Output device, output system, and output method

Abstract

In the present invention, forms of changes in a characteristic and an evaluation index of an entire system of a servo control device as a result of adjustment of the gain of the servo control device or a parameter of a filter or the like are obtained through simulation.　The present invention provides an output device provided to a servo control device that controls a motor for driving a shaft of a machine tool, a robot, or an industrial machine, the output device comprising: a parameter adjustment unit that uses a frequency characteristic of an entire system of the servo control device before adjustment and a control parameter of the servo control device during the adjustment to update the control parameter; a frequency characteristic calculation unit that uses the updated control parameter to calculate the frequency characteristic of the entire system of the servo control device during the adjustment; an evaluation index calculation unit that calculates an index for evaluating the frequency characteristic of the entire system of the servo control device during the adjustment; and an output unit that outputs the frequency characteristic and the evaluation index during the adjustment.

Description

Output device, output system, and output method

[0001] The present disclosure relates to an output device provided for a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine, an output system including the output device, and an output method.

[0002] Patent document 1 describes a technology that provides a user with the process of adjusting parameters such as gain or filter during or after machine learning when using machine learning to adjust the gain or filter parameters of a servo control device that controls the motor of a machine tool, robot, or industrial machine.

[0003] Patent Literature 1 describes an output device that acquires parameters during or after machine learning and converts the parameters into user-friendly information for output. Specifically, Patent Literature 1 describes an output device that includes: an information acquisition unit that acquires, from a machine learning device that performs machine learning on a servo control device that controls a servo motor that drives an axis of a machine tool, robot, or industrial machine, parameters or a first physical quantity of a component of the servo control device that is currently being learned or has been learned; and an output unit that outputs at least one of the acquired first physical quantity and a second physical quantity calculated from the acquired parameters, the time response characteristics of the component of the servo control device, and the frequency response characteristics of the component of the servo control device, where the time response characteristics and the frequency response characteristics are calculated using the parameters, the first physical quantity, or the second physical quantity. The physical quantity is, for example, the center frequency, bandwidth, and attenuation coefficient of a filter.

[0004] Japanese Patent Application Laid-Open No. 2020-67874

[0005] A user may want to know how the frequency characteristics and evaluation index of the entire system of a servo control device change when a control parameter such as the gain or filter of the servo control device is adjusted. However, it takes a lot of time to repeat the process of adjusting the control parameters such as the gain and filter and actually determining the frequency characteristics. Therefore, when adjusting the control parameters such as the gain and filter, it is desirable to use a simulation to determine how the frequency characteristics and evaluation index of the entire system of the servo control device change.

[0006] A first representative aspect of the present disclosure is an output device provided for a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine, the output device comprising: a frequency characteristic storage unit that stores frequency characteristics of the entire system of the servo control device before adjustment; a parameter storage unit that stores control parameters before adjustment; a parameter adjustment unit that adjusts the control parameters using the frequency characteristics of the entire system of the servo control device before adjustment and the control parameters before adjustment; a frequency characteristic calculation unit that calculates frequency characteristics of the entire system of the servo control device being adjusted using the adjusted control parameters; an evaluation index calculation unit that calculates an index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted; and an output unit that outputs the frequency characteristics of the entire system of the servo control device being adjusted and the evaluation index.

[0007] A representative second aspect of the present disclosure is an output system including: the output device of the first aspect described above; a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine; and a frequency characteristic measuring device that measures the frequency characteristics of the entire system of the servo control device before adjustment.

[0008] A third representative aspect of the present disclosure is an output method in which a computer serving as an output device provided for a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine executes the following processes: a process of saving the frequency characteristics of the entire system of the servo control device before adjustment; a process of adjusting the control parameters using the frequency characteristics of the entire system of the servo control device before adjustment and the control parameters of the servo control device before adjustment; a process of calculating the frequency characteristics of the entire system of the servo control device being adjusted using the adjusted control parameters; a process of calculating an index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted; and a process of outputting the frequency characteristics of the entire system of the servo control device and the evaluation index during the adjustment.

[0009] 7 is a block diagram showing the configuration of an output system including an output device according to an embodiment of the present disclosure. FIG. 1 is a block diagram showing the configuration of a servo control device. FIG. 2 is a block diagram showing a filter configured with a plurality of notch filters connected in series. FIG. 3 is a diagram showing a display screen of a liquid crystal display device that displays frequency characteristics, control parameters, and evaluation indices. FIG. 4 is a diagram showing an example of a display of frequency characteristics during adjustment that are judged as "NG" and are displayed in the display area of ​​the display screen. FIG. 5 is a diagram showing an example of a display of control parameters and evaluation indices before and during adjustment that are judged as "OK" and are displayed in the display area of ​​the display screen. FIG. 6 is a diagram showing an example of a display of control parameters and evaluation indices before and during adjustment that are judged as "OK" and are displayed in the display area of ​​the display screen. FIG. 7 is a diagram showing a configuration of a mechanical model of a two-inertia system. FIG. 8 is a block diagram showing the configuration of a machine learning device. FIG. 9 is a diagram showing a reference model of a servo control device having ideal characteristics without resonance. FIG. 10 is a characteristic diagram showing the frequency characteristics of input / output gains of a reference model servo control device and a servo control device before and after learning. FIG. 11 is a Bode diagram showing an example of frequency characteristics. FIG. 12 is a flowchart showing the operation of an output device. FIG. 13 is a diagram showing the display screen of a liquid crystal display device that adds the frequency characteristics of a filter to the display screen shown in FIG. 1 is a diagram showing the frequency characteristics of a filter before adjusting the filter coefficient; FIG. 2 is a diagram showing the frequency characteristics of a servo control device before adjusting the filter coefficient; FIG. 3 is a diagram showing the frequency characteristics of a filter while adjusting the coefficient of a first-stage filter; FIG. 4 is a diagram showing the frequency characteristics of a servo control device while adjusting the coefficient of a first-stage filter; FIG. 5 is a diagram showing the frequency characteristics of a filter while adjusting the coefficient of a second-stage filter; FIG. 6 is a diagram showing the frequency characteristics of a filter while adjusting the coefficients of the first and second-stage filters; FIG. 7 is a diagram showing the frequency characteristics of a servo control device while adjusting the coefficients of the first and second-stage filters.

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration of an output system including an output device according to an embodiment of the present disclosure. As shown in FIG. 1, the output system 10 includes an output device 100 that outputs the frequency characteristics and an evaluation index of the entire system of the servo control device 200, the servo control device 200, and a frequency characteristic measurement device 300 that measures the frequency characteristics of the entire system of the servo control device 200 based on input and output signals of the servo control device 200. The frequency characteristic measurement device 300 may be included in the output device 10. The output device 100 may be included in the servo control device 200. The output device 100 uses the frequency characteristics of the entire system output from the frequency characteristic measurement device 300 to adjust control parameters of the components of the servo control device 200, such as at least one of a gain parameter and a filter parameter, and outputs the frequency characteristics and an evaluation index of the entire system of the servo control device 200 being adjusted. Prior to describing the output device 100, the servo control device 200 and the frequency characteristic measurement device will be described.

[0011] (Servo Control Device 200) Fig. 2 is a block diagram showing the configuration of the servo control device. As shown in Fig. 2, the servo control device 200 includes a subtractor 201, a controller 202, a filter 203, and a controlled object 204.

[0012] The subtractor 201 calculates the difference between the input control command and the feedback value from the controlled object 204, and outputs the difference as a deviation to the controller 202. For example, the control command is a speed command, the feedback value is a speed feedback value from a motor included in the controlled object 204, and the deviation is a speed deviation.

[0013] A sine wave signal with a changed frequency is input to the subtractor 201 as a control command. The sine wave signal with a changed frequency is input from a higher-level device, but the servo control device 200 may also include a frequency generation unit that generates a sine wave signal with a changed frequency. The control command and feedback value are input to the frequency characteristic measurement unit 300. The controller 202 applies an integral gain k ｉ The integral of the deviation is multiplied by the proportional gain k ｐThe controller 202 adds the value obtained by multiplying the value by a gain k and outputs the result as a torque command to the filter 203. The controller 202 is, for example, a speed controller.

[0014] For example, multiple notch filters are used as the filter 203. A machine tool, robot, or industrial machine driven by a motor has multiple resonance points, and resonance may increase in the servo control device 200. Using multiple notch filters can reduce the multiple resonances. A torque command is output from the filter 203 to the control target 204.

[0015] 3 is a block diagram showing a filter configured by connecting multiple notch filters in series. In FIG. 3, when there are k resonance points (k is a natural number of 2 or greater), filter 203 is configured by connecting m filters 203-1 to 203-m in series (m is a natural number of 2 or greater, m<k). Each of m filters 203-1 to 203-m corresponds to a different frequency band. In the following description, filter 203 is assumed to be configured by three filters 203-1, 203-2, and 203-3 connected in series.

[0016] The controlled object 204 is a motor, or a mechanical part of a machine tool, robot, or industrial machine. The motor is driven based on a torque command. If the feedback value from the controlled object 204 is velocity feedback from the motor, the feedback value can be a velocity detection value obtained by integrating a rotational angular velocity position detected by a rotary encoder (not shown) provided on the motor.

[0017] (Frequency Response Measurement Apparatus) The frequency response measurement apparatus 300 uses a control command (sine wave) that serves as an input signal and a feedback (sine wave) that serves as an output signal to determine the frequency characteristics of the amplitude ratio (input / output gain) between the input signal and the output signal and the phase delay for each frequency defined by the control command, and outputs the determined frequency characteristics to the output apparatus 100. The determined frequency characteristics are closed-loop frequency characteristics Pc. The frequency response measurement apparatus 300 also calculates open-loop frequency characteristics Po from the frequency characteristics Pc, and outputs the closed-loop frequency characteristics Pc and the open-loop frequency characteristics Po to the output apparatus 100. The open-loop frequency characteristics Po can be determined using the closed-loop frequency characteristics Pc by Po = Pc / (1 - Pc).

[0018] The above has described the servo control device 200 and the frequency characteristic measurement device 300. Next, the output device 100 will be described.

[0019] 1, the output device 100 includes an information acquisition unit 101, an information output unit 102, a frequency characteristic storage unit 103, a parameter storage unit 104, a control unit 105, a parameter adjustment unit 106, a frequency characteristic calculation unit 107, an evaluation index calculation unit 108, and an output unit 109. When the control parameters of the servo control device 200 are not changed, the information output unit 102 does not need to be provided.

[0020] The information acquisition unit 101 acquires the closed-loop frequency characteristic Pc and the open-loop frequency characteristic Po from the frequency characteristic measurement device 300. The control unit 105 stores the acquired closed-loop frequency characteristic Pc and open-loop frequency characteristic Po in the frequency characteristic storage unit 103.

[0021] The information output unit 102 sets the optimum control parameters obtained by the parameter adjustment unit 106 in the servo control device 200. The control parameters are the integral gain k ｉ , proportional gain k ｐ , gain k, and the attenuation bandwidth d and center frequency ω of the transfer function of the filter 203. ｎ , and attenuation damping ratio ζ.

[0022] The frequency characteristic storage unit 103 stores the closed-loop frequency characteristic Pc and the open-loop frequency characteristic Po. The parameter storage unit 104 stores the control parameters of the servo control device 200 before adjustment, i.e., the integral gain k of the controller 202. ｉ , proportional gain k ｐ , gain k, and the attenuation bandwidth d and center frequency ω of the transfer function of the filter 203. ｎ , and the attenuation damping rate ζ.

[0023] The control unit 105 controls the information acquisition unit 101 , the information output unit 102 , the frequency characteristic storage unit 103 , the parameter storage unit 104 , the parameter adjustment unit 106 , the frequency characteristic calculation unit 107 , the evaluation index calculation unit 108 , and the output unit 109 .

[0024] The parameter adjustment unit 106 reads the open-loop frequency characteristic Po from the frequency characteristic storage unit 103 via the control unit 105. The parameter adjustment unit 106 also reads the control parameters before adjustment from the parameter storage unit 104 via the control unit 105. The parameter adjustment unit 106 then adjusts the control parameters based on the read-out open-loop frequency characteristic Po and the read-out control parameters, and outputs the control parameters being adjusted to the frequency characteristic calculation unit 107. The parameter adjustment unit 106 also adjusts the control parameters again based on the frequency characteristics calculated by the frequency characteristic calculation unit 107, and outputs the read-adjusted control parameters to the frequency characteristic calculation unit 107.

[0025] The parameter adjusting unit 106 can determine whether or not to adjust the control parameters again, for example, based on whether or not the resonance of the frequency characteristics calculated by the frequency characteristics calculating unit 107 has been suppressed.

[0026] The frequency characteristic calculation unit 107 calculates the open-loop and closed-loop frequency characteristics of the servo control device 200 by simulation using the control parameters adjusted by the parameter adjustment unit 106, the control parameters before adjustment read from the parameter storage unit 104 via the control unit 105, and the open-loop frequency characteristic Po read from the frequency characteristic storage unit 103 via the control unit 105, and outputs the calculated open-loop frequency characteristic or the calculated open-loop and closed-loop frequency characteristics to the parameter adjustment unit 106. The frequency characteristic calculation unit 107 performs a similar operation when the control parameters are adjusted again. If the frequency characteristic calculation unit 107 stores the control parameters before adjustment and the open-loop frequency characteristic Po, it is not necessary to read out the control parameters before adjustment and the open-loop frequency characteristic Po via the control unit 105. A method for calculating the frequency characteristic of the servo control device 200 by simulation using the adjusted control parameters will be described later.

[0027] The control parameters are optimized by repeating the adjustment of the control parameters and the calculation of the frequency characteristics between the parameter adjustment unit 106 and the frequency characteristic calculation unit 107. The parameter adjustment unit 106 outputs the optimized control parameters to the information output unit 102.

[0028] The evaluation index calculation unit 108 calculates an evaluation index based on the open-loop and closed-loop frequency characteristics calculated by the frequency characteristic calculation unit 107. The evaluation index is, for example, at least one of a gain margin, a phase margin, and a high-frequency maximum gain calculated based on the open-loop frequency characteristics, and a closed-loop maximum gain calculated based on the closed-loop frequency characteristics.

[0029] The gain margin indicates how much margin there is until the gain reaches 1 (0 dB) when the phase delay of the loop gain is 180 degrees. The phase margin indicates how much the phase difference is above 0 degrees (180 degrees in phase amount) at the frequency where the gain is 0 dB (i.e., it indicates the phase difference).

[0030] The closed loop maximum gain indicates the maximum value of the gain characteristic of the closed loop frequency response. The high frequency region maximum gain indicates the maximum value of the gain characteristic of the closed loop frequency response at frequencies of 1000 Hz or higher. The control band (response band) indicates the frequency at which the phase delay in the closed loop response is 90°.

[0031] The output unit 109 includes a display unit such as a liquid crystal display device, a printer, etc. The output unit 109 outputs the frequency characteristics calculated by the frequency characteristics calculation unit 107, the control parameters being adjusted, and the evaluation index calculated by the evaluation index calculation unit 108.

[0032] 1, the output unit 109 includes a drawing unit 1091 and a display unit 1092 when the frequency characteristics and the like are to be displayed on a display unit such as a liquid crystal display device. The drawing unit 1091 creates a gain diagram and a phase diagram of the frequency characteristics and displays them in a display area 110A of the display screen of the display unit 1092 in Fig. 4 (to be described later). The drawing unit 1091 also creates a table of the control parameters and evaluation indexes and displays them in a display area 110B of the display screen of the display unit 1092 in Fig. 4 (to be described later).

[0033] The output unit 109 may store multiple frequency characteristics, control parameters, and evaluation indexes before and during adjustment, and output multiple frequency characteristics, control parameters, and evaluation indexes. The drawing unit 1091 can display multiple frequency characteristics, control parameters, and evaluation indexes side by side on a single display screen on the display unit 1092. When displaying multiple frequency characteristics, the drawing unit 1091 can overlay the changing frequency characteristics on the display unit 1092 by changing the display method, such as color, line type (solid line, dotted line, chain line, etc.), and line thickness. When displaying multiple frequency characteristics, the drawing unit 1091 can display the changing frequency characteristics as a video on the display unit 1092. The drawing unit 1091 can create a diagram or video overlaid on the display screen showing the progression of multiple frequency characteristics during adjustment, and display it on the display unit 1092.

[0034] 4 is a diagram showing a display screen of a liquid crystal display device that displays frequency characteristics, control parameters, and evaluation indexes. The display screen 110 includes a display area 110A for displaying frequency characteristics, and a display area 110B.

[0035] 5 is a diagram showing frequency characteristics during adjustment that are judged as "NG" and are displayed in the display area 110A. The solid curves in the gain diagram and phase diagram in FIG. 5 indicate the frequency characteristics of the open loop, and the dotted curves indicate the frequency characteristics of the closed loop. The dashed-dotted line in FIG. 5 indicates the maximum value of the gain of the closed loop frequency characteristics. f1, f2, and f3 shown by the two-dot chain lines in FIG. 5 are the center frequencies ω of the filters 203-1, 203-2, and 203-3. ｎ Indicates the location of.

[0036] FIG. 6 shows the control parameters and evaluation indexes displayed in the display area 110B before and during adjustment. The control parameters and evaluation indexes "before adjustment" in FIG. 6 are evaluation indexes calculated using the control parameters stored in the parameter storage unit 104 and the open-loop frequency characteristic Po calculated by the frequency characteristic measurement device 300. The control parameters and evaluation indexes "during adjustment" in FIG. 6 are evaluation indexes calculated by the evaluation index calculation unit 108 using the control parameters adjusted by the parameter adjustment unit 106 and the open-loop and closed-loop frequency characteristics calculated using the adjusted control parameters. The open-loop and closed-loop frequency characteristics used to calculate the evaluation indexes in FIG. 6 correspond to the open-loop and closed-loop frequency characteristics shown by the solid and dotted lines in FIG. 5. The values ​​in brackets [ ] in the "before adjustment" evaluation index column in FIG. 6 are the required stability indexes set in advance before adjustment. The frequency characteristics in Fig. 5 are rated "NG" because the values ​​of the gain margin, phase margin, closed-loop maximum gain, and high-frequency region maximum gain, which are the stability indexes during adjustment in Fig. 6, are "-3," "-30," "6," and "-16," respectively, and do not satisfy the required stability indexes of "6," "30," "5," and "-20," respectively. In Fig. 6, the control band is unstable and is therefore not shown.

[0037] Fig. 7 is a diagram showing frequency characteristics during adjustment that are judged to be "OK" and are displayed in display area 110A. Fig. 8 is a diagram showing control parameters and evaluation indexes before and during adjustment and that are displayed in display area 110B. The explanation of the frequency characteristics in Fig. 7 and the control parameters and evaluation indexes before and during adjustment in Fig. 8 will be omitted to the extent that it overlaps with the explanation of the frequency characteristics in Fig. 5 and the control parameters and evaluation indexes before and during adjustment in Fig. 6.

[0038] The frequency characteristics in Figure 7 are rated "OK" because the values ​​of the gain margin, phase margin, closed-loop maximum gain, and high-frequency region maximum gain, which are the stability indexes during adjustment in Figure 8, are "8," "50," "3," and "-18," respectively, and the values ​​of the gain margin, phase margin, closed-loop maximum gain, high-frequency region maximum gain, gain margin, and phase margin are "6," "30," "5," and "-20," respectively.

[0039] (Method of determining frequency characteristics of servo control device by simulation) At the time of the first adjustment, the parameter adjustment unit 106 reads out the open-loop frequency characteristic Po from the frequency characteristic storage unit 103 via the control unit 105. The parameter adjustment unit 106 also reads out the control parameters before adjustment from the parameter storage unit 104 via the control unit 105. The parameter adjustment unit 106 then adjusts the control parameters based on the read-out open-loop frequency characteristic Po and the read-out control parameters, and outputs the control parameters under adjustment to the frequency characteristic calculation unit 107. At the time of the second or subsequent adjustment, the parameter adjustment unit 106 readjusts the control parameters based on the frequency characteristics output from the frequency characteristic calculation unit 107, and outputs the control parameters under readjustment to the frequency characteristic calculation unit 107.

[0040] During the initial adjustment, the frequency characteristic calculation unit 107 calculates the frequency characteristic CF of the open loop from the controller 202 to the filter 203 of the servo control device 200 using the transfer function of the servo control device 200 that uses the control parameters before adjustment. １Furthermore, the frequency characteristic calculation unit 107 calculates the frequency characteristic CF of the open loop from the controller 202 to the filter 203 of the servo control device 200 using the transfer function from the controller 202 to the filter 203 of the servo control device 200, which is obtained by using the adjusted control parameters output from the parameter adjustment unit 106. ２ The frequency characteristic calculation unit 107 calculates the frequency characteristic CF １ has already been calculated, the open-loop frequency characteristic CF from the controller 202 to the filter 203 of the servo control device 200 is calculated using the transfer function from the controller 202 to the filter 203 of the servo control device 200 using the adjusted control parameters output from the parameter adjustment unit 106. ２ Calculate only.

[0041] The frequency characteristic calculation unit 107 calculates the frequency characteristic CF １ , frequency characteristic CF ２ Based on the open loop frequency characteristic Po acquired from the frequency characteristic storage unit 103, the open loop frequency characteristic Eo of the input / output gain and phase delay of the motor control unit 100 is calculated using the following equation 1 (shown as equation 1). The frequency characteristic calculation unit 107 calculates the closed loop frequency characteristic Ec using the open loop frequency characteristic Eo by Ec=Eo / (1+Eo). The frequency characteristic Eo of the input / output gain and phase delay of the motor control unit 100 is calculated by the above formula 5, that is, Eo=CF ２ -CF １ +Po, but the calculation performed by the frequency characteristic prediction unit 403 is Eo=(CF ２ -CF １ )+Po,Eo=(Po-CF １ ) +CF ２ , E = (Po + CF ２ )-CF １ Either of the above is acceptable.

[0042] The gain characteristic and phase characteristic of the general open loop frequency characteristic C of the servo control device 200 are expressed by the following equation 2 (hereinafter, equation 2). The open loop transfer function of the servo control device 200 is indicated by Go(s). In equation 2, L Ｏ(ω) is the gain characteristic of the open loop frequency response, Φ Ｏ (ω) indicates the phase characteristic of the open loop frequency characteristic.

[0043] The open loop transfer function Go(s) of the servo control device 200 is given by the formula Go(s)=C(s)F(s)P(s), where C(s) is the transfer function of the controller, F(s) is the transfer function of the filter, and P(s) is the transfer function of the controlled object. The closed loop transfer function Gc(s) of the servo control device 200 is given by the formula Go(s) / (1+Go(s)). The gain characteristic L of the frequency characteristic of the closed loop Ｃ (ω), the phase characteristic of the closed loop frequency response Φ Ｃ (ω) is L in Equation 2 Ｏ (ω) and Go(jω) Ｃ (ω) and G Ｃ (jω) and Φ Ｏ (ω) and Go(jω) as Φ Ｃ (ω) and G Ｃ (jω) to obtain the value.

[0044] The frequency characteristic CF of the open loop from the controller 202 to the filter 203 of the servo control device 200 １ is the frequency characteristic when at least one control parameter of the transfer function C(s) of the controller and the transfer function F(s) of the filter is the control parameter before adjustment. ２ is the frequency characteristic when at least one of the control parameters, the transfer function of the controller C(s) and the transfer function of the filter F(s), is the control parameter being adjusted. １ and frequency characteristics CF ２ The difference is the integral gain k of the transfer function C(s), which will be described later. ｉ , proportional gain k ｐ and gain k, as well as the attenuation bandwidth d and center frequency ω of the transfer function F(s). ｎ and at least one of the damping rate and the damping coefficient are different.

[0045] The transfer function C(s) of the controller, the transfer function F(s) of the filter, and the transfer function P(s) of the controlled object are respectively expressed as follows. The transfer function C(s) of the controller 202 is expressed by the following Equation 3 (Equation 3 below). In Equation 3, k ｉ is the integral gain, k ｐ denotes the proportional gain, and k denotes the gain.

[0046] When a filter is composed of three filters, the transfer function F(s) of the filter is F(s) = F １ (s)F ２ (s)F ３ (s) is calculated as follows. １ (s), F ２ (s) and F ３ (s) shows the transfer functions of the three filters 203-1 to 203-3. The transfer function F of the notch filter as the filter 203-1 is １ (s) is expressed by the following equation 4 (the following equation 4). The transfer functions of the filters 203-2 and 203-3 are also expressed by the following equation 4. Here, the coefficient d in equation 4 is the attenuation bandwidth, and the coefficient ω ｎ is the central angular frequency, and the coefficient ζ is the damping rate.

[0047] When a two-inertia mechanical model as shown in FIG. 9 is used, the transfer function P(s) of the controlled object 204 is expressed as a transfer function from the motor position to the load position by Equation 5 (hereinafter referred to as Equation 6). In Equation 5, J Ｌ is the load inertia, K ｍ is the spring constant, C ｍ indicates the damper constant. The two-inertia mechanical model shown in FIG. 9 is described, for example, in "Research on Low-Frequency Vibration Suppression Control Using a Two-Inertia System Model for the Feed Axis of an NC Machine Tool," Journal of the Japan Society for Precision Engineering, Vol. 82, No. 8, pp. 745-750, 2016.

[0048] (Modification in which the parameter adjustment unit is configured with a machine learning device) The parameter adjustment unit 106 is configured with a machine learning device, and machine learning (hereinafter referred to as learning) can be used when adjusting the control parameters based on the frequency characteristics calculated by the frequency characteristic calculation unit 107. Learning by the machine learning device is performed before shipment, but re-learning may be performed after shipment. The machine learning device repeatedly adjusts the control parameters to find optimal values ​​for the control parameters.

[0049] The configuration and operation of the machine learning device will be described in further detail below. Prior to describing each functional block included in the machine learning device, the basic mechanism of reinforcement learning will first be described. An agent (corresponding to the machine learning device in this embodiment) observes the state of the environment, selects an action, and the environment changes based on that action. As the environment changes, some kind of reward is given, and the agent learns to select a better action (decision-making). While supervised learning indicates a completely correct answer, the reward in reinforcement learning is often a fragmented value based on partial changes in the environment. For this reason, the agent learns to select an action that maximizes the total reward over the future.

[0050] In this way, reinforcement learning learns appropriate actions based on the interaction of the actions with the environment, i.e., it learns a learning method to maximize future rewards. In this embodiment, this means that it is possible to acquire actions that will have an impact on the future, such as selecting action information to suppress vibrations at the machine end.

[0051] Any learning method can be used for reinforcement learning, but the following explanation will be given taking as an example the case of using Q-learning, which is a method of learning the value Q(S, A) of selecting action A in a certain environmental state S. The purpose of Q-learning is to select, as the optimal action, the action A with the highest value Q(S, A) from among the actions A that can be taken in a certain state S.

[0052] However, when Q-learning first begins, the correct value Q(S, A) for a combination of state S and action A is not known at all. Therefore, the agent selects various actions A in a certain state S, and learns the correct value Q(S, A) by selecting the better action based on the reward given for each action A at that time.

[0053] Furthermore, since we want to maximize the total rewards that will be obtained in the future, we aim to ultimately achieve Q(S, A) = E[Σ(γt)rt]. Here, E[ ] represents the expected value, t is time, γ is a parameter called the discount rate (described later), rt is the reward at time t, and Σ is the total at time t. The expected value in this equation is the expected value when the state changes according to the optimal action. However, since it is unknown what the optimal action is in the Q-learning process, reinforcement learning is performed while exploring by performing various actions. The update equation for such value Q(S, A) can be expressed, for example, by the following equation 6 (hereinafter referred to as equation 6).

[0054]

[0055] In the above equation 6, St represents the state of the environment at time t, and At represents the action at time t. Action At changes the state to St+1. rt+1 represents the reward obtained by that change in state. Furthermore, the term with max is the Q value multiplied by γ when action A with the highest Q value known at that time is selected under state St+1. Here, γ is a parameter with a range of 0<γ≦1 and is called the discount rate. Furthermore, α is a learning coefficient with a range of 0<α≦1.

[0056] The above-mentioned formula 6 represents a method for updating the value Q(St, At) of an action At in state St based on the reward rt+1 returned as a result of trial At. This update formula indicates that if the value Q(St, At) of the action At in state St, the value maxa Q(St+1, A) of the best action in the next state St+1 resulting from the action At, is greater than the value Q(St, At), Q(St, At) is increased; conversely, if it is smaller, Q(St, At) is decreased. In other words, the value of a certain action in a certain state is brought closer to the value of the best action in the next state resulting from that action. However, while this difference depends on the discount rate γ and the reward rt+1, essentially, the value of the best action in a certain state is propagated to the value of the action in the state immediately prior to that state.

[0057] Here, in Q-learning, there is a method of creating a table of Q(S, A) for all state-action pairs (S, A) and then performing learning. However, there are cases where the number of states is too large to calculate the Q(S, A) values ​​for all state-action pairs, and it takes a long time for Q-learning to converge.

[0058] Therefore, a well-known technology called DQN (Deep Q-Network) may be used. Specifically, the value function Q may be configured using an appropriate neural network, and the value Q(S, A) may be calculated by approximating the value function Q with an appropriate neural network by adjusting the parameters of the neural network. By using DQN, it is possible to shorten the time required for Q-learning to converge. Note that DQN is described in detail in, for example, the following non-patent document.

[0059] <Non-patent literature> "Human-level control through deep reinforcement learning", by Volodymyr Mnih1 [online], [searched on January 17, 2017], Internet <URL: http: / / files.davidqiu.com / research / nature14236.pdf>

[0060] The machine learning device performs the Q learning described above. Specifically, the machine learning device takes the control parameters of the servo control device 200 and the frequency characteristic output from the frequency characteristic calculation unit 107 as state S, and learns the value Q of selecting adjustment of the control parameters of the servo control device 200 related to state S as action A. The control parameters of the servo control device 200 are the coefficient k of the transfer function of the controller 202. ｉ , coefficient k ｐ , and coefficient k, as well as each coefficient d and coefficient ω of the transfer function of the filter 203 ｎ , and at least one value of the coefficient ζ.

[0061] The machine learning device observes state information S including frequency characteristic Eo calculated using Equation 1 based on the control parameters of the servo control device 200, and determines action A. The machine learning device receives a reward each time it performs action A. The machine learning device, for example, searches by trial and error for the optimal action A that maximizes the total reward over the future. In this way, the machine learning device can select the optimal action A (i.e., the control parameters of the servo control device 200) for the state S including frequency characteristic Eo calculated based on the control parameters of the servo control device 200.

[0062] In other words, by selecting an action A that maximizes the value of Q from among the actions A that are applied to the control parameters of the servo control device 200 relating to a certain state S based on the value function Q learned by the machine learning device, it is possible to select an action A (i.e., the control parameters of the servo control device 200) that minimizes the vibration of the machine end caused by executing the machining program.

[0063] Fig. 10 is a block diagram showing the configuration of a machine learning device. To perform the above-described reinforcement learning, as shown in Fig. 10, machine learning device 400 includes state information acquisition unit 401, learning unit 402, behavior information output unit 403, value function storage unit 404, and optimized behavior information output unit 405. Learning unit 402 includes reward output unit 4021, value function update unit 4022, and behavior information generation unit 4023.

[0064] The state information acquisition unit 401 acquires a state S including frequency characteristics from the frequency characteristic calculation unit 107 based on the control parameters of the servo control device 200. This state information S corresponds to the environmental state S in Q-learning. The state information acquisition unit 401 outputs the acquired state information S to the learning unit 402.

[0065] The control parameters of the servo control device 200 at the time when Q learning is first started are generated in advance by the user and stored in the parameter storage unit 104. In this embodiment, the control parameters of the servo control device 200 generated by the user are adjusted to optimum values ​​by reinforcement learning. If the operator has adjusted a machine tool or the like in advance, the adjusted values ​​may be used as initial values ​​for machine learning.

[0066] 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.

[0067] The reward output unit 4021 is a unit that calculates a reward when action A is selected under a certain state S. When the control parameters of the servo control device 200 are adjusted, the reward output unit 4021 compares the measured or calculated input-output gain Gs with the input-output gain Gb of a preset reference model for each frequency. The reward output unit 4021 provides a negative reward when the measured or calculated input-output gain Gs is greater than the input-output gain Gb of the reference model. On the other hand, when the measured or calculated input-output gain Gs is equal to or less than the input-output gain Gb of the reference model, the reward output unit 4021 provides a positive reward when the phase lag decreases, a negative reward when the phase lag increases, and a zero reward when the phase lag remains unchanged.

[0068] First, the operation of the reward output unit 4021 to give a negative reward when the measured or calculated input-output gain Gs is greater than the input-output gain Gb of the reference model will be described with reference to Figures 11 and 12. The reward output unit 4021 stores a reference model of the input-output gain. The reference model is a model of a servo control device having ideal characteristics without resonance. The reference model can be calculated, for example, from the inertia Ja, torque constant Kt, proportional gain Kp, integral gain KI, and derivative gain KD of the model shown in Figure 11. The inertia Ja is the sum of the motor inertia and the machine inertia.

[0069] FIG. 12 is a characteristic diagram showing the frequency characteristics of the input / output gain between the servo control device of the reference model and the servo control device 200 before and after learning. As shown in the characteristic diagram of FIG. 12, the reference model has region A, which is a frequency region where the ideal input / output gain is equal to or greater than a certain input / output gain, for example, -20 dB or greater, and region B, which is a frequency region where the ideal input / output gain is less than the certain input / output gain. In region A of FIG. 12, the ideal input / output gain of the reference model is shown by curve MC1 (thick line). In region B of FIG. 12, the ideal virtual input / output gain of the reference model is shown by curve MC11 (thick dashed line), and the input / output gain of the reference model is a constant value, shown by line MC12 (thick line). In regions A and B of FIG. 12, the curves of the input / output gain with the servo control unit before and after learning are shown by curves RC1 and RC2, respectively.

[0070] In region A, the reward output unit 4021 provides a first negative reward when the pre-learning curve RC1 of the measured input / output gain exceeds the ideal input / output gain curve MC1 of the reference model. In region B, which exceeds the frequency at which the input / output gain becomes sufficiently small, the impact on stability is small even if the pre-learning input / output gain curve RC1 exceeds the ideal virtual input / output gain curve MC11 of the reference model. Therefore, as described above, in region B, the input / output gain of the reference model uses the straight line MC12 of a constant input / output gain (e.g., -20 dB) rather than the ideal gain characteristic curve MC11. However, if the pre-learning measured input / output gain curve RC1 exceeds the constant input / output gain curve MC12, there is a possibility of instability, so the first negative value is provided as a reward.

[0071] Next, an operation of the reward output unit 4021 to determine a reward based on information about the phase delay when the measured or calculated input-output gain Gs is equal to or less than the input-output gain Gb of the reference model will be described. In the following description, the phase delay, which is a state variable related to the state information S, will be denoted as D(S), and the phase delay, which is a state variable related to the state S' resulting from a change from the state S due to the action information A (adjustment of the control parameters of the servo control device 200), will be denoted as D(S').

[0072] The reward output unit 4021 may determine a reward based on information about the phase delay, for example, as follows. Note that the method for determining a reward based on information about the phase delay is not particularly limited to the method described below. The reward is determined based on whether the frequency at which the phase delay is 180 degrees increases, decreases, or remains the same when transitioning from state S to state S'. Here, the case where the phase delay is 180 degrees is discussed, but it is not limited to 180 degrees and other values ​​may be used. For example, when the phase delay is shown in the phase diagram of the Bode diagram shown in FIG. 13, if the curve changes so that the frequency at which the phase delay is 180 degrees decreases (toward X2 in FIG. 13) when transitioning from state S to state S', the phase delay increases. On the other hand, if the curve changes so that the frequency at which the phase delay is 180 degrees increases (toward X1 in FIG. 13) when transitioning from state S to state S', the phase delay decreases.

[0073] Therefore, when the frequency at which the phase lag is 180 degrees decreases when changing from state S to state S', the phase lag D(S) is defined as < phase lag D(S'), and the reward output unit 4021 sets the reward value to a second negative value. The absolute value of the second negative value is smaller than the first negative value. On the other hand, when the frequency at which the phase lag is 180 degrees increases when changing from state S to state S', the phase lag D(S) is defined as > phase lag D(S'), and the reward output unit 4021 sets the reward value to a positive value. Furthermore, when the frequency at which the phase lag is 180 degrees does not change when changing from state S to state S', the phase lag D(S) is defined as = phase lag D(S'), and the reward output unit 4021 sets the reward value to zero.

[0074] Note that, when the phase lag D(S') of state S' after action A is performed is defined as being larger than the phase lag D(S) in the previous state S, the negative value may be increased in accordance with the ratio. For example, in the above-described method, the negative value may be increased in accordance with the degree to which the frequency has decreased. Conversely, when the phase lag D(S') of state S' after action A is performed is defined as being smaller than the phase lag D(S) in the previous state S, the positive value may be increased in accordance with the ratio. For example, in the first method described above, the positive value may be increased in accordance with the degree to which the frequency has increased.

[0075] The value function update unit 4022 updates the value function Q stored in the value function storage unit 404 by performing 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 value calculated as described above. The value function Q may be updated by online learning, batch learning, or mini-batch learning. Online learning is a learning method in which the value function Q is updated immediately each time a certain action A is applied to the current state S, causing the state S to transition to a new state S'. Batch learning is a learning method in which learning data is collected by repeatedly applying a certain action A to the current state S, causing the state S to transition to a new state S', and the value function Q is updated using all of the collected learning data. Mini-batch learning is a learning method that is intermediate between online learning and batch learning, in which the value function Q is updated each time a certain amount of learning data is accumulated.

[0076] The behavior information generation unit 4023 selects behavior A in the Q-learning process for the current state S. In order to perform an operation (corresponding to behavior A in Q-learning) to change the control parameters of the servo control device 200 in the Q-learning process, the behavior information generation unit 4023 generates behavior information A and outputs the generated behavior information A to the behavior information output unit 403. More specifically, the behavior information generation unit 4023, for example, incrementally adds or subtracts the control parameters of the servo control device 200 included in behavior A to the control parameters of the servo control device 200 included in state S.

[0077] Then, the behavior information generation unit 4023 may apply an increase or decrease in the control parameters of the servo control device 200 to transition to state S', and if a positive reward (a reward with a positive value) is returned, the behavior information generation unit 4023 may take a measure to select the next behavior A' so that the measured phase delay is smaller than the previous phase delay, such as incrementally adding or subtracting from the control parameters of the servo control device 200 in the same way as the previous action.

[0078] Conversely, if a negative reward (a reward with a negative value) is returned, the behavioral information generation unit 4023 may take measures to select the next behavior A', for example, by incrementally subtracting or adding to the control parameters of the servo control device 200 in the opposite direction to the previous action, so that if the measured or calculated input / output gain is greater than the input / output gain of the reference model, the difference in input gain is smaller than the previous time, or so that the measured phase delay is smaller than the previous phase delay.

[0079] The behavior information output unit 403 is a part that transmits the behavior information A output from the learning unit 402 to the frequency characteristic calculation unit 107. As described above, the frequency characteristic calculation unit 107 changes the current state S, i.e., the currently set control parameters of the servo control device 200, based on this behavior information, thereby transitioning to the next state S' (i.e., the changed control parameters of the servo control device 200).

[0080] 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. The value function Q stored in the value function storage unit 404 is updated by a value function update unit 4022. The value function Q stored in the value function storage unit 404 may also be shared with other machine learning devices 400. If the value function Q is shared by multiple machine learning devices 400, reinforcement learning can be performed in a distributed manner among the machine learning devices 400, thereby improving the efficiency of reinforcement learning.

[0081] The optimization behavior information output unit 405 generates behavior information A (hereinafter referred to as "optimization behavior information") for causing the servo control device 200 to perform an operation that maximizes the value Q(S, A), based on the value function Q updated by the value function update unit 4022 through Q-learning. More specifically, the optimization behavior information output unit 405 acquires the value function Q stored in the value function storage unit 404. This value function Q has been updated by the value function update unit 4022 through Q-learning, as described above. The optimization behavior information output unit 405 then generates behavior information based on the value function Q and outputs the generated behavior information to the servo control device 200. This optimization behavior information includes information for correcting the control parameters of the servo control device 200, similar to the behavior information output by the behavior information output unit 403 during Q-learning.

[0082] In the servo control device 200, the control parameters are corrected based on this behavior information. Through the above operations, the machine learning device 400 can optimize the control parameters of the servo control device 200 and operate to suppress vibrations at the machine end.

[0083] As described above, by using the machine learning device 400, it is possible to simplify the adjustment of the control parameters of the servo control device 200. When the filter 203 is configured by three filters 203-1 to 203-3, the machine learning device 400 adjusts the control parameters (coefficient ξ, coefficient ω ｎ、 For each of the coefficients d, the optimum value for attenuating the resonance point is sequentially determined by machine learning.

[0084] Next, the operation of the output device 100 in this embodiment will be described with reference to the flowchart of Fig. 14. Fig. 14 is a flowchart showing the operation of the output device.

[0085] In step S11 , the control unit 105 stores in the frequency characteristic storage unit 103 the closed-loop frequency characteristics and open-loop frequency characteristics of the input / output gain and the phase delay output from the frequency characteristic measurement device 300 .

[0086] In step S12 , the control unit 105 reads the open-loop frequency characteristics from the frequency characteristic storage unit 103 , and reads the control parameters before adjustment from the parameter storage unit 104 .

[0087] In step S13 , the parameter adjusting unit 106 adjusts the control parameters and outputs the adjusted control parameters to the frequency characteristic calculating unit 107 .

[0088] In step S14, the frequency characteristic calculation unit 107 calculates the open loop frequency characteristic of the servo control device 200 using the adjusted control parameters by simulation, and also calculates the closed loop frequency characteristic.

[0089] In step S15, the evaluation index calculation unit 108 calculates an evaluation index based on both the open-loop frequency characteristics and the closed-loop frequency characteristics.

[0090] In step S16, the output unit 109 displays the frequency characteristics and the evaluation index. In step S17, the parameter adjustment unit 106 determines whether or not to readjust. If readjustment is to be performed, the process returns to step S13. If readjustment is not to be performed, the process ends.

[0091] According to the present embodiment described above, when adjusting the control parameters of a servo control device, it is sufficient to measure the frequency characteristics once, and the frequency characteristics after adjusting the control parameters can be obtained by simulation. According to the present embodiment, by checking a plurality of frequency characteristics and / or a plurality of evaluation indices of the frequency characteristics, it is possible to easily compare the frequency characteristics and / or evaluation indices of the frequency characteristics being adjusted and to determine the control parameters to be applied.

[0092] (First Modification) The output unit 109 may add and display the frequency characteristics of the filters on the display screen 110 shown in Fig. 4. The frequency characteristic calculation unit 107 calculates the frequency characteristics of the filters using the adjusted control parameters and outputs them to the output unit 109. The frequency characteristics of a filter made up of three notch filters are expressed by Equation 7 (Equation 7 below). L F (ω) is the gain characteristic of the filter frequency response, Φ F(ω) indicates the phase characteristic of the frequency characteristic of the filter. 15 is a diagram showing a display screen of a liquid crystal display device that displays the frequency characteristics of a filter in addition to the frequency characteristics, control parameters, and evaluation indexes. The display screen 110 includes a display area 110C in addition to display areas 110A and 110B that display the frequency characteristics. The position of display area 110B in FIG. 15 is different from the position of display area 110B shown in FIG. 4. Display area 110C displays the frequency characteristics of the filter.

[0093] 16 to 23 show examples of the frequency characteristics of the servo control device and the frequency characteristics of the filter before and during adjustment of the filter coefficients. The frequency characteristics of the servo control device are displayed in display area 110A, and the frequency characteristics of the filter are displayed in display area 110C.

[0094] Figures 16 and 17 are diagrams showing the frequency characteristics of the filter and the servo control device before the filter coefficients are adjusted. Figures 18 and 19 are diagrams showing the frequency characteristics of the filter and the servo control device during adjustment of the coefficient of the first-stage filter 203-1. Figures 20 and 21 are diagrams showing the frequency characteristics of the filter and the servo control device during adjustment of the coefficient of the second-stage filter 203-2. Figures 22 and 23 are diagrams showing the frequency characteristics of the filter and the servo control device during adjustment of the coefficients of the first-stage and second-stage filters 203-1 and 203-2.

[0095] 2 is not limited to being composed of a plurality of notch filters, and may be composed of, for example, one notch filter, or may be composed of three filters: a first-order low-pass filter, a second-order low-pass filter, and a notch filter. When the filter 203 is composed of three filters: a first-order low-pass filter, a second-order low-pass filter, and a notch filter, the transfer function F(s) of the filter 203 is expressed as F(s)=F １ＬＰＦ (s)F ２ＬＰＦ (s)F ｎ (s) The transfer function F １ＬＰＦ (s), F ２ＬＰＦ (s) and Fｎ (s) show the transfer functions of a first-order low-pass filter, a second-order low-pass filter, and a notch filter, respectively.

[0096] The output device 100 can be configured in the same manner as the present embodiment already described, except that the three filters 203-1 to 203-3 are replaced from three notch filters to a first-order low-pass filter, a second-order low-pass filter, and a notch filter.

[0097] Transfer function F １ＬＰＦ (s), F ２ＬＰＦ (s) and F ｎ (s) is expressed by the following equation 8 (the following equation 8): In equation 8, the coefficient d is the attenuation bandwidth, the coefficient ω ｎ denotes the angular center frequency, the coefficient ζ denotes the damping rate, and T denotes the time constant.

[0098] The frequency characteristics (gain characteristics and phase characteristics) of a filter composed of a first-order low-pass filter, a second-order low-pass filter, and a notch filter are expressed as F １ (jω), F ２ (jω) and F ３ (jω) to F １ＬＰＦ (jω), F ２ＬＰＦ (jω) and F ｎ The frequency characteristic of the filter is displayed in a display area 110C of the display screen shown in FIG.

[0099] The components included in the output device 100 of each of the above-described embodiments can be realized by hardware, software, or a combination thereof. Here, "realized by software" means that the components are realized by a computer reading and executing a program. To realize the components included in the output device 100 by software or a combination thereof, the output device 100 includes a central processing unit (CPU) or other such processing device. The processing unit functions as an execution unit. The output device 100 also includes a secondary storage device such as a hard disk drive (HDD) that stores various control programs, such as application software or an operating system (OS), and a main storage device such as a random access memory (RAM) that stores data temporarily required for the processing device to execute the programs.

[0100] The output device 100 has an arithmetic processing unit that reads application software or an OS from the auxiliary storage device, and then loads the loaded application software or OS into the main storage device, while performing arithmetic processing based on the application software or OS. Furthermore, based on the results of this arithmetic processing, the output device 100 controls various pieces of hardware. This realizes the functional blocks of this embodiment.

[0101] The components included in the output device 100 can be realized by hardware including electronic circuits, etc. When the output device 100 is configured by hardware, some or all of the functions of the components included in the output device 100 can be configured by integrated circuits (ICs), such as application specific integrated circuits (ASICs), gate arrays, field programmable gate arrays (FPGAs), and complex programmable logic devices (CPLDs).

[0102] 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, programmable ROMs (PROMs), erasable PROMs (EPROMs), flash ROMs, and random access memories (RAMs)). The program may also be supplied to a computer by various types of transitory computer-readable media.

[0103] The effect of the above-described embodiments and modifications is that it is possible to determine by simulation how the characteristics and evaluation indexes of the entire system of the servo control device change by adjusting parameters such as the gain or filter of the servo control device.

[0104] Although the present disclosure has been described above, the present disclosure is not limited to the individual embodiments and modifications described above. Various additions, substitutions, changes, partial deletions, etc. are possible to these embodiments and modifications within the scope of the gist of the present disclosure, or within the scope of the gist of the present disclosure derived from the content of the claims and their equivalents. Furthermore, these embodiments and modifications can also be implemented in combination. For example, in the above-described embodiments, the order of each operation and the order of each process are shown as examples and are not limited to these.

[0105] The following supplementary notes are further disclosed regarding the above-described embodiments and modifications: (Supplementary Note 1) An output device (100) provided in a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine, comprising: a frequency characteristic saving unit (103) that saves frequency characteristics of the entire system of the servo control device before adjustment, a parameter memory unit (104) that stores control parameters before adjustment, a parameter adjusting unit (106) that adjusts the control parameters using the frequency characteristics of the entire system of the servo control device before adjustment and the control parameters before adjustment, a frequency characteristic calculating unit (107) that calculates frequency characteristics of the entire system of the servo control device being adjusted using the adjusted control parameters, an evaluation index calculating unit (108) that calculates an index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted, and an output unit (109) that outputs the frequency characteristics of the entire system of the servo control device being adjusted and the evaluation index.

[0106] (Supplementary Note 2) The output device according to Supplementary Note 1, wherein the frequency characteristic calculation unit (107) calculates the frequency characteristic of a filter included in the servo control device using the adjusted control parameters, and the output unit (109) outputs the frequency characteristic of the filter.

[0107] (Supplementary Note 3) The output device according to Supplementary Note 1 or 2, wherein the output unit (109) includes a display unit (1092) that displays the plurality of frequency characteristics being adjusted and the plurality of evaluation indexes for evaluating the plurality of frequency characteristics, side by side or overlapping, or as a moving image on a display screen.

[0108] (Supplementary Note 4) The output device according to Supplementary Note 3, wherein the output unit includes a drawing unit (1091) that creates a diagram or a video in which transitions of the plurality of frequency characteristics being adjusted are superimposed on the display screen, and the display unit (1092) displays the diagram or the video created by the drawing unit.

[0109] (Supplementary Note 5) The output device according to any one of Supplementary Notes 1 to 4, wherein the parameter adjustment unit (106) is a machine learning device (400) that iteratively adjusts the control parameter to find an optimal value of the control parameter.

[0110] (Supplementary Note 6) The output device according to any one of Supplementary Notes 1 to 5, comprising a frequency characteristic measuring device (300) for measuring the frequency characteristic of the entire system of the servo control device before adjustment.

[0111] (Supplementary Note 7) An output system comprising: an output device (100) according to any one of Supplementary Notes 1 to 6; a servo control device (200) that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine; and a frequency characteristic measuring device (300) that measures the frequency characteristics of the entire system of the servo control device before adjustment.

[0112] (Supplementary Note 8) An output method in which a computer as an output device provided for a servo control device (200) that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine executes the following processes: a process of saving the frequency characteristics of the entire system of the servo control device before adjustment; a process of adjusting the control parameters using the frequency characteristics of the entire system of the servo control device before adjustment and the control parameters of the servo control device before adjustment; a process of calculating the frequency characteristics of the entire system of the servo control device being adjusted using the adjusted control parameters; a process of calculating an index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted; and a process of outputting the frequency characteristics of the entire system of the servo control device and the evaluation index during the adjustment.

[0113] REFERENCE SIGNS LIST 10 Output system 100 Output device 101 Information acquisition unit 102 Information output unit 103 Frequency characteristic storage unit 104 Parameter storage unit 105 Control unit 106 Parameter adjustment unit 107 Frequency characteristic calculation unit 108 Evaluation index calculation unit 109 Output unit 110 Display screen 200 Servo control device 201 Subtractor 202 Controller 203 Filter 204 Control target 300 Frequency characteristic measurement device 400 Machine learning device

Claims

1. An output device provided for a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine, a frequency characteristic storage unit that stores the frequency characteristics of the entire system of the servo control device before adjustment; a parameter storage unit that stores the control parameters before adjustment; a parameter adjusting unit that adjusts the control parameters using a frequency characteristic of the entire system of the servo control device before adjustment and the control parameters before adjustment; a frequency characteristic calculation unit that calculates the frequency characteristic of the entire system of the servo control device being adjusted using the adjusted control parameters; an evaluation index calculation unit that calculates an evaluation index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted; an output unit that outputs the frequency characteristic of the entire system of the servo control device and the evaluation index during the adjustment; An output device comprising:

2. the frequency characteristic calculation unit calculates the frequency characteristic of a filter included in the servo control device using the adjusted control parameters; The output device according to claim 1 , wherein the output section outputs the frequency characteristics of the filter.

3. 2. The output device according to claim 1, wherein the output unit includes a display unit that displays the plurality of frequency characteristics being adjusted and the plurality of evaluation indexes that evaluate the plurality of frequency characteristics, respectively, side by side or overlapping on a display screen, or displays them as a moving image.

4. 4. The output device according to claim 3, wherein the output unit includes a drawing unit that creates a diagram or video overlaid on the display screen showing the progression of the plurality of frequency characteristics being adjusted, and the display unit displays the diagram or video created by the drawing unit.

5. The output device according to claim 1 , wherein the parameter adjustment unit is a machine learning device that repeatedly adjusts the control parameters to find optimal values ​​of the control parameters.

6. 5. The output device according to claim 1, further comprising a frequency characteristic measuring device for measuring the frequency characteristic of the entire system of said servo control device before adjustment.

7. an output device according to any one of claims 1 to 4; a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine; a frequency characteristic measuring device for measuring the frequency characteristics of the entire system of the servo control device before adjustment; An output system comprising:

8. A computer as an output device provided for a servo control device that controls a motor that drives an axis of a machine tool, a robot, or an industrial machine, A process of storing the frequency characteristics of the entire system of the servo control device before adjustment; a process of adjusting the control parameters using a frequency characteristic of the entire system of the servo control device before adjustment and the control parameters of the servo control device before adjustment; a process of calculating a frequency characteristic of the entire system of the servo control device being adjusted using the adjusted control parameters; a process of calculating an evaluation index for evaluating the frequency characteristics of the entire system of the servo control device being adjusted; a process of outputting the frequency characteristics of the entire system of the servo control device and the evaluation index during the adjustment; Execute the output method.