DESCRIPTION
FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
[See section 10, Rule 13]
MOTOR CONTROL DEVICE AND ELECTRIC POWER STEERING DEVICE;
MITSUBISHI ELECTRIC CORPORATION, A CORPORATION ORGANISED AND
EXISTING UNDER THE LAWS OF JAPAN, WHOSE ADDRESS IS 7-3,
MARUNOUCHI 2-CHOME, CHIYODA-KU, TOKYO 100-8310, JAPAN
THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE
INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED.
2
DESCRIPTION
Title of the Invention: MOTOR CONTROL DEVICE AND ELECTRIC
POWER STEERING DEVICE
5
TECHNICAL FIELD
[0001]
The disclosure of the present application relates to
an electric motor control device, and to an electric power
10 steering device using the motor control device.
BACKGROUND ART
[0002]
In general, with an electric motor of multiple phases,
a failure detection device is provided for the purpose of
15 detecting a malfunction or failure. When the failure
detection device detects a failure, a drive operation of the
motor is halted in order not to allow the motor performing
the operations which are not intended.
[0003]
20 However, in a case in which it is possible to continue
a drive operation of a motor in a failure being detected, it
is desirable to continue a drive operation thereof; however,
there arises a problem in that a conventional failure
detection device cannot distinguish go/no-go determination
25 of the continuity in a drive operation of a motor with
3
respect to a failure being detected, so that a drive
operation of the motor results in being halted at a timepoint when the failure is detected.
[0004]
5 In order to solve the problem, such a motor control
device is proposed in Patent Document 1 that determination
is performed on go/no-go determination of the continuity in
a drive operation of a motor with respect to a failure being
detected by means of a failure detection device, so that a
10 drive operation of the motor is continued when it is possible
to do so; and, in the proposed motor control device, when
electric-current value abnormality is detected from detected
electric-currents detected by electric current detection
means provided for a motor through electric-current value
15 abnormality detection means, a drive operation of the motor
is continued by changing over a detected electric-currents
to an estimation electric-current in a case in which the
electric-current value abnormality is a failure due to the
electric current detection means in itself.
20 RELATED ART DOCUMENT
Patent Document
[0005]
Patent Document 1: Japanese Patent Publication No.
5092538
25
4
SUMMARY OF THE INVENTION
Problems to be Solved by the Invention
[0006]
As described above, in a motor control device disclosed
5 in Patent Document 1, electric current detection means is
defined as a subject matter in a failure detection device
disclosed, and, at a time when the electric current detection
means detects abnormality of a detected electric-current
value(s), determination is performed so that a drive
10 operation of a motor can be continued only when the
abnormality occurrence cause is due to the electric current
detection means in itself, so that there arises a problem in
that the determination cannot be performed on go/no-go
determination of the continuity in a drive operation of the
15 motor depending on the magnitude of an influence in which a
failure being caused gives to an output (system) of the motor
control device.
[0007]
The present disclosure in the application concerned
20 has been directed at solving those problems or issues as
described above, an object of the disclosure is to obtain an
electric motor control device which is appropriately capable
of performing go/no-go determination of the continuity in a
drive operation of a motor even when abnormality is caused
25 in output torque of the motor.
5
Means for Solving the Problems
[0008]
An electric motor control device disclosed in the
disclosure of the application concerned comprises: an
5 electric-current instruction calculation unit for
determining target electric-current values of a motor of
multiple phases; a detected phase-current calculation unit
for obtaining detected electric-current values by detecting
electric currents of phases each flowing through the motor;
10 a control calculation unit for calculating electric-current
instructions to control motor electric-currents each on the
basis of the target electric-current values and on that of
the detected electric-current values; an electrical angle
calculation unit for calculating, from a rotor’s rotational
15 position of the motor, an electrical angle being a phase
difference between a rotational position of the rotor of the
motor and a coil of the motor; an abnormal torque estimation
unit for estimating abnormality of torque of the motor from
the target electric-current values, the detected electric20 current values and the electrical angle; and a cutoff
management unit for halting a drive operation of the motor
if the torque thereof is abnormal.
Effects of the Invention
[0009]
25 According to the motor control device disclosed in the
6
disclosure of the application concerned, it is possible to
appropriately perform go/no-go determination of the
continuity in a drive operation of a motor even when
abnormality is caused in output torque of the motor.
5 BRIEF DESCRIPTION OF DRAWINGS
[0010]
FIG. 1 is a block diagram illustrating an electric
power steering device provided with an electric motor control
device according to Embodiment 1;
10 FIG. 2 is a diagram illustrating the general outlines
of the operations of the electric power steering device
according to Embodiment 1;
FIG. 3 is a diagram schematically showing an overview
of a time chart of target electric-currents of a motor and
15 detected electric-currents thereof in operational processing
of the electric power steering device according to Embodiment
1;
FIG. 4 is a block-line diagram illustrating, by way of
one example, a processing configuration of an abnormal torque
20 estimation unit in the electric power steering device
according to Embodiment 1;
FIG. 5 is a flow chart showing processing contents in
the abnormal torque estimation unit of the electric power
steering device according to Embodiment 1;
25 FIG. 6 is a flow chart showing calculation processing
7
contents of electric current deviations in the abnormal
torque estimation unit of the electric power steering device
according to Embodiment 1;
FIG. 7 is a flow chart showing the processing for
5 selecting a minimum value of electric current deviation in
the abnormal torque estimation unit of the electric power
steering device according to Embodiment 1;
FIG. 8 is a flow chart showing the processing for
selecting a maximum value of electric current deviation in
10 the abnormal torque estimation unit of the electric power
steering device according to Embodiment 1;
FIG. 9 is a diagram schematically showing an overview
of a time chart in a case in which deviations are caused
between target electric-currents of a motor and detected
15 electric-currents thereof in operational processing of the
electric power steering device according to Embodiment 1;
FIG. 10 is a diagram schematically showing an overview
of a time chart of electrical angles  in operational
processing of the electric power steering device according
20 to Embodiment 1;
FIG. 11 is a block-line diagram illustrating, by way
of another example, a processing configuration of the
abnormal torque estimation unit in the electric power
steering device according to Embodiment 1;
25 FIG. 12 is a flow chart showing processing contents in
8
an abnormal torque estimation-value calculation process of
FIG. 11;
FIG. 13 is a flow chart showing internal processing of
a cutoff management unit in the electric power steering
5 device according to Embodiment 1;
FIG. 14 is a diagram showing, by way of one example,
a hardware configuration of the motor control device
according to the embodiments each;
FIG. 15 is a block diagram illustrating an electric
10 power steering device provided with an electric motor control
device according to Embodiment 2;
FIG. 16 is a flow chart showing processing contents of
an open failure determination unit in the electric power
steering device according to Embodiment 2;
15 FIG. 17 is a flow chart showing processing contents of
phase-U open-failure determination processing in the open
failure determination unit of the electric power steering
device according to Embodiment 2;
FIG. 18 is a flow chart showing processing contents of
20 phase-V open-failure determination processing in the open
failure determination unit of the electric power steering
device according to Embodiment 2;
FIG. 19 is a flow chart showing processing contents of
phase-W open-failure determination processing in the open
25 failure determination unit of the electric power steering
9
device according to Embodiment 2;
FIG. 20 is a flow chart showing internal processing of
a cutoff management unit in the electric power steering
device according to Embodiment 2;
5 FIG. 21 is a schematic diagram in which a motor is
shown as a simple resistor model;
FIG. 22 is a schematic diagram in which the motor is
shown as a simple resistor model at a time when a phase-U
open failure is caused;
10 FIG. 23 is a schematic diagram of target three-phase
currents at a normal time; and
FIG. 24 is a schematic diagram of detected three-phase
currents at a time when a phase-U open failure is caused.
EMBODIMENTS FOR CARRYING OUT THE INVENTION
15 [0011]
Hereinafter, the explanation will be made referring to
the drawings for preferred exemplary embodiments of an
electric motor control device, and for an electric power
steering device provided with the motor control device.
20 [0012]
Embodiment 1.
FIG. 1 is an overall diagram illustrating, by way of
one exemplary embodiment, a configuration of an electric
power steering device according to Embodiment 1; and the
25 electric power steering device is provided with an electric
10
motor control device 100 including an electric-current
instruction calculation unit 3 for controlling output torque
of an electric motor 10 which is connected to a gear 11 and
which outputs steering assist torque for assisting the
5 steering of an operator or driver, on the basis of detection
torque Td being outputted by a torque detector 2 for
detecting steering torque, from the driver, being acted onto
a steering wheel 1 of an automotive vehicle on which the
driver steers and for externally outputting the detected
10 torque as an electrical signal (the detection torque Td), so
that vehicle’s wheels 12 are steered in accordance with
steering torque of the driver and with the steering assist
torque in which the motor 10 outputs. The electric-current
instruction calculation unit 3 determines target electric15 current values corresponding to motor electric-current
values of the motor 10.
[0013]
The motor 10 is a multiple phase motor made of, for
example, a surface-permanent-magnet three-phase brushless
20 motor of a star connection in which one end-terminal of a
phase-U coil, that of a phase-V coil and that of a phase-W
coil are mutually connected one another, and the other endterminal of the each coil is connected to an electric motor
driving device 7 by way of an electric motor current cutoff
25 unit 9, so that, by means of the motor driving device 7,
11
electric motor drive-operation voltages are applied to the
individual coils, and so, electric currents flow through the
individual coils in accordance with the voltages applied to
the individual phases.
5 [0014]
A rotor’s rotational position detector 13 detects a
rotor’s rotational position of the motor 10, and an
electrical angle calculation unit 14 calculates, from the
rotor’s rotational position detected by the rotor’s
10 rotational position detector 13, an electrical angle  being
a phase difference between the rotor’s rotational position
and the three-phase coils.
[0015]
The motor driving device 7 is connected to the motor
15 10 by way of the motor current cutoff unit 9. And, the motor
driving device is constituted of, for example: two fieldeffect transistors QHu and QLu connected in series with each
other, and, connected in parallel with the series-connected
circuit, a series-connected circuit of two field-effect
20 transistors QHv and QLv; and also in a similar manner, a
series-connected circuit of field-effect transistors QHw and
QLw. In the motor driving device, pulse width modulation
(PWM) signals being outputs from a control calculation unit
4 are converted by an FET-gate drive unit 6 being, for
25 example, an FET driver IC into voltage levels capable of
12
performing gate drives of the field-effect transistors QHu,
QLu, QHv, QLv, QHw and QLw each within the motor driving
device 7 connected at the subsequent stage. In accordance
with gate drive signals being output signals from the FET5 gate drive unit 6 after the conversion, these field-effect
transistors (FETs) QHu, QLu, QHv, QLv, QHw and QLw are driven,
whereby predetermined phase voltages are applied to the
individual coils of the motor 10 connected at the subsequent
stage, so that torque is produced in the motor 10 by flowing
10 phase currents through the motor 10.
[0016]
Moreover, the motor driving device 7 is provided with
shunt resistors 8u, 8vand 8w for detecting the phase currents.
[0017]
15 A detected phase-current calculation unit 15
calculates phase currents Iu, Iv and Iw flowing through the
individual phase coils of the motor 10 in accordance with
values detected by the shunt resistors 8u, 8v and 8w.
[0018]
20 The electric-current instruction calculation unit 3
calculates a q-axis electric-current instruction and a daxis electric-current instruction being electric current
instructions on the rotational two axes (d-axis and q-axis)
from detection torque Td detected by means of the torque
25 detector 2, from a vehicle speed signal Sv acquired by means
13
of an external device or apparatus, and from a motor
revolution number calculated by using an electrical angle 
from the electrical angle calculation unit 14. The q-axis
electric-current instruction is indicated as a target q-axis
5 current Iq_t, and the d-axis electric-current instruction is
indicated as a target d-axis current Id_t.
[0019]
When an output signal from a cutoff management unit 17
is not at a forced PWM-signal turn-off instruction INoff,
10 the control calculation unit 4 carries out publicly known
control-calculation processing, for example, PI control
processing by using a target q-axis current Iq_t and a target
d-axis current Id_t each of which is an output signal from
the electric-current instruction calculation unit 3 and by
15 using detected electric-currents Iq_m and Id_m in which
individual phase detected electric-currents Iu, Iv and Iw
being output signals of the detected phase-current
calculation unit 15 are performed through three- to twophase transformation into the electric currents on the
20 rotational two axes (d-axis and q-axis); and the control
calculation unit forms a phase-U PWM signal S-PWM(U), a
phase-V PWM signal S-PWM(V) and a phase-W PWM signal S-PWM(W)
corresponding to the processing-calculation results, which
are outputted into an FET-gate-driving cutoff unit 5 at the
25 subsequent stage. When an output signal from the cutoff
14
management unit 17 is at a forced PWM-signal turn-off
instruction INoff, the control calculation unit forms a
phase-U PWM signal S-PWM(U), a phase-V PWM signal S-PWM(V)
and a phase-W PWM signal S-PWM(W) by which each phase of the
5 motor 10 is not energized without depending on a target qaxis current Iq_t and a target d-axis current Id_t each of
which is an output signal from the electric-current
instruction calculation unit 3 and without depending on the
detected electric-currents Iq_m and Id_m in which individual
10 phase detected electric-currents Iu, Iv and Iw being output
signals of the detected phase-current calculation unit 15
are performed through three- to two-phase transformation
into the electric currents on the rotational two axes (daxis and q-axis), so that the control calculation unit
15 outputs such a phase-U PWM signal S-PWM(U), such a phase-V
PWM signal S-PWM(V) and such a phase-W PWM signal S-PWM(W)
into the FET-gate-driving cutoff unit 5 at the subsequent
stage.
[0020]
20 The motor current cutoff unit 9 is inserted in an
electric current path between the motor driving device 7 and
the motor 10, and the motor current cutoff unit interrupts
the electric current path by obeying a motor current cutoff
instruction INci from the cutoff management unit 17, whereby
25 output torque of the motor 10 is made to 0 (zero).
15
[0021]
The FET-gate-driving cutoff unit 5 is placed in a path
between the control calculation unit 4 and the FET-gate drive
unit 6; and the FET-gate-driving cutoff unit is provided
5 with the function to forcefully turn off a phase-U PWM signal
S-PWM(U), a phase-V PWM signal S-PWM(V) and a phase-W PWM
signal S-PWM(W) into the FET-gate drive unit 6 by obeying an
FET-gate-driving cutoff instruction INcg from the cutoff
management unit 17, and, for example, by halting gate drives
10 of an FET-gate driver IC being the FET-gate drive unit 6, so
that phase voltage supplies into the coils each of the motor
10 are halted in accordance with the function.
[0022]
The electric-current instruction calculation unit 3
15 calculates a target q-axis current Iq_t and a target d-axis
current Id_t being electric current instructions on the
rotational two axes (d-axis and q-axis) from at least
detection torque Td detected by means of the torque detector
2, from a vehicle speed signal Sv acquired by means of an
20 external device or apparatus, and from a motor revolution
number calculated by using an electrical angle  from the
electrical angle calculation unit 14.
[0023]
A battery voltage detection unit 20 is provided with
25 the function to detect a voltage of a battery 18 whose
16
voltage is inputted into the motor 10; and a battery voltage
being detected by the battery voltage detection unit is
inputted into the cutoff management unit 17.
[0024]
5 A battery voltage cutoff unit 19 is provided with the
function to interrupt a battery voltage being applied to the
motor driving device 7 by obeying a battery voltage cutoff
instruction INcv from the cutoff management unit 17.
[0025]
10 An abnormal torque estimation unit 16 is provided with
the function to estimate abnormal torque of the motor 10 in
accordance with a target q-axis current Iq_t and a target daxis current Id_t each being outputs of the electric-current
instruction calculation unit 3, phase currents Iu, Iv and Iw
15 each being outputs of the detected phase-current calculation
unit 15, and an electrical angle  being an output of the
electrical angle calculation unit 14. Depending on the
function, at a time when a failure or failures, for example,
such a short circuit of a field-effect transistor(s) and an
20 open failure thereof, and/or a short-to-power fault of a
gate drive signal(s), a short-to-ground fault thereof and
the like are caused in the FET-gate drive unit 6 and the
motor driving device 7, phase currents flowing through the
individual phase coils of the motor 10 become abnormal,
25 whereby it is feared that an influence may be exerted with
17
respect to output torque of the motor 10, and that, depending
on the magnitude of the degree of influence, abnormal events
may be brought about in such cases as vibration with respect
to the steering wheel 1, the production of steering force in
5 a reverse steering direction in response to the steering
direction of an operator or driver (reverse or reciprocal
assist), and the like. Hereinafter, the explanation will be
made for an influence on output torque of the motor 10 in a
case in which a surface-permanent-magnet three-phase
10 brushless motor is used for.
[0026]
In general, in a case of a surface permanent magnet
type motor, output torque thereof can be given by Expression
(1) using the number of pole pairs, P, of the motor, and
15 using an actual q-axis current Iq_r flowing through the motor,
and the maximum value of armature interlinkage magnetic-flux,
, by permanent magnets thereof.
[0027]
[Expression Figure-1]
20
[0028]
According to Expression (1), the number of pole pairs,
P, of a motor and the maximum value of armature interlinkage
18
magnetic-flux, , by permanent magnets thereof are fixed
values defined by the motor being used, and thus, it can be
known that a factor which exerts a direct influence on the
variation of output torque values during the rotation of the
5 motor is an actual q-axis current Iq_r flowing through the
motor.
[0029]
An actual q-axis current Iq_r flowing through the motor
10 means a q-axis current at a time when actual individual
10 phase currents Iu_r, Iv_r and Iw_r flowing through the
individual phases of the motor 10 are performed through
three- to two-phase transformation; and so, when an electric
current deviation is caused between the actual q-axis current
Iq_r and a target q-axis current Iq_t owing to some kind of
15 factor under the condition in which the torque is produced
in the motor 10 in accordance with the aforementioned target
q-axis current Iq_t, a torque value corresponding to the
electric current deviation is superimposed on the output
torque of the motor 10, whereby abnormal torque is caused in
20 the motor 10, so that steering torque of an operator or
driver changes due to the abnormal torque being caused.
[0030]
The abnormal torque estimation unit 16 calculates
individual phase target currents Iu_t, Iv_t and Iw_t of the
25 motor 10 in accordance with a target q-axis current Iq_t and
19
a target d-axis current Id_t where the electric-current
instruction calculation unit 3 outputs, and with an
electrical angle  where the electrical angle calculation
unit 14 outputs; and the abnormal torque estimation unit
5 estimates abnormal torque on the basis of individual phase
detected electric-currents Iu, Iv and Iw calculated by the
detected phase-current calculation unit 15. When electric
current deviations between individual phase target currents
and individual phase detected electric-currents are defined
10 as Iu, Iv and Iw, the following Expression (2), Expression
(3) and Expression (4) are held, respectively. However, it
is presumed that detection errors are not included in the
individual phase detected electric-currents Iu, Iv and Iw.
[0031]
15 [Expression Figure-2]
[0032]
[Expression Figure-3]
20 [0033]
[Expression Figure-4]
20
[0034]
Expression (5) described below is a conversion
expression of three- to two-phase transformation.
5 [0035]
[Expression Figure-5]
[0036]
As described above, a d-axis current does not exert an
10 influence on output torque of the motor 10, whereas a q-axis
current appears in its influence on the output torque of the
motor 10. Thus, the explanation will be made hereinafter
for the q-axis current related to the output torque of the
motor 10.
15 [0037]
By substituting Expression (2) through Expression (4)
described above into Expression (5), and expressing it by
dividing into the terms including deviations of q-axis
current and the terms not including the deviations thereof,
20 Expression (6) can be obtained.
[0038]
21
[Expression Figure-6]
[0039]
There results in showing that the first half portion
5 of Expression (6) described above indicates a target q-axis
current Iq_t from the electric-current instruction
calculation unit 3, and that the latter half portion thereof
indicates the deviation from a target q-axis current Iq_t
obtained in the electric-current instruction calculation
10 unit 3. Thus, when the deviation between the target q-axis
current Iq_t of the electric-current instruction calculation
unit 3 and a q-axis current calculated from individual phase
currents of the motor 10 having been calculated by the
detected phase-current calculation unit 15 is defined as
15 “Iq,” Expression (7) described below is given.
[0040]
[Expression Figure-7]
[0041]
20 By combining the three trigonometric functions of
22
Expression (7) described above, and transforming it into one
trigonometric function, the following Expression (8) is
derived.
[0042]
5 [Expression Figure-8]
[0043]
According to Expression (8) described above, in a case
in which an electric current deviation is caused between
10 individual phase target currents of the motor 10 calculated
in accordance with a target q-axis current Iq_t of the
electric-current instruction calculation unit 3 and with an
electrical angle  where the electrical angle calculation
unit 14 outputs, and individual phase detected electric15 currents calculated by the detected phase-current
calculation unit 15, there results in existing a possibility
that the deviation of Iq_amp is caused in an actual q-axis
current in which electric currents flowing through the motor
10 are performed through three- to two-phase transformation
20 at the maximum with respect to the target q-axis current Iq_t
of the electric-current instruction calculation unit 3, as
given by Expression (9) described below. That is to say,
23
there results in existing the possibility that actual torque
where the motor 10 outputs in accordance with actual
electric-currents, i.e. the torque corresponding to “Iq_amp”
is outputted as abnormal torque at the maximum with respect
5 to the torque of the motor 10 outputted in accordance with
a target q-axis current Iq_t where the electric-current
instruction calculation unit 3 outputs.
[0044]
[Expression Figure-9]
10
[0045]
Next, the range will be shown in which the
aforementioned Iq_amp is capable of taking on. When a minimum
value of electric current deviations Iu, Iv and Iw is
15 defined as an electric-current deviation minimum-value Imin,
and difference values between the parameter Imin and each of
Iu, Iv and Iw are defined as electric-current deviation
difference values Iu, Iv and Iw, Expression (10),
Expression (11) and Expression (12) are held as described
20 below, respectively.
[0046]
[Expression Figure-10]
24
[0047]
[Expression Figure-11]
5 [0048]
[Expression Figure-12]
[0049]
In addition, because the aforementioned electric10 current deviation minimum-value Imin is the minimum value
among the electric current deviations Iu, Iv and Iw, a
difference value from the parameter Imin becomes 0 (zero) in
any one among Expression (10) through Expression (12)
described above.
15 [0050]
In accordance with Expression (10) through Expression
(12), and with Expression (9) described above, Expression
(13) in which the aforementioned Iq_amp is expressed by using
Iu, Iv and Iw can be obtained as described below.
25
[0051]
[Expression Figure-13]
[0052]
5 In accordance with electric-current deviations Iu, Iv
and Iw, and with an electric-current deviation minimum-value
Imin, it is indicated that a maximum value among electriccurrent deviation difference values Iu, Iv and Iw is
defined as an electric-current deviation difference maximum10 value Imax; and herein, by presuming that the correlation
among the values of Iu, Iv and Iw being electric current
deviations is given as “Iu  Iw  Iv,” Expression (14),
Expression (15) and Expression (16) can be obtained as
described below, respectively.
15 [0053]
[Expression Figure-14]
[0054]
[Expression Figure-15]
26
[0055]
[Expression Figure-16]
5 [0056]
By substituting Expression (14) and Expression (15)
described above into Expression (13), and by performing
expressional transformation where “IDiff = Imax  Imin,”
Expression (17) described below can be obtained.
10 [0057]
[Expression Figure-17]
[0058]
In Expression (17) described above, the condition
15 where the aforementioned Iq_amp takes on the maximum is at
the time of “Iw = 0,” or at that of “Iw = IDiff,” so that
the aforementioned Iq_amp is given by Expression (18)
described below.
[0059]
27
[Expression Figure-18]
[0060]
In addition, the condition where the aforementioned
5 Iq_amp takes on the minimum is at the time of “Iw = IDiff /
2,” so that the aforementioned Iq_amp at this time is given
by Expression (19) described below.
[0061]
[Expression Figure-19]
10
[0062]
According to Expression (18) and Expression (19)
described above, the range in which the aforementioned Iq_amp
can take on is given by Expression (20) described below by
15 using individual phase target currents of the motor 10
calculated in accordance with a target q-axis current Iq_t of
the electric-current instruction calculation unit 3 and with
an electrical angle  where the electrical angle calculation
unit 14 outputs, and using parameter “IDiff” being the
20 difference between parameter “Imin” being the minimum value
of electric current deviations Iu, Iv and Iw between
28
individual phase detected electric-currents calculated by
the detected phase-current calculation unit 15, and
parameter “Imax” being the maximum value of the electric
current deviations therebetween.
5 [0063]
[Expression Figure-20]
[0064]
Therefore, the abnormal torque estimation unit 16 can
10 estimate a maximum value (abnormal torque) of steering-shaft
converted abnormal torque capable of being produced in the
motor 10, Temu, according to Expression (21) described below,
by giving a torque constant Km [Nm/A] which is the
correlation between a q-axis current flowing through the
15 motor 10 and its torque being outputted, and giving a motor
reduction ratio Ggear of the gear 11.
[0065]
[Expression Figure-21]
20 [0066]
Because the torque constant Km [Nm/A] and the motor
29
reduction ratio Ggear of the gear 11 are constants each, an
output of the abnormal torque estimation unit 16 may be
defined as parameter “IDiff” being a value which corresponds
to the abnormal torque. In the explanation stated above,
5 the explanation has been made for the parameter Iu as the
minimum value Imin exactly as Expression (14) described
above; however, even when parameter Iv takes on the minimum
value Imin, or even when parameter Iw takes on the minimum
value Imin, the range in which Expression (20) described
10 above is capable of taking on does not change, so that
Expression (21) does not also change.
[0067]
Abnormality of an electric power steering device is
defined in accordance with a steering torque change-quantity
15 of an operator or driver. Hereinafter, the explanation will
be made for a method for determining abnormality of the
electric power steering device by means of abnormal torque
estimated by the abnormal torque estimation unit 16.
[0068]
20 Firstly, the explanation will be made briefly for the
operations of the electric power steering device. When an
operator or driver steers the steering wheel 1, the vehicle’s
wheels 12 are steered, so that pieces of road-surface load
torque are produced in the vehicle’s wheels 12 each of which
25 try to return toward their midpoints. At a time when the
30
motor 10 is not outputting its steering assist torque, these
pieces of road-surface load torque act as a steering torque
of the driver. The steering torque of the driver is detected
by means of the torque detector 2; and then, the steering
5 torque being detected is inputted into the electric-current
instruction calculation unit 3, so that a target q-axis
current Iq_t and a target d-axis current Id_t are outputted
therefrom. The target q-axis current Iq_t and the target daxis current Id_t are inputted into the control calculation
10 unit 4; and then, the control calculation unit 4 controls
electric currents flowing through the motor 10 by means of
the FET-gate drive unit 6 and the motor driving device 7 so
that the values in which phase currents detected by the
detected phase-current calculation unit 15 are transformed
15 into a q-axis current and a d-axis current coincide with the
target q-axis current Iq_t and the target d-axis current Id_t.
The motor 10 outputs its torque in accordance with a q-axis
component of the electric currents flowing through the motor.
The output torque of the motor 10 becomes, by way of the
20 gear 11, steering assist torque which acts to assist the
steering torque by the driver.
[0069]
When steering torque by the driver is defined as “Th,”
and the relationship of steering assist torque produced by
25 an output of the torque detector 2 is defined in a
31
proportional relationship of a simple proportional gain “G”
for brevity of explanation, and when road-surface load torque
is defined as “Ta,” a relational expression of Expression
(22) described below can be obtained.
5 [0070]
[Expression Figure-22]
[0071]
The explanation will be made for a change of steering
10 torque of an operator or driver as a result of causing
abnormality in the FET-gate drive unit 6 and/or the motor
driving device 7, at a time when an electric current does
not flow through the motor 10 exactly as a target q-axis
current Iq_t of the electric-current instruction calculation
15 unit 3 through the motor 10. In a case in which an electric
current does not flow through the motor 10 exactly as a
target q-axis current Iq_t, steering-shaft converted torque
of output torque of the motor 10 causes the deviation of “T”
with respect to steering-shaft converted torque in a case in
20 which the electric current flows through the motor 10 exactly
as the target q-axis current Iq_t. Due to the deviation being
caused, steering torque of the driver changes to “Th'.”
Steering torque of the driver having been changed is detected
32
by means of the torque detector 2, and the steering torque
being detected is inputted into the electric-current
instruction calculation unit 3, so that a target q-axis
current Iq_t is calculated, and so, target output torque of
5 the motor 10 takes on “GTh” in accordance with steering
shaft conversion. Because output torque of the motor 10 due
to abnormality in accordance with the steering shaft
conversion has the deviation of “T” with respect to the
target output torque, the output torque of the motor 10 in
10 accordance with the steering shaft conversion takes on “GTh
 T.” On the other hand, the road-surface load torque Ta
does not change, and thus, a relational expression of
Expression (23) described below can be obtained.
[0072]
15 [Expression Figure-23]
[0073]
With respect to Expression (23) described above,
Expression (22) described above is substituted so that the
20 road-surface load torque Ta is cancelled, and expressional
transformation is performed, whereby the following
Expression (24) can be obtained.
[0074]
33
[Expression Figure-24]
[0075]
The left-hand side of Expression (24) described above
5 indicates a difference value at a time when steering torque
of a driver changes from a state of “Th” to that of “Th'”
because of abnormal torque T in accordance with steering
shaft conversion which is caused in the motor 10, and thus,
it is possible to express the left-hand side thereof as the
10 quantity of change in steering torque, Th, of the driver at
the time of causing the abnormal torque T in the motor 10
in accordance with the steering shaft conversion, so that
the relational expression can be given by Expression (25)
described below.
15 [0076]
[Expression Figure-25]
[0077]
By using Expression (24) and Expression (25) described
20 above, a relational expression between steering-shaft
converted abnormal torque T being caused in the motor 10
34
and a steering torque change-quantity Th of a driver can be
obtained by Expression (26) described below.
[0078]
[Expression Figure-26]
5
[0079]
Meanwhile, as a result of causing abnormality in the
FET-gate drive unit 6 and/or the motor driving device 7,
steering-shaft converted abnormal torque of the motor 10,
10 which may possibly be caused at a time when an electric
current(s) does not flow through the motor 10 exactly as a
target q-axis current Iq_t of the electric-current
instruction calculation unit 3 through the motor 10, is
exactly given as in Expression (21). Therefore, a steering
15 torque change-quantity Th of the driver which may possibly
be caused due to the abnormality is given by Expression (27)
described below, by using abnormal torque Temu estimated by
the abnormal torque estimation unit 16.
[0080]
20 [Expression Figure-27]
35
[0081]
Thus, when a steering torque change-quantity of the
driver in which determination is performed so that an
electric power steering device is “abnormal” is defined as
5 “Thth,” abnormality of the electric power steering device
can be determined, in accordance with abnormal torque Temu
estimated by the abnormal torque estimation unit 16, by
Expression (28) described below.
[0082]
10 [Expression Figure-28]
[0083]
Expression (21) is substituted into “Temu” of
Expression (28) described above, and expressional
15 transformation is performed, whereby, by using parameter
IDiff which is calculated in the process of estimating
abnormal torque by the abnormal torque estimation unit 16,
Expression (29) described below can be obtained by which the
determination is performed on an electric power steering
20 device so that it is abnormal.
[0084]
[Expression Figure-29]
36
[0085]
In accordance with Expression (29) described above, it
is possible to give a relational expression between a
5 steering torque change-quantity Th and the maximum
deviation IDiff, so that it becomes possible to indirectly
estimate abnormal torque by calculating an electric-current
deviation difference maximum-value IDiff and by monitoring
it.
10 [0086]
FIG. 2 is a diagram illustrating the general outlines
of the operations of the electric power steering device
according to Embodiment 1. Steering torque of an operator
or driver is acquired in a steering torque detection process
15 201 by means of the torque detector 2 shown in FIG. 1, and,
by means of the electric-current instruction calculation
unit 3, a target q-axis current Iq_t and a target d-axis
current Id_t are calculated in an electric-current
instruction calculation process 301 in accordance with the
20 steering torque (detection torque Td shown in FIG. 1)
detected in the steering torque detection process 201, and
with a vehicle speed signal Sv shown in FIG. 1. The steering
torque detection process 201 and the electric-current
37
instruction calculation process 301 are repeatedly carried
out on a period t0 basis.
[0087]
On the basis of a detection signal from the rotor’s
5 rotational position detector 13, an electrical angle is
detected in an electrical angle detection process 1401 by
means of the electrical angle calculation unit 14. By means
of the detected phase-current calculation unit 15, phase
currents of the individual phases are calculated in a
10 detected phase-current detection process 1501. In a
detected electric-currents’ three- to two-phase
transformation process 401 of the control calculation unit
4, a detected q-axis current Iq_m and a detected d-axis
current Id_m are calculated in accordance with an output of
15 the electrical angle detection process 1401 and with that of
the detected phase-current detection process 1501. In a
target q-axis voltage/target d-axis voltage calculation
process 402, a target q-axis voltage is calculated so that
a target q-axis current Iq_t of the electric-current
20 instruction calculation process 301 is made coincident with
a detected q-axis current Iq_m of the detected electriccurrents’ three- to two-phase transformation process 401
each other, by using the processing of, for example, a PI
feedback control or the like. In addition, a target d-axis
25 voltage is calculated so that a target d-axis current Id_t of
38
the electric-current instruction calculation process 301 is
made coincident with the detected d-axis current Id_m of the
detected electric-currents’ three- to two-phase
transformation process 401 each other, by using the
5 processing of, for example, the PI feedback control or the
like. The target q-axis voltage and the target d-axis
voltage described above are transformed in accordance with
target voltages’ two- to three-phase transformation 403 into
three-phase voltages to be applied to the motor 10, and, in
10 a PWM output process 404, PWM signals are outputted into the
FET-gate-driving cutoff unit 5 shown in FIG. 1. The
electrical angle detection process 1401, the detected phasecurrent detection process 1501, the detected electriccurrents’ three- to two-phase transformation process 401,
15 the target q-axis voltage/target d-axis voltage calculation
process 402, the target voltages’ two- to three-phase
transformation 403 and the PWM output process 404 are
repeatedly carried out on a period t1 basis which is shorter
than the period t0 described above.
20 [0088]
FIG. 3 is a diagram schematically showing an overview
of a time chart of target q-axis currents Iq_t and detected
q-axis currents Iq_m in the processing having been explained
by FIG. 2. As described before, it is shown in the figure
25 that a target q-axis current Iq_t is updated in its value on
39
the period t0 basis, and that applied voltages of the motor
10 are controlled on the period t1 basis so that a detected
q-axis current Iq_m is made coincident with the target q-axis
current Iq_t each other. Therefore, at a time-point when the
5 target q-axis current Iq_t is updated, the deviation is caused
between a target q-axis current Iq_t(n) and a detected q-axis
current Iq_m(k); meanwhile thereat, the deviation between a
target q-axis current Iq_t(n  1) before time t0 and the
detected q-axis current Iq_m(k) becomes smaller. Because it
10 can be similarly considered also for the relationship between
a target d-axis current Id_t(n  1) before the time t0 and a
detected d-axis current Id_m(k), it can be considered that
the deviation becomes the smallest. In addition, because
three-phase voltages Vu(k  1), Vv(k  1) and Vw(k  1) to be
15 applied to the motor 10 are calculated in the processing
which is carried out on a period t1 basis by using an
electrical angle of (k  1) for detected phase currents
Iu(k), Iv(k) and Iw(k) at a time-point when the Iq_t(n) is
updated, and because phase currents are made flowed in
20 accordance with these three-phase voltages, it can be
considered that the deviations each between target phase
currents Iu_t, Iv_t and Iw_t, calculated by using a target qaxis current Iq_t(n  1) before time t0 and a target d-axis
current Id_t(n  1) before the time t0 and by using the
25 electrical angle (k  1) before time t1, and detected phase
40
currents Iu(k), Iv(k) and Iw(k) are small. Because of this,
it is preferable to carry out, after having carried out the
electric-current instruction calculation process 301, the
processing of abnormal torque estimation 1601 in the abnormal
5 torque estimation unit 16 by defining as inputs the target
phase currents Iu_t, Iv_t and Iw_t calculated by using the
target q-axis current Iq_t(n  1) at the time t0 before and
the target d-axis current Id_t(n  1) thereat, and by using
the electrical angle (k  1) before the time t1, and also
10 by defining as inputs the detected phase currents Iu(k),
Iv(k) and Iw(k).
[0089]
The explanation will be made referring to FIG. 4 for
an example of the processing of abnormal torque estimation
15 1601. As illustrated in the figure, the processing of a
two- to three-phase transformation process 1602 and that of
an abnormal torque estimation-value calculation process 1603
are carried out.
[0090]
20 In the two- to three-phase transformation process 1602,
an electrical angle (k  1) of the electrical angle
detection process 1401, and a target q-axis current Iq_t(n 
1) of the electric-current instruction calculation process
301 and a target d-axis current Id_t(n  1) thereof are
25 inputted, so that their two- to three-phase transformation
41
is carried out by using Expression (30) described below, and
that a target phase-U current Iu_t, a target phase-V current
Iv_t and a target phase-W current Iw_t are outputted.
[0091]
5 [Expression Figure-30]
[0092]
In the abnormal torque estimation-value calculation
process 1603, an electric-current deviation difference
10 maximum-value IDiff corresponding to an abnormal torque
estimation value Tae is calculated in accordance with a
target phase-U current Iu_t of the two- to three-phase
transformation process 1602, a target phase-V current Iv_t
thereof and a target phase-W current Iw_t thereof, and with
15 a phase-U detected electric-current Iu(k) of the detected
phase-current calculation unit 15, a phase-V detected
electric-current Iv(k) thereof and a phase-W detected
electric-current Iw(k) thereof.
[0093]
20 The explanation will be made referring to FIG. 5 for
processing contents in the abnormal torque estimation-value
calculation process 1603. At Step S101 shown in FIG. 5, the
42
calculation of electric current deviations (Iu, Iv and Iw)
is carried out. The detailed processing of Step S101 is
constituted of Steps S201, S202 and S203 shown in FIG. 6;
and a phase-U detected electric-current Iu(k), a phase-V
5 detected electric-current Iv(k) and a phase-W detected
electric-current Iw(k) are individually subtracted in each
phase from a target phase-U current Iu_t, a target phase-V
current Iv_t and a target phase-W current Iw_t shown in FIG.
4, so that electric current deviations Iu, Iv and Iw are
10 calculated, respectively.
[0094]
At Step S102 shown in FIG. 5, minimum-value and
maximum-value selection of electric current deviations
calculated is carried out on an every phase basis. The
15 detailed processing of minimum-value selection at Step S102
is exactly shown as in FIG. 7; and first, the aforementioned
electric current deviation Iu is substituted as a minimum
value of electric current deviation at Step S301 in FIG. 7,
and a minimum value of electric current deviation and the
20 aforementioned electric current deviation Iv are compared
with each other at Step S302. When the electric current
deviation Iv is smaller than the minimum value of electric
current deviation, the minimum value of electric current
deviation is updated at Step S303 to the electric current
25 deviation Iv. When an electric current deviation Iv is
43
larger than a minimum value of electric current deviation at
Step S302, a minimum value of electric current deviation and
the aforementioned electric current deviation Iw are
compared with each other at Step S304. When the electric
5 current deviation Iw is smaller than the minimum value of
electric current deviation, the aforementioned minimum value
of electric current deviation is updated at Step S305 to the
electric current deviation Iw. When the electric current
deviation Iw is larger than a minimum value of electric
10 current deviation at Step S304, no processing is carried out,
so that the minimum value of electric current deviation
remains as the aforementioned electric current deviation Iu.
[0095]
In addition, the detailed processing of maximum-value
15 selection at Step S102 of FIG. 5 is exactly shown as in FIG.
8; and first, the aforementioned electric current deviation
Iu is substituted as a maximum value of electric current
deviation at Step S401 in FIG. 8, and a maximum value of
electric current deviation and the aforementioned electric
20 current deviation Iv are compared with each other at Step
S402. When the electric current deviation Iv is larger than
the maximum value of electric current deviation, the maximum
value of electric current deviation is updated at Step S403
to the electric current deviation Iv. When an electric
25 current deviation Iv is smaller than a maximum value of
44
electric current deviation at Step S402, a maximum value of
electric current deviation and the aforementioned electric
current deviation Iw are compared with each other at Step
S404. When the electric current deviation Iw is larger than
5 the maximum value of electric current deviation, the maximum
value of electric current deviation is updated at Step S405
to the electric current deviation Iw. When an electric
current deviation Iw is smaller than a maximum value of
electric current deviation at Step S404, no processing is
10 carried out, so that the maximum value of electric current
deviation remains as the aforementioned electric current
deviation Iu.
[0096]
Next, at Step S103 shown in FIG. 5, the calculation of
15 an electric-current deviation difference maximum-value IDiff
is carried out. To be specific, a minimum value of electric
current deviation acquired at Step S102 is subtracted from
a maximum value of electric current deviation acquired at
Step S102, so that an electric-current deviation difference
20 maximum-value IDiff is calculated.
[0097]
The explanation described above is based on the premise
that, as shown in the schematic time chart of FIG. 3,
detected electric-currents are to coincide with target
25 electric-currents before time t0, at a time when a target q-
45
axis current is updated by means of an electric current
control for performing the feedback of a detected q-axis
current Iq_m and that of a detected d-axis current Id_m through
the control calculation unit 4 with respect to a target q5 axis current Iq_t and a target d-axis current Id_t being
updated on an every t0 basis.
However, as far as a feedback control is applied, it
is unavoidable to cause the deviations between the target
electric-currents and the detected electric-currents due to
10 a sharp change of a target electric-current(s) and/or an
influence of disturbance. FIG. 9 is a diagram schematically
showing, by way of one example, an overview of a time chart
at a time when such an exemplary case described above is
caused. For example, it is shown that, at a time-point “A,”
15 a detected electric-current Iq_m does not coincide with a
target q-axis current Iq_t(n  1) before time t0, but does
reach between a target q-axis current Iq_t(n  2) further
before the time t0 and the target q-axis current Iq_t(n  1)
before the time t0. Hereinbefore, the explanation has been
20 made for the q-axis current; however, it can be similarly
considered also for the d-axis current.
[0098]
In addition, because applied voltages of phases each
to the motor 10 change in accordance with the rotation at a
25 time when a rotational speed of the motor 10 is fast,
46
response delays are included in electric currents of the
individual phases in the motor with respect to changes of
the applied voltages of phases each due to the rotation of
the motor 10. FIG. 10 is a diagram schematically showing,
5 by way of one example, an overview of a time chart of
electrical angle’s changes at a time when a rotational speed
of the motor 10 is fast. For example, at a time-point “A,”
an electrical angle is at a value of (k), and so, target
phase currents are not coincident with detected phase
10 currents by using the electrical angle thereat even when the
target phase currents are calculated by performing two- to
three-phase transformation on a target q-axis current Iq_t
and on a target d-axis current Id_t. Therefore, it is
appropriately understood that, for an electrical angle being
15 used in the two- to three-phase transformation, parameter
(k  1) before time t1 and/or parameter (k  2) further
before the time t1 are used.
[0099]
As described above, it is preferable to calculate
20 targeted individual phase currents in accordance with the
combinations of target electric-currents and electrical
angles described as follows:
the combination of a target q-axis current Iq_t(n  1)
and a target d-axis current Id_t(n  1), and an electrical
25 angle (k  1);
47
the combination of a target q-axis current Iq_t(n  2)
and a target d-axis current Id_t(n  2), and the electrical
angle (k  1);
the combination of the target q-axis current Iq_t(n 
5 1) and the target d-axis current Id_t(n  1), and an
electrical angle (k  2); and
the combination of the target q-axis current Iq_t(n 
2) and the target d-axis current Id_t(n  2), and the
electrical angle (k  2).
10 [0100]
FIG. 11 is a diagram showing, by way of one example,
another exemplary embodiment of the abnormal torque
estimation unit 16 based on the aforementioned statements.
In two- to three-phase transformation processes 1606, in FIG.
15 11, an electrical angle (k  1), an electrical angle (k 
2), a target q-axis current Iq_t(n  1), a target d-axis
current Id_t(n  1), a target q-axis current Iq_t(n  2) and
a target d-axis current Id_t(n  2) are inputted. And then,
the respective two- to three-phase transformation processes
20 are carried out in accordance with the four combinations of
the target electric-currents and the electrical angles
described above, and respective maximum values and minimum
values of individual phase target currents are calculated,
so that a target phase-U current maximum-value Iu_t_Max, a
25 target phase-U current minimum-value Iu_t_Min, a target phase-
48
V current maximum-value Iv_t_Max, a target phase-V current
minimum-value Iv_t_Min, a target phase-W current maximum-value
Iw_t_Max and a target phase-W current minimum-value Iw_t_Min are
outputted. In an abnormal torque estimation-value
5 calculation process 1607, parameter “IDiff” corresponding to
an abnormal torque estimation value Tae is outputted, by
undergoing through predetermined processing, in accordance
with the target phase-U current maximum-value Iu_t_Max, the
target phase-U current minimum-value Iu_t_Min, the target
10 phase-V current maximum-value Iv_t_Max, the target phase-V
current minimum-value Iv_t_Min, the target phase-W current
maximum-value Iw_t_Max and the target phase-W current minimumvalue Iw_t_Min, and in accordance with a phase-U detected
electric-current Iu(k), a phase-V detected electric-current
15 Iv(k) and a phase-W detected electric-current Iw(k).
[0101]
Next, the explanation will be made for the abnormal
torque estimation-value calculation process 1607 of FIG. 11.
Basic flows of processing contents in the abnormal torque
20 estimation-value calculation process 1607 are exactly shown
as in FIG. 5 and FIG. 6, similarly to the exemplary
embodiment of FIG. 4; and first, the calculation of electric
current deviations (Iu, Iv and Iw) is carried out at Step
S101. FIG. 12 shows the flows of detailed processing
25 contents with respect to those of FIG. 6, which are common
49
processing in conjunction with a phase-U, a phase-V and a
phase-W each; and so, the processing is carried out in the
order of the phase-U, the phase-V and the phase-W, so that
an electric current deviation Iu, an electric current
5 deviation Iv and an electric current deviation Iw are
acquired. First, at Step S501, a detected phase-X current
Ix is compared with a target phase-X current maximum-value
Ix_t_Max being an output signal of the two- to three-phase
transformation processes 1606. When a detected phase-X
10 current Ix is larger than a target phase-X current maximumvalue Ix_t_Max, an electric current deviation Ix takes on a
value in which the target phase-X current maximum-value
Ix_t_Max is subtracted from the detected phase-X current Ix at
Step S502.
15 [0102]
When a detected phase-X current Ix is smaller than a
target phase-X current maximum-value Ix_t_Max at Step S501,
the detected phase-X current Ix is compared with a target
phase-X current minimum-value Ix_t_Min at Step S503, so that,
20 when the detected phase-X current Ix is smaller than the
target phase-X current minimum-value Ix_t_Min, an electric
current deviation Ix takes on a value in which the target
phase-X current minimum-value Ix_t_Min is subtracted from the
detected phase-X current Ix at Step S504; and, when the
25 detected phase-X current Ix is larger than the target phase-
50
X current minimum-value Ix_t_Min, the electric current
deviation Ix takes on “0” at Step S505. Note that, in the
detected phase-X current Ix, the target phase-X current
maximum-value Ix_t_Max, the target phase-X current minimum5 value Ix_t_Min and the electric current deviation Ix,
parameter “X” designates any one of U, V and W, and parameter
“x,” any one of u, v and w, respectively.
[0103]
Next, minimum-value and maximum-value selection of
10 electric current deviations at Step S102 shown in FIG. 5 is
carried out. The detailed processing of minimum-value
selection at Step S102 of FIG. 5 is exactly shown as in FIG.
7, similarly to the exemplary embodiment of FIG. 3; and first,
the electric current deviation Iu is substituted into a
15 minimum value of electric current deviation at Step S301
shown in FIG. 7, and a minimum value of electric current
deviation and the aforementioned electric current deviation
Iv are compared with each other at Step S302. When the
electric current deviation Iv is smaller than the minimum
20 value of electric current deviation, the minimum value of
electric current deviation is updated at Step S303 to the
aforementioned electric current deviation Iv. When an
electric current deviation Iv is larger than a minimum value
of electric current deviation at Step S302, a minimum value
25 of electric current deviation and the electric current
51
deviation Iw are compared with each other at Step S304.
When the electric current deviation Iw is smaller than the
minimum value of electric current deviation, the
aforementioned minimum value of electric current deviation
5 is updated at Step S305 to the electric current deviation
Iw. In addition, when the electric current deviation Iw
is larger than a minimum value of electric current deviation
at Step S304, no processing is carried out, so that the
minimum value of electric current deviation remains as the
10 electric current deviation Iu.
[0104]
The detailed processing of maximum-value selection at
Step S102 of FIG. 5 is exactly shown as in FIG. 8, similarly
to the exemplary embodiment of FIG. 3; and first, the
15 electric current deviation Iu is substituted into a maximum
value of electric current deviation at Step S401 shown in
FIG. 8, and a maximum value of electric current deviation
and the electric current deviation Iv are compared with each
other at Step S402. When the electric current deviation Iv
20 is larger than the maximum value of electric current
deviation, the maximum value of electric current deviation
is updated at Step S403 to the aforementioned electric
current deviation Iv. When an electric current deviation
Iv is smaller than a maximum value of electric current
25 deviation at Step S402, a maximum value of electric current
52
deviation and the electric current deviation Iw are compared
with each other at Step S404. When the electric current
deviation Iw is larger than the maximum value of electric
current deviation, the maximum value of electric current
5 deviation is updated at Step S405 to the electric current
deviation Iw. In addition, when an electric current
deviation Iw is smaller than a maximum value of electric
current deviation at Step S404, no processing is carried out,
so that the maximum value of electric current deviation
10 remains as the aforementioned electric current deviation Iu.
[0105]
Next, the calculation of an electric-current deviation
difference maximum-value IDiff at Step S103 in FIG. 5 is
carried out. To be specific, the minimum value of electric
15 current deviation acquired at Step S102 is subtracted from
the maximum value of electric current deviation similarly to
the exemplary embodiment of FIG. 3, so that an electriccurrent deviation difference maximum-value IDiff is
calculated.
20 [0106]
As described above, it can be known that an abnormal
torque estimation value Tae being an output signal of the
abnormal torque estimation unit 16 in FIG. 1, namely, the
electric-current deviation difference maximum-value IDiff is
25 calculated.
53
[0107]
Next, the explanation will be made for the processing
of the cutoff management unit 17 in FIG. 1. The cutoff
management unit 17 is provided with the function to perform
5 determination whether the motor 10 is to be brought about in
a motor drive-operation state or brought about in a motor
drive-operation halt state, by using a value of electriccurrent deviation difference maximum-value IDiff being an
abnormal torque value estimated by the abnormal torque
10 estimation unit 16, a battery voltage Vba detected by the
battery voltage detection unit 20, and a motor revolution
number calculated from an electrical angle  detected by the
electrical angle calculation unit 14. And, only when
determination of a drive-operation halt state is performed,
15 the cutoff management unit is provided with the function to
transmit a forced PWM-signal turn-off instruction INoff
toward the control calculation unit 4, and also to transmit
an FET-gate-driving cutoff instruction INcg toward the FETgate-driving cutoff unit 5, and a battery voltage cutoff
20 instruction INcv toward the battery voltage cutoff unit 19.
[0108]
In addition, FIG. 13 shows processing flows of the
cutoff management unit 17; and first, at Step S601,
determination is performed whether or not a motor revolution
25 number calculated from an electrical angle  obtained from
54
the electrical angle calculation unit 14 is at a revolutionnumber threshold value being set in advance or less.
[0109]
In a surface permanent magnet type motor, its induced
5 voltages proportional to the number of revolutions of the
motor, i.e. a motor revolution number, are generated.
Electric currents capable of being flowed through the motor
10 are limited by voltage differences between the induced
voltages in accordance with the rotation of the motor and a
10 battery voltage. Due to the phenomenon, line currents
flowing through the motor are reduced with respect to
instruction electric-currents calculated from detection
torque detected by the torque detector 2, whereby it is
feared that individual phase detected electric-currents Iu,
15 Iv and Iw result in having electric current deviations with
respect to targeted individual phase currents Iu_t, Iv_t and
Iw_t being expected values, and that a value of electriccurrent deviation difference maximum-value IDiff being a
calculation result of the abnormal torque estimation unit 16
20 takes on a threshold value being set or more, so that
erroneous determination may be induced. For this reason, it
can be known that, as a monitoring condition, Step S601 being
the monitoring condition of a motor revolution number is set
in the processing flows of FIG. 13. When a motor revolution
25 number exceeds a threshold value of the number of revolutions,
55
i.e. a revolution-number threshold value, being set in
advance, a drive operation of the motor 10 is not halted.
[0110]
It is so arranged that, by obeying a battery voltage
5 Vba detected by the battery voltage detection unit 20, the
revolution-number threshold value being the monitoring
condition of the motor revolution number undergoes a variable
operation; and so, for example, the revolution-number
threshold value is determined in such a manner that the
10 number of revolutions to reach at a voltage saturation is
prepared on a battery voltage basis as a map in advance, and
a battery voltage Vba detected in a certain constant period
is checked against the map. In addition, after having taken
an operationally capable voltage(s) of system into
15 consideration, it may be adopted that, without carrying out
the variable operation of a revolution-number threshold
value in accordance with a battery voltage Vba, a revolutionnumber fixed-value which does not induce erroneous
determination is used for the revolution-number threshold
20 value.
[0111]
When a motor revolution number is at a revolutionnumber threshold value or less at Step S601 of FIG. 13, an
electric-current deviation difference maximum-value IDiff is
25 compared with a threshold value at Step S602. The threshold
56
value is given by the right-hand side of Expression (28)
described above, which takes on a value calculated, by using
a steering torque change-quantity Th being set in advance
corresponding to an adapted system, and by obeying a torque
5 constant (torque/electric-current conversion coefficient) Km
[Nm/A], a motor reduction ratio Ggear, an output gain G of
the motor 10 with respect to detection torque, and a safety
requirements and/or a safety target(s).

We Claim :
[claim 1] A motor control device, comprising:
5 an electric-current instruction calculation unit for
determining target electric-current values of a motor of
multiple phases;
a detected phase-current calculation unit for
obtaining detected electric-current values by detecting
10 electric currents of phases each flowing through the motor;
a control calculation unit for calculating electriccurrent instructions to control motor electric-currents each
on a basis of the target electric-current values and on that
of the detected electric-current values;
15 an electrical angle calculation unit for calculating,
from a rotational position of a rotor of the motor, an
electrical angle being a phase difference between a
rotational position of the rotor and a coil of the motor;
an abnormal torque estimation unit for estimating
20 abnormality of torque of the motor from the target electriccurrent values, the detected electric-current values and the
electrical angle; and
a cutoff management unit for halting a drive operation
of the motor if the torque thereof is abnormal.
25
117
[claim 2] The motor control device as set forth in claim 1,
wherein the cutoff management unit halts a drive operation
of the motor when determination is performed so that a change
of torque estimated by the abnormal torque estimation unit
5 is abnormal.
[claim 3] The motor control device as set forth in claim 1
or claim 2, wherein the cutoff management unit does not halt
a drive operation of the motor, when a motor revolution
10 number calculated from an electrical angle obtained by the
electrical angle calculation unit exceeds a revolutionnumber threshold value being set in advance.
[claim 4] The motor control device as set forth in claim 3,
15 wherein the revolution-number threshold value is changed in
accordance with a voltage inputted into the motor.
[claim 5] The motor control device as set forth in claim 1,
further comprising:
20 a detected phase-current abnormality detection unit
for detecting abnormality in the detected phase-current
calculation unit; and
an open failure determination unit for determining, by
using a detected phase current of the detected phase-current
25 calculation unit and a voltage instruction of the control
118
calculation unit, a phase in which an open failure is caused
at a time when torque of the motor is estimated being
abnormal by the abnormal torque estimation unit, and also at
a time when the detected phase-current abnormality detection
5 unit determines abnormality in the detected phase-current
calculation unit, wherein
the cutoff management unit continues or halts a drive
operation of the motor in accordance with a determination
result of the open failure determination unit.
10
[claim 6] The motor control device as set forth in claim 5,
wherein
the open failure determination unit determines that an
open failure is caused in a selected phase in a case in which
15 a detected phase current of a selected phase from the
detected phase-current calculation unit is at a
predetermined value defined in advance or less, and also an
absolute value of the voltage instruction of a selected phase
from the control calculation unit is larger than an absolute
20 value of the voltage instruction of another phase from the
control calculation unit; or
the open failure determination unit determines that an
open failure is caused in a selected phase in a case in which
an absolute value of a phase current of a selected phase
25 from the detected phase-current calculation unit is at a
119
predetermined value defined in advance or less, and also any
one of absolute values of phase currents of other phases
from the detected phase-current calculation unit is larger
than a predetermined value defined in advance.
5
[claim 7] The motor control device as set forth in claim 5
or claim 6, wherein the detected phase-current abnormality
detection unit calculates a total sum-value of individual
phase detected electric-current values detected by the
10 detected phase-current calculation unit, and determines that
a detected phase current of the detected phase-current
calculation unit is abnormal, when the total sum-value is at
a predetermined threshold value defined in advance or more.
15 [claim 8] The motor control device as set forth in any one
of claims 5 through 7, wherein the cutoff management unit
does not perform a continuity process in a drive operation
of the motor nor a halt process therein, when a motor
revolution number calculated from an electrical angle
20 obtained by the electrical angle calculation unit exceeds a
revolution-number threshold value being set in advance.
[claim 9] The motor control device as set forth in claim 8,
wherein the revolution-number threshold value is varied in
25 accordance with a voltage inputted into the motor.
120
[claim 10] The motor control device as set forth in any one
of claims 5 through 9, wherein, in a case in which an open
failure is determined by the open failure determination unit,
5 and also a case in which at least an open failure of a single
phase only is caused, the cutoff management unit is provided
with a function to continue a motor drive operation by other
phases in which an open failure of the single phase only is
not caused, and a function to halt a motor drive operation
10 when an open failure of the single phase only is not caused.
[claim 11] An electric power steering device, comprising:
the motor control device as set forth in any one of
claims 1 through 10, wherein
15 the electric power steering device gives steering
assist torque by the motor with respect to a vehicle wheel
of an automotive vehicle.