WO2008053772A1 - Motor control device, and its control method - Google Patents

Abstract

Provided are a motor control device, which can estimate an inertial moment highly precisely within a small working range and even in case the gradient of a low frequency is not fixed due to the influence of friction or control, and a control method for the device. The motor control device comprises a speed signal generating unit (2) for generating a speed signal from a position signal, and a current control unit (1) for controlling a motor current on the basis of a torque command. Further comprised are a test torque command generating unit (3) for generating a test torque command containing multiple frequency components, a frequency characteristic calculating unit (4) for calculating the frequency characteristics from a response of the speed signal to the test torque command, and a mechanical parameter calculating unit (6) for calculating a mechanical parameter from the frequency characteristics.

Description

 Specification

 Motor control device and control method thereof

 Technical field

 The present invention relates to a motor control device that can estimate a machine model.

 Background art

 Conventionally, in order to identify the moment of inertia of a machine, a certain range is actually driven by the operation of a specified pattern, and the inertia moment is obtained based on the torque command and speed output at that time. (Patent Document 1)

 In addition, the moment of inertia may be obtained based on the low frequency slope of the frequency characteristics. (Patent Literatures 2 and 3)

 By using the value of this moment of inertia, it is possible to optimally adjust the parameters of the motor control device, and to improve the machine characteristics by using it in a feedforward controller or an observer model that performs vibration suppression control. Was made.

 Patent Document 1: JP-A-9 182479

 Patent Document 2: Japanese Patent Laid-Open No. 2002-304219

 Patent Document 3: Japanese Unexamined Patent Publication No. 2003-79174

 Disclosure of the invention

 Problems to be solved by the invention

[0003] In the conventional machine model estimation device, in the case of the prior art of Patent Document 1, since a certain range is actually driven by the operation of the specified pattern, the identification operation is performed when the machine operation range is not obtained. There were cases where it was not possible. In addition, since the conventional techniques of Patent Documents 2 and 3 use the method of obtaining the moment of inertia based on the low-frequency slope of the frequency characteristics, sufficient identification accuracy is obtained when the low-frequency slope is not constant due to the effects of friction and control. In some cases, this was not possible.

The present invention has been made in view of such problems, and a motor control device capable of estimating the moment of inertia with high accuracy even when the low-frequency gradient is not constant due to the influence of friction and control in a small operating range. An object is to provide a control method thereof. Means for solving the problem

[0004] In order to solve the above problem, the present invention is configured as follows.

 The invention described in claim 1 is a motor control device comprising: a speed signal generation unit that generates a speed signal from a position signal; and a current control unit that controls a motor current based on a torque command! ! /, A test torque command generation unit that generates a test torque command including a number of frequency components, an actual machine frequency characteristic calculation unit that calculates a real machine frequency characteristic from the response of the speed signal to the test torque command, And a machine section that generates machine parameters from actual machine frequency characteristics.

 According to a second aspect of the present invention, in the motor control device according to the first aspect, a position control unit that generates a speed command from the position command and the position signal, the speed command, the speed signal force, and the torque command And a speed control unit to be generated. The invention according to claim 3 is the motor control apparatus according to claim 1, wherein the actual frequency characteristics include a resonance frequency and its amplitude, and an anti-resonance frequency and its amplitude.

 According to a fourth aspect of the invention, in the motor control device according to the first and second aspects of the present invention, the mechanical parameter generator repeatedly changes the total moment of inertia and braking coefficient of the two-inertia formula model repeatedly to slightly change the two-inertia formula model frequency. The characteristics substantially match the actual machine frequency characteristics, and the total inertia moment and the braking coefficient are determined as parameters.

 The invention according to claim 5 is the motor control device according to claim 4, wherein the two-inertia mathematical formula model is expressed by equation (1), and the gain of the intermediate frequency between the antiresonance frequency and the resonance frequency is expressed by equation (2). It is characterized by that.

[0005] Country (1)

ここで Jは総慣性モーメント、 ω は共振周波数、 ω は反共振周波数、 ωはテスト周波[0006] [Equation 2]

Where J is the total moment of inertia, ω is the resonance frequency, ω is the anti-resonance frequency, and ω is the test frequency

H L

The number, ζ is the braking coefficient, and S is the Laplace operator.

 According to a sixth aspect of the present invention, in the motor control device according to the fifth aspect, the machine parameter generation unit has a gain in a range from the anti-resonance frequency to the resonance frequency of the actual machine frequency characteristic and the two-inertia mathematical formula model frequency characteristic. In comparison, if the gain of the actual machine frequency characteristic is large, the total moment of inertia of the two-inertia mathematical formula model is slightly increased, and if it is small, it is slightly decreased, and the process is repeated until they substantially coincide.

 The invention according to claim 7 is the motor control device according to claim 6, wherein the magnitude of the gain is determined by comparing gain areas from an anti-resonance frequency to a resonance frequency.

 The invention according to claim 8 is the motor control apparatus according to claim 6, wherein the initial value of the total moment of inertia is expressed by equation (3).

[Equation 3]

j ₌ 2Α½ · (¾¾ ( ₃ )

( _ff )

Where Y is the gain in terms of db of the anti-resonance frequency of the actual machine frequency characteristics, and Y is the resonance frequency.

 L H

db conversion gain.

 The invention according to claim 9 is the motor control apparatus according to claim 6, characterized in that the initial value of the total inertia moment is expression (5).

] = (Ω / ω) / (ω · Υ) (4)

 Η L L Μ

Where Υ is the gain when ω = ^ (ω · ω) of the actual frequency characteristics.

 M H L

The invention according to claim 10 is a control of a motor control device comprising: a speed signal generation unit that generates a speed signal from a position signal; and a current control unit that controls a motor current based on a torque command! Compare the actual machine frequency characteristics with the two-inertia equation model frequency characteristics by inputting a test torque command containing a large number of frequency components to the current control unit and determining the actual machine frequency characteristics. A step of repeatedly finely correcting the total inertia moment and the braking coefficient so that the gain of the actual machine frequency characteristics and the two-inertia formula model frequency characteristics match, and if the gains substantially match, the total moment of inertia and the number of braking systems are Parameters And a step of determining as a special feature.

 The invention's effect

 [0008] According to the invention described in claim 1, when the moment of inertia is identified using the frequency analysis result in the open loop, the frequency characteristic in the low frequency region is lowered due to viscous friction or the influence of the controller. It is possible to provide a motor control device that can accurately identify such objects.

 According to the second aspect of the present invention, when the moment of inertia is identified using the frequency analysis result in the closed loop configuration controlled by the controller, the frequency characteristics in the low frequency region due to the influence of viscous friction and the controller. It is possible to provide a motor control device that can identify with high accuracy even for objects that fall.

 According to the inventions of claims 3 to 9, it is possible to provide a motor control device capable of estimating the moment of inertia with high accuracy even when the low frequency inclination is not constant due to the influence of friction and control within a small operating range.

 According to the invention of claim 10, it is possible to provide a control method for a motor control device that can estimate the moment of inertia with high accuracy even when the inclination of the low frequency is not constant due to the influence of friction and control within a small operating range.

 Brief Description of Drawings

 [0009] FIG. 1 is a block diagram showing a configuration of the present invention.

 FIG. 2 is a block diagram showing the configuration of the present invention.

 FIG. 3 is a flowchart showing the method of the present invention.

 [Fig. 4] Diagram showing the curve fitting method of the present invention.

 [Fig.5] Diagram showing identification results by conventional method

 [Fig. 6] Diagram showing the identification results according to the present invention.

 Explanation of symbols

[0010] 1 Current control unit

 2 Speed signal generator

 3 Test torque command generator

4 Frequency characteristics calculator 5 Machine parameter calculator

 6 Machine parameters

 7 Position controller

 8 Speed controller

 11 Motor

 12 Position detector

 13 Machine

 21 Frequency response obtained by measurement

 22 Frequency response of two-inertia model

 23 Starting point of evaluation

 24 Evaluation end point

 25 Frequency characteristics of rigid system model

 BEST MODE FOR CARRYING OUT THE INVENTION

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

 Example 1

FIG. 1 is a block diagram showing the configuration of the first exemplary embodiment of the present invention. In the figure, 1 is a current control unit, 2 is a speed signal generation unit, 3 is a test torque command generation unit, 4 is a frequency characteristic calculation unit, 5 is a machine model calculation unit, 6 is a machine model, 11 is a motor, and 12 is a position. Detector 13 is a machine. The current control unit 1 converts the torque command into a current command, performs a PID control process on the current deviation between the current command and the motor current, generates a voltage command, drives the power converter by PWMing the voltage command, and driving the motor To supply power. The speed signal generation unit 2 generates a speed signal by taking the time difference of the position signals of the position detector coupled to the motor. The test torque command generation unit 3 generates a torque command including many frequency components and inputs the torque command to the current control unit 1 when the motor control device is not in the normal operation mode but in the test mode. The frequency characteristic calculation unit measures the speed signal when the test torque command is input to the current control unit 1, and calculates the frequency characteristic. The machine model calculation unit 5 estimates the combination of resonance frequency and anti-resonance frequency as the frequency of peaks and valleys in the frequency characteristics, extracts some candidates, and discriminates the two-inertia system to be modeled. FIG. 3 shows a flowchart of the method of the present invention. As shown in the figure, the method of the present invention performs processing in seven steps, steps 1-7. In Step 1, a test torque command including many frequency components is input to the current control unit, and the response of the speed signal is measured. In step 2, the frequency characteristics of the machine are calculated based on the input test torque command and the response of the measured speed signal. In Step 3, detect the valleys and valleys from the frequency characteristic calculation result. A combination of resonance frequency and antiresonance frequency as a two-inertia model is estimated, and several candidates are extracted. In step 4, depending on the objective, a set of two inertial systems to be modeled is determined. For example, when obtaining the moment of inertia of the rigid mode of the entire system, select the lowest frequency. For the purpose of adjusting damping control for high-frequency vibration, select the resonance frequency in question and the corresponding anti-resonance frequency. If there are many resonance frequencies and automatic discrimination is difficult, or if the target frequency has been set directly! /, Select the resonance frequency and anti-resonance frequency manually in step 5 if necessary. In Step 6, the model is adjusted by fitting a curve of the two-inertia model to the selected combination of resonant and anti-resonant frequencies. In step 7, in addition to the resonance frequency and antiresonance frequency, the moment of inertia and vibration damping are identified from the model after evaluation. The dotted line portion 8 is a portion corresponding to Patent Document 3.

 [0014] Hereinafter, step 6 and step 7 will be described in detail.

 Assuming that the resonance frequency ω Η, the anti-resonance frequency co L, the damping ζ, and the total moment of inertia J, the transfer function from the torque to the speed of the two-inertia system model is expressed by equation (1).

 Expressing this in the frequency domain as a db-converted gain Η (ω), it is expressed by equation (2) as a function of the frequency ω. Assuming that the attenuation ζ is sufficiently small, the intermediate point between the resonance frequency ω 反 and the anti-resonance frequency ω L is approximately expressed as equation (5).

[0015] [Equation 4]

If the anti-resonance frequency gain YL and the resonance frequency gain YH obtained by frequency response calculation are used, the total moment of inertia J can be calculated by equation (3). Gain is the average of several points It is good to use.

 [0016] Approximately obtained by Equation (5) and ζ = 0 as the initial value. The parameter is identified by fitting the curve using the model using Equation (2) and the frequency response calculation result obtained by measurement. To do. Various methods, such as the least square method and genetic algorithm, can be applied to curve fitting. Here, the area comparison between the antiresonance frequency and the resonance frequency as shown in Fig. 4 and the midpoint comparison will be explained. The figure shows the actual machine frequency characteristic 1 obtained by measurement, the two-inertia formula model frequency characteristic 2, the evaluation start point 3, and the evaluation end point 4. 3 corresponds to the anti-resonance frequency and 4 corresponds to the resonance frequency. The adjustment conditions are as follows.

 First, the damping coefficient ζ is adjusted under the condition of Equation (6), assuming the gain Υ (ω) obtained by frequency characteristic calculation.

ここで、 δは任意の調整値、また、総慣性モーメント Jについては、以下の条件で調 整する。 [0017] [Equation 5]

Here, δ is an arbitrary adjustment value, and the total moment of inertia J is adjusted under the following conditions.

[0018] [Equation 6]

 Here, σ is an arbitrary adjustment value. When the model is limited to the two-inertia system, the total inertia moment can be easily obtained by the condition of Equation (7), so the attenuation coefficient can be obtained by judging the coincidence of the frequency characteristics obtained by the model and measurement. .

Figures 3 and 4 show examples of specific identification results when the inertial moment of the rigid body mode of the entire system is obtained. Fig. 3 shows the results of identification using the conventional method. FIG. 6 is a diagram showing an identification result according to the present invention. For the measured frequency characteristic 1, the model identified by curve fitting is 5 for the conventional example and 2 for the model according to the present invention. In contrast to the machine with a true inertia moment of 60 times the true value of the rigid body mode, the conventional method was identified as 70 times the average slope of the low frequency region. 55. Example 2 shows that the moment of inertia can be identified with higher accuracy than the conventional method.

 FIG. 2 is a block diagram showing the configuration of the second embodiment of the present invention. Position control unit 7 and speed control unit 8 are added to the first embodiment!

 In addition, to identify the total moment of inertia, the frequency at the midpoint between the resonance frequency and the antiresonance frequency can be selected as ω = ^ (ω * ω). at frequency ω when ζ = 0

 M L Η Μ

 2 The gain Η '(ω) of the inertial model is expressed by equation (8).

 Η '(ω) = I G' (] ω) | = (ΐ /] / ω) * ^ (ω / ω) (8)

 L Η L

 If the gain at the midpoint of the actual machine frequency characteristics is YM, the total moment of inertia J is expressed by equation (9).

 ] = (Ω / ω) / ω / Υ (9)

 Η L L Μ

 If you use this as an initial value and identify it.

 [0020] Since the inertia moment to be obtained uses the frequency and gain of the resonance point and anti-resonance point! /, Since the spring constant and inertia moment of each spring element are also required in modeling of the multi-inertia mechanism, Industrial applicability that can be applied to optimum filter settings

[0021] The motor control device of the present invention has few! /, The influence of friction and control is large in the operating range! /, And the moment of inertia can be estimated with high precision in some cases, so that general industrial machines such as robots and machine tools can be estimated. We can expect application to. The present invention can also be applied as a force inertia moment estimation device premised on being incorporated in a motor control device.

Claims

The scope of the claims  [1] A motor control device having a speed signal generation unit that generates a speed signal from a position signal and a current control unit that controls a motor current based on a torque command!  A test torque command generating unit that generates a test torque command including a number of frequency components; an actual machine frequency characteristic calculating unit that calculates an actual machine frequency characteristic from a response of the speed signal to the test torque command;  A motor control device comprising: a machine parameter generation unit that generates a machine parameter from the actual machine frequency characteristic.  2. A position control unit that generates a speed command from the position command and the position signal, and a speed control unit that generates the torque command from the speed command and the speed signal. The motor control apparatus described.  3. The motor control device according to claim 1, wherein the actual machine frequency characteristic includes a resonance frequency and its amplitude, and an anti-resonance frequency and its amplitude.  [4] The machine parameter generator repeatedly changes the total moment of inertia and braking coefficient of the inertial mathematical model repeatedly so that the inertial mathematical model frequency characteristics substantially match the actual machine frequency characteristics, and the total inertia moment and the braking coefficient 3. The motor control device according to claim 1, wherein the parameter is determined as a parameter. 5. The motor control device according to claim 4, wherein the inertial equation model is expressed by equation (1), and a gain at an intermediate frequency between the antiresonance frequency and the resonance frequency is expressed by equation (2). ) ² (1) s I ω _Η- }

 Where J is the total moment of inertia, ω is the resonance frequency, ω is the anti-resonance frequency, and ω is the test frequency  H L  The number, ζ is the braking coefficient, and S is the Laplace operator. [6] The machine parameter generation unit compares the gain in the range from the anti-resonance frequency to the resonance frequency of the actual machine frequency characteristic and the inertia system mathematical model frequency characteristic, and if the gain of the actual machine frequency characteristic is large, the inertia system Slightly increased the moment of inertia of the mathematical model And if it is less, it is reduced slightly and repeated until it substantially matches. 5. The motor control device according to 5. 7. The motor control device according to claim 6, wherein the magnitude of the gain is determined by comparing gain areas from an antiresonance frequency to a resonance frequency. 8. The motor control device according to claim 6, wherein the initial value of the total moment of inertia is expressed by equation (3). j ₍₃₎

 Where Υ is the db equivalent gain of the anti-resonance frequency of the actual machine frequency characteristics, and Y is the resonance frequency  L H  db conversion gain.  9. The motor control device according to claim 6, wherein an initial value of the total moment of inertia is Equation (5).  ] = (Ω / ω) / (ω · Υ) (4)  Η L L Μ  Where Υ is the gain when ω = ^ (ω · ω) of the actual frequency characteristics.  M H L  [10] A method for controlling a motor control device including a speed signal generation unit that generates a speed signal from a position signal and a current control unit that controls a motor current based on a torque command! A step of inputting a test torque command including a frequency component of  The step of comparing the actual machine frequency characteristic and the two-inertia formula mathematical model frequency characteristic, and the step of repeatedly finely correcting the total inertia moment and the braking coefficient so that the gains of the actual machine frequency characteristic and the two-inertia formula mathematical model frequency characteristic are matched. When,  A control method for a motor control device, comprising: a step for determining the total moment of inertia and the number of braking systems as parameters when the gains substantially coincide with each other.