Electricity: motive power systems – Positional servo systems – Unwanted harmonic or voltage component elimination...
Reexamination Certificate
2002-10-18
2004-11-23
Duda, Rina (Department: 2837)
Electricity: motive power systems
Positional servo systems
Unwanted harmonic or voltage component elimination...
C318S632000, C318S638000, C318S560000, C318S603000, C318S609000, C318S610000
Reexamination Certificate
active
06822415
ABSTRACT:
TECHNICAL FIELD
The present invention relates to a method of and a device for controlling an electric motor to actuate, through a transmitting mechanism, a movable member of a machine which has the movable member and an immovable member for supporting the movable member.
BACKGROUND ART
FIG. 1
is a block diagram of a first conventional electric motor control device.
According to the first conventional electric motor control device, servo operation command
15
as an input item is determined without recognizing mechanical vibration characteristics of a machine which has a movable member
7
and an immovable member
8
, and motion command signal
9
is sent to servo device
3
, which sends motion command signal
9
as operation command
12
to electric motor
5
, which causes transmitting mechanism
6
to move movable member
7
. If the electric motor control device cannot sufficiently perform the servo function, then servo operation command
15
is changed on a trial-and-error basis.
The first conventional electric motor control device needs a very long period of time to determine an optimum servo operation command.
FIG. 2
is a block diagram of a second conventional electric motor control device.
According to the second conventional electric motor control device, analyzing device
31
′, input device
32
, and output device
34
are added to the first conventional electric motor control device. Motion command signal
9
generated by analyzing device
31
′, is sent as an analog signal to servo device
3
, which sends motion command signal
9
as operation command
12
to electric motor
5
, which causes transmitting mechanism
6
to move movable member
7
. Rotation detector
4
sends rotation detector signal
10
through servo device
3
to analyzing device
31
′. Analyzing device
31
′ performs a fast Fourier transform on motion command signal
9
and rotation detector signal
10
to calculate frequency characteristics, determines analytical result
35
, and determines servo operation command
15
based on analytical result
35
.
According to the second conventional electric motor control device, as shown in
FIG. 3
, since motion command signal
9
generated by analyzing device
31
′ has frequency components up to maximum frequency component frmax in excess of maximum measured frequency component fq, rotation detector signal
10
and analytical result
35
suffer an aliasing error representing components outside of the measured frequency range which are introduced when digital sampling is carried out. Therefore, the second conventional electric motor control device fails to determine accurate frequency characteristics.
Problems of the second conventional electric motor control device will be described in detail below.
As shown in
FIG. 3
, motion command signal
9
generated by analyzing device
31
′ has frequency components up to maximum frequency component frmax which include those frequency components in excess of maximum measured frequency component fq. When motion command signal
9
having the frequencies shown in
FIG. 3
is used, if mechanical resonances f
4
, f
5
are present at frequencies higher than maximum measured frequency component fq and lower than maximum frequency component frmax, then motion command signal
9
excites mechanical resonances f
4
, f
5
outside of the measured frequency range, and the components of mechanical resonances f
4
, f
5
are contained in rotation detector signal
10
. Because mechanical resonances f
4
, f
5
have frequencies higher than maximum measured frequency fq, if digital sampling is carried out, then an aliasing error occurs to cause mechanical resonances f
4
, f
5
to be observed apparently as f
4
′, f
5
′. Since analytical result
35
represents solid-line components with broken-line components added thereto, no proper frequency characteristics can be evaluated. When a signal having a frequency higher than maximum measured frequency fq is processed for digital sampling, an aliasing error occurs which causes a true high-frequency waveform to be recognized in error as a low-frequency waveform. The relationship between sampling interval &Dgr;t and maximum measured frequency fq is a known fact referred to as the sampling theorem, and is expressed by the equation (1) below. As a result, frequency characteristics including components that are not actually present are output as shown in FIG.
5
.
f
q
=
1
2
×
Δ
t
(
1
)
For measuring the frequency characteristics of a conventional electric motor control device, it is necessary to have on hand an expensive instrument such as an FFT analyzer.
When an electric motor is operated, a movable member connected thereto is moved. The movable member of a load machine changes its characteristics depending on its position, causing a shift in the resonance frequency and the anti-resonance frequency which lower the accuracy with which to measure the frequency characteristics. In order to increase the amount of data to be measured for the purpose of averaging the data, it is necessary to collect data over a long period of time or carry out a plurality of operations and measurements. However, these requirements tend to cause problems in that the movable member moves increased distances and the measurement accuracy is further lowered, as shown in FIG.
7
. Specifically, the position of the electric motor is greatly displaced from the start position due to the measurement, and hence the movable member is moved, changing the characteristics of the load machine. Consequently, the accuracy with which to measure the frequency characteristics is lowered, as when a peak is split as shown in FIG.
6
.
FIG. 8
is a block diagram of a third conventional electric motor control device. The third conventional electric motor control device is different from the second conventional electric motor control device in that it has FFT analyzer
41
and signal generator
42
in place of analyzing device
31
′, input device
32
, and output device
34
of the second conventional electric motor control device.
The third conventional electric motor control device has FFT analyzer
41
and signal generator
42
in order to perform an electric motor control process in view of the characteristics of the machine. Motion command signal
43
generated by signal generator
42
is sent to servo device
3
, which sends motion command signal
43
as control signal
12
to electric motor
5
, which causes transmitting mechanism
6
to move movable member
7
. Rotation detector
4
transmits rotation detector signal
10
via servo device
3
to FFT analyzer
41
. FFT analyzer
41
receives motion command signal
43
from signal generator
42
and rotation detector signal
44
from FFT analyzer
41
, and carries out a fast Fourier transform to calculate frequency characteristics. The operator reads an anti-resonance frequency and a resonance frequency from the calculated frequency characteristics, and determines servo operation command
15
based on the read frequencies. The operator needs to manually enter servo operation command
15
into servo device
3
. Consequently, it has been customary for the operator to adjust the electric motor control device with a large expenditure of labor and time.
Heretofore, there have been various methods of tuning a mechanical control system having a flexible structure which is approximated by a two-inertia system. For example, Japanese laid-open patent publication No. 10-275003 discloses a vibration suppressing apparatus of a two-inertia resonance system for estimating a mechanical load speed and a disturbance torque through a obserber and suppressing vibrations based on the estimated mechanical load information in controlling a two-inertia system. The disclosed vibration suppressing apparatus has produced good results.
However, the conventional vibration suppressing apparatus has been problematic in that since parameters of the obserber and parameters of an PI (proportional plus integral) controller are adjusted indi
Ide Kozo
Inoki Keisei
Izumi Tetsuro
Kaku Soki
Komiya Takehiko
Duda Rina
Kabushiki Kaish Yaskawa Denki
Knobbe Martens Olson & Bear LLP
LandOfFree
Motor controller does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Motor controller, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Motor controller will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3304500