WO2023067787A1 - Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage - Google Patents

Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage Download PDF

Info

Publication number
WO2023067787A1
WO2023067787A1 PCT/JP2021/039045 JP2021039045W WO2023067787A1 WO 2023067787 A1 WO2023067787 A1 WO 2023067787A1 JP 2021039045 W JP2021039045 W JP 2021039045W WO 2023067787 A1 WO2023067787 A1 WO 2023067787A1
Authority
WO
WIPO (PCT)
Prior art keywords
unit
margin
gain
closed curve
output
Prior art date
Application number
PCT/JP2021/039045
Other languages
English (en)
Japanese (ja)
Inventor
亮太郎 恒木
Original Assignee
ファナック株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ファナック株式会社 filed Critical ファナック株式会社
Priority to JP2023554199A priority Critical patent/JPWO2023067787A1/ja
Priority to PCT/JP2021/039045 priority patent/WO2023067787A1/fr
Publication of WO2023067787A1 publication Critical patent/WO2023067787A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P29/00—Arrangements for regulating or controlling electric motors, appropriate for both AC and DC motors

Definitions

  • the present invention relates to a setting support device that supports a user in setting the stability margin of a servo control device, a control system including the setting support device, and a setting support method.
  • Japanese Unexamined Patent Application Publication No. 2002-100002 describes a setting support device that supports setting of a plurality of control parameters used in control processing in a motor control device.
  • a setting support device changes at least one value of a control parameter to control a servo driver to actually perform a test operation, or to perform a test operation by simulation using a virtual model. It is described that an operation instructing unit and a performance index calculation unit that calculates a performance index of control by the servo driver according to the result of the test operation are provided.
  • Patent Document 1 also describes that a velocity proportional gain and a position proportional gain can be used as control parameters, and a phase margin can be used as a performance index.
  • Patent Document 2 describes a design support device that supports the design of a DC power feeding system in which power is supplied from a DC power supply to a plurality of servo devices including an inverter circuit and an electric motor via a DC bus.
  • a design support device controls a plurality of servo units via a DC bus based on system information of a DC power supply system and operation pattern information indicating operation patterns of each of a plurality of servo units included therein.
  • Time-series current data indicating the time-varying pattern of the total current value supplied to the device is generated, and the output impedance Zo(s) on the power supply side of the DC power supply system and the input impedance Zin(s) on the load side of the DC power supply system s) and Zin(s), which is also a function of the value of the current flowing through the DC bus, the DC power supply system in which the current flowing through the DC bus is the maximum value of the generated time-series current data. is output, and the operation pattern information is changed based on the information.
  • Patent Document 2 also describes that the design support device displays a Nyquist diagram corresponding to the actual operation of the DC power supply system.
  • the stability margin during gain adjustment of the servo control device is often evaluated using a plurality of indices such as the gain margin, the phase margin, and the maximum gain of the closed loop gain.
  • indices such as the gain margin, the phase margin, and the maximum gain of the closed loop gain.
  • it is required to determine these indices in some form, but expert knowledge is required to set them appropriately. Therefore, when the user sets the stability margin of the servo control device, it is desired to easily set and change the stability margin.
  • a first aspect of the present disclosure is a setting assistance device that assists a user in setting a stability margin of a servo control device, a closed curve drawing unit that draws a closed curve on the complex plane that includes ( ⁇ 1, 0) on the complex plane and passes through the gain margin and the phase margin; a changing unit that changes the gain margin and the phase margin based on a user's operation; and a closed curve enlarging/reducing unit that expands or contracts the closed curve in conjunction with the change amount of the gain margin and the phase margin by the changing unit.
  • a second aspect of the present disclosure is the setting support device of (1) above;
  • the control system includes at least one filter coefficient and a feedback gain adjuster provided in the servo control device.
  • the computer a process of setting a first gain margin and a first phase margin, which are reference stability margins; A process of drawing a closed curve on the complex plane including ( ⁇ 1, 0) on the complex plane and passing through the first gain margin and the first phase margin on the complex plane; a process of changing the first gain margin and the first phase margin to a second gain margin and a second phase margin based on a user's operation; expanding or contracting the closed curve in conjunction with the amount of change from the first gain margin and the first phase margin to the second gain margin and the second phase margin; is a setting support method for executing
  • the stability margin when the user sets the stability margin of the servo control device, the stability margin can be easily set and changed.
  • FIG. 10 is a diagram showing a display screen of a display section displaying a complex plane and a slide bar; 3 is a partially enlarged view of the complex plane showing a unit circle on the complex plane and circles forming closed curves;
  • FIG. 10 is a diagram showing a display screen of the display section displaying a slide bar and a circle on the complex plane when the slider is moved; 3 is a partially enlarged view of a complex plane showing a unit circle on the complex plane and two circles forming closed curves;
  • FIG. 1 is a block diagram showing a control system of one embodiment of the present disclosure.
  • the control system 10 includes a servo control section 100 , a frequency generation section 200 , a frequency characteristic calculation section 300 , an adjustment section 400 and a setting support section 500 .
  • the servo control unit 100 corresponds to a servo control device
  • the setting support unit 500 corresponds to a setting support device.
  • One or more of the frequency generation section 200 , the frequency characteristic calculation section 300 , the adjustment section 400 and the setting support section 500 may be provided in the servo control section 100 .
  • the frequency characteristic calculator 300 may be provided within the adjuster 400 .
  • the setting support unit 500 may be provided within the adjustment unit 400 .
  • the servo control unit 100 includes a subtractor 110, a speed control unit 120, a filter 130, a current control unit 140, and a motor 150.
  • the subtractor 110, the speed control section 120, the filter 130, the current control section 140, and the motor 150 constitute a closed speed feedback loop servo system.
  • the motor 150 a linear motor that moves linearly, a motor that has a rotating shaft, or the like can be used.
  • the target driven by the motor 150 is, for example, a mechanical part of a machine tool, a robot, or an industrial machine.
  • Motor 150 may be provided as part of a machine tool, robot, industrial machine, or the like.
  • Control system 10 may be provided as part of a machine tool, robot, industrial machine, or the like. The details of the configuration of the servo control unit 100 will be described later.
  • the frequency generation unit 200 outputs a sine wave signal as a speed command to the subtractor 110 and the frequency characteristic calculation unit 300 of the servo control unit 100 while changing the frequency.
  • the frequency characteristic calculator 300 uses the speed command (sine wave) as an input signal generated by the frequency generator 200 and the detection as an output signal output from a rotary encoder (not shown) provided in the motor 150. Using the velocity (sine wave) or the integration of the detected position (sine wave) that becomes the output signal output from the linear scale, the amplitude ratio of the input signal and the output signal ( Input/output gain) and phase delay are obtained and output to the adjustment section 400 .
  • the setting support unit 500 assists the user in setting the values of the gain margin and phase margin (stability margin) of the open loop circuit of the servo control unit 100, and calculates a closed curve passing through the gain margin and the phase margin on the complex plane.
  • Outputs image data containing An open loop circuit is composed of the speed controller 120, the filter 130, the current controller 140, and the motor 150 shown in FIG.
  • the adjustment unit 400 uses a closed curve passing through the gain margin and the phase margin included in the image data output from the setting support unit 500 and the input/output gain (amplitude ratio) and the phase delay output from the frequency characteristic calculation unit 300.
  • One or both of the integral gain K1v and the proportional gain K2v of the velocity control unit 120 are adjusted so that the gain margin and the phase margin of the servo control unit 100 are equal to or greater than the values set by the user using the obtained Nyquist locus. and/or the coefficients ⁇ c , ⁇ , ⁇ of the transfer function of filter 130 are adjusted.
  • the servo control section 100 comprises a subtractor 110, a speed control section 120, a filter 130, a current control section 140 and a motor 150.
  • the subtractor 110 obtains the difference between the input speed command and the speed feedback detected speed, and outputs the difference to the speed control unit 120 as a speed deviation.
  • the speed control unit 120 adds the value obtained by multiplying the speed deviation by the integral gain K1v and integrating it, and the value obtained by multiplying the speed deviation by the proportional gain K2v, and outputs the result to the filter 130 as a torque command.
  • the speed control section 120 serves as a control section that sets a feedback gain.
  • the filter 130 is a filter that attenuates a specific frequency component, such as a notch filter, low-pass filter, or band-stop filter.
  • a machine such as a machine tool having a mechanical section driven by the motor 150 has a resonance point, and the resonance may increase in the servo control section 100 .
  • Resonance can be reduced by using a filter such as a notch filter.
  • the output of filter 130 is output to current control section 140 as a torque command.
  • Equation 1 (shown as Equation 1 below) represents the transfer function F(s) of the notch filter as filter 130 .
  • the parameters indicate the coefficients ⁇ c , ⁇ , ⁇ .
  • Current control unit 140 generates a current command for driving motor 150 based on the torque command, and outputs the current command to motor 150 .
  • the motor 150 is a linear motor
  • the position of the movable portion is detected by a linear scale (not shown) provided on the motor 150
  • the detected speed value is obtained by differentiating the detected position value, and the detected speed is obtained.
  • the value is input to subtractor 110 as velocity feedback. If the motor 150 has a rotating shaft, the rotation angle position is detected by a rotary encoder (not shown) provided on the motor 150, and the speed detection value is input to the subtractor 110 as speed feedback.
  • FIG. 2 is a block diagram showing a configuration example of the setting support unit.
  • the setting support unit 500 includes a reference stability margin setting unit 501 , a closed curve drawing unit 502 , a display unit 503 , a stability margin changing unit 504 and a closed curve scaling unit 505 as changing units.
  • a reference stability margin setting unit 501 sets a reference stability margin (gain margin and phase margin) (hereinafter referred to as a reference stability margin).
  • the reference stability margin may be set to a predetermined value in advance, or may be arbitrarily set by the user.
  • the closed curve drawing unit 502 draws a unit circle whose circumference passes through ( ⁇ 1, 0) on the complex plane 5031 of FIGS. draw a closed curve that intersects with this unit circle and includes (-1, 0) on the complex plane.
  • the center of the circle is on the real axis in FIG. 3, it does not have to be on the real axis.
  • the point where the circle and the real axis intersect is the gain margin, and the point where the circle and the unit circle intersect is the phase margin.
  • a closed curve may be a closed curve other than a circle, such as a rhombus, a rectangle, or an ellipse. In the following description, it is assumed that the closed curve is a circle.
  • the display unit 503 displays a complex plane 5031, a slide bar 5032, and an end button 5033 on the display screen 503A shown in FIG.
  • FIG. 3 shows a display screen 503A of the display unit 503.
  • the complex plane 5031 is output from the closed curve drawing unit 502 as image data.
  • the display unit 503 is composed of a liquid crystal display device, and the slide bar 5032 serves as an operation unit for the stability margin changing unit 504, which will be described later.
  • FIG. 4 is a partially enlarged view of the complex plane 5031, showing a unit circle on the complex plane and circles that form closed curves.
  • the slider 5032A of the slide bar 5032 shown in FIG. 3 and the stability margin are interlocked, and the user changes the stability margin (gain margin and phase margin) by moving the slider 5032A of the slide bar 5032 shown in FIG. can do.
  • the slider 5032A moves left and right in FIG. 3, the direction in which the slider moves is not particularly limited, and the slider may move up and down.
  • a stability margin changing unit 504 is a changing unit that changes the gain margin and the phase margin based on the user's operation, and is incorporated as part of the display unit 503 .
  • the stability margin changing unit 504 is, for example, a slide bar.
  • the slide bar displays a small mark (slider) indicating the current position on the bar-shaped area on the display screen, and slides it in the bar-shaped area. It is an input element operated by In FIG. 3, as described above, the stability margin change unit 504 creates the slide bar 5032 as an operation unit and displays it on the display screen 503A. When the user moves the slider 5032A of the slide bar 5032 shown in FIG.
  • the amount of movement of the slider 5032A corresponds to the amount of change between the gain margin and the phase margin, and by moving the slider 5032A, the gain margin and the phase margin can be changed.
  • the stability margin change unit 504 may be provided separately from the display unit 503 , and in this case, the slide bar 5032 is displayed on a display screen of a display device separate from the display unit 503 .
  • the stability margin changing unit 504 may be an operation tool that slides and operates a knob-like slider used in a machine control panel. When the stability margin changing unit 504 is an operating instrument, the stability margin changing unit 504 is provided separately from the display unit 503 .
  • the closed curve scaling unit 505 scales up or down the diameter of the circle passing through the stability margin (gain margin and phase margin) with the center of the circle as the reference point, corresponding to the amount of movement of the slider 5032A.
  • the reference point for enlargement or reduction may be another point on the real axis instead of the center of the circle.
  • the user moves the slider 5032A of the slide bar 5032 from the center to the left in FIG. 5 when emphasizing the stability of the servo system. from the center to the right side of FIG.
  • the gain margin is 6 dB and the phase margin is 30 degrees.
  • the gain margin is 10 dB and the phase margin is 45 degrees as the limit values for stability
  • the gain margin is 4.5 dB and the phase margin is 25 degrees as the limit values for responsiveness.
  • a closed curve drawing unit 502 draws a circle with an enlarged diameter on a complex plane, and a display unit 503 displays a complex plane with a circle with an enlarged diameter drawn on the display screen and a slide bar with a slider moved to the left in FIG. to display.
  • the user looks at the complex plane on which a circle with an enlarged diameter is drawn displayed on the display screen of the display unit 503, and moves the slider 5032A of the slide bar 5032 to the left or right to change the stability margin again.
  • the closed curve drawing unit 502 outputs to the adjustment unit 400 the image data regarding the complex plane including the circle that becomes the closed curve and the unit circle.
  • the closed curve drawing unit 502 draws a unit circle whose circumference passes through (-1, 0) on the complex plane 5031 as shown in FIG. , and based on two reference stability margins (gain margin and phase margin), two first and second circles that intersect this unit circle can also be drawn.
  • the reference stability margin setting unit 501 reduces the stability margin for cutting feed of the machine tool and increases the stability margin for rapid traverse.
  • the closed curve drawing unit 502 draws a circle for cutting feed as the first circle and a circle for fast feed as the second circle. Then, when the user moves the slider 5032A of the slide bar 5032, the closed curve scaling unit 505 changes the radii of the first circle and the second circle.
  • the centers of the first and second circles are common, but the centers of the first and second circles may be different.
  • the closed-curve drawing unit 502 does not limit the number of circles forming a closed curve to two, and may increase the number to three or more as necessary. By using a plurality of closed curves in this way, a closed curve suitable for each mode can be used in a plurality of modes of the servo control device, for example, cutting feed and rapid feed.
  • the adjustment unit 400 acquires from the closed-curve drawing unit 502 the image data on the complex plane including the circle that forms the closed curve and the unit circle. Furthermore, the adjustment unit 400 obtains the open loop frequency characteristic H(j ⁇ ) using the input/output gain (amplitude ratio) and the phase delay output from the frequency characteristic calculation unit 300, and creates the open loop frequency characteristic H(j ⁇ ).
  • the Nyquist trajectory is drawn on the complex plane containing the closed circle and the unit circle.
  • FIG. 7 is a diagram showing the Nyquist locus, the unit circle, and the circle passing through the gain and phase margins drawn on the complex plane.
  • the adjustment unit 400 adjusts the integral gain K1v and the proportional gain K2v of the velocity control unit 120 and the transfer function coefficients ⁇ c , ⁇ , ⁇ of the filter 130 so that the Nyquist locus does not pass through the inside of the circle.
  • the velocity feedback loop consists of the subtractor 110 and the transfer function H open loop circuit.
  • the open loop circuit is composed of the speed controller 120, the filter 130, the current controller 140, and the motor 150 shown in FIG. 1, as already explained.
  • the closed loop frequency characteristic G(j ⁇ 0 ) is c ⁇ e j ⁇ .
  • the adjustment unit 400 drives the servo control unit 100 using a speed command (sine wave) whose frequency changes based on the integral gain K1v, the proportional gain K2v, and the coefficients ⁇ c , ⁇ , and ⁇ .
  • the input/output gain and phase delay are obtained from the frequency characteristic calculator 300 .
  • the adjustment section 400 uses the input/output gain and phase delay obtained from the frequency characteristic calculation section 300 to draw the open loop frequency characteristic H(j ⁇ ) on the complex plane to create a Nyquist locus.
  • the Nyquist locus in the initial state is obtained by driving the servo control unit 100 using a velocity command (sine wave) based on the integral gain K1v and proportional gain K2v set by the user and the coefficients ⁇ c , ⁇ , and ⁇ . be done.
  • the Nyquist locus during adjustment is obtained by correcting the integral gain K1v and proportional gain K2v and/or the coefficients ⁇ c , ⁇ , ⁇ and driving the servo control unit 100 using a speed command (sine wave).
  • step S11 the reference stability margin setting unit 501 sets the stability margins (gain margin and phase margin) that serve as the reference stability reference.
  • step S12 the closed curve drawing unit 502 draws a unit circle whose circumference passes through ( ⁇ 1, 0) on the complex plane, and draws the unit circle Draw a circle that is a closed curve that intersects with .
  • step S13 the display unit 503 displays on the display screen of the display unit 503 the complex plane drawn by the closed curve drawing unit 502, which indicates the unit circle and the closed curve circle, the slide bar, and the end button.
  • step S14 the closed curve scaling unit 505 determines whether the user has moved the slider of the slide bar.
  • the closed curve enlarging/reducing unit 505 proceeds to step S15 when the slider of the slide bar has moved, and proceeds to step S16 when the slider of the slide bar has not moved.
  • step S15 the closed curve scaling unit 505 scales up or down the diameter of a circle centered on the reference point and passing through the stability margin (gain margin and phase margin) in accordance with the amount of movement of the slider, and returns to step S12.
  • step S ⁇ b>16 the closed-curve drawing unit 502 outputs to the adjustment unit 400 image data on the complex plane including the circle that forms the closed curve and the unit circle.
  • the stability margin can be easily changed by using the stability margin changing section such as the slide bar of the setting support section.
  • FIG. 9 is a block diagram showing a modification of the control system in which the adjustment section 400 shown in FIG. 1 is replaced with a machine learning section 400A.
  • Control system 10A is the same as control system 10 shown in FIG.
  • the learning performed by the machine learning unit 400A is not particularly limited to reinforcement learning.
  • the present invention can also be applied when performing supervised learning. is.
  • An agent corresponding to the machine learning unit 400A in this embodiment observes the state of the environment, selects a certain action, and changes the environment based on the action. As the environment changes, some kind of reward is given, and the agent learns to choose better actions (decisions). While supervised learning shows perfect correct answers, rewards in reinforcement learning are often fragmentary values based on partial changes in the environment. Thus, the agent learns to choose actions that maximize the total reward over the future.
  • Q-learning any learning method can be used.
  • the purpose of Q-learning is to select the action A with the highest value Q(S, A) from among possible actions A in a certain state S as the optimum action.
  • the agent selects various actions A under a certain state S, and selects a better action based on the reward given to the action A at that time, thereby obtaining the correct value Q(S , A).
  • Equation 2 S t represents the state of the environment at time t, and A t represents the behavior at time t.
  • Action At causes the state to change to St +1 .
  • r t+1 represents the reward obtained by changing that state.
  • the term with max is the product of ⁇ times the Q value when action A with the highest Q value known at that time is selected under state St+1 .
  • ⁇ is a parameter that satisfies 0 ⁇ 1 and is called a discount rate.
  • ⁇ is a learning coefficient and is in the range of 0 ⁇ 1.
  • Equation 2 described above represents a method of updating the value Q ( St , At ) of the action At in the state St based on the reward rt +1 returned as a result of the trial At .
  • This update formula is based on the value max a Q(St +1 , A ) of the best action in the next state St +1 by the action A t rather than the value Q( St, At ) of the action A t in the state St. If it is larger, Q(S t ,A t ) is increased, and if it is smaller, Q(S t ,At ) is decreased. That is, it brings the value of one action in one state closer to the value of its best action in the next state. However, the difference varies depending on the discount rate ⁇ and reward r t+1 . Basically, the value of the best action in a certain state propagates to the value of the action in the previous state. It is a system that goes on.
  • DQN Deep Q-Network
  • the value Q (S, A) is calculated by approximating the value function Q with an appropriate neural network. You may make it calculate a value.
  • DQN it is possible to shorten the time required for Q-learning to converge. DQN is described in detail in, for example, the following non-patent literature.
  • Non-Patent Literature> Human-level control through deep reinforcement learning, by Volodymyr Mnih1 [online], [searched January 17, 2017], Internet ⁇ URL: http://files.davidqiu.com/research/nature14236.pdf>
  • the machine learning unit 400A performs the Q-learning described above. Specifically, the machine learning unit 400A calculates the integral gain K1v and the proportional gain K2v of the speed control unit 120, the values of the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130, and Assuming that the input/output gain (amplitude ratio) and the phase delay are in a state S, the integral gain K1v and proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130 related to the state S are: Learn a value Q that selects value adjustment as action A.
  • the machine learning unit 400A Based on the integral gain K1v and proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130, the machine learning unit 400A generates the speed command, which is a sinusoidal wave whose frequency changes as described above. is used to drive the servo control unit 100, the state information S including the input/output gain and phase delay for each frequency obtained from the frequency characteristic calculation unit 300 is observed, and action A is determined.
  • the machine learning unit 400A receives a reward each time action A is performed.
  • the machine learning unit 400A for example, searches for the optimum action A by trial and error so that the total future reward is maximized.
  • the machine learning unit 400A can generate a sine wave whose frequency changes based on the integral gain K1v and the proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130.
  • Optimal action A that is, , the integral gain K1v and the proportional gain K2v of the speed control unit 120, and the optimum coefficients ⁇ c , ⁇ , ⁇ ) of the transfer function of the filter 130 can be selected.
  • the integral gain K1v and proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , of the transfer function of the filter 130 related to a certain state S Servo control unit 100 generated by executing a program that generates a sinusoidal signal whose frequency changes by selecting the action A that maximizes the value of Q from among the actions A applied to ⁇ . select an action A (that is, the integral gain K1v and proportional gain K2v of the speed control unit 120, and/or the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130) such that the stability margin of is equal to or greater than a predetermined value. It becomes possible to
  • FIG. 10 is a block diagram showing the configuration of the machine learning section 400A.
  • the learning unit 402 includes a reward output unit 4021, a value function update unit 4022, and an action information generation unit 4023.
  • the state information acquisition unit 401 performs servo control using a speed command (sine wave) based on the integral gain K1v and the proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130.
  • the state S including the input/output gain (amplitude ratio) and the phase delay obtained by driving the unit 100 is acquired from the frequency characteristic calculation unit 300 .
  • This state information S corresponds to the environmental state S in Q-learning.
  • the state information acquisition unit 401 also acquires image data on a complex plane including a circle that is a closed curve and a unit circle from the closed curve drawing unit 502 .
  • the state information acquisition unit 401 outputs to the learning unit 402 the acquired state information S and the image data related to the complex plane including the circle forming the closed curve and the unit circle.
  • the integral gain K1v and proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130 at the time when the Q learning is first started are generated by the user in advance.
  • the user-created initial settings of the integral gain K1v and proportional gain K2v of the speed control unit 120 and/or the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130 are optimized by reinforcement learning. adjust to what When the machine tool is adjusted in advance by the operator, the integral gain K1v, the proportional gain K2v, and the coefficients ⁇ c , ⁇ , and ⁇ may be machine-learned using the adjusted values as initial values.
  • the learning unit 402 is a part that learns the value Q(S, A) when a certain action A is selected under a certain environmental state S.
  • the reward output unit 4021 of the learning unit 402 is a part that obtains a reward when action A is selected under a certain state S.
  • FIG. 4021 is a part that obtains a reward when action A is selected under a certain state S.
  • the reward output unit 4021 acquires image data relating to a complex plane including a closed curve circle and a unit circle from the state information acquisition unit 401 .
  • the reward output unit 4021 uses the input/output gain and the phase delay obtained from the state information obtaining unit 401 to draw the open-loop frequency characteristic H(j ⁇ ) on the obtained complex plane to create a Nyquist trajectory. Since the method of creating the Nyquist locus has already been explained in the explanation of the operation of the adjustment unit 400, it will be omitted here. In this way, the complex plane shown in FIG. 7 is obtained showing the circle passing through the Nyquist locus, the unit circle, and the gain and phase margins.
  • the Nyquist locus in the initial state is obtained by driving the servo control unit 100 using a velocity command (sine wave) based on the integral gain K1v and proportional gain K2v set by the user and the coefficients ⁇ c , ⁇ , and ⁇ . be done.
  • the Nyquist trajectory in the process of Q-learning is obtained by correcting the integral gain K1v and the proportional gain K2v and/or the coefficients ⁇ c , ⁇ , ⁇ and driving the servo control unit 100 using a speed command (sine wave). be done.
  • the radius of the circle is assumed to be radius r
  • the shortest distance between the circle and the Nyquist locus is assumed to be d.
  • the shortest distance d is the shortest distance between the center of the circle and the Nyquist locus, but it is not limited to this, and may be the shortest distance between the outer periphery of the circle and the Nyquist locus, for example.
  • the reward output unit 4021 gives a negative reward when the shortest distance d is smaller than the radius r (d ⁇ r) and the Nyquist locus passes inside the closed curve. On the other hand, the reward output unit 4021 gives a zero value reward when the shortest distance d is equal to or greater than the radius r (d ⁇ r) and the Nyquist locus does not pass inside the circle.
  • machine learning unit 400A prevents the Nyquist locus from passing through the inside of the circle, and makes the gain margin and phase margin equal to or greater than the values set by the user.
  • the proportional gain K2v and the coefficients ⁇ c , ⁇ , ⁇ of the transfer function of filter 130 are searched by trial and error.
  • whether or not the Nyquist locus passes through the closed circle is determined based on the shortest distance between the circle and the Nyquist locus. Well, for example, it may be determined by whether or not the Nyquist locus touches or intersects with the outer circumference of a circle that is a closed curve.
  • the reward output unit 4021 gives a reward that is equal to or greater than the gain margin and the phase margin determined by the user and that the feedback gain is as large as possible.
  • the reward output unit 4021 performs servo control using a speed command (sine wave) based on the integral gain K1v, the proportional gain K2v, and the coefficients ⁇ c , ⁇ , and ⁇ .
  • a Bode diagram is created from the closed loop input/output gain (amplitude ratio) and phase delay obtained by driving the unit 100 .
  • FIG. 11 shows an example of a closed-loop Bode diagram.
  • the cutoff frequency is, for example, the frequency at which the gain characteristic of the Bode diagram is -3 dB or the frequency at which the phase characteristic is -180 degrees. In FIG. 11, the cutoff frequency is the frequency at which the gain characteristic is ⁇ 3 dB.
  • the reward output unit 4021 determines the reward so that the cutoff frequency is increased. Specifically, the reward output unit 4021 corrects the integral gain K1v, the proportional gain K2v, and/or the coefficients ⁇ c , ⁇ , and ⁇ , and when the state S before correction changes to state S′, the cutoff frequency A reward is determined depending on whether fcut is large, the same, or small.
  • the cutoff frequency fcut in the state S is denoted as fcut(S)
  • the cutoff frequency fcut in the state S' is denoted as fcut(S').
  • the integral gain K1v and proportional gain K2v of the velocity control unit 120 and/or the filter are searched by trial and error. As the cutoff frequency fcut increases, the feedback gain increases and the response speed increases.
  • the reward output unit 4021 corrects the integral gain K1v, the proportional gain K2v, and/or the coefficients ⁇ c , ⁇ , and ⁇ , and when the state S before correction becomes the state S′, the evaluation function A reward is determined depending on whether the value of f becomes smaller, the same, or larger.
  • the value of the evaluation function f in the state S is described as f(S)
  • the value of the evaluation function f in the state S' is described as f(S').
  • the cut frequency of the Bode diagram of the closed loop shown in FIG. 11 increases.
  • the integral gain K1v and proportional gain K2v of the speed control unit 120 and the filter 130 are calculated so that the value of the evaluation function f becomes small.
  • the coefficients ⁇ c , ⁇ , ⁇ of the transfer function of are searched for by trial and error. As the value of the evaluation function f decreases, the feedback gain increases and the response speed increases.
  • the Nyquist trajectory approaches a closed curve determine the reward.
  • the reward output unit 4021 corrects the integral gain K1v, the proportional gain K2v, and/or the coefficients ⁇ c , ⁇ , and ⁇ , and when the state S before correction changes to state S′, the circle A reward is determined depending on whether the shortest distance d between the center and the Nyquist locus is small, the same, or large.
  • the shortest distance d in state S is described as d(s)
  • the shortest distance d in state S' is described as d(s').
  • the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or the coefficient ⁇ c of the transfer function of the filter 130 are obtained so that the Nyquist locus passes on the circle or approaches the circumference of the circle.
  • ⁇ , ⁇ are searched by trial and error.
  • the feedback gain increases and the response speed increases.
  • the method of determining the reward based on the information of the shortest distance d is not limited to the above method, and other methods can be applied.
  • the reward output unit 4021 determines the reward so as to suppress the resonance above the gain margin and the phase margin determined by the user. A method of determining the reward by comparing the open-loop characteristics and the reference model will be described below.
  • the operation of the reward output unit 4021 to give a negative reward when the input/output gain for each frequency in the generated frequency characteristics is greater than the input/output gain of the reference model will be described below with reference to FIGS. 12 and 13. .
  • the reward output unit 4021 stores a reference model of input/output gains.
  • the reference model is a model of a servo control section that has ideal characteristics without resonance.
  • the reference model can be calculated from the inertia Ja, torque constant Kt , proportional gain Kp , integral gain KI , and differential gain KD of the model shown in FIG. 12, for example.
  • Inertia Ja is the sum of motor inertia and mechanical inertia.
  • FIG. 13 is a characteristic diagram showing frequency characteristics of input/output gains of the servo control unit of the reference model and the servo control unit 100 before and after learning.
  • the reference model has an area FA, which is a frequency area with an ideal input/output gain above a certain input/output gain, for example, ⁇ 20 dB or above, and a frequency area below the certain input/output gain. and a region FB which is a frequency region where In area FA of FIG. 13, the ideal input/output gain of the reference model is indicated by curve MC 1 (thick line). In the region FB of FIG.
  • curve MC 11 the ideal virtual input/output gain of the reference model
  • straight line MC 12 the input/output gain of the reference model
  • curves RC 1 and RC 2 indicate curves of input/output gains with the servo control unit before and after learning, respectively.
  • the reward output unit 4021 outputs a negative value when the pre-learning curve RC1 of the input/output gain for each frequency in the created frequency characteristics exceeds the ideal input/output gain curve MC1 of the reference model. Reward. In the region FB exceeding the frequency at which the input/output gain becomes sufficiently small, even if the input/output gain curve RC1 before learning exceeds the ideal virtual input/output gain curve MC11 of the reference model, the stability is not affected. become smaller. Therefore, in the area FB, as described above, the input/output gain of the reference model uses a straight line MC12 of a constant input/output gain (eg, -20 dB) instead of the ideal gain characteristic curve MC11 . However, if the input/output gain curve RC1 measured before learning exceeds the fixed input/output gain straight line MC12 , it may become unstable, so a negative value is given as a reward.
  • a constant input/output gain eg, -20 dB
  • the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130 are adjusted.
  • the gain and phase change depending on the bandwidth fw of the filter 130
  • the gain and phase change depending on the attenuation coefficient k of the filter 130 can be adjusted by adjusting the coefficient of the filter 130 .
  • the reward output unit 4021 outputs the negative reward as a value function. Output to the update unit 4022 . If the reward output unit 4021 gives a positive reward when the shortest distance d is equal to or greater than the radius r (d ⁇ r) and the Nyquist locus does not pass through the circle, It outputs the reward to the value function updating unit 4022 . When the reward output unit 4021 gives a reward with three examples considering response speed or an example considering resonance, the reward is added to the reward with a positive value given when the Nyquist locus does not pass inside the circle. to the value function updating unit 4022.
  • the positive value reward given when the Nyquist locus does not pass through the inside of the circle is the reward given in the three examples considering the response speed or the example considering the resonance. can be given weights that make them more important than The reward output unit 4021 has been described above.
  • the value function updating unit 4022 performs Q-learning based on the state S, the action A, the state S′ when the action A is applied to the state S, and the reward obtained as described above,
  • the value function Q stored in the value function storage unit 404 is updated.
  • the value function Q may be updated by online learning, batch learning, or mini-batch learning.
  • Online learning is a learning method in which, by applying a certain action A to the current state S, the value function Q is updated immediately each time the state S transitions to a new state S'.
  • batch learning learning data is collected by applying a certain action A to the current state S, and by repeating the transition of the state S to a new state S′. This is a learning method in which the value function Q is updated using learning data.
  • mini-batch learning is a learning method intermediate between online learning and batch learning, in which the value function Q is updated every time learning data is accumulated to some extent.
  • the action information generation unit 4023 selects action A for the current state S in the process of Q learning.
  • the action information generation unit 4023 performs an operation (Q-learning (equivalent to the action A in )), the action information A is generated and the generated action information A is output to the action information output unit 403 .
  • the action information generation unit 4023 generates, for example, the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or each coefficient ⁇ c , ⁇ , of the transfer function of the filter 130 included in the state S.
  • the integral gain K1v and proportional gain K2v of the speed control unit 120 and the transfer function coefficients ⁇ c , ⁇ , and ⁇ of the filter 130 included in the action A may be incrementally added or subtracted from ⁇ .
  • All of the integral gain K1v and proportional gain K2v of the speed control unit 120 and the coefficients ⁇ c , ⁇ , and ⁇ of the filter 130 may be corrected, or some of the coefficients may be corrected.
  • the action information A may be generated, and the generated action information A may be output to the action information output unit 403 in order to perform the operation of correcting the attenuation coefficient ⁇ .
  • the action information generation unit 4023 uses a greedy method for selecting the action A' having the highest value Q(S, A) among the currently estimated values of the action A, or a random action with a certain small probability ⁇ .
  • A' is selected, and otherwise, a known method such as the ⁇ -greedy method of selecting the action A' with the highest value Q(S, A) may be used to select the action A'.
  • the action information output unit 403 is a part that transmits the action information A output from the learning unit 402 to the speed control unit 120 and the filter 130 .
  • the filter 130 determines the current state S, that is, the currently set integral gain K1v and proportional gain K2v of the speed control unit 120, and/or the coefficients ⁇ c , ⁇ , By finely modifying ⁇ , a transition is made to the next state S' (that is, the modified integral gain K1v and proportional gain K2v of the speed control unit 120 and/or each coefficient of the filter 130).
  • the value function storage unit 404 is a storage device that stores the value function Q.
  • the value function Q may be stored as a table (hereinafter referred to as an action value table) for each state S and action A, for example.
  • Value function Q stored in value function storage unit 404 is updated by value function update unit 4022 .
  • the value function Q stored in the value function storage unit 404 may be shared with another machine learning unit 400A. If the value function Q is shared by a plurality of machine learning units 400A, it is possible to perform reinforcement learning in a distributed manner in each machine learning unit 400A, so that the efficiency of reinforcement learning can be improved. Become.
  • the optimized behavior information output unit 405 selects an operation that maximizes the value Q(S, A) by the speed control unit 120 and filter 130.
  • Behavior information A (hereinafter referred to as “optimization behavior information”) is generated. More specifically, the optimized behavior information output unit 405 acquires the value function Q stored in the value function storage unit 404. FIG. This value function Q is updated by the value function updating unit 4022 performing Q learning as described above. Then, the optimized behavior information output unit 405 generates behavior information based on the value function Q, and outputs the generated behavior information to the filter 130 .
  • this optimization action information includes the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or the transfer function of the filter 130.
  • Information that modifies the coefficients ⁇ c , ⁇ , ⁇ is included.
  • the speed control unit 120 corrects the integral gain K1v and the proportional gain K2v based on this action information, and the filter 130 corrects each coefficient ⁇ c , ⁇ , ⁇ of the transfer function based on this action information.
  • the machine learning unit 400A optimizes the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130 through the above operation, and the servo control unit 100 can be operated so that the stability margin of is greater than or equal to a predetermined value.
  • the above operation optimizes the integral gain K1v and the proportional gain K2v of the speed control unit 120 and/or the coefficients ⁇ c , ⁇ , and ⁇ of the transfer function of the filter 130, thereby increasing the stability margin of the servo control unit 100.
  • the gain of the speed control unit 120 and parameter adjustment of the filter 130 can be simplified.
  • control system 10 or 10A, the setting support unit 500 or the machine learning unit 400A is provided with an arithmetic processing device such as a CPU (Central Processing Unit).
  • the control systems 10 and 10A also include auxiliary storage devices such as HDDs (Hard Disk Drives) storing various control programs such as application software or OSs (Operating Systems), and an arithmetic processing unit for executing programs. It also has a main memory such as a random access memory (RAM) for storing temporarily needed data.
  • main memory such as a random access memory (RAM) for storing temporarily needed data.
  • the arithmetic processing unit reads the application software or OS from the auxiliary storage device, and develops the read application software or OS in the main storage device, Arithmetic processing is performed based on these application software or OS. Also, based on the result of this calculation, various hardware included in each device is controlled. This implements the functional blocks of the present embodiment. In other words, this embodiment can be realized by cooperation of hardware and software.
  • the machine learning unit 400A has a large amount of computation associated with machine learning
  • a personal computer is equipped with a GPU (Graphics Processing Units), and a technology called GPGPU (General-Purpose computing on Graphics Processing Units) is used to convert the GPU into a machine.
  • GPGPU General-Purpose computing on Graphics Processing Units
  • High-speed processing can be achieved by using it for arithmetic processing associated with learning.
  • multiple computers equipped with such GPUs are used to construct a computer cluster, and the multiple computers included in this computer cluster perform parallel processing.
  • Each component included in the above control system 10 or 10A can be realized by hardware, software, or a combination thereof. Further, the setting support method performed by cooperation of each component included in the setting support device can also be realized by hardware, software, or a combination thereof.
  • “implemented by software” means implemented by a computer reading and executing a program.
  • 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 discs), CD-ROMs (Read Only Memory), CD-Rs, CD-R/ W, including semiconductor memory (eg, mask ROM, PROM (programmable ROM), EPROM (erasable PROM), flash ROM, RAM (random access memory));
  • the program may also be delivered to the computer on various types of transitory computer readable medium.
  • FIG. 14 is a block diagram showing an example of configuring a filter by directly connecting a plurality of filters.
  • filter 130 when there are m resonance points (where m is a natural number of 2 or more), filter 130 is configured by connecting m filters 130-1 to 130-m in series.
  • Optimal values of the coefficients ⁇ c , ⁇ , and ⁇ of the m filters 130-1 to 130-m are obtained by machine learning.
  • FIG. 15 is a block diagram showing another configuration example of the control device.
  • the control device 10B shown in FIG. 15 differs from the control system 10 shown in FIG. It is connected to n adjusting sections 400-1 to 400-n, and has a frequency generating section 200 and a frequency characteristic calculating section 300, respectively.
  • the n adjustment units 400-1 to 400-n are connected to the n setting support units 500-1 to 500-n.
  • Setting support units 500-1 to 500-n have the same configuration as setting support unit 500 shown in FIG.
  • Servo control units 100-1 to 100-n correspond to servo control devices, respectively, and setting support units 500-1 to 500-n correspond to setting support devices, respectively.
  • the frequency generator 200 and the frequency characteristic calculator 300 may be provided outside the servo controllers 100-1 to 100-n.
  • the servo control section 100-1 and the adjustment section 400-1 are paired in a one-to-one manner and are communicably connected.
  • Servo control units 100-2 to 100-n and adjustment units 400-2 to 400-n are connected in the same manner as servo control unit 100-1 and adjustment unit 400-1.
  • n sets of servo control units 100-1 to 100-n and adjustment units 400-1 to 400-n are connected via network 600, but servo control unit 100-1 100-n and adjustment units 400-1 to 400-n, the servo control unit and machine learning unit of each set may be directly connected via a connection interface.
  • These n sets of servo control units 100-1 to 100-n and adjustment units 400-1 to 400-n may be installed in the same factory, for example, or may be installed in different factories. good too.
  • the network 600 is, for example, a LAN (Local Area Network) built in a factory, the Internet, a public telephone network, or a combination thereof.
  • the servo control units 100-1 to 100-n and the adjustment units 400-1 to 400-n are connected as one-to-one pairs so as to be communicable. may be communicably connected to a plurality of servo control units via the network 600 .
  • a distributed processing system may be employed in which each function of one adjustment unit is appropriately distributed to a plurality of servers.
  • each function of one adjustment unit may be realized by using a virtual server function or the like on the cloud.
  • each adjustment may be shared. By doing so, it becomes possible to construct a more optimal model.
  • a setting support device e.g., setting support unit 500 that assists a user in setting a stability margin of a servo control device (e.g., servo control unit 100), a closed curve drawing unit (for example, a closed curve drawing unit 502) that draws a closed curve on the complex plane that includes ( ⁇ 1, 0) on the complex plane and passes through the gain margin and the phase margin; a changing unit (for example, a stability margin changing unit 504) that changes the gain margin and the phase margin based on a user's operation;
  • a setting support device comprising: a closed curve enlarging/reducing unit (for example, a closed curve enlarging/reducing unit 505) that expands or contracts the closed curve in conjunction with the change amount of the gain margin and the phase margin by the changing unit.
  • a setting support device according to any one of (1) to (4) above;
  • a control system comprising at least one filter coefficient and a feedback gain adjuster provided in the servo controller. According to this control system, the user can easily set and change the stability margin when setting the stability margin of the servo control device.
  • the adjustment unit a state information acquisition unit that acquires state information including at least one of the filter coefficient and the feedback gain, and input/output gain and input/output phase delay of the servo control device; an action information output unit that outputs action information including adjustment information for at least one of the coefficient and the feedback gain included in the state information; a reward output unit that obtains and outputs a reward based on whether the Nyquist trajectory calculated from the input/output gain and the input/output phase delay passes through the inside of the closed curve output from the setting support device; a value function updating unit that updates a value function based on the reward value output by the reward output unit, the state information, and the behavior information;
  • the control system according to (7) above, which is a machine learning unit comprising:
  • the computer a process of setting a first gain margin and a first phase margin, which are reference stability margins;
  • a configuration assistance method that performs According to this setting support method, the user can easily set and change the stability margin when setting the stability margin of the servo control device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Feedback Control In General (AREA)

Abstract

Selon la présente invention, un utilisateur peut régler et modifier facilement la marge de stabilité d'un dispositif de servocommande.　Un dispositif de soutien de réglage, destiné à soutenir un utilisateur dans le réglage d'une marge de stabilité d'un dispositif de servocommande, comprend : une unité de dessin de courbe fermée destinée à dessiner, sur un plan complexe, une courbe fermée renfermant (-1, 0) sur le plan complexe et passant à travers une marge de gain et une marge de phase ; une unité de modification destinée à modifier la marge de gain et la marge de phase en fonction d'un actionnement de l'utilisateur ; et une unité de mise à l'échelle de courbe fermée destinée à élargir ou à contracter la courbe fermée conjointement avec la quantité de marge de gain et de marge de phase modifiée par l'unité de modification. Le dispositif de soutien de réglage peut comprendre en outre une unité de réglage de marge de stabilité de référence destinée à régler des marges de stabilité de référence, c'est-à-dire la marge de gain et la marge de phase, et à délivrer les marges de stabilité de référence réglées à l'unité de dessin de courbe fermée.
PCT/JP2021/039045 2021-10-22 2021-10-22 Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage WO2023067787A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2023554199A JPWO2023067787A1 (fr) 2021-10-22 2021-10-22
PCT/JP2021/039045 WO2023067787A1 (fr) 2021-10-22 2021-10-22 Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/039045 WO2023067787A1 (fr) 2021-10-22 2021-10-22 Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage

Publications (1)

Publication Number Publication Date
WO2023067787A1 true WO2023067787A1 (fr) 2023-04-27

Family

ID=86058031

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/039045 WO2023067787A1 (fr) 2021-10-22 2021-10-22 Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage

Country Status (2)

Country Link
JP (1) JPWO2023067787A1 (fr)
WO (1) WO2023067787A1 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019168777A (ja) * 2018-03-22 2019-10-03 三菱重工業株式会社 制御系設計支援装置、制御系設計支援方法、及び制御系設計支援プログラム
WO2020153488A1 (fr) * 2019-01-24 2020-07-30 オムロン株式会社 Dispositif d'aide à la conception, procédé d'aide à la conception et programme d'aide à la conception

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019168777A (ja) * 2018-03-22 2019-10-03 三菱重工業株式会社 制御系設計支援装置、制御系設計支援方法、及び制御系設計支援プログラム
WO2020153488A1 (fr) * 2019-01-24 2020-07-30 オムロン株式会社 Dispositif d'aide à la conception, procédé d'aide à la conception et programme d'aide à la conception

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MAEDA, YOSHIHIRO: "Using FB compensator design based on circle conditions to improve disturbance compensation", PROCEEDINGS OF THE 2012 ANNUAL MEETING OF THE IEEJ, 5 March 2012 (2012-03-05) *
NAKAMURA, KAZUKI.: "A Study on Controller Design Method to Achieve High Control Bandwidth Using Frequency Response Data Sets", PROCEEDINGS OF THE 59TH SCI MEETING, 20 May 2015 (2015-05-20) *

Also Published As

Publication number Publication date
JPWO2023067787A1 (fr) 2023-04-27

Similar Documents

Publication Publication Date Title
CN111103849B (zh) 输出装置、控制装置、以及学习参数的输出方法
JP6538766B2 (ja) 機械学習装置、サーボモータ制御装置、サーボモータ制御システム、及び機械学習方法
US11592789B2 (en) Output device, control device, and method for outputting evaluation functions and machine learning results
JP6748135B2 (ja) 機械学習装置、サーボ制御装置、サーボ制御システム、及び機械学習方法
JP6474456B2 (ja) 機械学習装置、サーボ制御システム及び機械学習方法
US11256220B2 (en) Machine learning device, control device and machine learning method
JP6901450B2 (ja) 機械学習装置、制御装置及び機械学習方法
US11914334B2 (en) Output device, control device and method of outputting evaluation function value
JP2019008472A (ja) 機械学習装置、サーボ制御装置、サーボ制御システム、及び機械学習方法
US11029650B2 (en) Machine learning device, control system, and machine learning method
US20200005499A1 (en) Output device, control device, and evaluation function value output method
JP2020119295A (ja) フィルタの係数を最適化する機械学習システム、制御装置及び機械学習方法
CN111552237B (zh) 机器学习装置、控制装置、以及机器学习的搜索范围的设定方法
US11243501B2 (en) Machine learning device, control system, and machine learning
JP7469476B2 (ja) 制御支援装置、制御システム及び制御支援方法
WO2023067787A1 (fr) Dispositif de soutien de réglage de marge de stabilité, système de commande et procédé de soutien de réglage
JP7381751B2 (ja) 制御支援装置、制御装置及び制御支援方法
JP7022261B1 (ja) 周波数特性予測装置及び周波数特性予測方法
JP7509866B2 (ja) 機械学習装置、制御装置及び機械学習方法
JP6740290B2 (ja) 機械学習装置、制御装置、及び機械学習方法
WO2022065190A1 (fr) Dispositif d'aide à la commande, système de commande et procédé de réglage de filtres
WO2023181418A1 (fr) Dispositif de régulation pour réguler un paramètre de commande, système de commande et procédé de régulation de paramètre de commande

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21961436

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2023554199

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE