JP4697663B2 - Electric motor control device and filter setting method thereof - Google Patents

Description

  The present invention relates to a motor control device used in a positioning device such as a machine tool or an industrial robot, and more particularly to a motor control device having a filter setting function for suppressing mechanical resonance and a filter setting method thereof.

In an electric motor control device such as a machine tool, since the resonance between the electric motor and the machine affects the control performance, a filter for absorbing the resonance is mounted on the control device side. Various methods for setting this filter have been studied.
In the feed axis parameter adjustment system of the machine tool as the first conventional example, a mathematical model is created and various constants are identified based on the measurement result of the machine movement, and the parameters of the control system are changed based on the actual parameters. It is implemented by simulation (simulation) on a mathematical model, not on a machine. (For example, refer to Patent Document 1).

  Also, in the servo system notch filter automatic adjustment device, which is the second conventional example, first, the test operation is performed with the digital notch filter disconnected, the frequency of the operation signal from the servo system is analyzed, and the notch filter is incorporated. The center frequency is determined, and each coefficient of the digital notch filter is obtained by a coefficient setting unit and substituted into the digital notch filter. (For example, refer to Patent Document 2).

First, the first conventional example will be described.
FIG. 9 shows a block diagram of a feed axis parameter adjusting system of the first conventional example.
Here, the CNC device 101, the servo controller 103, the motor 104, and the machine tool 10
A measuring device 106 for measuring the relative movement of the spindle and the table is attached to a machine tool feed drive system consisting of 5, and further equipped with a parameter adjustment system 100 (hereinafter referred to as an adjustment system).
In the figure, only the feed forward gain and the position loop gain are shown among the parameters to be adjusted. The other three parameters are used for processing inside the CNC 101.
The adjustment system 100 uses the servo command from the CNC and the measurement data from the measuring device as input signals, and changes the CNC parameters and starts operation as output signals.
The adjustment system includes a feedforward processing unit 107, a position control device 108, and a machine model 109 that simulates a machine.

FIG. 10 is a diagram showing an adjustment procedure of the feed axis parameter adjustment system of the first conventional example.
In step S1, a predetermined machining program is executed on the actual machine, and the relative motion between the spindle and the table is measured.
In step S2, the coefficient of the mathematical model (hereinafter, machine model) is identified from the measurement data.
In step S3, simulation (simulation) is performed using this machine model, and acceleration / deceleration parameters are adjusted so that the accuracy falls within an allowable value.
In step S4, the accuracy is confirmed on the actual machine using this parameter.

FIG. 11 is an explanatory diagram of a machine model of the feed axis parameter adjusting system of the first conventional example.
The model A is a mathematical model having two frequencies, and may be a high-order linear model like the model B or a discrete model like the model C.

The setting operation will be described below.
First, in step S1, the changeover switch 102 is set to the contact a, and the machine is operated with a predetermined program. The operating parameters should be within a safe range and the machine will vibrate easily. Here, the motion trajectory of the machine is measured with a measuring instrument and input to the parameter adjustment system.
The parameter adjustment system analyzes the measured data and identifies the parameters of the machine model.
Next, in step S2, the servo switch is given to the adjustment system by setting the changeover switch 102 to the contact point b, and simulation (simulation) is performed in the adjustment system. If the servo command is the same as when the machine model is identified, the same motion trajectory can be obtained.
In step S3, parameters are adjusted using a predetermined algorithm based on this data. For example, the feed forward gain is changed to a certain value, simulated again (simulation), and changed to a more appropriate value compared to before the change. When changing the parameters used in the CNC, after changing the CNC parameters, a servo command is received from the CNC and simulation (simulation) is performed.
In the final step S4, the accuracy is confirmed on the actual machine using this parameter.
Thus, in the first conventional example, each parameter is adjusted while performing simulation (simulation) based on a mathematical model.

Next, a second conventional example will be described.
FIG. 12 is a block diagram of a notch filter automatic adjusting apparatus as a second conventional example.
In FIG. 12, the notch filter automatic adjustment device includes a servo motor 201 that drives a load device such as a machine tool, and a servo amplifier 202 that controls the servo motor 201.
The speed command input unit 203 is configured so that a speed command signal VCM1 from a controller (not shown) or a speed command signal VCM2 from a function generator 204 can be selected by a switch 205.

The deviation between the speed command signal VCM1 or VCM2 and the speed detection signal VTG is input to the speed amplifier 206.
The servo amplifier 202 is provided with a digital notch filter 207 so that the notch filter 207 can be selected by switching the switch 208.
The model of the notch filter 207 is a discrete transfer function obtained by bilinear transformation of the transfer function of the analog notch filter as shown in the figure.
FIG. 13 is a block diagram showing the configuration of the center frequency setting unit and the coefficient setting unit of the notch filter automatic adjustment apparatus as the second conventional example.
The center frequency setting unit 209 includes signal storage means 209a, frequency analysis means 209b, amplitude ratio calculation means 209c, amplitude ratio storage means 209d, maximum amplitude ratio search means 209e, command frequency count means 209f, and center frequency determination means 209g. Is done.
The servo amplifier 202 is provided with a center frequency setting unit 209 and a count generator 210.
The setting operation will be described below.
First, the switch 208 selects the B terminal to select a state in which the notch filter 207 is not incorporated, and the switches 211 and 212 are turned on to make a closed loop configuration, and the center frequency setting operation is performed.
The function generator 204 sequentially outputs speed command signals VCM2 having different frequencies, and the center frequency setting unit 209 receives the speed command signal VCM2 and the speed detection signal VTG from the function generator 204. This is subjected to frequency analysis, and an unnecessary frequency component outside the velocity loop response frequency (cutoff frequency) band is detected and determined as the center frequency fc of the digital notch filter.

  Next, the signal storage means 209a stores the data of the speed detection signal VTG according to a predetermined sampling period. The frequency analysis means 209b performs frequency analysis on this and measures the amplitude value of the speed detection signal VTG for each sampling.

The amplitude ratio calculating means 209c calculates the ratio (amplitude ratio R) between the amplitude of VTG from the frequency analyzing means 209b and the amplitude of the speed command signal VCM2 for each frequency.
The amplitude ratio R and the corresponding frequency component are sequentially stored in the amplitude ratio storage means 209d, and the maximum amplitude ratio search means 209e searches for the maximum amplitude ratio Rc.
The command frequency counting means 209f recognizes the current frequency of the speed command signal VCM2 by counting the number of frequency changes of the speed command signal VCM2.

Next, the center frequency determining means 209g determines whether or not the maximum amplitude ratio Rc is greater than the reference amplitude ratio Rref. If the maximum amplitude ratio Rc is greater than the reference amplitude ratio Rref, the frequency corresponding to the maximum amplitude ratio Rc is determined. The component is determined as the center frequency fc of the notch filter 207, and the information is output to the coefficient setting unit 210. If the maximum amplitude ratio Rc is equal to or less than the reference amplitude ratio Rref, the setting of the coefficient in the coefficient setting unit 210 is stopped.
Once the center frequency fc is determined, the coefficient of the notch filter is determined by determining the sharpness. The sharpness is determined based on the center frequency fc, and the coefficient is determined from the coefficient data table 210b.

Thus, in the second conventional example, the response signal in the closed loop configuration with the notch filter removed is analyzed, and the center frequency and sharpness of the notch filter are set.
Japanese Patent Publication No. 2004-188541 (FIGS. 1, 2, and 3) Japanese Patent No. 2744731 (FIGS. 1 and 2)

The feed axis parameter adjustment system of the first conventional example adjusts parameters by simulation (simulation) with a machine model in the adjustment system based on the initial test result.
However, since measured values cannot be used at the stage of parameter adjustment, there is a problem that simulation (simulation) examination considering a portion that cannot be modeled cannot be performed.
Also, if the portion that could not be modeled has a large effect on the performance of the entire apparatus, the simulation (simulation) study is meaningless.
Further, obtaining the machine model from the response data obtained as a result of operating the apparatus has a problem in that it requires work of a skilled expert and takes time for modeling.

The notch filter automatic adjustment device of the second conventional example determines the center frequency of the notch filter by frequency analysis of the speed detection signal obtained by driving the servo motor using the measurement speed command signal.
However, since there is no means for confirming in advance the condition determined by frequency analysis, there is a problem that the determined condition must be set in an actual machine and evaluated.

  The present invention has been made in view of such problems, and an object of the present invention is to make it possible to efficiently set an optimum filter by utilizing an actual measurement value in a parameter examination stage in filter setting of an electric motor control device. And

In order to solve the above problem, the present invention is configured as follows.
The invention described in claim 1 includes a command device that generates a command signal, a controller that receives the command signal and outputs a drive command, a drive force filter unit that performs a filtering process on the drive command, and the drive force And a current control unit that drives an electric motor connected to the machine using an output of the filter unit as an input, and the driving force filter unit configured to feed back an operation amount of the machine to the controller A filter setting unit that calculates a parameter of the measured value, and the filter setting unit turns off the feedback and measures the input of the current control unit and the operation amount, and the actually measured transfer characteristic of the mechanical system including the motor and the machine A mechanical characteristic calculation unit that calculates a parameter, a filter constant estimation unit that estimates a parameter to be set in the driving force filter unit from the mechanical system actually measured transmission characteristic, and the drive Using a driving force filter unit model in which the filter unit is numerically modeled and the parameters are applied, and a measured transfer characteristic of the mechanical system, a simulation characteristic calculation unit that simulates a combined operation characteristic, and an evaluation by the simulation characteristic calculation unit And an actual machine characteristic confirmation unit that obtains a combined operation characteristic using an actual machine.

Further, in the invention according to claim 2, the filter constant estimation unit selects a candidate from the resonance frequency estimation unit that estimates a resonance frequency from the actually measured transfer characteristic of the mechanical system, and selects the candidate from the resonance frequency. A filter estimator for calculating the reflected parameter is provided.
In the invention according to claim 3, the simulation characteristic calculation unit uses the driving force filter unit model in which the parameter is set and the measured transmission characteristic of the mechanical system, and A filter effect simulation unit that evaluates the combined operation characteristics of the mechanical system by simulation, a controller model obtained by numerically modeling the controller, the driving force filter unit model, and the measured transmission characteristics of the mechanical system are used. And a characteristic calculation unit that simulates a combined operation characteristic of the electric motor control device and the mechanical system.

Further, in the invention according to claim 4, the actual machine characteristic confirmation unit measures the input and the operation amount of the driving force filter unit which has turned off the feedback and sets the parameter, and the driving force filter unit and A combination of the motor control device and the mechanical system using the filter effect calculation unit for obtaining the combined operation characteristic of the mechanical system by an actual machine, and further using the combined operation characteristic evaluated by the controller model and the filter effect calculation unit. And a second characteristic calculator that evaluates the operating characteristic by simulation.
The filter setting unit further includes a filter model setting unit, and the filter model setting unit calculates a mechanical transfer characteristic model that is a numerical model of the mechanical system by numerical calculation. A calculation unit that estimates a parameter to be set in the driving force filter unit from the mechanical system transfer characteristic model, the driving force filter unit model in which the parameter is set, and the mechanical system transfer characteristic model. A mechanical system transfer characteristic model filter effect simulation section for simulating a combined operation characteristic of the driving force filter section and the mechanical system, the controller model, the driving force filter section model, and the mechanical system transmission characteristic model, And a model characteristic calculation unit that evaluates the combined operation characteristic of the electric motor control device and the mechanical system by simulation. It is characterized in that.
According to a sixth aspect of the present invention, in the third, fourth, and fifth aspects, the characteristic calculation unit (26), the second characteristic calculation unit, and the model characteristic calculation unit are configured to provide a closed loop frequency characteristic command. One of the response, the disturbance response of the closed loop frequency characteristic, and the open loop frequency characteristic is calculated.
The invention according to claim 7 is characterized in that the operation amount is a position, speed or acceleration detected by a detector attached to the machine or the electric motor.
Further, in the invention according to claim 8, the operation amount is detected at a plurality of locations of the machine or the electric motor, and mechanical system measured transfer characteristics at the plurality of locations are calculated, and at least one of the plurality of locations is calculated. It is characterized by feedback.

The invention according to claim 9 is characterized in that the operation amount detects a sound pressure generated from the machine.
According to a tenth aspect of the present invention, the drive command is a torque command when the electric motor is a rotary motor, and the drive command is a thrust command when the electric motor is a translational motor. It is said.

In the invention according to claim 11, the motor control device includes two or more of the controller, the driving force filter unit, and the current control unit, and the filter setting unit has a function of two or more systems. It is characterized by.
The invention according to claim 12 is characterized in that the command signal is one of a swept sine wave, an M-sequence signal, and a random wave.
The invention according to claim 13 is an input device for setting a parameter in the driving force filter unit, an output device for observing the mechanical measured transmission characteristics and the combined operating characteristics, and the mechanical measured transmission characteristics. And a storage device for storing the operation characteristics of the combination, and the output device includes a sound pressure reproducing device.

The invention described in claim 14 is a command device that generates a command signal, a controller that receives the command signal and outputs a drive command, a driving force filter unit that filters the drive command, An electric motor control device configured to feed back an operation amount of the machine to the controller, the current control unit driving an electric motor connected to the machine using an output of the driving force filter unit as an input; When calculating parameters of the filter unit, cutting off the feedback, measuring the input of the current control unit and the operation amount, and calculating an actual measurement transfer characteristic of a mechanical system including the electric motor and the machine; Estimating a parameter to be set in the driving force filter unit from the measured transmission characteristics of the system, and a driving force filter unit model that numerically models the driving force filter unit And the measured transmission characteristics of the mechanical system, applying the parameters, evaluating the combined operation characteristics by simulation, and applying the parameters evaluated in the step of evaluating the combined operation characteristics by simulation And a step of evaluating the combined operation characteristics with an actual machine.
In the invention according to claim 15, when estimating the parameter, the resonance frequency is estimated from the actually measured transfer characteristic of the mechanical system, a candidate is selected from the resonance frequency, and the parameter reflecting this is selected. It is characterized by calculating.
In the invention according to claim 16, when evaluating the combined operation characteristic by the simulation, the driving force filter is used by using the driving force filter unit model in which the parameters are set and the measured transmission characteristic of the mechanical system. Using a controller model obtained by numerically modeling the controller, the driving force filter unit model, and an actually measured transfer characteristic of the mechanical system, The combined operation characteristics of the motor control device and the mechanical system are evaluated by simulation.

Further, in the invention according to claim 17, when evaluating the combined operation characteristic by the actual machine, the input of the driving force filter unit in which the feedback is turned off and the parameter is set and the operation amount are measured, and the operation amount is measured. A combination operation characteristic of the driving force filter unit and the mechanical system is evaluated by an actual machine, and the motor controller is used by using the controller model, and a combination operation characteristic of the driving force filter unit and the mechanical system evaluated by the actual machine. The combined operation characteristics of the mechanical system are evaluated by simulation.
The invention according to claim 18 is a step of calculating a mechanical system transfer characteristic model which is a numerical model of the mechanical system by numerical calculation, and estimating a parameter to be set in the driving force filter unit from the mechanical system transfer characteristic model. Using the driving force filter unit model and the mechanical system transfer characteristic model in which the parameters are set, and evaluating the combined operation characteristics of the driving force filter unit and the mechanical system in simulation, Using the controller model, the driving force filter unit model, and the mechanical system transfer characteristic model to further evaluate the combined operation characteristics of the electric motor control device and the mechanical system by simulation. It is characterized by that.
According to a nineteenth aspect of the present invention, in the sixteenth, seventeenth and eighteenth aspects, when the combined operation characteristics of the motor control device and the mechanical system are evaluated by simulation, a command response of a closed loop frequency characteristic, a closed loop frequency One of characteristic disturbance response and one loop frequency characteristic is calculated.

According to the first aspect of the present invention, the mechanical characteristic calculation means calculates the actual transmission characteristic of the mechanical system, calculates the parameter of the driving force filter unit from the mechanical system actual transmission characteristic, and uses the driving force filter unit model. Thus, the combination operation characteristic can be evaluated by simulation, and finally the combination operation characteristic can be evaluated by an actual machine, so that the optimum parameter of the driving force filter unit can be determined.
According to the second aspect of the invention, it is possible to extract the resonance frequency based on the actually measured value and calculate the parameter reflecting this.
According to the third aspect of the present invention, the driving force filter unit model is used to evaluate the combined operating characteristics in a simulated manner before setting the driving force filter unit. Validity can be confirmed. Further, it is possible to determine the parameters of the driving force filter unit while predicting the effect.

According to the invention described in claim 4, the validity of the parameter is further improved by applying the parameter confirmed in the simulation evaluation of claim 3 to the actual machine for evaluation.
According to the invention described in claim 5, at the design stage where there is no actual machine, the mechanical system transfer characteristic model is calculated by calculation, the parameters of the driving force filter unit are calculated, and the simulation is performed using the driving force filter unit model. By confirming the operation characteristics step by step, the optimum parameters of the driving force filter unit can be determined.
Further, according to the invention described in claim 6, it is possible to grasp various characteristics at the time of combination with a simulation or an actual machine, and comprehensive evaluation can be performed.
Further, according to the seventh aspect, since the operation amount of various units can be used, it can be used for various configurations.
According to the eighth aspect of the present invention, since the mechanical system transfer characteristics at a plurality of locations of the machine can be calculated, more detailed mechanical system transfer characteristics can be grasped.
According to the ninth aspect of the present invention, since the sound pressure can be detected, the ambiguity that relies on the human ear can be eliminated and the data can be utilized for scientific and quantitative analysis.
According to the invention of claim 10, a rotary motor or a translational motor can be used for the electric motor. Thereby, the motor according to various situations can be utilized.
According to the eleventh aspect of the present invention, even in the case of a multi-axis motor control device, optimal processing can be performed by performing the same processing.
According to the twelfth aspect of the present invention, a swept sine wave, an M-sequence signal, or a random wave can be selected as the command signal, so that signals according to various situations can be used for grasping various characteristics.
According to the invention of claim 13, various calculation results can be clearly observed by the output device. In addition, sound pressure playback devices can be used to hear and confirm sound even when the physical sound pressure does not match the sense of hearing with the human ear. The filter effect can be confirmed. In the input device, various conditions and parameter settings can be performed, and various conditions and settings can be stored in the storage device, and work can be performed at different dates and places.

  According to the invention of claim 14, the measured transmission characteristic of the mechanical system is calculated, the parameter of the driving force filter unit is calculated from the measured transmission characteristic of the mechanical system, and the combined operation is simulated by using the driving force filter unit model. Since the characteristics can be evaluated and finally the combined operation characteristics can be evaluated with an actual machine, the optimum parameter of the driving force filter unit can be determined.

According to the invention described in claim 15, it is possible to extract the resonance frequency based on the actually measured value and calculate a parameter reflecting this.
According to the invention described in claim 16, the driving force filter unit model is used to simulate the combination of the driving force filter unit and the mechanical system and the control device and the mechanical system, so that the parameters of the driving force filter unit are valid. Sex can be confirmed. Further, it is possible to determine the parameters of the driving force filter unit while predicting the effect.

According to the seventeenth aspect of the present invention, the parameters evaluated in the sixteenth aspect can be applied to the actual machine for evaluation.
According to the invention described in claim 18, in the design stage where there is no actual machine, the mechanical system transfer characteristic model is calculated, the parameters of the driving force filter unit are calculated, the combined operation in the simulation using the driving force filter unit model, the actual machine The combination operation can be performed step by step to determine the optimum parameters of the driving force filter unit.
According to the nineteenth aspect of the invention, various characteristics when combined with a simulated or actual machine can be grasped, and comprehensive evaluation can be performed.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a configuration diagram of an electric motor control device showing a first embodiment of the present invention.
In FIG. 1, 3 is a controller, 4 is a commander, 6 is a current control unit, 8 is an adder, 9 is a switch, 10 is a driving force filter unit, and 50 is a filter setting unit. It is configured.
The filter setting unit 50 is a set of filter setting functions, and includes a mechanical characteristic calculation unit 11, a filter constant estimation unit 33, a simulated operation calculation unit 33, and an actual machine operation confirmation unit 32.
The filter constant estimation unit 33 includes a resonance frequency estimation unit 24 and a filter estimation unit 25. The simulation operation calculation unit 33 includes a characteristic calculation unit 26 and a filter effect simulation unit 13. The actual machine operation confirmation unit 32 includes , A filter effect calculation unit 14 and a second characteristic calculation unit 27.
The output device 18, the input device 20, and the storage device 19 are attached devices for the filter setting unit.
An electric motor 1 driven by this electric motor control device is fixed to a machine 5, and a detection unit 2 is attached to the machine. Here, the electric motor 1 is a translational linear motor.
When the present invention is compared with the first conventional example and the second conventional example, the mechanical characteristic calculation unit 11 for calculating the transfer characteristic of the mechanical system is similar, but the content of the implementation is different, and a filter that simulates using the measured mechanical characteristic The difference is that the effect simulation unit 13, the characteristic calculation unit 26, the filter effect calculation unit 14 that confirms the characteristic using an actual machine, and the second characteristic calculation unit 27 are provided.

FIG. 2 is a flowchart of the motor control apparatus showing the first embodiment of the present invention.
The outline of each step will be described with reference to FIG.
STEP 11 is a step of calculating a measured transmission characteristic of the mechanical system, and grasps a measured transmission characteristic of the mechanical system including the electric motor 1 and the machine 5.
STEP 12 is a step of estimating the resonance frequency, and estimates the resonance frequency included in the mechanical system actually measured transfer characteristic obtained in STEP 11.
STEP 13 is a step of calculating a parameter of the driving force filter unit, and selects a resonance frequency candidate and calculates a parameter reflecting this.

STEP 14 is a step of simulating the combined operation of the driving force filter unit and the mechanical system, and evaluates the combined operation characteristics of the driving force filter unit and the mechanical system by simulation using the driving force filter unit model.
STEP 15 is a step of evaluating the combined operation characteristics of the motor control device and the mechanical system by simulating the combined operation characteristics of the motor control device and the mechanical system using the driving force filter unit model and the controller model. To evaluate.

STEP 16 is a step of evaluating the combination of the driving force filter unit and the mechanical system transmission characteristics with an actual machine, and calculates the combined characteristics of the driving force filter unit and the mechanical system by applying the above parameters.
STEP 17 is a step of evaluating the combined operation characteristics of the motor control device and the mechanical system by simulation, and using the combined characteristics of the driving force filter unit and the mechanical system obtained in STEP 16 and the controller model, the motor control device And the composite characteristics of the mechanical system.
Here, after STEPs 11, 14, and 16, the next step may be skipped as necessary.

Each procedure will be described below in order.
First, STEP 11 is a step in which the mechanical characteristic calculation unit 11 is involved, and is a step of calculating actual transmission characteristics of the mechanical system including the electric motor and the machine. Actual measurement of the mechanical system including the electric motor 1 and the machine 5 is performed using an actual electric motor control device. Know the transfer characteristics.
In FIG. 1, switches 9 a, 9 b, 9 c, adders 8 a, 8 b, command unit 4, current control unit 6, electric motor 1, detector 2, machine 5, and mechanical characteristic calculation unit 11 are involved.
The feedback loop is cut open by the switch 9c, the output of the command unit 4 is switched to the adder 8a side by the switch 9a, and the driving force filter unit 10 is set not to be used by the switch 9b.
The command unit 4 generates a command signal of the swept sine wave C, and when the controller 3 gives a thrust command to the current control unit 6, the motor 1 operates together with the machine 5, and the detection unit 2 calculates the operation amount r of the machine 5. To detect.

The mechanical characteristic calculation unit 11 receives the command signal and the operation amount r, calculates them, and calculates the transfer characteristic H (ω) of the mechanical system including the electric motor 1 and the machine 5.
For example, frequency analysis is performed using FFT (Fast Fourier Transform) to obtain a spectrum. Here, the auto power spectrum and the cross spectrum are calculated, and the auto power spectrum G ′ _CC (ω) and the cross spectrum G ′ _rC ( From (ω), the transmission characteristic H (ω) of the mechanical system is obtained by the equation (1).
Actually, the auto power spectrum G ′ _CC (ω) and the cross spectrum G ′ _rC (ω) obtained by averaging the data acquisition repeatedly are used.
The transmission characteristic of the mechanical system obtained by actual measurement is hereinafter referred to as the actual measurement transmission characteristic of the mechanical system.

  Here, the actually measured transfer characteristic H of the mechanical system may be recorded in the storage device 19 or may be observed by the output device 18.

Step STEP12 is a step in which the resonance frequency estimation unit 24 is involved and estimates the resonance frequency from the mechanical system measured transfer characteristic. Here, the resonance frequency included in the mechanical system measured transfer characteristic obtained in STEP11 is estimated.
Although the mechanical system actually measured transfer characteristic has irregularities, the apex of the convex is resonance, and the frequency is the resonance frequency.
Step STEP13 is a step in which the filter estimation unit 25 is involved and calculates a parameter of the driving force filter unit, and selects a resonance frequency candidate and calculates a parameter reflecting this.
Here, the frequency range is determined so as not to interfere with the parameter (gain) of the controller 3, and the resonance frequency candidate is selected from the one having a large convex amplitude of the mechanically measured transfer characteristic.

  Step STEP14 is a step in which the filter effect simulation unit 13 is involved and evaluates the combined operation characteristics of the driving force filter unit and the mechanical system by simulation. Here, the actual measured transmission characteristic and the driving force filter unit obtained in STEP11 are used. The combined characteristic with the driving force filter unit model that is modeled is calculated. In actual motor control devices, the number of resonance frequencies and the setting method are limited, so the driving force filter unit model is applied. Note that the driving force filter unit model used here simulates the actual driving force filter unit 10 up to the characteristics of discretization.

FIG. 3 is a diagram illustrating a relationship between characteristics of the mechanical system and the driving force filter unit according to the first embodiment. In FIG. 3, C is a command signal (swept sine wave) output from the command unit 4, f is a characteristic of the driving force filter unit 10, and H is a mechanical system transmission characteristic including the electric motor 1 and the machine 5.
The driving force filter unit 10 usually has a function of a notch filter or a low-pass filter. The characteristic f1 of the notch filter is expressed by Expression (2), and the characteristic f2 of the first-order lag low-pass filter is expressed by Expression (3).

Here, s is a Laplace operator, Q is an attenuation term, and ωr is a notch filter target frequency. The notch filter has a characteristic of blocking the passage of the signal of the frequency ωr, and as the attenuation term Q value is increased, the indentation and the phase delay in the frequency characteristic of blocking the passage of the signal become steeper.

Here, Tf is a first-order lag low-pass filter time constant.
The combined characteristic H ′ of the mechanically measured transmission characteristic H and the driving force filter unit model is expressed by Equation (4).
H ′ = H · f ₁ · f ₂ formula (4)
Moreover, the effect can be confirmed by changing the parameters in the equations (2) and (3).
In FIG. 3, the example of the frequency characteristic of each part is shown under a block diagram.
The mechanical transfer characteristic shows a convex part that shows resonance, but the center frequency of the next notch filter is set according to this resonant frequency, and the mechanical part's transfer characteristic and notch filter combined characteristic show a convex part that shows resonance. The state of disappearing is shown.

Equations (2) and (3) may be calculated by Euler primary approximation and bilinear approximation.
In this example, the characteristics of the driving force filter unit model are the notch filter f1 and the first-order lag low-pass filter f2, but one or more notch filters or two or more low-pass filters may be provided. The filter may be a secondary filter, or may be provided with primary and secondary low-pass filters.
Such a simulation result can be confirmed by the output device 18.
Further, a change in characteristics when the parameters are changed can be recorded in the storage device 19.

STEP 15 is a step in which the characteristic calculation unit 26 is involved and simulates the combined operation characteristics of the motor control device and the mechanical system.
The mechanical system measured transfer characteristics obtained in STEP 11, the driving force filter unit model, and the controller model that is a numerical model of the controller are used, and the parameters obtained in STEP 13 are set in the driving force filter unit model. Then, the combined characteristics of the motor control device and the mechanical system are calculated.
In FIG. 1, switches 9a, 9b, 9c, adders 8a, 8b, command unit 4, controller 3, current control unit 6, electric motor 1, detector 2, and machine 5 are involved. The controller uses a numerical model, and the mechanical system uses the mechanical system measured transfer characteristic obtained in STEP11.

Here, the command response, disturbance response, and one-round open loop frequency characteristics, which are closed loop frequency characteristics, are calculated.
The command response which is one of the closed-loop frequency characteristics is shown in FIG. 1 from the commander 4, the switch 9 a gives the command signal C to the controller 3 side, and the driving force works via the controller 3 from the motor 1 to the machine. 5 through, the operation amount r from the detector 2, closing the switch 9c, a characteristic Z _R from the command signal C in a state that feedback to the controller 3 to the operating amount r.
When the mechanical transmission characteristic including the electric motor 1 and the machine 5 is H, the characteristic of the controller 3 is G, and the driving force filter unit 10 is f, the command response characteristic Zr is expressed by Equation (5).

Here, G: characteristics of the controller, H: mechanical system transmission characteristics of the electric motor and the machine, f: characteristics of the driving force filter section.
In addition, f is the characteristic f of the driving force filter unit, and includes the characteristics of a plurality of filters, and is given by equation (6).
f = f ₁ · f ₂ formula (6)
The disturbance response which is one of the closed loop frequency characteristics is the characteristic from the command signal C as a disturbance to the response r when the switch 9a gives the command signal C to the adder 8a and the switch 9c is closed in FIG. Zd, which is Equation (7).

Further, in FIG. 1, the open loop frequency characteristic Zo is obtained from the command device 4 by the switch 9 a giving the command signal C to the controller 3 side, and the driving force is acted via the controller 3 from the motor 1 to the machine 5. The characteristic Zo from the open loop command signal C to the response r when the switch 9c is opened up to the operation amount r from the detection unit 2 is expressed by Equation (8).
Zo = -G. (F.H) Formula (8)
In STEP15, the configuration of FIG. 1 is converted into a numerical model, and the characteristics of the equations (5), (6), and (7) are grasped using the mechanical system actually measured transfer characteristics.
Here, if the parameter of the controller model is changed, the G of the control system changes, so this effect can also be examined in advance.

STEP 16 is a step in which the filter effect calculation unit 14 is involved and the combined characteristics of the driving force filter unit and the mechanical system are evaluated with an actual machine.
In FIG. 1, switches 9a, 9b, 9c, adders 8a, 8b, command unit 4, current control unit 6, electric motor 1, detector 2, machine 5, driving force filter unit 10, and filter effect calculation unit 14 are involved.
Here, the parameter evaluated in STEP 15 is set in the driving force filter unit 10 so that the command signal C of the command device 4 is given to the driving force filter unit, and the command signal C and the operation amount r from the detection unit 2 are used. Similarly to the equation (4), the combined H ″ of the mechanical system shown in (9) and the driving force filter unit 10 is calculated.
H ″ = H · f ₁ · f ₂ formula (9)

STEP 17 is a step in which the second characteristic calculation unit 27 is involved and is a step for simulating the combined operation characteristics of the motor control device and the mechanical system. The combined characteristics of the driving force filter unit 10 and the mechanical system obtained in STEP 16 and the controller Calculate the combined characteristics with the model.
The second characteristic calculator 27 has substantially the same function as the characteristic calculator 26 used in STEP15. Here, the command response Z _R in Equation (5), the disturbance response Zd in Equation (7), and the open loop frequency characteristic Zo in Equation (8) shown in STEP 15 are calculated.
The difference from STEP 15 is that an actual driving force filter unit 10 is used instead of the driving force filter unit model.

In this embodiment, a swept sine wave is used for the command C, but a random wave or an M-sequence signal may be used.
Further, in the storage device 19, the mechanical system actual transmission characteristic H, the composite characteristic H ′ in which the driving force filter unit model acts on the mechanical system actual transmission characteristic, and the composite characteristic H in which the driving force filter unit 10 acts on the actual mechanical system. The effect can be confirmed by comparing "".
Further, the parameters of the driving force filter unit 10 may be changed by the input device 20.
The characteristic calculation unit 26 and the second characteristic calculation unit 27 may be shared because they have the same function although they have different inputs.
The switch 9b is a thing for implement | achieving the function to bypass without using the driving force filter part 10 or using it.
As an alternative to this, for example, as shown later in the second and third embodiments, a switch 9b is provided in front of the driving force filter unit 10, and the output of the driving force filter unit 10 and the driving force filter unit 10 are not used. It is also possible to receive the output of the controller 3 bypassed by the adder 8 and input this output to the current control unit 6. That is, it is only necessary to use the driving force filter unit 10 or to realize a bypass function without using it.

By executing each step as described above, it is possible to set an optimum parameter of the driving force filter unit 10 in the motor control device.
In FIG. 2, intermediate steps can be omitted as necessary.

FIG. 4 is a configuration diagram of an electric motor control device showing a second embodiment of the present invention.
4, 3 is a controller, 4 is a commander, 5 is a machine, 6 is a current control unit, 8 is an adder, 9 is a switch, 10 is a driving force filter unit, 15 is a position control unit, and 16 is a speed control. Reference numeral 17 denotes a conversion unit, which constitutes an electric motor control device.
Reference numeral 1 denotes an electric motor, which is a rotary motor, and translates the table via a ball screw. Reference numerals 2a, 2b, 2c, 2d,... 2m are detection units, which have a fully closed configuration having a detection unit 2a that detects the operation amount of the rotary motor and a detection unit 2b that detects the operation amount of the table. ing. The output of the detection unit 2 a is fed back to the speed control unit 16 via the conversion unit 17.
Further, detectors 2 c, 2 d,..., 2 m that are not fed back to the controller 3 are attached to the machine 5. The detection units 2c,... 2m are accelerometers, and the sound pressure detection unit 29 is a sound pressure meter.
Reference numeral 50 denotes a filter setting unit, which includes a mechanical characteristic calculation unit 11, a filter constant estimation unit 33, a simulated operation calculation unit 31, an actual machine operation confirmation unit 32, and a filter model setting unit 51.
The filter constant estimation unit 33 includes a resonance frequency estimation unit 24 and a filter estimation unit 25.
The simulated operation calculation unit 31 includes a filter effect simulation unit 13 and a characteristic calculation unit 26, and the actual machine operation confirmation unit 32 includes a second characteristic calculation unit 27 and a filter effect calculation unit 14.
The filter model setting unit 51 includes a calculation unit 23, a filter model estimation unit 36, a mechanical system transfer characteristic model filter effect simulation unit 22, and a model characteristic calculation unit 35.
Further, the output device 18, the input device 20, the storage device 19, the sound pressure reproducing device 30, and the filter setting unit 50 are attached devices.
In addition, although one line from the detection units 2a, 2b, 2c, 2d,..., 2m to the mechanical characteristic calculation unit 11 is omitted, each signal is individually connected in parallel.

  FIG. 5 is a diagram illustrating a mechanical system of the motor and the machine by a transfer function in the configuration diagram of the motor control device according to the second embodiment of the present invention. 5 omits the components from FIG. 4, and the mechanical system of the electric motor and the machine in FIG. 4 is represented as a mechanical system transfer characteristic regarding the detection units 2 a, 2 b, 2 c,. , Ha, Hb, Hc,..., Hm, Hn.

The difference from the first embodiment is that the electric motor 1 is a rotary motor and has a fully closed configuration including detection units 2a and 2b. The detection unit 2a detects the position of the machine 5 and feeds back to the position control unit 15. The detection unit 2 b detects the amount of rotation of the electric motor 1, forms a rotation speed via the conversion unit 17, and feeds it back to the speed control unit 17.
Moreover, the point which is equipped with the accelerometer which is the detection part 2c and ... 2m which detect the other operation amount of the machine 5, and the sound pressure meter of the sound pressure detection part 29 differs from the 1st Example.
Further, the controller 3 is disassembled into a position control unit 15 and a speed control unit 16 and a conversion unit 17 is added, which is different from the first embodiment.

FIG. 6 is a flowchart of filter setting of the motor control device according to the second embodiment.
The part different from the first embodiment is a part provided with the procedures of STEP01, STEP02, STEP03, and STEP04, and STEP11 to STEP17 are the same as the first embodiment.
Here, STEP01 is a process of calculating a mechanical system transfer characteristic model which is a numerical model of the mechanical system by numerical calculation.
STEP 02 is a process of performing the same processing as STEP 12 and 13 and calculating a parameter to be applied to the driving force filter unit from the mechanical system transfer characteristic model obtained in STEP 01.
In STEP 03, the driving force filter unit model is used for the mechanical system transfer characteristic model obtained in STEP 01 to simulate the combination of the driving force filter unit and the mechanical system.
STEP 04 is a process of simulating the combined characteristics of the motor control device and the mechanical system using the mechanical system transfer characteristic model and the driving force filter unit model obtained in STEP 01.

Each procedure will be described below in order.
In steps 01 to 04, since an actual motor control device does not actually exist, all numerical models are used.
STEP01 is a step of calculating a mechanical system transfer characteristic model by numerical calculation, and the calculation unit 23 is related thereto.
In this, the machine's natural frequency and vibration mode can be obtained by incorporating the machine and physical properties of the machine as a model, and defining and solving various conditions such as boundary conditions according to the actual machine.
If the natural frequency (resonance frequency) and vibration mode are obtained, the machine model can be converted into a numerical value, so that the motor 1 and the mechanical system transfer characteristic model 21 of the machine can be output.

  As shown in FIG. 4, even when the sound pressure meter of the sound pressure detection unit 29 is used, the sound characteristics can be quantified by the finite element method.

As shown in FIG. 4, since it has a plurality of detectors 2a, 2b, 2c,..., 2m and a sound pressure detector 29, when actually measured, one electric motor 1 and detector 2a, 2b, 2c,..., 2m and the mechanical transmission characteristics between the sound pressure detector 29 are obtained.
Each element in the broken line in FIG. 5 is a mechanical transmission characteristic Ha, Hb,..., Hm, Hn corresponding to the characteristic from the actual electric motor 1 to the detection unit. Characteristic models ha, hb,... Hm, hn can be obtained.
Thus, a simulated mechanical transmission characteristic can be obtained even when there is no actual machine.

In STEP02, the filter model estimation unit 36 is involved, and a resonance frequency candidate is calculated from the mechanical transmission characteristics obtained in the previous step.
As in STEP 13, a parameter reflecting the resonance frequency is set from the model of the mechanical system transfer characteristic obtained in the previous step.
STEP03 is a step of simulating the result of applying the driving force filter unit (10) to the mechanical system transfer characteristic model, involving the mechanical system transfer characteristic model filter effect simulation unit 22, and the mechanical system transfer characteristic model ha, hb ,..., Hm, hn are applied to the driving force filter unit model to perform a process for examining the situation in advance.
The output of the mechanical system transfer characteristic model filter effect simulation unit 22 is a characteristic h′a, h ′ that simulates the result of applying the driving force filter unit model 12 to the mechanical system transfer characteristic models ha, hb,. b, h′c,..., h′m, h′n, which is the expression (10) similar to the expression (4).
h ′ _k = h _k · f ₁ · f ₂ Formula (10)

  Here, k = a, b, c,..., N, f1, and f2 are characteristics of the driving force filter unit model.

Since the case where the driving force filter unit 10 is used for the mechanical system transfer characteristic model (ha, hb,..., Hn) is estimated, the state of the entire machine can be examined in advance.
Further, the parameters of the driving force filter unit 10 such as the attenuation term Q of the notch filter shown in the equation (2) and the first-order lag low-pass filter time constant Tf shown in the equation (3) are changed. The effect can be examined in advance.

  In this example, it is obtained by the finite element method. However, a model of concentrated mass and spring may be calculated by the calculation unit 23, or a result obtained by manual calculation from the model may be input. Can be expressed numerically and expressed as a mechanical system transfer characteristic model and can be input.

  In STEP 04, the model characteristic calculation unit 35 is involved, and a combined characteristic of the mechanical system transfer characteristic model, the driving force filter unit model, and the controller model is calculated in a step of simulating the combined characteristics of the motor control device and the mechanical system.

  If the respective characteristics of the position control unit 15, the speed control unit 16, and the conversion unit 17 are G1, G2, and b, the command response Zra is expressed by Equation (11) in the case of the operation amount ra obtained from the detection unit 2a.

  Here, C: command signal, ra: output of the detection unit 2a, G1: position control unit characteristic, G2: speed control characteristic, b: conversion unit characteristic, ha: mechanical system transfer characteristic from the motor to the detection unit 2a, hb : Mechanical system transfer characteristics from the electric motor to the detector 2b.

  In the case of the operation amount rb obtained from the detection unit 2b, the command response Zrb is expressed by Equation (12).

  In the case of the operation amount ra obtained from the detection unit 2a, the disturbance response Zda is expressed by Equation (13).

  In the case of the response rb obtained from the detection unit 2b, the command response Zdb is expressed by Expression (14).

Further, the open loop frequency characteristic Zo is expressed by Equation (15).
Zo = − {G ₁ · (f · h _b ) + b · (f · h _a )} · G ₂ formula (15)
In the above formula, by applying the numerical model and calculating the combined characteristics of the motor control device and the mechanical system, it is possible to examine in advance at the design stage where there is no actual machine. If it can be judged that the performance is sufficient, the actual machine is manufactured and the next step is performed.

Next, steps using an actual machine will be described.
As in the first embodiment, STEP 11 grasps the mechanical transmission characteristics of the mechanical system including the electric motor 1 and the machine 5 using the actual electric motor control device.
In the second embodiment, since the plurality of detection units and the sound pressure detection unit 29 are provided, a plurality of mechanical system transfer characteristics can be obtained. That is, the mechanical characteristic calculation unit 11 includes the swept sine wave C that is a command signal of the command unit 4 and the machine operation amounts r1, r2, r3 detected by the detection units 2a, 2b, 2c,. .., Rm, a plurality of mechanical system measured transfer characteristics Ha, Hb, Hc,..., Hm, Hn are calculated.
Similarly, the actual measurement characteristic Hn of the sound generated by the vibration of the machine 5 can be calculated from the command signal of the command device 4 and the sound pressure rn detected by the detection unit 29.
Here, the description will be continued by including Hn in the measured transmission characteristics of the mechanical system.
It should be noted that the unit system may be changed by converting the movement amounts r1, r2, r3,..., Rm, and the sound pressure rn through the conversion unit 17, and further, after calculating the mechanical system actually measured transfer characteristics, the unit system You may perform the calculation which converts.

In STEP 12, as in the first embodiment, the resonance frequency estimator 24 detects the resonance frequency from the mechanical measured transmission characteristics Ha, Hb, Hc,..., Hm, Hn obtained in step STEP11.
When the vibration mode of the machine is complicated, there is a case where the resonance frequency cannot be determined by measuring one point of the mechanical system transfer characteristic. However, since a plurality of mechanical system transfer characteristics can be used here, the determination of the resonance frequency is reduced. There is also an effect. Moreover, sound measurement can also contribute to discrimination.

In STEP 13, as in the first embodiment, the filter estimation unit 25 is involved, and the parameters of the driving force filter unit 10 are calculated. A resonance frequency candidate is selected from the resonance frequencies obtained in STEP 12. The parameter that reflects is calculated.
Here, only the detection units 2a and 2b are fed back to the controller 3, but filter setting candidates can be calculated even for resonances that cannot be detected by the detection units 2a and 2b.

As in STEP 1, the STEP 14 involves the filter effect simulation unit 13, and here, the result of applying the driving force filter unit model that models the driving force filter unit to the mechanical transmission characteristics obtained in STEP 11 Is calculated.
Here, it is possible to simulate the case where the driving force filter unit 10 is used for a plurality of mechanical system transmission characteristics, and it is possible to evaluate effects at various locations that cannot be obtained with one mechanical system transmission characteristic H.
The result can be observed by the output device 18. Furthermore, Hm including sound characteristics can be handled in the same manner as the others, and the effect of the driving force filter unit 10 on sound can be evaluated.
The characteristics of the sound may be observed in the form of a graph on the output device 18 or may be confirmed with the sound pressure reproducing device 30 by audibly distinguishing it with the human ear and simulating the filter effect.
The time series data of the sound is stored in the storage device 19 and input to the filter effect simulation unit 13, and the output is reproduced by the sound pressure reproduction device 30, so that the effect of the filter on the sound can be obtained. You can check with your ears.

  Further, the calculated mechanical system actually measured transfer characteristics Ha, Hb, Hc,..., Hm may be converted into time series data by inverse Fourier transform and graphed on the output device 18.

As in the first embodiment, STEP 15 includes a characteristic calculation unit 26, and includes a mechanical system actually measured transfer characteristic obtained in STEP 11, a driving force filter unit model, and a controller model that is a numerical model of the controller. Used to simulate a combination of a motor control device and a mechanical system.
Here, in equations (11), (12), (13), (14), and (15) shown in step STEP03, the mechanical system transmission characteristic model h is replaced with the mechanical system actual transmission characteristic H, and the driving force filter unit A driving force filter unit model is applied to f, and a combined characteristic of the controller model, the driving force filter unit model, and the actual measured transmission characteristic of the mechanical system is calculated.
Note that the detection units 2c,... 2m and the sound pressure detection unit 29 are not fed back to the controller 3 and are therefore independent of the characteristics of the control system.

As in the first embodiment, STEP 16 involves the filter effect calculation unit 14 and calculates the result of applying the driving force filter unit to the mechanical system actually measured transfer characteristic obtained in STEP 11.
Here, the switch 9b may be set and processed so that the driving force filter unit 10 acts.
Here, the output of the filter effect calculation unit 14 can be confirmed by the output device 18.
Further, in the storage device 19, the effect can be confirmed by comparing the case where the driving force filter unit model of STEP 14 is applied with the case where the driving force filter unit 10 of STEP 16 is applied.

With respect to Hn including the sound characteristics, the characteristics that the driving force filter unit 10 has acted can be confirmed by the output device 18, so that the ambiguity that relies on the human ear can be eliminated and the sound reduction can be grasped.
Similarly to the first embodiment, the setting parameter of the driving force filter unit 10 may be changed to a setting value different from that of the driving force filter unit model by the input device 20.

In STEP 17, the second characteristic calculation unit 27 is involved, and the combination of the motor control device and the mechanical system is simulated using the driving force filter unit 10 obtained in STEP 16, the mechanical system transmission characteristics, and the controller model.
Here, the combined characteristics of the driving force filter unit 10 and the mechanical transmission characteristic calculated in STEP 16 are used, and the position control unit 15 of the controller 3 and the characteristic G2 of the speed control unit 16 are shown in STEP 15 using a model. Using the equations (11), (12), (13), (14), and (15), the command responses Zra, Zrb, the disturbance responses Zda, Zdb, and the open loop frequency characteristic Zo are calculated.

Thus, by executing STEPs 11 to 17, the optimum parameters of the driving force filter unit 10 can be obtained.
In FIG. 6, intermediate steps may be omitted or the previous step may be returned as necessary.

FIG. 7 is a block diagram of an electric motor control device showing a third embodiment of the present invention.
In FIG. 7, 1a and 1b are motors, 2a, 2b and 2c are detectors, 3a and 3b are controllers, 4 is a commander, 5 is a machine, and this constitutes a motor control device.
In FIG. 7, for the sake of simplification, the controller 3 has an external appearance of hardware including various functional units including the driving force filter unit 10 and the switch 9 shown in the first and second examples. It is shown as a diagram. That is, although it is set as the controller 3, it comprises the controller 3, the driving force filter unit 10, the switch 9 and the adder shown in the first and second embodiments, and similarly constitutes a motor control device. And
Reference numeral 50 denotes a filter setting unit, which includes a mechanical characteristic calculation unit 11, a filter constant estimation unit 33, a simulated operation calculation unit 31, an actual machine operation confirmation unit 32, and a filter model setting unit 51.
The filter constant estimation unit 33 includes a resonance frequency estimation unit 24 and a filter estimation unit 25.
The simulated operation calculation unit 31 includes a filter effect simulation unit 13 and a characteristic calculation unit 26, and the actual machine operation confirmation unit 32 includes a second characteristic calculation unit 27 and a filter effect calculation unit 14.
The filter model setting unit 51 includes a calculation unit 23, a filter model estimation unit 36, a mechanical system transfer characteristic model filter effect simulation unit 22, and a model characteristic calculation unit 35.
Further, the output device 18, the input device 20, and the storage device 19 are accessory devices of the filter setting unit 50.
The electric motors 1a and 1b are rotary motors, and the rotary motor translates the table through a ball screw. Further, the Y-axis has a fully closed configuration having a detection unit 2a for detecting the operation amount of the rotary motor 1a and a detection unit 2c for detecting the operation amount of the table.
The X-axis has a semi-closed configuration having a detection unit 2b that detects the operation amount of the rotary motor 1b.

FIG. 8 is a block diagram of the electric motor control apparatus showing the third embodiment of FIG. 7, in which the mechanical system of the electric motor and the machine is represented by a transfer function. In FIG. 8, components are omitted from FIG.
As mechanical system transfer characteristics including the motors 1a and 1b and the machine 5, mechanical system transfer characteristics Ha and a obtained by the drive and detection unit 2a of the motor 1a, and Hc and a obtained by the drive and detection unit 2c of the motor 1a. In addition to the mechanical system transmission characteristics Hb, b obtained by the drive and detection unit 2b of the motor 1b, the mechanical system transmission characteristics Hb, a obtained by the drive and detection unit 2b of the motor 1a, and the drive and detection unit 2a of the motor 1b There are mechanical system transmission characteristics Ha, b obtained, and mechanical system transmission characteristics Hc, b obtained by driving the motor 1b and the detection unit 2c.

The difference from the first embodiment is that it has a multi-axis configuration with two electric motors 1 and three controllers 3 and three detectors 2.
The one axis has a fully closed configuration in which the electric motor 1 has a rotary motor and includes detection units 2a and 2c, and the other axis has the detection unit 2b in which the electric motor 1 has a rotary motor. This is a semi-closed configuration.
Further, the controller 3 is disassembled into a position control unit 15 and a speed control unit 16 and a conversion unit 17 is added, which is different from the first embodiment.
The mechanical characteristic calculator 11 and the filter effect calculator 14 have two systems of inputs.

The difference from the second embodiment is that it has a multi-axis configuration with two electric motors 1 and two controllers 3 and three detectors 2. Further, the point that the one axis is a fully closed configuration is the same as that of the second embodiment, but the other axis is provided, and the rotary motor motor 1 and the detection unit 2b are provided in a semi-closed configuration. It is.
Another difference is that a plurality of accelerometers, a sound pressure detector 29 and a sound pressure reproducing device 30 are not provided in addition to the three detectors 2 that feed back to the controller 3.
In addition, since the procedure is the same as that in FIG. 1 of the first embodiment, it is different in that it does not have elements related to steps STEP01 to STEP04 calculated by the mechanical system transfer characteristic model.

Hereinafter, the operation of each step will be described in order.
First, STEP 11 is a step in which the mechanical characteristic calculation unit 11 is involved and calculates a mechanical system actual transmission characteristic of a mechanical system including an electric motor and a machine. The actual electric motor control device is used to grasp the mechanical system transmission characteristic.
Here, because of the multi-axis configuration, six mechanical system transfer characteristics are measured from the three detectors 2.
The Y-axis feedback loop is cut by the switch 9c to open the loop, the output of the command unit 4 is switched to the adder 8a side by the switch 9a, and the switch 9b on the adder 8a side does not use the driving force filter unit 10. Set.
When the command device 4 generates the sweep sine wave Ca and gives it to the current control unit 6a as the torque command Ta, the command device 4 operates together with the machine 5 to which the electric motor 1a is added. The detection units 2a and 2c detect the operation amounts ra and rc of the operated electric motor 1a and machine 5.
For the other axis, when the X axis is opened by the switch 9f and the swept sine wave Cb is not received and no torque command is given to the current control unit 6b, or if Cb = 0 is given, the motor 1a operates. The detection unit 2b also obtains a response rb.
Thereby, the mechanical characteristic calculation unit 11 obtains the mechanical transmission characteristics Ha, a, Hc, a, Hb, a from the torque command Ta and the operation amounts ra, rb, rc.

Next, the axis to be operated is switched.
The X-axis feedback loop is cut by the switch 9f to open the loop, the output of the command unit 4 is switched to the adder 8c side by the switch 9d, and the switch 9e on the adder 8c side does not use the driving force filter unit 10. Set.
When the command device 4 generates the swept sine wave Cb and gives it to the current control unit 6b as the torque command Tb, it operates together with the machine 5 to which the electric motor 1b is added. The detection unit 2b detects the operation amount rb of the operated electric motor 1b.
For the other axis, when the YX axis is opened by the switch 9c and the swept sine wave Ca is not received and no torque command is given to the current control unit 6a, or if Ca = 0 is given, the motor 1b operates. The detectors 2a and 2c also obtain responses ra and rc.
Thereby, the mechanical transmission characteristics Ha, b, Hc, b, Hb, b are obtained from the torque command Tb and the operation amounts ra, rb, rc.
The mechanical transfer characteristic H obtained here may be recorded in the storage device 19, and the mechanical transfer characteristic H may be observed with the output device 18.

Next, STEP 12 will be described.
This is a step in which the resonance frequency estimation unit 24 is involved, and is a step of estimating the resonance frequency from the mechanical system transfer characteristic, and a process of detecting resonance included in the mechanical system measured transfer characteristic of the electric motor 1 and the machine 5 obtained in STEP 11. I do.
In mechanical measured transmission characteristics Ha, a, Hc, a, Hb, a and Ha, b, Hc, b, Hb, b obtained in step STEP11, irregularities appear, and the convex vertices are resonances. The magnitude (height) of the resonance differs depending on each mechanical system measured transmission characteristic, but each mechanical system measured transmission characteristic Ha, a, Hc, a, Hb, a and Ha, b, Hc, b, Hb, b Therefore, the resonance frequency estimation unit 24 can estimate the resonance frequency.

The next STEP 13 is a step in which the filter estimator 25 is involved, selects candidates from the resonance frequencies and calculates the parameters of the driving force filter unit 10, and selects the resonance frequency candidates. Calculate the reflected parameters.
Here, the frequency range is determined so as not to interfere with the parameter (gain) of the controller 3, and the mechanical system transfer characteristic having a large resonance frequency is selected.

The next STEP 14 is a step in which the filter effect simulation unit 13 is involved and simulates the combined characteristics of the driving force filter unit and the mechanical system. Here, the driving force filter unit is modeled on the mechanical system actually measured transmission characteristic obtained in STEP 11. The result of applying the converted driving force filter unit model is calculated. Here, processing is performed using the parameters obtained in STEP13.
That is, the filter effect simulation unit 13 drives the parameters obtained in STEP 13 to the mechanical measured transmission characteristics Ha, a, Hc, a, Hb, a and Ha, b and Ha, b, Hc, b, Hb, b. Simulates the case of applying the force filter part model.

Unlike the first and second embodiments, since the mechanical measured transmission characteristics Hb, a and Ha, b, Hc, b are also obtained, the detection of the Y axis when the X axis motor 1b is operated. With respect to the influence on the parts 2a and 2c and the influence on the X-axis detection part 2b when the Y-axis motor 1a operates, the effect of the presence or absence of the filter can be examined in advance.
For this reason, it is possible to study the setting of a filter that is effective in removing inter-axis interference.
If the calculated result is graphed and compared in the output device 18, it can be confirmed how effective the driving force filter unit 10 is with respect to the characteristics of the single axis or the characteristics between the XY axes.

STEP 15 is a step in which the calculation unit 26 is involved and simulates the combined characteristics of the motor control device and the mechanical system. The composite characteristic is calculated using the mechanical system transfer characteristic obtained in STEP 11, the driving force filter unit model, and the controller model which is a numerical model of the controller.
The Y axis of the fully closed configuration uses the equations (11), (12), (13), (14), and (15) shown in the second embodiment, and the X axis of the semi closed configuration uses the first embodiment. Formulas (5), (7), and (8) shown in the example are used. The mechanical system transfer characteristics ha in the equations (11), (12), (13), and (14) may be replaced with Ha, a, hb with Hc, a, and the response rb of the detector 2b with rc. Further, in the present embodiment, as in the second embodiment, the characteristic G of the controller 3 is divided into the characteristic G1 of the position control unit and the characteristic G2 of the speed control unit, so that the equations (5) and (7) , (8) are rewritten into equations (16), (17), and (18).
The command response Zr is expressed by equation (16), the disturbance response Zd is expressed by equation (17), and the loop-opening loop frequency characteristic Zo is expressed by equation (18).

Zo = − (G1 + b) · G2 · (f · Hb, b) Equation (18)
Here, the characteristic calculation unit 26 measures the measured mechanical transmission characteristics Ha, a, Hc, a, Hb, b, the driving force filter unit model characteristic f, the position control unit characteristic G1, and the speed control unit characteristic. G2 is used to calculate a combined characteristic of the electric motor control device and the mechanical system actual transmission characteristic.

STEP 16 is a step in which the filter effect calculation unit 14 is involved, and is a step of confirming the combined characteristics of the driving force filter unit and the mechanical system transmission characteristics. The driving force filter unit is applied and evaluated by an actual machine. The driving force filter unit 10 actually acts, and the parameters determined in STEP 13 are set in the driving force filter unit 10 for execution.
The switches 9b and 9e are set so that the command signal C (Ca, Cb) is given to the current control units 6a and 6b via the driving force filter units 10a and 10b, and the filter effect calculation unit 14 is actually driven. A mechanical system transfer characteristic H ″ acted by the force filter unit 10 is calculated.

The next STEP 17 is a step in which the second characteristic calculation unit 27 is involved, and is a step for simulating the combination characteristic of the motor control device and the mechanical system. The combination of the driving force filter unit 10 obtained in STEP 16 and the actual measured transmission characteristic of the mechanical system The composite characteristics are calculated using the characteristics and the controller model.
Here, the command response Zr of the closed loop frequency characteristic, the disturbance response Zd of the closed loop frequency characteristic, and the one-round open loop frequency characteristic Zo are obtained, and the operation in the actual machine is evaluated.
As described above, the optimum parameter of the driving force filter unit is obtained, and it is evaluated by simulation and realization, so that the filter setting of the motor control device can be realized.

In the third embodiment, in step STEP11, when measuring the mechanical transmission characteristics Ha, a, Hc, a, Hb, a and Ha, b, Hc, b, Hb, b, the motors 1a and 1b are measured. However, if the two types of commands C of the commander 4 are uncorrelated with each other, the motors 1a and 1b may be operated simultaneously and measured.
This method will be described below.
For example, the motor 1a is operated with a random wave Ca, the motor 1b is operated with an M-sequence wave Cb, and responses ra, rb, rc are obtained from the three detectors 2a, 2b, 2c, and their mechanical system transfer characteristics. Is calculated. Here, the mechanical characteristic calculation unit 11 obtains the mechanical system transfer characteristic H from the auto power spectrum and the cross spectrum obtained by performing frequency analysis using FFT (Fast Fourier Transform) and averaging.
The mechanical system transfer characteristic H is the relationship of the equation (19) shown as a simulation.

If the command signal Ca, Cb on the left side is converted to a square matrix and multiplied by the inverse matrix, the matrix of the mechanical system transfer characteristic can be solved. Therefore, Ha, a, Hc, a, Hb, It can be seen that a, Ha, b, Hc, b, and Hb, b are obtained.

Where A *; A complex conjugate
[A] ^-1 ; Shows the inverse matrix of [A].

That is, from the averaged auto power spectrum and cross spectrum, the matrix of the mechanical system transfer characteristic H can be solved by Equation (21).

Where G'ra, ca; cross-spectrum averaged between response ra and command signal Ca
G'rb, ca: averaged cross spectrum of response rb and command signal Ca
G'rc, ca; averaged cross spectrum of response rc and command signal Ca
G'ra, cb; averaged cross spectrum of response ra and command signal Cb
G'rb, cb: averaged cross spectrum of response rb and command signal Cb
G'rc, cb; averaged cross spectrum of response rc and command signal Cb
G'ca, ca; averaged auto power spectrum of command signal Ca
G'ca, cb; averaged cross spectrum of command signal Ca and command signal Cb
G'cb, ca; averaged cross spectrum of command signal Cb and command signal Ca
G′cb, cb; shows an averaged auto power spectrum of the command signal Cb.

  Filter parameters are obtained from mechanical system transfer characteristics obtained by measurement with an actual machine, and an optimum filter can be set by evaluating the numerical model and the actual machine. These functions can be applied to all motor control devices.

1 is a configuration diagram of an electric motor control device showing a first embodiment of the present invention. Flowchart of filter setting of the motor control device showing the first embodiment of the present invention. The figure which shows the relationship of the characteristic of the mechanical system which shows 1st Example of this invention, and a driving force filter part. Configuration diagram of an electric motor control apparatus showing a second embodiment of the present invention. The figure which represented the mechanical system of the motor and the machine with the transfer function in the block diagram of the motor control apparatus which shows 2nd Example of this invention. Flowchart of filter setting of motor control apparatus showing second embodiment of the present invention. Configuration diagram of an electric motor control apparatus showing a third embodiment of the present invention. The figure which represented the mechanical system of the motor and the machine with the transfer function in the block diagram of the motor control apparatus which shows 3rd Example of this invention. Block diagram of feed axis parameter adjustment system of first conventional example Explanatory drawing of the machine model of the feed axis parameter adjustment system of the first conventional example The figure which shows the adjustment procedure of the feed axis parameter adjustment system of the machine tool of the 1st prior art example Block diagram of the second conventional notch filter automatic adjustment device The block diagram which shows the structure of the center frequency setting device and coefficient setting device of the notch filter automatic adjustment method of the 2nd prior art example

Explanation of symbols

1 Electric motor (1a, 1b)
2 detectors (2a, 2b, 2c, ..., 2m)
3 Controller (3a, 3b)
4 Commander 5 Machine 6 Current control part (6a, 6b)
8 Adder (8a, 8b)
9 switches (9a, 9b, 9c, 9d, 9e, 9f)
10 Driving force filter section (10a, 10b)
DESCRIPTION OF SYMBOLS 11 Mechanical characteristic calculation part 13 Filter effect simulation part 14 Filter effect calculation part 15 Position control part (15a, 15b)
16 Speed controller (16a, 16b)
17 Conversion unit (17a, 17b)
18 output device 19 storage device 20 input device 22 mechanical characteristic model filter effect simulation unit 23 calculation unit 24 resonance frequency estimation unit 25 filter estimation unit 26 characteristic calculation unit 27 second characteristic calculation unit 29 sound pressure detection unit 30 sound pressure reproduction device 31 simulation characteristic calculation unit 32 actual machine characteristic confirmation unit 33 filter constant estimation unit 36 filter model estimation unit 50 filter setting unit 51 filter model setting unit
(Conventional)
DESCRIPTION OF SYMBOLS 100 Parameter adjustment system 101 CNC apparatus 102 Changeover switch 103 Servo controller 104 Motor 105 Machine tool 106 Measuring instrument 107 Feedforward process part 108 Position control apparatus.
109 Machine model
(Conventional)
201 Servo motor 202 Servo amplifier 203 Speed command input unit 204 Function generator 205 Switch 206 Speed amplifier 207 Digital notch filter 208 Switch 209 Center frequency setting unit 210 Coefficient setting unit 211, 212 Switch 213 Current command setting unit 214 Amplifier

Claims (19)

A command device for generating a command signal;
A controller that receives the command signal and outputs a drive command;
A driving force filter unit that filters the drive command;
A current control unit that drives an electric motor connected to the machine using the output of the driving force filter unit as an input;
In the motor control device configured to feed back the operation amount of the machine to the controller,
A filter setting unit for calculating parameters of the driving force filter unit;
The filter setting unit
Cutting off the feedback, measuring the input of the current control unit and the operation amount, and calculating a measured transmission characteristic of a mechanical system including the electric motor and the machine;
A filter constant estimator for estimating a parameter to be set in the driving force filter from the measured transmission characteristics of the mechanical system;
Using the driving force filter unit model that numerically models the driving force filter unit and applying the parameters, and the measured transmission characteristics of the mechanical system, a simulation characteristic calculation unit that simulates combined operation characteristics,
Applying the parameters evaluated by the simulated characteristic calculation unit, an actual machine characteristic confirmation unit that obtains a combined operation characteristic by an actual machine,
An electric motor control device comprising: The filter constant estimator is
A resonance frequency estimator for estimating the resonance frequency from the measured transmission characteristics of the mechanical system;
The motor control device according to claim 1, further comprising: a filter estimation unit that selects a candidate from the resonance frequency and calculates the parameter reflecting the candidate. The simulated characteristic calculator
Using the driving force filter unit model and the measured transmission characteristic of the mechanical system, a filter effect simulation unit that simulates a combined operation characteristic of the driving force filter unit and the mechanical system,
A characteristic that simulates a combined operation characteristic of the motor control device and the mechanical system using a controller model obtained by numerically modeling the controller, the driving force filter unit model, and an actual measurement transmission characteristic of the mechanical system. The motor control device according to claim 1, further comprising: an arithmetic unit. The actual machine characteristic confirmation unit
A filter effect calculation unit that turns off the feedback and measures the input and operation amount of the driving force filter unit, and obtains a combined operation characteristic of the driving force filter unit and the mechanical system by an actual machine;
And a second characteristic calculator that simulates the combined operation characteristics of the motor control device and the mechanical system using the controller model and the combined operation characteristics evaluated by the filter effect calculation unit. The electric motor control device according to claim 1, wherein The filter setting unit further includes a filter model setting unit,
The filter model setting unit
A calculation unit that calculates a mechanical transfer characteristic model that is a numerical model of the mechanical system by numerical calculation;
A filter model estimation unit for estimating a parameter to be set in the driving force filter unit from the mechanical system transfer characteristic model;
Using the driving force filter unit model and the mechanical system transfer characteristic model in which the parameters are set, a mechanical characteristic model filter effect simulation unit that simulates a combined operation characteristic of the driving force filter unit and the mechanical system,
Using the controller model, the driving force filter unit model, and the mechanical system transfer characteristic model, and a model characteristic calculation unit that simulates a combined operation characteristic of the electric motor control device and the mechanical system. The motor control device according to claim 1.   The characteristic calculation unit, the second characteristic calculation unit, and the model characteristic calculation unit calculate at least one of a closed-loop frequency characteristic command response, a closed-loop frequency characteristic disturbance response, and an open-loop frequency characteristic. The motor control device according to claim 3, wherein the motor control device is a motor control device.   The motor control device according to claim 1, wherein the operation amount is a position, speed, or acceleration detected by a detector attached to the machine or the motor.    The operation amount is detected at a plurality of locations of the machine or the electric motor, mechanical system measured transmission characteristics at the plurality of locations are calculated, and at least one of the plurality of locations is fed back. The motor control device described.   The motor control device according to claim 1, wherein the operation amount detects a sound pressure generated from the machine.   2. The motor control device according to claim 1, wherein the drive command is a torque command when the motor is a rotary motor, and a thrust command when the motor is a translational motor.   The said motor control apparatus is provided with two or more of the said controllers, the said driving force filter part, and the said current control part, The said filter setting part has a function of two or more systems, The Claim 1 characterized by the above-mentioned. Electric motor control device.   The motor control device according to claim 1, wherein the command signal is one of a swept sine wave, an M-sequence signal, and a random wave. An input device for setting parameters in the driving force filter unit;
An output device for observing the measured transmission characteristics and the combined operation characteristics of the mechanical system;
A storage device for storing the measured transmission characteristics of the mechanical system and the combined operation characteristics;
The motor control device according to claim 1, wherein the output device includes a sound pressure reproducing device. A command device that generates a command signal, a controller that receives the command signal and outputs a drive command, a driving force filter unit that filters the drive command, and an output of the driving force filter unit as an input to the machine An electric motor control device configured to feed back an operation amount of the machine to the controller, and a current control unit that drives a connected electric motor;
When calculating the parameters of the driving force filter unit,
Cutting off the feedback, measuring the input of the current control unit and the operation amount, and calculating an actual transmission characteristic of a mechanical system including the electric motor and the machine;
Estimating a parameter to be set in the driving force filter unit from the measured transmission characteristics of the mechanical system;
Using the driving force filter unit model obtained by numerically modeling the driving force filter unit and the measured transmission characteristics of the mechanical system, applying the parameters, and evaluating the combined operation characteristics by simulation;
Applying the parameters evaluated in the step of evaluating the combined operation characteristics by the simulation, and evaluating the combined operation characteristics by an actual machine,
A filter setting method for an electric motor control device, comprising:   15. When estimating the parameter, a resonance frequency is estimated from an actually measured transfer characteristic of the mechanical system, a candidate is selected from the resonance frequency, and the parameter reflecting the parameter is calculated. The filter setting method of the electric motor control apparatus of description. When evaluating the combined operation characteristics by the simulation,
Using the driving force filter part model and the measured transmission characteristics of the mechanical system, the combined operation characteristics of the driving force filter part and the mechanical system are evaluated by simulation,
Using the controller model obtained by numerically modeling the controller, the driving force filter unit model, and the measured transmission characteristics of the mechanical system, the combined operation characteristics of the motor control device and the mechanical system are evaluated by simulation. The filter setting method for an electric motor control device according to claim 14, wherein: When evaluating the combined operation characteristics using the actual machine, the feedback is turned off, the input of the driving force filter unit and the operation amount are measured, and the combined operation characteristics of the driving force filter unit and the mechanical system are evaluated using the actual machine. And
The combined operation characteristics of the motor control device and the mechanical system are evaluated by simulation using the controller model, and the combined operation characteristics of the driving force filter unit and the mechanical system evaluated by an actual machine. Item 15. A filter setting method for an electric motor control device according to Item 14. Calculating a mechanical transfer characteristic model that is a numerical model of the mechanical system by numerical calculation;
Estimating a parameter to be set in the driving force filter unit from the mechanical system transfer characteristic model;
Using the driving force filter unit model and the mechanical characteristic model in which the parameters are set, and evaluating a combined operation characteristic of the driving force filter unit and the mechanical system by simulation;
Using the controller model, the driving force filter unit model, and the mechanical system transfer characteristic model to evaluate the combined operation characteristics of the motor control device and the mechanical system by simulation,
15. The filter setting method for an electric motor control device according to claim 14, further comprising: When evaluating the combined operation characteristics of the motor control device and the mechanical system by simulation,
19. The motor control device according to claim 16, wherein at least one of a command response of a closed loop frequency characteristic, a disturbance response of the closed loop frequency characteristic, and a single loop frequency characteristic is calculated. Filter setting method.