DIAGNOSTIC DEVICE, MOTOR DRIVE DEVICE AND DIAGNOSIS METHOD
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present disclosure relates to a diagnostic device,
a motor drive device, and a diagnostic method for diagnosing
a gear state.
2. Description of Related Art
Abnormality of a gear due to wear is one of the causes
of failure of machines such as machine tools (for example, steel
rolling mills).
JP-A-2018-73327 describes a technique capable of
detecting the gear state. In the technique described in
JP-A-2018-73327, the abnormality is diagnosed by performing
diagnostic drive on a drive unit driven by a motor and obtaining
evaluation data such as backlash of the drive unit, and then
comparing the value of the evaluation data with a preset value.
The set value is set by adding a predetermined ratio to the
value of the evaluation data calculated in the first drive or
the predetermined drive for diagnosis.
In the industrial world, it may be necessary to perform
gear abnormality diagnosis on an existing machine which does
not have the gear abnormality diagnosis function. However,
in the technique described in JP-A-2018-73327, in order to
3
perform abnormality diagnosis, it is necessary to obtain the
set value from the evaluation data calculated in the first drive
or the predetermined drive for diagnosis, but in the case of
the existing machine, it is possible that the gear has already
worn, so it is difficult to obtain an appropriate set value
to be used as a criterion for abnormality diagnosis. Therefore,
it is difficult to diagnose the gear state for existing
machines.
An object of the present invention is to provide a
diagnostic device, a motor drive device, and a diagnostic
method capable of diagnosing a gear state even in an existing
machine.
SUMMARY OF THE INVENTION
A diagnostic device according to an aspect of the present
disclosure is a diagnostic device which diagnoses a state of
a gear which operates according to rotation of a motor and
includes a rotation speed acquisition unit which acquires
rotation speed of the motor, a current acquisition unit which
acquires a motor current according to a torque current of the
motor, a generation unit which specifies a plurality of time
points at which a predetermined change occurs in a feature value
relating to the rotation speed as reference time points and
generates, for each of the reference time points, time-series
data of the motor current in a predetermined period according
4
to the reference time point, a calculation unit which
calculates a degree of similarity between respective
time-series data, and a diagnostic unit which diagnoses the
state of the gear based on the degree of similarity.
According to the present invention, it is possible to diagnose
a gear state even for an existing machine.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a configuration diagram illustrating a drive
system according to a first example of the present disclosure;
FIGS. 2A to 2D are diagrams illustrating an example of
a relationship between a motor angular velocity command value
and a motor current;
FIGS. 3A to 3C are diagrams for explaining the reason
why current pulsation occurs when a gear abnormality occurs;
FIGS. 4A to 4D are diagrams illustrating another example
of the relationship between the motor angular velocity command
value and the motor current;
FIGS. 5A to 5C are diagrams for explaining the reason
why current pulsation occurs when a gear abnormality occurs;
FIG. 6 is a flowchart illustrating an example of an
operation of a time series data generation unit;
FIG. 7 is a flowchart illustrating the example of the
operation of the time series data generation unit;
FIG. 8 is a flowchart illustrating the example of the
5
operation of the time series data generation unit;
FIGS. 9A and 9B are diagrams illustrating an example of
time-series motor current data;
FIG. 10 is a flowchart illustrating an example of an
operation of a similarity calculation unit;
FIGS. 11A and 11B are diagrams illustrating an example
of a current waveform of time-series motor current data during
the operation illustrated in FIG. 10;
FIG. 12 is a flowchart for illustrating an example of
an operation of an abnormality diagnostic unit;
FIG. 13 is a configuration diagram illustrating a drive
system according to a second example of the present disclosure;
FIGS. 14A to 14D are diagrams illustrating an example
of an angular velocity command;
FIGS. 15A to 15D are diagrams illustrating another
example of the angular velocity command;
FIGS. 16A to 16D are diagrams illustrating another
example of the angular velocity command;
FIG. 17 is a flowchart illustrating another example of
the operation of the time series data generation unit; and
FIG. 18 is a configuration diagram illustrating a drive
system according to a third example of the present disclosure.
DESCRIPTION OF EMBODIMENTS
Hereinafter, examples of the present disclosure will be
6
described with reference to the drawings.
First Example
First, an abnormality diagnostic device according to a
first example of the present disclosure will be described with
reference to FIGS. 1 to 12.
FIG. 1 is a configuration diagram illustrating a drive
system according to the first example. A drive system 1
illustrated in FIG. 1 includes an abnormality diagnostic device
100, a rotating machine unit 200, a motor drive device 300,
and a Programmable Logic Controller (PLC) 400.
First, the rotating machine unit 200 will be described.
The rotating machine unit 200 is applied to a machine
(for example, a steel rolling mill) which requires gears. The
rotating machine unit 200 includes a motor 201, a first output
shaft 202, a gear 203, a second output shaft 204, and a rotating
operation portion 205.
The motor 201 is rotationally driven according to a drive
signal from the motor drive device 300. The rotor of the motor
201 is connected to the first output shaft 202. The gear 203
is connected to the first output shaft 202 and the second output
shaft 204. The second output shaft 204 is connected to the
rotating operation portion 205.
When the motor 201 is rotationally driven, the first
output shaft 202 rotates in association with the rotation of
the rotor of the motor 201. The gear 203 decelerates or
7
accelerates the rotation of the first output shaft 202 and
transmits the decelerated or accelerated rotation to the second
output shaft 204. The rotating operation portion 205 rotates
in association with the rotation of the second output shaft
204. As a result, a predetermined operation (for example, the
rolling operation of the steel rolling mill) of the machine
is realized.
Next, the motor drive device 300 will be described.
The motor drive device 300 includes an encorder 301, a
motor drive circuit 302, and a controller 303.
The encorder 301 is connected to the rotor of the motor
201 of the rotating machine unit 200 and detects a rotation
value related to the rotation of the motor 201. The encorder
301 specifically detects the rotation angle (rotation
position) of the motor 201 as a rotation value. The encorder
301 transmits a rotation detection signal indicating the
detected rotation angle to the controller 303. When the motor
201 is configured by an induction machine or the like and
position control is not performed, a rotation speed detector
which detects the rotation speed of the motor 201 as a rotation
value may be provided instead of the encorder 301.
The motor drive circuit 302 is a drive circuit which
drives the motor 201 by supplying a voltage or current drive
signal to the motor 201. The motor drive circuit 302 includes
a current detector (not illustrated). The motor drive circuit
8
302 detects the current supplied to the motor 201 as the torque
current of the motor 201 using the current detector and outputs
a current detection signal indicating the detected torque
current to the controller 303.
Based on the rotation detection signal from the encorder
301, the current detection signal from the motor drive circuit
302, and the angular velocity command signal from the PLC 400,
the controller 303 generates a current command signal
indicating a current command value, which is a command value
for the current flowing through the motor 201, and outputs the
current command signal to the motor drive circuit 302. The
angular velocity command signal indicates an angular velocity
command value which is a command value for the angular velocity
of the motor 201. The motor 201 may be driven without a sensor
which does not use a rotation detection unit such as the
encorder 301. In this case, the controller 303 may calculate
the current command value based on the predicted value of the
rotation angle or the rotation speed of the motor 201 calculated
by the controller 303 itself, instead of the rotation detection
signal from the encorder 301. The controller 303 may also
calculate the angular velocity command value by itself.
Next, the PLC 400 will be described.
The PLC 400 is a host system of the abnormality diagnostic
device 100 and the motor drive device 300. The PLC 400 outputs
an angular velocity command signal indicating an angular
9
velocity command value of the motor 201 to the controller 303
and the abnormality diagnostic device 100 in order to control
the rotation speed of the rotating operation portion 205 of
the rotating machine unit 200.
Next, the abnormality diagnostic device 100 will be
described.
The abnormality diagnostic device 100 is a diagnostic
device which diagnoses the state of the gear 203 which operates
according to the rotation of the motor 201 of the rotating
machine unit 200. The abnormality diagnostic device 100
includes a rotation speed acquisition unit 101, a current
acquisition unit 102, a load state acquisition unit 103, a time
series data generation unit 104, a temporary storage unit 105,
a similarity calculation unit 106, an abnormality diagnostic
unit 107, and a display unit 108. Each unit of the abnormality
diagnostic devices 100 may be realized by the processor reading
a program that defines the operation of the processor and
executing the read program.
The rotation speed acquisition unit 101 acquires the
rotation speed of the motor 201 and temporarily stores it. In
this example, the rotation speed acquisition unit 101 acquires
the angular velocity command value indicated by the angular
velocity command signal from the PLC 400 as the rotation speed
of the motor 201. The rotation speed acquisition unit 101 may
acquire the angular velocity command value from the controller
10
303. Further, the rotation speed acquisition unit 101 may
acquire the rotation speed of the motor 201 based on the
rotation angle from the encorder 301 instead of the angular
velocity command value. When a rotation speed detector is
provided instead of the encorder 301, the rotation speed
acquisition unit 101 may acquire the rotation speed from the
rotation speed detector.
The current acquisition unit 102 acquires a motor current
corresponding to the torque current of the motor 201 and
temporarily stores the motor current. The motor current is
specifically the torque current itself or a current having a
correlation with the torque current. In this example, the
current acquisition unit 102 acquires the motor current
indicated by the current detection signal from the motor drive
circuit 302. The current acquisition unit 102 may acquire the
current command value indicated by the current command signal
output by the controller 303 as the motor current.
The load state acquisition unit 103 acquires a load state
which is a state of a load applied to the rotating operation
portion 205. The load state may indicate a numerical value
indicating the degree of load on the rotating operation portion
205 or may indicate the presence or absence of a load on the
rotating operation portion 205. For example, when the
rotating machine unit 200 is applied to a steel rolling mill,
a state in which a steel sheet is rolled is a loaded state with
11
a load and a state in which the steel sheet is not rolled is
an unloaded state without a load.
In this example, the PLC 400 detects the load state of
the rotating operation portion 205 and outputs the load state
to the load state acquisition unit 103. The load state
acquisition unit 103 acquires the load state from the PLC 400.
Further, a sensor (not illustrated) for detecting the load
state may be provided near the rotating operation portion 205
and the load state acquisition unit 103 may acquire the load
state from the sensor. Examples of this sensor include a sensor
which detects whether a load (for example, a steel plate
material in the case of a steel rolling mill) exists near the
rotating operation portion 205. Further, the load state
acquisition unit 103 may calculate the motor current according
to the change in the rotation speed acquired by the rotation
speed acquisition unit 101, and then the load state acquisition
unit 103 may estimate the load state of the rotating operation
portion 205 by comparing the motor current with the motor
current acquired by the current acquisition unit 102.
The time series data generation unit 104 generates
time-series motor current data with which the state of the gear
203 can be diagnosed based on the rotation speed acquired by
the rotation speed acquisition unit 101, specifically, the
angular velocity command value. Specifically, the time series
data generation unit 104 specifies a plurality of time points
12
at which a predetermined change occurs in the feature value
related to the angular velocity command value acquired by the
rotation speed acquisition unit 101, as a reference time point,
and then the time series data generation unit 104 generates,
for each reference time point, time-series data of the motor
current acquired by the current acquisition unit 102 in a
predetermined period according to the reference time point as
time-series motor current data.
In this example, the feature value relating to the
angular velocity command value is an angular acceleration
obtained by differentiating the angular velocity command value
with respect to time, and the predetermined period is assumed
to be a period having a predetermined length after the time
point at which the angular acceleration changes, which is the
reference time point. Also, the time series data generation
unit 104 identifies two time points of change in angular
acceleration, and then the time series data generation unit
104 generates time-series motor current data corresponding to
each of the two time points of change in angular acceleration,
as first time-series motor current data and second time-series
motor current data.
The temporary storage unit 105 temporarily stores a
plurality of time-series motor current data generated by the
time series data generation unit 104. In particular, the
temporary storage unit 105 temporarily stores the first
13
time-series motor current data and the second time-series motor
current data as first time-series motor current data 501 and
second time-series motor current data 502, respectively.
The similarity calculation unit 106 is a calculation unit
which calculates and outputs the similarity between the first
time-series motor current data 501 and the second time-series
motor current data 502 stored in the temporary storage unit
105.
The abnormality diagnostic unit 107 is a diagnostic unit
which diagnoses the state of the gear 203 based on the degree
of similarity from the similarity calculation unit 106.
The display unit 108 displays the diagnosis result of
the abnormality diagnostic unit 107. The display unit 108 may
be provided outside the abnormality diagnostic device 100. In
this case, the abnormality diagnostic device 100 includes an
interface for outputting the diagnosis result to the outside,
instead of the display unit 108.
FIGS. 2A to 2D are diagrams for illustrating an example
of the relationship between the angular velocity command value
of the motor 201 and the motor current of the motor 201.
FIG. 2A illustrates the change over time of the angular
velocity command value. In the example of FIGS. 2A to 2D, the
angular velocity command value decreases until a time point
t1, increases from the time point t1 to a time point t2, and
decreases after the time point t2. That is, the gear 203
14
decelerates to the time point t1, accelerates from the time
point t1 to the time point t2, and decelerates after the time
point t2.
FIG. 2B illustrates the change over time of the angular
acceleration which is a differential value obtained by
differentiating the angular velocity command value with
respect to time. In the example of FIGS. 2A to 2D, the angular
acceleration is negative until the time point t1, changes
positive at the time point t1, is positive until the time point
t2, and changes negative at the time point t2. The angular
acceleration before the change at the time point t1 is set as
“A1A” and the angular acceleration after the change at the time
point t1 is set as “A1B”. Further, the angular acceleration
before the change at the time point t2 is set as “A2A” and the
angular acceleration after the change at the time point t2 is
set as “A2B”. In the example of FIGS. 2A to 2D, the angular
acceleration “A1B” is equal to the angular acceleration “A2A”
and the absolute value of the angular acceleration “A2B” is
smaller than the absolute value of the angular acceleration
“A1B”.
FIGS. 2C and 2D illustrate the motor current change over
time. Specifically, FIG. 2C illustrates the motor current
change over time when the gear is normal, the gear 203 being
in a normal state, and FIG. 2D illustrates the motor current
change over time when the gear is abnormal, the gear 203 being
15
in an abnormal state. As illustrated in FIGS. 2C and 2D, when
the angular acceleration is negative, the motor current becomes
negative, and when the angular acceleration is positive, the
motor current becomes positive.
Further, when the gear is abnormal, current pulsation
occurs in the motor current after the time points t1 and t2 at
which the sign of the angular acceleration of the gear 203 is
reversed (changed), as illustrated in FIG. 2D. Specifically,
at the time point t1 + t1, the motor current has the current
pulsation with a pulsation amplitude (maximum pulsation
amplitude) I1, and at the time point t2 + t2, the motor current
has the current pulsation with a pulsation amplitude I2. In
those current pulsations (current pulsation after the time
point t1 and current pulsation after the time point t2), there
is a difference between the times t1 and t2 from the time points
t1 and t2 at which the sign of the angular acceleration is
reversed until the current pulsation occurs, and further, there
is a difference between the pulsation amplitudes I1 and I2.
FIGS. 3A to 3C are diagrams for explaining the reason
why the current pulsation occurs when the gear is abnormal.
FIGS. 3A to 3C schematically illustrate two-dimensionally the
state of each gear 203 at the time immediately before (before
the change in angular acceleration) the time point t1 in FIGS.
2A to 2D, t1 + t1, and t2 + t2. In FIGS. 3A to 3C, the gear
203 includes a motor side gear 701 connected to the first output
16
shaft 202, and a rotating operation portion side gear 702
connected to the second output shaft 204 and meshing with the
motor side gear 701.
Immediately before the time point t1, the motor side gear
701 rotates counterclockwise and the rotating operation
portion side gear 702 rotates clockwise. Then, at the time
point t1, the sign of the angular acceleration of the motor
side gear 701 is reversed and the angular velocity of the motor
side gear 701 changes. On the other hand, since the rotating
operation portion side gear 702 does not change in angular
velocity due to inertia of the rotating operation portion side
gear 702 and the rotating operation portion 205, the motor side
gear 701 does not come in contact with the rotating operation
portion side gear 702.
After that, at the time point t1 + t1, an upward tooth
surface 701A of the motor side gear 701 and the downward tooth
surface 702A of the rotating operation portion side gear 702
collide, and due to the collision, torque is applied to the
motor side gear 701 and the rotating operation portion side
gear 702. The torque applied to the motor side gear 701 is
transmitted to the motor 201 via the first output shaft 202,
causing current pulsation in the motor current.
Similarly, at the time point t2, the sign of the angular
acceleration of the motor-side gear 701 is reversed, and the
angular velocity of the motor side gear 701 changes. On the
17
other hand, since the angular velocity of the rotating
operation portion side gear 702 does not change due to inertia
of the rotating operation portion side gear 702 and the rotating
operation portion 205, the motor side gear 701 and the rotating
operation portion side gear 702 do not come into contact with
each other.
After that, at the time point t2 + t2, a downward tooth
surface 701B of the motor side gear 701 and a downward tooth
surface 702B of the rotating operation portion side gear 702
collide, and due to the collision, torque is applied to the
motor side gear 701 and the rotating operation portion side
gear 702. The torque applied to the motor side gear 701 is
transmitted to the motor 201 via the first output shaft 202,
causing current pulsation in the motor current.
In the operation described above, when the gear 203 is
normal and a backlash 703 of the gear 203 is sufficiently small,
the non-contact time in which the motor side gear 701 and the
rotating operation portion side gear 702 do not come into
contact after the sign reversal of the angular acceleration
of the gear 203 is extremely short and the difference in angular
velocity between the motor side gear 701 and the rotating
operation portion side gear 702 at the time of collision is
small. Therefore, the torque generated by the collision is
small, and the current pulsation is negligibly small. On the
other hand, when the gear 203 is abnormal and the backlash of
18
the gear 203 is large, the non-contact time is longer than that
in the case where the gear 203 is normal, and thus the difference
in angular velocity between the motor side gear 701 and the
rotating operation portion side gear 702 at the time of
collision becomes large. Therefore, the torque generated by
the collision is large, thus and the current pulsation is
significantly large.
The reason why there is a difference between time t1
from the time point t1 to the current pulsation and time t2
from the time point t2 to the current pulsation when the gear
203 is abnormal and the backlash of the gear 203 is large will
be described. The magnitude of the angular acceleration after
the time point t2 is smaller than the angular acceleration after
the time point t1 and before the time point t2. Therefore,
compared with the non-contact time in the case after the time
point t1, the non-contact time in the case after the time point
t2 at which the angular acceleration of the gear 203 is small
becomes longer. Therefore, the time t2 from the time point
t2 until the current pulsation occurs is longer than the time
t1 from the time point t1 until the current pulsation occurs.
That is, the relationship of “t1  t2” is satisfied.
The reason why there is a difference between the
pulsation amplitude I1 of the current pulsation after the time
point t1 and the pulsation amplitude I2 of the current
pulsation after the time point t2 when the gear 203 is abnormal
19
and the backlash of the gear 203 is large will be described.
The difference in angular velocities between the motor side
gear 701 and the rotating operation portion side gear 702 is
smaller after the time point t2 when the angular acceleration
of the gear 203 is small than after the time t1 when the angular
acceleration of the gear 203 is large. Therefore, the torque
generated by the collision between the motor side gear 701 and
the rotating operation portion side gear 702 is also smaller
after the time point t2 when the angular acceleration of the
gear 203 is small than after the time point t1 when the angular
acceleration of the gear 203 is large. Therefore, the
pulsation amplitude I1 of the current pulsation after the time
point t1 is larger than the pulsation amplitude I2 of the
current pulsation after the time point t2. That is, the
relationship of “I1  I2” is satisfied.
FIGS. 4A to 4D are diagrams for explaining another
example of the relationship between the angular velocity
command value of the motor 201 and the motor current of the
motor 201.
FIG. 4A illustrates the change over time of the angular
velocity command value. In the example of FIGS. 4A to 4D, the
angular velocity command value decreases until the time point
t1 and increases from the time point t1 to the time point t2
and after the time point t2. Further, the increase rate of the
angular velocity command value is greater after the time point
20
t2 than between the time point t1 and the time point t2.
FIG. 4B illustrates the change over time of the angular
acceleration which is a differential value obtained by
differentiating the angular velocity command value with
respect to time. In the example of FIGS. 4A to 4D, the angular
acceleration is negative until the time point t1, changes to
positive at the time point t1, and becomes even larger at the
time point t2. Therefore, the angular acceleration “A2B” after
changing at the time point t2 is larger than the angular
acceleration “A1B” after changing at the time point t1.
FIGS. 4C and 4D illustrate the change over time of the
motor current. Specifically, FIG. 4C illustrates the change
over time of the motor current when the gear is normal and FIG.
4D illustrates the change over time of the motor current when
the gear is abnormal. When the gear is abnormal, current
pulsation occurs in the motor current after the time point t1
when the angular acceleration of the gear 203 changes from
negative to positive, as illustrated in FIG. 4D. Specifically,
the current pulsation occurs at the time point t1 + t1.
FIGS. 5A to 5C are diagrams for explaining the reason
why the current pulsation occurs when the gear is abnormal.
FIGS. 5A to 5C schematically illustrate two-dimensionally the
state of each gear 203 at the time immediately before (before
the change in angular acceleration) the time point t1 in FIGS.
4A to 4D, t1 + t1, and t2 + t2. In FIGS. 5A to 5C, the gear
21
203 includes the motor side gear 701 and the rotating operation
portion side gear 702, as in the example of FIGS. 3A to 3C.
Immediately before the time point t1, the motor side gear
701 is rotating counterclockwise and the rotating operation
portion side gear 702 is rotating clockwise. Then, at the time
point t1, the sign of the angular acceleration of the motor
side gear 701 is reversed and the angular velocity of the motor
side gear 701 changes. On the other hand, since the rotating
operation portion side gear 702 does not change in angular
velocity due to inertia of the rotating operation portion side
gear 702 and the rotating operation portion 205, the motor side
gear 701 and the rotating operation portion side gear 702 do
not come in contact with each other. Therefore, according to
the same principle as described with reference to FIGS. 3A to
3C, at the time of gear abnormality, at the time point t1 +
t1, the upward tooth surface 701A of the motor side gear 701
collides with the downward tooth surface 702A of the rotating
operation portion side gear 702, and the collision causes the
current pulsation in the motor current.
On the other hand, at the time point t2 when the sign
of the angular acceleration does not reverse, the direction
of the angular acceleration of the motor side gear 701 does
not change, so the contact surface between the motor side gear
701 and the rotating operation portion side gear 702 does not
change. Therefore, after the time point t2, the current
22
pulsation due to the collision of the gear 203 does not occur
even when the gear is abnormal.
FIGS. 6 to 8 are flowcharts for explaining an example
of the operation of the time series data generation unit 104.
In Step S101 of FIG. 6, the time series data generation
unit 104 reads the angular velocity command value from the
rotation speed acquisition unit 101, the motor current from
the current acquisition unit 102, and the load state of the
rotating operation portion 205 from the load state acquisition
unit 103, and then the time series data generation unit 104
proceeds to Step S102.
In Step S102, the time series data generation unit 104
calculates the angular acceleration of the gear 203 by
differentiating the angular velocity command value and
proceeds to Step S103.
In Step S103, the time series data generation unit 104
compares the amount of change in angular acceleration with a
predetermined change threshold value and determines whether
the relationship of “amount of change in angular acceleration
 change threshold value” is satisfied. When the relationship
of “amount of change in angular acceleration  change threshold
value” is satisfied (Step S103: YES), the time series data
generation unit 104 determines that the angular acceleration
has changed and proceeds to Step S104. On the other hand, when
the relationship of “amount of change in angular acceleration
23
 change threshold value” is not satisfied (Step S103: NO),
the time series data generation unit 104 determines that the
angular acceleration has not changed and returns to Step S101.
In Step S104, the time series data generation unit 104
determines whether the load state of the rotating operation
portion 205 satisfies the predetermined condition. In the
following, the load state indicates a load degree, which is
a numerical value indicating the degree of load applied to the
rotating operation portion 205 and the time series data
generation unit 104 compares the load degree of the rotating
operation portion 205 with the load threshold value and
determines whether the relationship of “load degree of rotating
operation portion 205  load threshold value” is satisfied.
When the relationship of “load degree of rotating
operation portion 205  load threshold value” is satisfied
(Step S104: YES), the time series data generation unit 104
determines that the load state of the rotating operation
portion 205 does not satisfy the predetermined condition and
returns to Step S101. On the other hand, when the relationship
of “load degree of rotating operation portion 205  load
threshold value” is not satisfied (Step S104: NO), the time
series data generation unit 104 determines that the load state
of the rotating operation portion 205 satisfies the
predetermined condition and proceeds to Step S105. When the
load state indicates the presence or absence of load, if there
24
is a load, the time series data generation unit 104 determines
that the load state of the rotating operation portion 205 does
not satisfy the predetermined condition and returns to Step
S101, whereas when there is no load, the time series data
generation unit 104 determines that the load state of the
rotating operation portion 205 satisfies the predetermined
condition and proceeds to Step S105.
In Step S105, the time series data generation unit 104
specifies the time point when it is determined that the angular
acceleration has changed in Step S103 as a reference time point
(angular acceleration change time point) and the time-series
data of the motor current, the angular velocity command value,
and the load state in the predetermined period corresponding
to the reference time point is temporarily stored in the time
series data generation unit 104, and then the process proceeds
to Step S106. In this example, the predetermined period is
a period having a predetermined length T0 after the reference
time point. For example, when the time point t1 in FIGS. 2A
to 2D is specified as the reference time point, the
predetermined time period is the time period from the time point
t1 to the time point t1 + T0.
In Step S106, the time series data generation unit 104
compares the load degree within a predetermined period with
the load threshold value and determines whether the load degree
is always smaller than the load threshold value within the
25
predetermined period. When the load degree is always smaller
than the load threshold value within the predetermined period
(Step S106: YES), the time series data generation unit 104
determines that the load state always satisfies the
predetermined condition within the predetermined period and
proceeds to Step S107, and when the load degree may be equal
to or higher than the load threshold value within the
predetermined period (Step S106: NO), it is determined that
the load state may not satisfy the predetermined condition
within the predetermined period and the process returns to Step
S101.
In Step S107, the time series data generation unit 104
outputs the time-series data of the motor current stored in
Step S105 to the temporary storage unit 105 as the first
time-series motor current data for storage and proceeds to Step
S108. When standardizing the time-series motor current data
as described below, the time series data generation unit 104
may apply the angular acceleration before and after the
reference time point corresponding to the first time-series
motor current data to the first time-series motor current data
and store it in the temporary storage unit 105.
In Step S108, the time series data generation unit 104
temporarily stores the angular acceleration after (after the
reference time point specified in Step S105) determining that
the angular acceleration has changed in Step S103 as an first
26
angular acceleration A1 in the time series data generation unit
104 and proceeds to Step S109.
In Step S109, the time series data generation unit 104
determines whether the signs of the angular accelerations
before and after (after the reference time point specified in
Step S105) the time point when it is determined that the angular
acceleration has changed in Step S103 match each other. When
the signs of the angular accelerations match each other (Step
S109: YES), the time series data generation unit 104 proceeds
to Step S110 in FIG. 7, and when the signs of the angular
accelerations are different from each other (Step S109: NO),
the time series data generation unit 104 proceeds to Step S118
in FIG. 8.
The processing of Steps S110 to S115 in FIG. 7 is the
same as the processing of Steps S101 to 106 in FIG. 6. When
it is YES in Step S115, the process proceeds to Step S116.
In Step S116, the time series data generation unit 104
determines whether the signs of the angular accelerations
before and after (after the reference time point specified in
Step S114) the time point when it is determined that the angular
acceleration has changed in Step S112 match each other. When
the signs of the angular accelerations match each other (Step
S116: YES), the time series data generation unit 104 proceeds
to Step S117, and when the signs of the angular accelerations
are different from each other (Step S116: NO), the time series
27
data generation unit 104 returns to Step S110.
In Step S117, the time series data generation unit 104
outputs the time-series data of the motor current stored in
Step S114 to the temporary storage unit 105 as the second
time-series motor current data and finishes the processing.
When standardizing the time-series motor current data as
described below, the time series data generation unit 104 may
apply the angular acceleration before and after the reference
time point corresponding to the second time-series motor
current data to the second time-series motor current data and
store it in the temporary storage unit 105.
The processing of Steps S118 to S123 of FIG. 8 is the
same as the processing of Steps S101 to 106 of FIG. 6. When
it is YES in Step S123, the process proceeds to Step S124.
In Step S124, the time series data generation unit 104
temporarily stores the angular acceleration after (after the
reference time point specified in Step S122) determining that
the angular acceleration has changed in Step S120 as a second
angular acceleration A2 in the time series data generation unit
104 and proceeds to Step S125.
In Step S125, the time series data generation unit 104
determines whether the signs of the angular accelerations
before and after the time point (the reference time point
specified in Step S122) when it is determined that the angular
acceleration has changed in Step S120 match each other. When
28
the signs of the angular accelerations match each other (Step
S125: YES), the time series data generation unit 104 proceeds
to Step S127, and when the signs of those angular accelerations
are different from each other (Step S125: NO), the time series
data generation unit 104 proceeds to Step S126.
In Step S126, the time series data generation unit 104
compares the evaluation value, which is the absolute value of
the difference between the absolute value of the first angular
acceleration A1 and the absolute value of the second angular
acceleration A2, with the reference value and determines
whether the evaluation value is larger than the reference value.
That is, the time series data generation unit 104 determines
whether the relationship of “A1 - A2  reference value”
is satisfied. X indicates the absolute value of X. When the
relationship of “A1 - A2  reference value” is satisfied
(Step S127: YES), the time series data generation unit 104
proceeds to Step S127, and when the relationship of “A1 -
A2  reference value” is not satisfied (Step S127: NO), the
process returns to Step S118.
In Step S127, the time series data generation unit 104
outputs the time-series data of the motor current stored in
Step S122 to the temporary storage unit 105 as the second
time-series motor current data for storage and finishes the
processing. When standardizing the time-series motor current
data as described below, the time series data generation unit
29
104 may apply the angular acceleration before and after the
reference time point corresponding to the second time-series
motor current data to the second time-series motor current data
and store it in the temporary storage unit 105.
In the operation of the time series data generation unit
104 described above, one of the following two types of
combinations is generated as the time-series motor current data.
In the first combination, the time-series motor current data
corresponding to each of the first time point when the angular
acceleration changes and the sign of the angular acceleration
is different before and after the change and the second time
point when the angular acceleration changes and the signs of
the angular acceleration before and after the change are
generated. In the second combination, the time-series motor
current data corresponding to each of the first time point when
the angular acceleration changes and the sign of the angular
acceleration is different before and after the change and the
second time point when the angular acceleration changes and
the sign of the angular acceleration is different before and
after the change, and further the absolute value of the
difference between the absolute value of the angular
acceleration after the change and the absolute value of the
angular acceleration after the first time point is equal to
or greater than the reference value are generated.
The operation of the time series data generation unit
30
104 described above is merely an example and the operation is
not limited to this. For example, when the rotation speed of
the motor 201 is low, the detection accuracy of the encorder
301 or the rotation speed detector used in place of the encorder
301 may be low. When the field weakening control is performed
on the motor 201, the rotation speed increases, but in that
case, the motor current and the torque applied to the motor
201 may not be in a proportional relationship. Therefore, the
time series data generation unit 104 may generate and output
the time-series motor current data when the angular velocity
command value (rotation speed) is within the predetermined
range in the predetermined period.
FIGS. 9A and 9B are diagrams illustrating an example of
the time-series motor current data stored in the temporary
storage unit 105. In the example of FIGS. 9A and 9B, as the
first time-series motor current data 501, the time-series data
of the motor current from the time points t1 to t1 + T0 of FIGS.
2A to 2D is stored, and as the second time-series motor current
data 502, the time-series data of the motor current from the
time points t2 to t2 + T0 of FIGS. 2A to 2D is stored. Also,
FIG. 9A illustrates an example of the time-series motor current
data when the gear is normal, and FIG. 9B illustrates an example
of the time-series motor current data when the gear is abnormal.
As illustrated in FIGS. 9A and 9B, when the gear is
abnormal, current pulsations that cannot be confirmed when the
31
gear is normal occur in the first time-series motor current
data 501 and the second time-series motor current data 502.
FIG. 10 is a flowchart for explaining an example of the
operation of the similarity calculation unit 106. FIGS. 11A
to 11B are diagrams illustrating an example of the first
time-series motor current data 501 and the second time-series
motor current data 502 during the operation described in FIG.
10. FIG. 11A illustrates an example of the time-series motor
current data when the gear is normal and FIG. 11B illustrates
an example of the time-series motor current data when the gear
is abnormal.
In Step S201, the similarity calculation unit 106 reads
the first time-series motor current data 501 and the second
time-series motor current data 502 from the temporary storage
unit 105 and their corresponding angular accelerations before
and after the reference time point. Then, the similarity
calculation unit 106 calculates a difference G1 (= A1B - A1A)
between the angular accelerations before and after the
reference time point t1 corresponding to the first time-series
motor current data 501 and a difference G2 (= A2B - A2A) between
the angular accelerations before and after the reference time
point t2 corresponding to the second time-series motor current
data 502 and proceeds to Step S202. When there are three or
more time-series motor current data stored in the temporary
storage unit 105, the similarity calculation unit 106 reads
32
arbitrary two time-series motor current data from those
time-series motor current data as the first time-series motor
current data 501 and the second time-series motor current data
502, for example.
(a-1) of FIG. 11A and (b-1) of FIG. 11B illustrate the
first time-series motor current data 501 and the second
time-series motor current data 502 read in Step S201. The first
time-series motor current data 501 and the second time-series
motor current data 502 shown in (a-1) of FIG. 11A and (b-1)
of FIG. 11B are the same as the first time-series motor current
data 501 and the second time-series motor current data 502
illustrated in FIGS. 9A and 9B.
In Step S202, the similarity calculation unit 106 adjusts
the offset of the first time-series motor current data 501 and
the second time-series motor current data 502 such that the
initial values (values at time points t1 and t2) of the first
time-series motor current data 501 and the second time-series
motor current data 502 are set to 0 and proceeds to Step S203.
For example, when the value of the first time-series
motor current data 501 read in Step S201 is set as “I10”, the
value of the second time-series motor current data 502 as “I20”,
the value at the time point t1 of the first time-series motor
current data 501 as “I10(t1)”, and the value at the time point
t2 of the second time-series motor current data 502 as “I20(t2)”,
the similarity calculation unit 106 obtains the value “I11” of
33
the offset-adjusted first time-series motor current data 501
and the value “I21” of the second time-series motor current data
502 using the following equations (1) and (2).
I11 = I10 - I10(t1)(1)
I21 = I20 – I20(t2)(2)
(a-2) of FIG. 11A and (b-2) of FIG. 11B illustrate the
first time-series motor current data 501 and the second
time-series motor current data 502 which are offset-adjusted
in Step S202.
In Step S203, the similarity calculation unit 106
standardizes the sign and magnitude of each of the first
time-series motor current data 501 and the second time-series
motor current data 502 by the difference in angular
acceleration and proceeds to Step S204.
For example, the similarity calculation unit 106 obtains
the standardized value “I12” of the first time-series motor
current data 501 and the standardized value “I22” of the second
time-series motor current data 502 using the following
equations (3) and (4).
I12 = I11/G1 = I11/(A1B – A1A)(3)
I22 = I21/G2 = I21/(A2B – A2A)(4)
(a-3) of FIG. 11A and (b-3) of FIG. 11B illustrate the
first time-series motor current data 501 and the second
time-series motor current data 502 which are standardized in
Step S203.
34
In Step S204, the similarity calculation unit 106 aligns
(for example, t1 = t2 = t0) the time axes of the first time-series
motor current data 501 and the second time-series motor current
data 502, and then the degree of similarity between the first
time-series motor current data 501 and the second time-series
motor current data 502 is calculated and the process is
finished.
Next, a specific example of the degree of similarity will
be described.
In the first example, the similarity calculation unit
106 calculates Sum of Squared Difference (SSD) which is the
sum of squares of the difference between the time-series motor
current data 501 and the time-series motor current data 502
at each time point t as the degree of similarity. SSD can be
calculated using the following equation (5).
[Expression 1]
In the equation (5), the time point at which the angular
acceleration changes is set to t = 0, and the time width of
the time-series motor current data 501 and the time-series
motor current data 502 is set from t = 0 to T0. I12(t) is a
value at the time point t of the time-series motor current data
501 and I12(t) is a value at the time point t of the time-series
motor current data 502. The smaller the SSD value, the higher
35
the degree of similarity between the time-series motor current
data 501 and the time-series motor current data 502.
In a second example, the similarity calculation unit 106
calculates Sum of Absolute Difference (SAD) which is the sum
of absolute values of the differences between the time-series
motor current data 501 and the time-series motor current data
502 at each time point t as the degree of similarity. SAD can
be calculated using the following equation (6).
[Expression 2]
The smaller the SAD value, the higher the degree of
similarity between the time-series motor current data 501 and
the time-series motor current data 502.
In a third example, the similarity calculation unit 106
calculates Normalized Cross-Correlation (NCC), which is the
standardized cross-correlation between the time-series motor
current data 501 and the time-series motor current data 502,
as the degree of similarity. NCC can be calculated using the
following equation (7).
[Expression 3]
NCC takes a value in the range of -1 to 1. When NCC is
36
1, the time-series motor current data 501 and the time-series
motor current data 502 are matched or have the proportional
relationship, and when NCC is -1, -1 times the time-series motor
current data 501 and the time-series motor current data 502
are matched or have the proportional relationship. Therefore,
as the value of NCC deviates from 0, the degree of similarity
between the time-series motor current data 501 and the
time-series motor current data 502 increases. Further, the
closer the value of NCC is to 0, the lower the degree of
similarity between the time-series motor current data 501 and
the time-series motor current data 502.

CLAIMS
1. A diagnostic device which diagnoses a state of a gear
which operates according to rotation of a motor, comprising:
a rotation speed acquisition unit which acquires
rotation speed of the motor;
a current acquisition unit which acquires a motor current
according to a torque current of the motor;
a generation unit which specifies a plurality of time
points at which a predetermined change occurs in a feature value
relating to the rotation speed as reference time points and
generates, for each of the reference time points, time-series
data of the motor current in a predetermined period according
to the reference time point;
a calculation unit which calculates a degree of
similarity between respective time-series data; and
a diagnostic unit which diagnoses the state of the gear
based on the degree of similarity.
2. The diagnostic device according to claim 1, wherein
the feature value is an angular acceleration obtained
by differentiating the rotation speed with time, and
the predetermined period is a period having a
predetermined length after the reference time point.
3. The diagnostic device according to claim 2, wherein
53
the generation unit specifies, as the reference time
point, a first time point when the angular acceleration changes
and a sign of the angular acceleration is different before and
after the change and a second time point when the angular
acceleration changes and the sign of the angular acceleration
matches before and after the change.
4. The diagnostic device according to claim 2, wherein
the generation unit specifies, as the reference time
point, a first time point when the angular acceleration changes
and the sign of the angular acceleration is different before
and after the change and a second time point when the angular
acceleration changes and the sign of the angular acceleration
is different before and after the change, and further an
absolute value of a difference between an absolute value of
the angular acceleration after the change and an absolute value
of the angular acceleration after the first time point is
greater than the reference value.
5. The diagnostic device according to claim 1, further
comprising:
a command unit which controls the rotation speed of the
motor to cause the predetermined change in the feature value.
6. The diagnostic device according to claim 1, wherein
54
the calculation unit standardizes each time-series data
and calculates the degree of similarity based on the
standardized time-series data.
7. The diagnostic device according to claim 1, wherein
the calculation unit calculates, as the degree of
similarity, a comparison result obtained by comparing at least
one of a difference between the reference time point and a time
point when the motor current reaches a peak value within the
predetermined period, a difference between the motor current
at the reference time point and a peak value of the motor current
within the predetermined period, and peak values of the motor
current within the predetermined period in each time-series
data.
8. The diagnostic device according to claim 1, further
comprising:
a load acquisition unit which acquires a state of load
applied to the rotating operation portion linked with the gear,
wherein
the generation unit generates the time-series data when
the state of load satisfies a predetermined condition during
the predetermined period.
9. The diagnostic device according to claim 1, wherein
55
the generation unit generates the time-series data when
the rotation speed is within a predetermined range during the
predetermined period.
10. The diagnostic device according to claim 1, wherein
the diagnostic unit determines whether the time-series
data are similar to each other based on the degree of similarity
and outputs an alarm when the time series data is not similar
to each other.
11. A motor drive device, comprising:
the diagnostic device according to claim 1; and
a drive circuit for driving the motor.
12. A diagnosis method using a diagnostic device which
diagnoses a state of gear which operates according to rotation
of a motor, comprising:
acquiring rotation speed of the motor;
acquiring a motor current according to a torque current
of the motor;
specifying a plurality of time points at which a
predetermined change occurs in a feature value relating to the
rotation speed as reference time points and generating, for
each of the reference time points, time-series data of the motor
current in a predetermined period according to the reference
56
time point;
calculating a degree of similarity between respective
time-series data; and
diagnosing the state of the gear based on the degree of
similarity.