetermined magnification.

[0028] 11 is a control system diagram of the numerical control device 2 including the oscillation condition changing means 25. The same reference numerals as those described above indicate the same or equivalent parts. When unstable operation is detected after a predetermined time T or a predetermined cycle C has elapsed since the start of operation of the machine tool 1, the numerical control device 2 switches between and operates at least one of the speed gain changing means 21, learning condition changing means 22, learning control disabling means 23, and oscillation condition changing means 25 using a switching unit 26.

[0029] The swing command calculation means 7 calculates a swing command 40 from the movement command 30 and the swing conditions issued by the numerical control device 2. The learning controller 10 stores correction data 16, which is created by attenuating high frequency components of the position deviation, which is the deviation between the position information of the machine tool 1 and a superimposed command 41 in which the swing command 40 is superimposed on the movement command 30, using a band limiting filter 11, in learning memory 12, and compensates for the position deviation.

[0030] This embodiment also includes a steady-state position deviation deduction means 6 that applies the swing command 40 to the position deviation 50 and then subtracts the steady-state position deviation determined by the movement command 30 and the position gain 17. The learning control creates correction data 16 based on the position deviation 50 after the steady-state position deviation has been subtracted. The steady-state position deviation deduction means 6 also subtracts the steady-state position deviation using a high-pass filter. Furthermore, the steady-state position deviation deduction means 6 subtracts the steady-state position deviation estimated from the movement command 30.

[0031] 12 is a flowchart showing the procedure for stabilization control according to this embodiment. This flowchart corresponds to a setting in which control stabilization means 20 includes speed gain changing means 21, learning condition changing means 22, learning control disabling means 23, and machining condition changing means 24, and in which detection of unstable operation is started when a predetermined time T has elapsed since the start of operation of machine tool 1.

[0032] In step S1, operation of the machine tool 1 is started. In step S2, it is determined whether or not a predetermined time T has elapsed since operation of the machine tool 1 was started, and if the determination is affirmative, the process proceeds to step S3. On the other hand, if the determination is negative in step S2, the process returns to the determination in step S2.

[0033] In step S3, the unstable operation detection means 8 starts detecting unstable operation. In the following step S4, it is determined whether or not unstable operation has been detected, and if a positive determination is made, the process proceeds to step S5. In step S5, the speed gain is changed by the speed gain change means 21. If a negative determination is made in step S4, it is determined that the operation of the machine tool is stable, and the series of controls is terminated.

[0034] In the next step S6, it is determined whether or not unstable operation has been detected. If the determination in step S6 is affirmative, that is, if the unstable operation has not converged despite the change in the speed gain, the process proceeds to step S7, where the learning condition change means 22 changes the learning conditions.

[0035] In step S8, it is determined whether or not unstable operation has been detected. If the determination in step S8 is affirmative, that is, if the unstable operation has not converged despite the changes in the speed gain and the learning conditions, the process proceeds to step S9, where the learning control disabling means 23 disables the learning control.

[0036] In step S10, it is determined whether or not unstable operation has been detected. If the determination in step S10 is affirmative, that is, if the unstable operation does not converge despite the changes in the speed gain, the changes in the learning conditions, and the disabling of the learning control, the process proceeds to step S11, where the machining conditions are changed by the machining condition change means 24, and the series of control operations is terminated.

[0037] As described above, in this embodiment, each time an unstable operation is detected, the means included in the control stabilization means 20 are operated in the following order: speed gain change means 21 → learning condition change means 22 → learning control invalidation means 23 → machining condition change means 24, thereby making it possible to reliably converge the unstable operation. In the flowchart, if a negative determination is made in steps S6, S8, or S10, that is, if it is determined that the unstable operation has converged by operating the control stabilization means 20 after the unstable operation has been detected, the series of controls is terminated.

[0038] The numerical control device 2 can be configured with at least one control unit and a storage unit. Here, the control unit is a processor such as a CPU (Central Processing Unit) that realizes various functions by executing programs stored in the storage unit. The storage unit is composed of a storage device such as a ROM (Read Only Memory) or RAM (Random Access Memory) that stores an OS (Operating System) and application programs, a hard disk drive, an SSD (Solid State Drive), or the like that stores various other information.

[0039] According to the above-described numerical control device, there is provided a position information acquisition means for acquiring position information of the machine tool, a position deviation calculation means for calculating a position deviation which is the deviation between a movement command issued by the numerical control device and the position information, a learning controller for storing correction data in a learning memory in which high frequency components of the position deviation have been attenuated by passing it through a band limiting filter, and further for performing learning control of the position deviation by applying a phase advance to the correction data using a dynamic characteristic compensation element, and unstable operation detection means for detecting that the position deviation has exceeded a threshold value as the occurrence of unstable operation of the machine tool, and when unstable operation is detected by the unstable operation detection means after a predetermined time or predetermined period has elapsed since the start of operation of the machine tool, a numerical control method can be obtained in which at least one of speed gain change means for changing a speed gain, learning condition change means for changing the learning conditions of the learning controller, learning control disabling means for disabling the learning control, and machining condition change means for changing the machining conditions of the periodic operation pattern is switched between and operated by a switching unit.

[0040] The above-described numerical control device and numerical control method can be realized by hardware, software, or a combination of these. Here, "realized by software" means that the computer reads and executes a program.

[0041] The program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)).

[0042] Furthermore, although the above-described embodiments are preferred embodiments of the present invention, the scope of the present invention is not limited to only the above-described embodiments, and the present invention can be implemented in various modified forms within the scope that does not deviate from the gist of the present invention.

[0043] Although the present disclosure has been described in detail, the present disclosure is not limited to the above-described embodiments. Various additions, substitutions, modifications, partial deletions, etc. are possible to these embodiments without departing from the gist of the present disclosure or the spirit of the present disclosure derived from the claims and their equivalents. These embodiments can also be implemented in combination. For example, in the above-described embodiments, the order of each operation and each process is shown as an example and is not limited to these. In the above embodiment, the workpiece is rotated by a motor, but a configuration in which the cutting tool is rotated by a motor may also be used. Furthermore, the threshold calculated by the threshold calculation means may have different values ​​on the positive and negative sides.

[0044] The numerical control device and numerical control method disclosed herein can be applied to various control systems that operate based on repetitive commands, as well as to oscillating cutting and eccentric machining, which perform machining operations based on specific frequencies determined by machining conditions.

[0045] The following supplementary note is further disclosed regarding the above embodiment and modified examples: (Supplementary Note 1) A numerical control device (2) for controlling a machine tool (1) that operates according to a periodic operation pattern, comprising: position information acquisition means (3) for acquiring position information of the machine tool (1); position deviation calculation means (4) for calculating a position deviation which is a deviation between a movement command (30) issued by the numerical control device (2) and the position information; a learning controller (10) for storing correction data (16) in a learning memory (12) by attenuating high frequency components of the position deviation through a band limiting filter (11), and further for performing learning control of the position deviation by implementing phase advancement on the correction data (16) using a dynamic characteristic compensation element (13); and unstable operation detection means (8) for detecting that the position deviation has exceeded a threshold value (±L) as an occurrence of unstable operation of the machine tool (1), When unstable operation is detected by the unstable operation detection means (8) after a predetermined time (T) or a predetermined period (C) has elapsed since the start of operation of the machine tool (1), a numerical control device switches between and operates at least one of speed gain change means (21) that changes a speed gain, learning condition change means (22) that changes the learning conditions of the learning controller (10), learning control disabling means (23) that disables the learning control, and machining condition change means (24) that changes the machining conditions of the periodic operation pattern by a switching unit (26).

[0046] (Supplementary Note 2) A numerical control device (2) for controlling a machine tool (1) for machining a workpiece (W) with a cutting tool (B) comprises: position information acquisition means (3) for acquiring position information of the machine tool (1); swing command calculation means (7) for calculating a swing command (40) from a movement command (30) issued by the numerical control device (2) and a swing condition; a learning controller (10) for storing correction data (16) in a learning memory (12) to compensate for the position deviation, the correction data being a deviation between the position information and a superimposed command (41) in which the movement command (30) is superimposed with the swing command (40), by a band-limiting filter (11); and an unstable operation detection means (8) for detecting that the position deviation has exceeded a threshold value (±L) as an occurrence of unstable operation of the machine tool (1). When the unstable operation is detected after a predetermined time (T) or a predetermined period (C) has elapsed since the start of operation of the machine tool (1), a numerical control device switches between and operates at least one of a speed gain change means (21) that changes a speed gain, a learning condition change means (22) that changes a learning condition of the learning controller (10), a learning control disabling means (23) that disables the learning control, and an oscillation condition change means (25) that changes the oscillation condition using a switching unit (26).

[0047] (Note 3) The above-mentioned numerical control device comprises a steady-state position deviation deduction means (6) for deducting a steady-state position deviation determined by the movement command (30) and a position gain (17) after applying the swing command (40) to the position deviation, and the learning control creates the correction data (16) based on the position deviation (50) after deducting the steady-state position deviation.

[0048] (Note 4) In the above-mentioned numerical control device, the steady-state position deviation subtraction means (6) subtracts the steady-state position deviation by a high-pass filter.

[0049] (Supplementary Note 5) In the above numerical control device, the steady-state position deviation subtraction means (6) subtracts the estimated steady-state position deviation from the movement command (30).

[0050] (Supplementary Note 6) In the above-mentioned numerical control device, the speed gain changing means (21) changes at least one of the integral gain (34) and the proportional gain (35).

[0051] (Supplementary Note 7) In the above-mentioned numerical control device, the learning condition changing means (22) changes at least one of the cutoff frequency of the band-limiting filter (11) and the order or coefficient of the dynamic characteristic compensation element (13).

[0052] (Supplementary Note 8) In the above-mentioned numerical control device, the learning control disabling means (23) disables the learning control, and then compensates the speed command with a feedforward term (37) obtained by multiplying the differential value of the periodic operation pattern by a feedforward coefficient (31).

[0053] (Supplementary Note 9) In the above-mentioned numerical control device, the learning control disabling means (23) disables the learning control, and then compensates the speed command with a swing feedforward term (60) obtained by multiplying the differential value of the swing command (40) by a swing feedforward coefficient (28).

[0054] (Supplementary Note 10) In the above-mentioned numerical control device, the machining condition changing means (24) changes the frequency of the periodic operation pattern.

[0055] (Supplementary Note 11) In the above numerical control device, the swing condition changing means (25) changes at least one of the frequency and amplitude of the swing command (40).

[0056] (Supplementary Note 12) The above-mentioned numerical control device comprises: a position deviation storage means (5) for storing a position deviation, which is a deviation between the position information and the movement command (30); and a threshold calculation means (9) for calculating the threshold value (±L) based on the position deviation stored in the position deviation storage means (5).

[0057] (Supplementary Note 13) In the above-mentioned numerical control device, the threshold value calculation means (9) divides the position deviation (70) stored in the position deviation storage means (5) for each predetermined period, calculates the maximum value of the position deviation for each of the intervals, and calculates the threshold value (±L) by multiplying the minimum or average value of the maximum values ​​of the position deviation (70) by a predetermined magnification.

[0058] (Supplementary Note 14) A numerical control method for controlling a machine tool (1) that operates according to a periodic operation pattern, comprising: position information acquisition means (3) for acquiring position information of the machine tool (1); position deviation calculation means (4) for calculating a position deviation, which is a deviation between a movement command (30) issued by the numerical control device (2) and the position information; a learning controller (10) for storing correction data (16) in a learning memory (12) by attenuating high frequency components of the position deviation through a band limiting filter (11), and further for performing learning control of the position deviation by implementing phase advancement on the correction data (16) using a dynamic characteristic compensation element (13); and unstable operation detection means (8) for detecting that the position deviation has exceeded a threshold value (±L) as an occurrence of unstable operation of the machine tool (1), When unstable operation is detected by the unstable operation detection means (8) after a predetermined time (T) or a predetermined period (C) has elapsed since the start of operation of the machine tool (1), at least one of speed gain change means (21) that changes a speed gain, learning condition change means (22) that changes the learning conditions of the learning controller (10), learning control disabling means (23) that disables the learning control, and machining condition change means (24) that changes the machining conditions of the periodic operation pattern is switched between and operated by a switching unit (26).

[0059] REFERENCE SIGNS LIST 1 Machine tool 2 Numerical control device 3 Position information acquisition means 4 Position deviation calculation means 5 Position deviation storage means 6 Steady-state position deviation deduction means 7 Swing command calculation means 8 Unstable operation detection means 9 Threshold calculation means 10 Learning controller 11 Band-limiting filter 12 Learning memory 13 Dynamic characteristic compensation element 16 Correction data 17 Position gain 21 Speed ​​gain change means 22 Learning condition change means 23 Learning control invalidation means 24 Machining condition change means 25 Swing condition change means 26 Switching unit 28 Swing feedforward coefficient 30 Movement command 31 Feedforward coefficient 34 Integral gain 35 Proportional gain 37 Feedforward term 40 Swing command 41 Superposition command 50 Position deviation 60 Swing feedforward term B Cutting tool W Workpiece S1 Spindle S2 Feed axis S3 Cutting axis M1 Main axis motor M2 Feed axis motor M3 Cutting axis motor T Predetermined time C Predetermined cycle ±L Threshold

Claims

1. A numerical control device for controlling a machine tool that operates according to a periodic operation pattern, comprising: position information acquisition means for acquiring position information of the machine tool; position deviation calculation means for calculating a position deviation, which is the deviation between a movement command issued by the numerical control device and the position information; a learning controller that stores correction data in a learning memory in which high-frequency components of the position deviation have been attenuated through a band-limiting filter, and further performs learning control of the position deviation by applying a phase advance to the correction data using a dynamic characteristic compensation element; and unstable operation detection means that detects when the position deviation has exceeded a threshold value as the occurrence of unstable operation of the machine tool, and when unstable operation is detected by the unstable operation detection means after a predetermined time or predetermined period has elapsed since the start of operation of the machine tool, the numerical control device switches between and operates at least one of speed gain change means for changing the speed gain, learning condition change means for changing the learning conditions of the learning controller, learning control disable means for disabling the learning control, and machining condition change means for changing the machining conditions of the periodic operation pattern using a switching unit.

2. A numerical control device for controlling a machine tool that machines a workpiece with a cutting tool, comprising: position information acquisition means for acquiring position information of the machine tool; swing command calculation means for calculating a swing command from a movement command and swing conditions issued by the numerical control device; a learning controller that stores correction data in a learning memory by attenuating, using a band-limiting filter, high-frequency components of a position deviation, which is the deviation between the position information and an superimposed command in which the swing command is superimposed on the movement command, and compensates for the position deviation; and unstable operation detection means that detects when the position deviation exceeds a threshold value as the occurrence of unstable operation of the machine tool, and when the unstable operation is detected after a predetermined time or predetermined period has elapsed since the start of operation of the machine tool, the numerical control device is characterized in that at least one of speed gain change means for changing the speed gain, learning condition change means for changing the learning conditions of the learning controller, learning control disabling means for disabling the learning control, and swing condition change means for changing the swing conditions is switched between and operated by a switching unit.

3. A numerical control device according to claim 2, further comprising a steady-state position deviation subtraction means for subtracting a steady-state position deviation determined by the movement command and a position gain after applying the swing command to the position deviation, and wherein the learning control creates the correction data based on the position deviation after subtracting the steady-state position deviation.

4. A numerical control device according to claim 3, wherein said steady-state position deviation subtracting means subtracts said steady-state position deviation using a high-pass filter.

5. The numerical control device according to claim 3, wherein said steady-state position deviation subtracting means subtracts an estimated steady-state position deviation from said movement command.

6. A numerical control device according to any one of claims 1 to 5, wherein said speed gain changing means changes at least one of an integral gain and a proportional gain.

7. A numerical control device according to any one of claims 1 to 5, characterized in that the learning condition changing means changes at least one of the cutoff frequency of the band-limiting filter, the order or coefficient of the dynamic characteristic compensation element.

8. The numerical control device according to claim 1, characterized in that the learning control disabling means compensates the speed command with a feedforward term obtained by multiplying the differential value of the periodic operating pattern by a feedforward coefficient after disabling the learning control.

9. A numerical control device according to any one of claims 2 to 5, characterized in that the learning control disabling means compensates the speed command with a swing feedforward term obtained by multiplying the differential value of the swing command by a swing feedforward coefficient after disabling the learning control.

10. A numerical control device according to claim 1, wherein said machining condition changing means (24) changes the frequency of said periodic operation pattern.

11. A numerical control device according to any one of claims 2 to 5, characterized in that the swing condition changing means (25) changes at least one of the frequency and amplitude of the swing command (40).

12. A numerical control device as described in any one of claims 1 to 5, characterized in that it comprises: a position deviation storage means (5) that stores a position deviation, which is the deviation between the position information and the movement command (30); and a threshold calculation means (9) that calculates the threshold value (±L) based on the position deviation stored in the position deviation storage means (5).

13. A numerical control device as described in claim 12, characterized in that the threshold calculation means (9) divides the position deviation (70) stored in the position deviation storage means (5) into predetermined cycles, calculates the maximum value of the position deviation for each of the sections, and calculates the threshold value (±L) by multiplying the minimum or average value of the maximum values of the position deviation (70) by a predetermined magnification.

14. A numerical control method for controlling a machine tool (1) that operates according to a periodic operation pattern, comprising: position information acquisition means (3) for acquiring position information of the machine tool (1); position deviation calculation means (4) for calculating a position deviation, which is the deviation between a movement command (30) issued by the numerical control device (2) and the position information; a learning controller (10) for storing correction data (16) in a learning memory (12) by attenuating high-frequency components of the position deviation through a band-limiting filter (11), and further for performing learning control of the position deviation by implementing phase advancement on the correction data (16) using a dynamic characteristic compensation element (13); and unstable operation detection means (8) for detecting that the position deviation has exceeded a threshold value (±L) as an occurrence of unstable operation of the machine tool (1), When unstable operation is detected by the unstable operation detection means (8) after a predetermined time (T) or a predetermined period (C) has elapsed since the start of operation of the machine tool (1), at least one of speed gain change means (21) for changing a speed gain, learning condition change means (22) for changing learning conditions of the learning controller (10), learning control disabling means (23) for disabling the learning control, and machining condition change means (24) for changing the machining conditions of the periodic operation pattern is switched between and operated by a switching unit (26).

Images