JPS6082078A - Gain automatic correcting method of speed control system of motor - Google Patents

Abstract

PURPOSE:To automatically improve a servo characteristic of a speed control system when the inertial moment of a load is varied by varying the gain of the system in response to the quantity of variation of the speed signal of a motor for a sinusoidal speed command signal. CONSTITUTION:An inverter controller 14 controls an inverter 13 in response to a deviation between a speed command signal Nref and a speed feedback signal N. The gain KN of a speed amplifier is predetermined from parameters such as a gain of a speed control system, or an inertial moment of a motor in a microcomputer 15 in the initial state. When a tool or a work is mounted, the microcomputer vibrates the speed command signal in the prescribed width in sinusoidal wave state, the frequency of the vibration is varied to measure the inertial moment of a load from the followup of the amplitude or the phase, and the gain KN of the speed amplifier is corrected to KN' on the basis of the measured value.

Description

TECHNICAL FIELD The present invention relates to a speed control system for an electric motor, and more particularly to a method for improving the servo characteristics of a speed-side 111 system when the moment of inertia of a load fluctuates.

BACKGROUND OF THE INVENTION FIG. 1 is a block diagram of a general speed control 'Irl1 system of an electric motor. The speed command signal Nref output from the speed command section 1 is subtracted by the speed feedback signal Nfb from the speed detector 5 connected to the rotating shaft of the electric motor 4 by the scale calculator 6, and then input to the speed amplifier 2. The output becomes a torque command signal and drives the electric motor 4 by the torque control device B.

FIG. 2 is a block diagram of the speed control system shown in FIG. 1. K
N is the gain of speed amplifier 2 (''/v), KT is the command-torque conversion coefficient of torque controller B (Kg・m7'v)
), TL is load torque, GD2 is (motor 4 + load)
S is the Laplace operator, N is the motor speed, and KTa is the speed detection conversion coefficient (V/rpm) of the speed detector 5. Here, τ in the command-torque variation J9i constant of the torque controller 6 does not necessarily have to be linear.

From the block diagram of FIG. 2, the transfer function regarding speed is. Here, Ko=1- (2) IIIG. Therefore, the transient response of the speed control system is determined by the time constant T
. represented by. And this time constant T.

is proportional to the moment of inertia GD2 from equation (3), so
As the load component of the moment of inertia GD2 increases, the transient response worsens. Such a reaction of the moment of inertia GD2 occurs when an electric motor is used to move the main shaft 1) of a machine tool, for example, and a tool is changed during operation. Conventionally,
In normal applications, the gain KN of the speed amplifier 2 is set to the moment of inertia GD2 when the load is small, so that the motor continues to operate even if the moment of inertia GD2 increases. Therefore, as mentioned above, the moment of inertia l-C
When D2 becomes large, the transient characteristics deteriorate and the drive system becomes unstable during positioning (1i) control and P-work control.

Here, a phenomenon in which the drive system becomes unstable due to the moment of inertia (CD2) during positioning control will be explained. FIG. 6 is a block diagram during positioning control, in which a position control loop is added to the block diagram of FIG. 2. Pre
f is a motor position command signal, Pout is a motor position signal,
Kpc is the position detection mosquito displacement coefficient (v/rad
), 8 is an electric motor position signal P which is a position feedback signal from the electric motor position command signal P rcf. A subtractor for subtracting ut, K1 is the transfer function of the position-based 'l]1 device.

From the block diagram in Figure 6, the transfer function regarding position is: Kv = KO-KILlo HK p+, +
--(5).

The characteristic equation is YuO82+±S+]-〇・(6) K v K v
'1Δ]7−ζ, then 1. Z3T2.7 T22S' l-2ζT2S -11,= O・ ((
ii'.

In a servo system, it is said that the optimum performance is achieved when the damping rate ζ = 0.5, that is, Kv-±TO, and the reduction ψ↓・
When tζ becomes smaller, the overshoot L soil with respect to the step input becomes larger, and the culm becomes oscillatory.

Now, assume that the moment of inertia GD2 has increased by ΔGD2, and the time constant at this time is T. ', and the damping rate is ζ', then ζ' = 8 shields - 20, 5 (8), that is, the value of the damping rate ζ' becomes smaller than 0.5, and the system vibrates. become a target. Conventionally, the system is operated without changing the gain KN of the speed amplifier 2 even if the inertia motor (inertia motor) changes, so as mentioned above, the system becomes imitative and stable positioning control is difficult. Met.

OBJECTS OF THE INVENTION Accordingly, an object of the present invention is to provide a method for automatically improving the servo characteristics of a speed control system when the moment of inertia of a load varies.

DESCRIPTION OF THE INVENTION According to the present invention, the gain of the speed control system is changed in accordance with the change meter of the speed signal of the electric motor with respect to the sinusoidal speed command signal.

DESCRIPTION OF EMBODIMENTS The present invention will now be described with reference to drawings of embodiments. FIG. 4 shows a configuration N of a speed control system of an electric motor to which the automatic gain adjustment method for a speed control system of an 'I6 motor of the present invention is applied.

11 is electric (ijj machine, 12 is speed detector, 16 is electric l
F: 11 is an inverter which is a power circuit, and 14 is an inverter-controlled circuit which supplies the inverter 16 with a Beine's current. (Inverter 16 + Inverter 11, Monthly 61
1j times L° each 14) corresponds to (speed amplifier 2 + torque control 4! 41 device 6) in FIG. Further, a block diagram of a speed control system consisting of an electric motor 11, a vehicle speed detector 12, an inverter 16, and an inverter rb control circuit 14 is a microcomputer that outputs a command signal Nref. 16
converts the speed feedback signal N from the speed detector 12 from analog to digital to invert 1ff11
This is a converter for outputting to one circuit 14 and microcomputer 15.

In the initial state, the microcomputer 15 stores the gain of the speed control system and the electric motor moment GD2M.
The gain KN of the speed amplifier is determined in advance from parameters such as . Then, once the tool and/or workpiece is installed, no speed control is performed and the microcomputer 1
5 switches to the load inertia moment measurement mode, the load inertia moment GD2L is measured, and the speed amplifier gain K is determined according to the value of the measured inertia moment GD2L.
It is corrected to N and A. Thereafter, speed control is performed based on the gain KN' of this speed amplifier. This load loss V
The moment measurement mode is switched to not only when the tool and/or workpiece is first installed, but also when they are subsequently replaced.

The measurement of the moment of inertia GD2L of the load and the correction of the gain KN of the feed amplifier by the microcomputer 15 will be described below. As a method to examine the response of the servo system, the speed command number is vibrated in a sinusoidal manner with a constant amplitude.
There is a method of varying the frequency of the vibration and examining how the servo system follows it, and the characteristics of the servo system with respect to this frequency are called frequency characteristics. The diagram that represents this is called a Bode diagram. The microcomputer 15 is
Once the tool and/or workpiece is installed, the frequency characteristics of the speed control system are measured. Now, T is the time constant of the speed control system in the initial state. Then, the frequency characteristic is expressed as equation (1)
Since it can be approximated by a first-order delay as shown in Figure 5 (a) when expressed as a Bode diagram (frequency ω). In other words, the gain G is c = −2o+og(1+ωTo) −・−・・(
9).

Here, if j(ωTo), then G=-2QlogωTo (9).

The frequency at the point where the gain bends in the Bode diagram is called the cutoff frequency, and this cutoff frequency is ω. Then, from equation (9'). . -1/T.・Song...(1o).

Cutoff frequency ω of the speed control system from the frequency characteristics.

(moment of inertia of the load)
Changes in GD2L can be estimated.

For example, as shown in (1) in Figure 5, the cutoff frequency is 0.
of.

But ω. ', and the time constant at this time is T. ′
Then, it follows from equation (10). Therefore, the fluctuation characteristic a of the moment of inertia GD2 when a tool and/or negative running motion (moment of inertia GD2L) is attached is calculated from equations (3) and (11) by the speed amplifier gain ■ <4; = a-KN-- ---・-(13).

In the above, we focused on the gain of the speed amplifier from the cutoff frequency ωt, but the time constant T. ’ can also be intertwined. As shown in FIG. 5(b), if the gain at frequency ω-ω' is G', then from equation 0') G' = -2Q lo
gω/ To', . . . , (L4). Also, the gain G is expressed in decibels as the ratio B/A of the measured amplitude B of the speed signal N of the system to the amplitude A of the load command number Nref, and G'-, -20log and...
...05).

From equation (1・υ+ (15)), the time constant To' is To'-
上・dan ...0] 5) ω' A . The amplitude A, the measurement vibration 11 of the system B, and the frequency ω
' is the time constant T according to equation (I6). ', therefore, the same as the target) From equations (r2) and (13), the gain of the speed amplifier is set to 4, and Cain's sleeve is formed.

Figure 6 shows the old moment GD2□ of the load mentioned above.

5 is a flowchart in the case where the measuring method of '' is carried out by the speed control system of the dragon motor shown in FIG. 4.

In step 20, the microcomputer 15 sets the frequency ω' of the speed command signals Nr and 3f for frequency characteristic measurement. This frequency ω' is several tens of times the cutoff frequency when the speed control system to be measured is under no load (the moment of inertia GD2 is only the 0 width of the moment of inertia of the heart motor 11). By setting to such a value, it is possible to reduce the speed change of the electric motor 11 during measurement of the moment of inertia of the load (CD).

In step 21, the amplitude A of the speed command signal Nref is set. This amplitude A is the number -l-r at the rotational speed of the electric motor 11.
Approximately pm.

In step 22, the microcomputer 15 outputs a speed command signal Nref (=A sinω't).

'The heart motor 11 has a rotational speed N (= Bs1n(ω't+δ
)) Te vibrates in a sinusoidal manner. Here, B is the amplitude of the rotation speed N, and δ is the phase delay with respect to the speed command signal Nref. This amplitude B is determined in steps 23 to 27 below. To calculate the amplitude B (1) Rotation speed N = Bsin (ω't + δ)
Either measure N at the point where ω'L + δ - π/2, or (11) If the point where the rotational speed N = O, that is, the point delayed by δ from the speed command signal Nral, is the reference point. The number N=B sin ω't, and it is sufficient to find the rotational speed N at the point where ω't - πh. Here we will take the latter method.

In step 23, the analog/digital converted rotation speed N is read from the sex converter 16.

In step 24, it is determined whether the read rotation speed N is zero, and if it is not zero, the process returns to step 23 and the rotation speed N is reciprocally read. If it is zero (point p+ in Figure 7), proceed to the next step 25.
Proceed to.

In step 25, the timer is set to a time T-π/2ω' at which ω't=.

Proceed to step 27.

In step 27, the number of revolutions N is read. This rotation speed N is t
Since it is the value when =T-π/2ω' (point P2 in FIG. 7), it is one straight of the 4-wire φ suspension B.

In step 28, the frequency characteristics of the ArU control system are determined. That is, the frequency ω' of the speed command signal Nref (2, the amplitude A of the speed command signal Nref and the amplitude B of the number N per rotation)
The gain G', which is the ratio of

In step 2'9, the time constant T. ’. G=G', ω
−ω′, To−To′, then from the formula σ), G′ −−
201ogω'To' ・・・・・・・・・A is established, so from this 2 (17), the moment of inertia GD of the load
Time constant T when 2L increases. ’ will be blamed.

Equations f15) and (17) are both equations for calculating the gain G', but equation α5) can be said to be a theoretical equation and equation Qk is an experimental equation. Both formulas < 15), from αη -201og÷=-201ogω'To',
Time constant T. ' becomes '-1B (18) 7°-♂°W.

The no-load characteristics can be calculated in advance or measured. That is, the time constant T of the control system at this time. is the frequency ω′, electric! The amplitude of the rotation speed N of 11. Then, To-up/view (i9) ω' A .

From equations (18) and (19), the amplitude ratio B/BO of the rotation speed N of the electric motor 11 is
Therefore, the change in time constant T. ’ can also be set.

In step 30, 4 is calculated as the positive value of the gain of the speed amplifier in the speed control system, and this is set in the inverter control circuit 14. Time constant T determined in step 29. ' is the upper motive 1
It is the time constant when a load (41° unity moment GD2.) is connected to F11 (time constant T when loaded).
13) Entangled. Therefore, if the gain KN of the speed amplifier in the speed control system is set within this range, as shown by equation (3), even if the moment of inertia GD2 fluctuates by connecting or replacing the load, the speed will be controlled. The time constant of the system can be made constant.

As above, the speed control system is controlled by steps 20 to 30! I-5
Since the constant can be kept constant even if the moment of inertia of the load fluctuates, stable servo characteristics can be obtained in positioning control or PI control control.

In the second embodiment, the speed command signal is vibrated sinusoidally with a constant amplitude, the frequency of the vibration is varied, and the moment of inertia of the load is estimated from the followability of the amplitude.
In addition to this, there is another method of measuring the frequency characteristics of a servo system that uses phase characteristics. The method for estimating the moment of inertia of the load from the followability of the phase will be described below. The speed control system shown in FIG. 1 is a first-order lag system as described above, and the phase in the frequency characteristic is expressed by l G -m-1(ω'ro). Therefore, the time constant T can be determined by measuring the phase difference δ in FIG. ′ is determined by −plow δ To−□ ・・・・・・・(2υ ω). Then, as in the previous example, the formula (
12) From (13), we can set the gain of the speed amplifier to 4.

FIG. 8 is a flowchart of an embodiment of the method for estimating the moment of inertia of a load based on phase followability, as described above. In steps 3L32 and 33, the same processes as steps 20 and 21°22 in the flowchart of FIG. 6 are performed, respectively, and then in steps 33 to 38, the speed command signal Nref
” Determine the time t' from 0 to the next speed N=Q, and in step 39 calculate the phase difference δ-ωJ(,/) from this time t'. In step 40, calculate the time constant T.' from the equation (2υ). Finally, in step 41, the speed amplifier gain .

Effects of the Invention The present invention measures variations in the moment of inertia of the load based on changes in the speed signal of the motor in response to a sinusoidal speed command signal, and corrects the gain of the speed control system accordingly. When performing control or PI sub-control,
It is possible to prevent deterioration of transient characteristics due to fluctuations in the moment of inertia of the load and vibration during positioning.

In addition, in order to detect the moment of inertia of the load and correct the gain of the speed control system, the frequency of the speed command signal is set to several tens of times the cutoff frequency of the motor alone, so that the rotational speed of the motor almost oscillates. It can be done without. For example, - set the cutoff frequency of the single motive system to 5
0 (Hz) and the frequency of the speed command signal is 500 (Hz), which is 10 times that frequency, the amplitude of the vibration of the motor is approximately 1/
It becomes 10.

[Brief explanation of the drawing]

Figure 1 is a block diagram of a general speed control system for a heart motor, Figure 2 is its block diagram, Figure 6 is a block diagram when a positioning control system is added to Figure 1, and Figure 4 is A configuration diagram of an embodiment of an electric motor speed control system to which the method of the present invention is applied,
Figure 5 is the Bode diagram, Figure 6 is the speed command signal N,,1
, a diagram showing the relationship between the waveform of the rotational speed N and the tie actuation, Figure 7 is a flowchart 1 showing one example of measuring the moment of inertia of a load, and Figure 8 is another example of measuring the moment of inertia of a load. 3 is a flowchart showing an example. 11: Electric motor, 12: Speed detector, 13: Inverter, 14° inverter control circuit, 15: Microcomputer, 16: A/D converter. Patent applicant Yaskawa Electric Manufacturing Co., Ltd. Figure 1 Figure 2 Figure 1 Figure 4 Figure 7

Claims (1)

[Claims] An automatic gain correction method for a speed control system of an electric motor, characterized in that, in the speed control system of an electric motor, the gain of the speed control system is changed according to the amount of change in a speed signal of the electric motor with respect to a sinusoidal speed command signal.