WO2020194637A1

WO2020194637A1 - Control method and control device for electric vehicle

Info

Publication number: WO2020194637A1
Application number: PCT/JP2019/013468
Authority: WO
Inventors: 翔大野; 弘征小松; 藤原　健吾; 中島　孝
Original assignee: 日産自動車株式会社
Priority date: 2019-03-27
Filing date: 2019-03-27
Publication date: 2020-10-01

Abstract

In this control method for an electric vehicle: a basic torque command value is set on the basis of an operating state; a d-axis current command value and a first q-axis current command value with respect to the stator current, and an f-axis current command value with respect to the rotor current, are calculated on the basis of the basic torque command value and the vehicle information; a magnetic flux estimated value, which is an estimated value of magnetic flux generated in the rotor, is calculated on the basis of the d-axis current command value and the f-axis current command value; a first torque command value is calculated on the basis of the first q-axis current command value and the magnetic flux estimated value; calculation is performed to control vibration of the drive shaft torque transmission system of the electric vehicle, on the basis of the vehicle information, in relation to the first torque command value, whereby a vibration-damping torque command value is calculated; and a second q-axis current command value is calculated on the basis of the magnetic flux estimated value and the vibration-damping torque command value. The stator current and the rotor current are controlled on the basis of the second q-axis current command value, the d-axis current command value, and the f-axis current command value.

Description

Electric vehicle control method and control device

The present invention relates to a control method for an electric vehicle and a control device.

Conventionally, as a control method for an electric vehicle using a synchronous motor that uses a permanent magnet as a rotor as a power source, vibration suppression control that reduces torsional vibration of a drive shaft that connects the motor and drive wheels has been used.

However, in contrast to the above-mentioned synchronous motor in which the rotor magnetic flux generated in the motor is constant, in the field winding type synchronous motor which does not use a permanent magnet in the rotor, the above-mentioned vibration suppression control is performed because the rotor magnetic flux generated in the motor fluctuates. It is difficult to apply it as it is.

On the other hand, JP5939316B discloses a method of applying the above-mentioned vibration damping control to an induction motor in which the rotor magnetic flux fluctuates. However, what is disclosed in JP5939316B is a control method for applying the above vibration damping control to an induction motor by correcting the torque current based on the exciting current (γ-axis current). Therefore, even if the control method is applied to a field winding type synchronous motor in which it is necessary to consider the current flowing through the field winding of the rotor (f-axis current) and the d-axis current that generates reluctance torque. It is difficult to obtain a vibration damping effect.

An object of the present invention is to provide a technique for applying vibration damping control for reducing torsional vibration of a drive shaft connecting a motor and a drive wheel to a field winding type synchronous motor.

The method for controlling an electric vehicle according to one aspect of the present invention is to use an electric vehicle using a winding field synchronous motor as a drive source, which includes a rotor having a rotor winding and a stator having a stator winding. This is a control method for an electric vehicle that controls a stator current flowing through a stator winding and a rotor current flowing through a rotor winding. In the control method, a basic torque command value is set based on vehicle information, and based on the basic torque command value and vehicle information, a d-axis current command value for a stator current, a first q-axis current command value, and The f-axis current command value for the rotor current is calculated, and the magnetic flux estimated value, which is the estimated value of the torque generated in the rotor, is calculated based on the d-axis current command value and the f-axis current command value, and the first q The first torque command value is calculated based on the shaft current command value and the estimated magnetic flux value, and the vibration of the drive shaft torque transmission system of the electric vehicle is suppressed with respect to the first torque command value based on the vehicle information. The vibration damping torque command value is calculated, and the second q-axis current command value is calculated based on the magnetic flux estimated value and the vibration damping torque command value. Then, the stator current and the rotor current are controlled based on the second q-axis current command value, the d-axis current command value, and the f-axis current command value.

The embodiment of the present invention will be described in detail below together with the attached drawings.

FIG. 1 is a schematic configuration diagram of a vehicle system to which the electric vehicle control method of the first embodiment is applied. FIG. 2 is a flowchart showing a flow of processing performed by the electric motor controller. FIG. 3 is a diagram showing an example of an accelerator opening degree-torque table. FIG. 4 is a block diagram of the motor control system of the first embodiment. FIG. 5 is a block diagram of the q-axis current control unit. FIG. 6 is a block diagram of the d-axis current control unit. FIG. 7 is a block diagram of the f-axis current control unit. FIG. 8 is a block diagram of the vibration damping control calculation unit. FIG. 9 is a block diagram of the magnetic flux estimator. FIG. 10 is a block diagram of a reluctance torque equivalent magnetic flux estimator. FIG. 11 is a block diagram of the field magnetic flux estimator. FIG. 12 is a block diagram of a vibration damping torque command value calculator. FIG. 13 is a diagram for explaining the equation of motion of the electric vehicle. FIG. 14 is a diagram showing the characteristics of the bandpass filter H (s). FIG. 15 is a time chart showing the control result by the control method of the electric vehicle. FIG. 16 is a block diagram of the motor control system of the second embodiment. FIG. 17 is a block diagram of the f-axis current control unit. FIG. 18 is a block diagram of the f-axis F / F compensator. FIG. 19 is a block diagram of an f-axis current model. FIG. 20 is a block diagram of an f-axis current F / B model. FIG. 21 is a block diagram of the f-axis limit processing unit. FIG. 22 is another example of the block diagram of the f-axis limit processing unit. FIG. 23 is a block diagram of the f-axis F / B compensator. FIG. 24 is a block diagram of an f-axis robust compensator. FIG. 25 is a flowchart showing the control process of the motor. FIG. 26 is a block diagram of a reluctance torque equivalent magnetic flux estimator. FIG. 27 is a block diagram of the field magnetic flux estimator. FIG. 28 is a block diagram of the vibration damping control calculation unit. FIG. 29 is a time chart showing the control result by the control method of the electric vehicle.

<First Embodiment>
FIG. 1 is a block diagram showing a configuration example of a motor control system 100 to which the electric vehicle control method according to the embodiment of the present invention is applied. An electric vehicle is a vehicle that is provided with at least one winding field type synchronous motor (hereinafter, also simply referred to as a motor) as a part or all of the drive source of the vehicle and can run by the driving force of the motor. It includes electric vehicles and hybrid vehicles.

The battery 1 discharges the drive power of the winding field type synchronous motor 4 and charges the regenerative power of the motor 4.

The electric motor controller 2 (hereinafter, also simply referred to as a controller) is composed of, for example, a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), and an input / output interface (I / O interface). To. The controller 2 has a vehicle speed V, an accelerator opening degree θ, an electric angle θre of the motor 4, a stator current of the motor 4 (iu, iv, iw in the case of three-phase alternating current), and a rotor current (if) of the motor 4. Signals of various vehicle variables indicating the vehicle state such as are input as digital signals. The controller 2 generates a PWM signal for controlling the motor 4 based on the input signal. Further, the controller 2 generates a drive signal of the inverter 3 according to the generated PWM signal.

The inverter 3 is supplied from the battery 1 by turning on / off two switching elements (for example, power semiconductor elements such as IGBTs and MOS-FETs) provided for each phase in order to control the stator current. The direct current is converted to alternating current or inversely converted, and a desired current is passed through the motor 4. Further, in the inverter 3, two pairs (four in total) of switching elements (for example, power semiconductor elements such as IGBTs and MOS-FETs) are connected to both ends of the rotor winding in order to control the rotor current. By turning these on / off according to the drive signal, a desired current is passed through the rotor winding. However, when the direction of the current flowing through the rotor is only one direction, two of the two pairs of switching elements located diagonally may be replaced with diodes.

The winding field type synchronous motor 4 (hereinafter, simply referred to as “motor 4”) has a rotor having a rotor winding (field winding) and a stator having a stator winding (armature winding). It is a winding field type synchronous motor equipped with. When the motor control system 100 of the present embodiment is mounted on a vehicle, the motor 4 serves as a drive source for the vehicle. Although the details will be described later, the motor 4 is controlled by controlling the rotor current flowing through the rotor winding and the stator current flowing through the stator winding. The motor 4 generates a drive torque by the current supplied from the inverter 3, and transmits the drive force to the left and right drive wheels 9 via the speed reducer 5 and the drive shaft 8. Further, the motor 4 recovers the kinetic energy of the vehicle as electric energy by generating a regenerative driving force when the motor 4 is rotated by the drive wheels 9 while the vehicle is traveling. In this case, the inverter 3 converts the alternating current generated during the regenerative operation of the motor 4 into a direct current and supplies it to the battery 1.

The current sensor 7 detects the three-phase currents iu, iv, and iwa (stator current) flowing in the stator winding of the motor 4, and also detects the current if (rotor current) flowing in the rotor winding of the motor 4. To do. However, since the sum of the three-phase AC currents iu, iv, and iw is 0 for the stator current, the current of any two phases may be detected and the current of the remaining one phase may be obtained by calculation.

The rotation sensor 6 is, for example, a resolver or an encoder, and detects the rotor phase α of the motor 4.

FIG. 2 is a flowchart showing the flow of processing performed by the controller 2. The processes according to steps S201 to S204 are programmed in the controller 2 so as to be constantly executed at regular intervals while the vehicle system is running.

In step S201, a signal indicating the vehicle state is input to the controller 2. Here, the vehicle speed V (km / h), the accelerator opening θ (%), the electric angle θre of the motor 4, the motor rotation speed Nm (rpm) of the motor 4, and the currents iu, iv, if, if, which flow through the motor 4. Then, the DC voltage value V _dc (V) of the battery 1 is input.

The vehicle speed V (km / h) is acquired by communication from a meter (not shown), a vehicle speed sensor, or another controller such as a brake controller. Alternatively, the controller 2 obtains the vehicle speed v (m / s) by multiplying the rotor mechanical angular velocity ωm by the tire driving radius r and dividing by the gear ratio of the final gear, and changes from m / s to km / s. The vehicle speed V (km / h) is obtained by multiplying by the unit conversion coefficient (3600/1000).

The accelerator opening θ (%) is obtained from an accelerator opening sensor (not shown). The accelerator opening degree θ (%) may be obtained from another controller such as a vehicle controller (not shown).

The electric angle θ _re (rad) of the motor 4 is acquired from the rotation sensor 6. For the rotation speed N _m (rpm) of the motor 4, the electric angular velocity ω _re is divided by the pole pair number p of the electric motor to obtain the motor rotation speed detection value ω _m (rad / s) which is the mechanical angular velocity of the motor 4. It is obtained by multiplying the obtained motor rotation speed detection value ω _m by the unit conversion coefficient (60 / (2π)) from rad / s to rpm.

The currents iu, iv, if, and if (A) flowing through the motor 4 are acquired from the current sensor 7.

The DC current value V _dc (V) is detected by a voltage sensor (not shown) provided in the DC power supply line between the battery 1 and the inverter 3. The DC voltage value V _dc (V) may be detected by a signal transmitted from the battery controller (not shown).

In step S202, the motor torque command value calculation process is executed. In the motor torque command value calculation process, the motor torque command value (motor torque command value () is obtained by referring to the accelerator opening-torque table shown in FIG. 3 based on the operation information such as the accelerator opening θ and the vehicle speed V input in step S201. Basic torque command value) T _m ^* is set.

In step S203, the vibration damping control calculation process is executed. Specifically, the controller 2 is based on the motor torque command value T _m ^* set in step S202, without wasting the response of the drive shaft torque, and the driving force transmission system vibration (torsion vibration of the drive shaft 8 or the like). ) Is suppressed, the q-axis current command value I _q2 ^* , the d-axis current command value i _d1 ^* , and the f-axis current command value i _f1 ^* are calculated. The details of the vibration damping control calculation process will be described later.

In step S204, the current control calculation process is executed. In the current control arithmetic processing, the d-axis current i _d , the q-axis current i _q, and the f-axis current i _f are obtained by the q-axis current command value i _q2 ^* , d-axis current command value i _d1 ^*, and f-axis obtained in step S203. Current control is performed to match each with the current command value I _f1 ^* . The details of the current control calculation processing will be described below with reference to FIG.

FIG. 4 is a diagram showing a configuration example of the motor control system 100, and is a control block diagram of the current control calculation processing unit 2a provided by the controller 2 as one functional unit. The controller 2 uses the current control calculation processing unit 2a to execute the current control calculation processing according to step S204.

The current control calculation processing unit 2a includes a stator PWM conversion unit 401, a rotor PWM conversion unit 402, a look-ahead compensation unit 403, coordinate

conversion units

404 and 410, a non-interference control unit 405, and a q-axis current control unit. It includes a 406, a d-axis current control unit 407, an f-axis current control unit 408, a voltage command value calculation unit 409, and an A / D conversion unit 411.

The stator PWM conversion unit 401 performs PWM_Duty to the stator switching element included in the inverter 3 based on the three-phase voltage command values v _u ^* , v _v ^* , and v _w ^* output from the coordinate conversion unit 410 described later. Drive signals (high voltage element drive signals) D _uu ^* , D _ul ^* , D _vu ^* , D _vl ^* , D _wu ^* , D _wl ^* are generated and output to the inverter 3.

The rotor PWM conversion unit 402 generates PWM_Duty drive signals D _fu ^* and D _fl ^* for the rotor switching element included in the inverter 3 based on the f-axis voltage command value v _f ^* described later, and causes the inverter 3 to generate PWM_Duty drive signals D _fu ^* and D _fl ^*. Output.

The inverter 3 has an AC voltage v _u , v _v , for controlling the dq axis currents i _d and i _q flowing in the stator winding of the motor 4 based on the PWM_Duty drive signal generated by the stator PWM conversion unit 401. V _w is generated and supplied to the motor 4. Further, the inverter 3 generates an f-axis voltage v _f for controlling the f-axis current _if flowing in the rotor winding of the motor 4 based on the PWM_Duty drive signal generated by the rotor PWM conversion unit 402. It is supplied to the motor 4.

The current sensor 7 detects at least two phase currents, for example, u-phase current i _u and v-phase current i _v , among the three-phase AC currents supplied from the inverter 3 to the motor 4. The detected two-phase currents i _u and _iv are converted into digital signals by the A / D (analog / digital) conversion unit 411 and input to the coordinate conversion unit 404. The current sensor 7 detects the f axis current i _f fed from the inverter 3 to the motor 4. The detected f-axis current _if is converted into a digital signal by the A / D conversion unit 411 and output to the f-axis current control unit 408.

The look-ahead compensation unit 403 inputs the electric angle θ _re and the electric angular velocity ω _re, and adds the product of the electric angular velocity ω _re and the dead time of the control system to the electric angle θ _re to perform the look-ahead compensation. Calculate the electrical angle θ _re '. After pre-reading compensation, the electric angle θ _re'is output to the coordinate conversion unit 410.

The coordinate conversion unit 404 converts the three-phase AC coordinate system (uvw axis) to the orthogonal two-axis DC coordinate system (dq axis). Specifically, the coordinate conversion unit 404 performs coordinate conversion processing from the u-phase current i _u , the v-phase current i _v , the w-phase current i _w , and the electric angle θ _re using the following equation (1). By doing so, the d-axis current i _d and the q-axis current i _q are calculated.

The non-interference control unit 405 has an electric angular velocity ω _re , a d-axis current reference response _{id_ref} , a q-axis current reference response i _{q_ref} , and f output from the q-axis, d-axis, and f-axis current control units ₄₀₆ , ₄₀₇ , and ₄₀₈ . axis current nominal response i _{F_REF,} enter the differential value s · i _{d_ref} the d-axis current nominal response, and a partial value s · i _{F_REF} of f-axis current nominal response, d-axis, between the q-axis, and the f-axis The non-interference voltages v _{d_dcpl} , v _{q_dcpl} , and v _{f_dcpl} required to cancel the interference voltage are calculated using the voltage equation shown by the following equation (2). The following equation (2) is a voltage equation of the winding field type synchronous motor 4 which is the controlled object of the present invention.

However, each parameter of the above equation (2) is as follows. Note that s in the equation is a Laplace operator.
_id : d-axis current i _q : q-axis current i _f : f-axis current v _d : d-axis voltage v _q : q-axis voltage v _f : f-axis voltage L _d : d-axis inductance L _q : q-axis inductance L _f : F-axis inductance M: Mutual inductance between stator / rotor L _d ': d-axis dynamic inductance L _q ': q-axis dynamic inductance L _f ': f-axis dynamic inductance M': Controller / rotor Dynamic mutual inductance between R _a : Controller winding resistance R _f : Rotor winding resistance ω _re : Electric angular velocity

Here, if the non-interference control by the non-interference control unit 405 functions ideally, the voltage equation of the above equation (2) can be diagonalized as shown in the following equation (3).

According to the above equation (3), the characteristics from the voltage to the current of the d-axis, the q-axis, and the f-axis have a first-order delay as shown in the following equations (4), (5), and (6), respectively. ..

The q-axis current control unit 406 desires the q-axis current i _q , which is a measured value of the actual current (actual current), to the q-axis current command value (second q-axis current command value) i _q2 ^* without steady deviation. The first q-axis voltage command value v _{q_dsh} for _tracking with _{responsiveness} is calculated and output to the voltage command value calculation unit 409. Details of the q-axis current control unit 406 will be described later with reference to FIG.

d-axis current control unit 407, the actual current (actual current) measurements at a d-axis current i _d a d-axis current command value i _d1 ^* in steady-state error without first for follow a desired response The d-axis voltage command value v _{d_dsh} is calculated and output to the voltage command value calculation unit 409. Details of the d-axis current control unit 407 will be described later with reference to FIG.

f-axis current control section 408, the actual current (actual current) a measure of the f-axis current i _f the f-axis current command value i _f1 ^* the steady-state error without first for follow a desired response of The f-axis voltage command value v _{f_dsh} is calculated and output to the voltage command value calculation unit 409. Details of the f-axis current control unit 408 will be described later with reference to FIG. 7.

FIG. 5 is a diagram illustrating details of the q-axis current control unit 406 of the present embodiment. The q-axis current control unit 406 includes a control block 501,

gains

502 and 503, an integrator 504, a subtractor 505, and an adder 506.

Control block 501, q-axis current command value i _q2 ^* transfer characteristics of the first order lag of the response delay to simulate the actual current i _q for _{1 / (τ q s + 1} ) is. The control block 501 inputs the q-axis current command value i _q2 ^* and outputs the q-axis current norm response i _{q_ref} . Incidentally, the transfer characteristic _{1 / (τ q s + 1} ) in such a tau _q is a q-axis current nominal response time constant.

The gain 502 is a proportional gain K _pq and is represented by the following equation (7). The gain 502 takes the deviation between the q-axis current command value i _q2 ^* and the q-axis current i _q as an input, and outputs the value obtained by multiplying the input value by the proportional gain K _pq to the adder 506.

The gain 503 is a proportional gain K _iq and is represented by the following equation (8). The gain 503 takes the deviation between the q-axis current command value i _q2 ^* and the q-axis current i _q as an input, and outputs the value obtained by multiplying the input value by the proportional gain K _iq to the integrator 504. The output of the integrator 504 is input to the adder 506.

Then, the adder 506 calculates the first q-axis voltage command value V _{q_dsh} by adding the output of the gain 502 and the output of the integrator 504. As described above, the q-axis current control unit 406 sets the gains of the

gains

502 and 503 as shown in the above equations (7) and (8), whereby the q-axis current command value i _q2 ^* to the q-axis current i The transmission characteristics up to _q can be matched with the normative response represented by the following equation (9).

FIG. 6 is a diagram illustrating details of the d-axis current control unit 407 of the present embodiment. The d-axis current control unit 407 includes control blocks 601 and 602,

gains

603 and 604, an integrator 605, a subtractor 606, and an adder 607.

The control block 601 has a first-order delay transmission characteristic (d-axis current transmission characteristic) 1 / (τ _d s + 1) that simulates the response delay of the actual current (d-axis current _id ) with respect to the d-axis current command value _id1 ^* . .. Control block 601 receives as input the d-axis current command value i _d1 ^*, and outputs the d-axis current nominal response i _{d_ref.} Incidentally, the transfer characteristic _{1 / (τ d s + 1} ) in such a tau _d is the d-axis current nominal response time constant.

Control block 602, the transfer characteristic _{s / (τ d s + 1} ) for calculating a differential value of d-axis current nominal response i _{d_ref} against d-axis current command value i _d1 ^* is. Control block 602 receives as input the d-axis current command value i _d1 ^*, and outputs the d-axis current nominal response differential value s · i _{d_ref.}

The gain 603 is a proportional gain K _pd and is expressed by the following equation (10). Gain 603 inputs the deviation between the d-axis current command value i _d1 * and d-axis current i _d, and outputs a value obtained by multiplying a proportional gain K _pd input values to the adder 607.

The gain 604 is a proportional gain K _id and is represented by the following equation (11). Gain 604 inputs the deviation between the d-axis current command value i _d1 ^* and d-axis current i _d, and outputs a value obtained by multiplying a proportional gain K _id to the input value to the integrator 605. The output of the integrator 605 is input to the adder 607.

Then, the adder 607 calculates the first d-axis voltage command value v _{d_dsh} by adding the output of the gain 603 and the output of the integrator 605. As described above, the d-axis current control unit 407 sets the gains of the

gains

603 and 604 as shown in the above equations (10) and (11), whereby the d-axis current command value i _d1 ^* to the d-axis current i The transmission characteristics up to _d can be matched with the normative response represented by the following equation (12).

FIG. 7 is a diagram illustrating details of the f-axis current control unit 408 of the present embodiment. The f-axis current control unit 408 includes control blocks 701 and 702,

gains

703 and 704, an integrator 705, a subtractor 706, and an adder 707.

The control block 701 has a first-order delay transmission characteristic 1 / (τ _f s + 1) that simulates a response delay of the actual current _if with respect to the f-axis current command value i _f1 ^* . The control block 701 receives the f-axis current command value if _f1 ^* as an input, and outputs the f-axis current reference response _{if_ref} . Incidentally, the transfer characteristic _{1 / (τ f s + 1} ) in such a tau _f is the f-axis current nominal response time constant.

The control block 702 is a transmission characteristic s / (τ _f s + 1) for calculating the differential value of the f-axis current normative response _{if_ref} with respect to the f-axis current command value if _f1 ^* . The control block 702 receives the f-axis current command value if _f1 ^* as an input, and outputs the f-axis current norm response differential value s · _{if_ref} .

The gain 703 is a proportional gain K _pf and is represented by the following equation (13). Gain 703 inputs the deviation between the f-axis current command value i _f1 ^* and f-axis current i _f, and outputs a value obtained by multiplying a proportional gain K _pf the input value to the adder 707.

The gain 704 is a proportional gain K _if and is represented by the following equation (14). Gain 704 inputs the deviation between the f-axis current command value i _f1 ^* and f-axis current i _f, and outputs a value obtained by multiplying a proportional gain K _{an if} the input value to the integrator 705. The output of the integrator 705 is input to the adder 707.

Then, the adder 707 calculates the first f-axis voltage command value v _{f_dsh} by adding the output of the gain 703 and the output of the integrator 705. As described above, the f-axis current control unit 408 sets the gains of the

gains

703 and 704 as in the above equations (13) and (14), whereby the f-axis current command value i _f1 ^* to the f-axis current i The transmission characteristics up to _f can be matched with the normative response represented by the following equation (15).

The explanation will be continued by returning to FIG. The voltage command value calculation unit 409 has a first q-axis voltage command value v _{q_dsh} and a first d, which are outputs of the q-axis current control unit 406, the d-axis current control unit 407, and the f-axis current control unit 408. The shaft voltage command value v _{d_dsh} and the first f-axis voltage command value v _{f_dsh} are corrected by using the non-interference voltages v _{q_dcpl} , v _{d_dcpl} , and v _{f_dcpl} which are the outputs of the non-interference control unit 405 (in this embodiment). to add. Then, the voltage command value calculation unit 409 outputs the second q-axis voltage command value v _q ^* and the second d-axis voltage command value v _d ^* obtained by the correction to the coordinate conversion unit 410. , The second f-axis voltage command value v _f ^* is output to the rotor PWM converter 402.

The coordinate conversion unit 410 converts the orthogonal 2-axis DC coordinate system (dq-axis) rotating at the electric angular velocity ω _re into the 3-phase AC coordinate system (uvw phase). Specifically, the coordinate conversion unit 410 uses the input second d-axis voltage command value v _d ^* , the second q-axis voltage command value v _q ^* , and the pre-reading compensation post-electric angle θ _re '. By performing the coordinate conversion process using the following equation (16), the voltage command values v _u ^* , v _v ^* , and v _w ^* of each uvw phase are calculated.

Subsequently, the details of the vibration damping control process executed in step S203 (see FIG. 2) will be described.

FIG. 8 is a control block diagram of the vibration damping control calculation unit 2b provided in front of the current control calculation processing unit 2a shown in FIG. 1 as a functional unit of the controller 2. That is, the output from the vibration damping control calculation unit 2b is input to the current control calculation processing unit 2a. The controller 2 uses the vibration damping control calculation unit 2b to execute the vibration damping control process according to step S203.

The vibration damping control calculation unit 2b includes a first current command value calculator 801, a magnetic flux estimator 802, a first torque command value calculator 803, a vibration damping torque command value calculator 804, and a second q. It is configured to include an axial current command value calculator 805.

The first current command value calculator 801 inputs the motor torque command value T _m ^* , the motor rotation speed (mechanical angular speed) ω _rm, and the DC voltage V _dc, and the q-axis current command values i _q1 ^* , d. Calculate the shaft current command value i _d1 ^* and the f-axis current command value i _f1 ^* . The first current command value calculator 801 includes each of the q-axis current command value i _q1 ^* , the d-axis current command value i _d1 ^* , and the f-axis current command value if _f1 ^* , and the motor torque command value (basic torque command). Value) T _m ^* , motor rotation speed (mechanical angular speed) ω _rm , and map data that defines the relationship with the DC voltage V _dc are stored in advance, and each value is calculated by referring to the map data. The calculated q-axis current command value i _q1 ^* is output to the first torque command value calculator 803, and the d-axis current command value i _d1 ^* and the f-axis current command value i _f1 ^* are the magnetic flux estimator 802. Is output to.

FIG. 9 is a control block diagram of the magnetic flux estimator 802. The magnetic flux estimator 802 includes a reluctance torque equivalent magnetic flux estimator 901, a field magnetic flux estimator 902, and an adder 903.

Reluctance torque equivalent flux estimator 901 inputs the d-axis current command value i _d1 ^*, calculates the reluctance equivalent flux estimation value [phi] r ^. Field flux estimator 902 inputs the f-axis current command value i _f1 ^*, calculates the magnetic field flux estimate .phi.f ^. Then, the adder 903 calculates the magnetic flux estimated value φ ^ by adding the reluctance equivalent magnetic flux estimated value φr ^ and the field magnetic flux estimated value φf ^.

FIG. 10 is a control block diagram of the reluctance torque equivalent magnetic flux estimator 901. The relaxation torque equivalent magnetic flux estimator 901 includes a phase lead compensator 1001 and a multiplier 1002.

The phase lead compensator 1001 has a transmission characteristic (d-axis current transmission characteristic (see control block 601)) in which the q-axis current response is phase-advance compensated with respect to the transmission characteristic of the first-order delay simulating the d-axis current response delay. τ _q s + 1) / (τ _d s + 1). The phase lead compensator 1001 outputs a value obtained by performing phase lead compensation using the transfer characteristic (τ _q s + 1) / (τ _d s + 1) on the d-axis current command value _id1 ^* to the multiplier 1002.

The multiplier 1002 calculates the output of the phase lead compensator 1001 multiplies the difference L _d -L _q and d-axis inductance L _d and q-axis inductance L _q, the reluctance torque equivalent flux estimation value [phi] r ^ To do. As the d-axis inductance L _d and the q-axis inductance L _q , the values at arbitrary operating points (representative operating points) of the motor 4 may be used, or may be obtained by referring to the map data stored in advance. The reluctance torque generated in the rotor by the dq axis currents i _d and i _q is expressed by the following equation (17). Therefore, it is possible to define the following formula (17) in the section (L _{_d} -L _q) i _d and reluctance torque equivalent flux. However, p _n is the logarithm of the motor 4.

The vibration suppression control calculation unit 2b uses the reluctance torque equivalent magnetic flux estimated value φr ^ in which the phase advance compensation of the q-axis current response is performed by the phase advance compensator 1001, so that the q-axis current in consideration of the q-axis current response delay is taken into consideration. The command value (second q-axis current command value) i _q2 ^* can be calculated.

FIG. 11 is a control block diagram of the field magnetic flux estimator 902. The field magnetic flux estimator 902 includes a phase lead compensator 1101 and a multiplier 1102.

The phase lead compensator 1101 advances and compensates for the q-axis current response with respect to the transmission characteristic of the first-order delay simulating the f-axis current response delay (see control block 701) (τ _q s + 1) / (τ _f ). s + 1). Phase lead compensator 1101 outputs a value obtained by performing phase-lead compensation using the f-axis current command value i _f1 ^* the transfer characteristic _{(τ q s + 1) /} (τ f s + 1) to the multiplier 1102.

The multiplier 1102 calculates the field magnetic flux estimated value φf ^ by multiplying the output of the phase lead compensator 1101 by the mutual inductance M _f between the stator and the rotor. The mutual inductance M _f may be a value at an arbitrary operating point (representative operating point) of the motor 4, or may be obtained by referring to map data stored in advance.

The vibration suppression control calculation unit 2b uses the field magnetic flux estimated value φf ^ in which the phase advance compensator 1101 compensates for the advance of the q-axis current response, so that the q-axis current command value in consideration of the q-axis current response delay. i _q2 ^* can be calculated. Hereinafter, the description will be continued by returning to FIG.

The first torque command value calculator 803 multiplies the q-axis current command value i _q1 ^* , the estimated magnetic flux value φ ^, and the number of pole pairs p _n of the motor 4 to obtain the first torque command value (vibration suppression). Pre-control torque command value) Calculate T _m1 ^* . The calculated first torque command value T _m1 ^* is output to the vibration damping torque command value calculator 804.

The vibration damping torque command value calculator 804 performs feed forward control, which is a filter process for removing the natural vibration frequency component of the drive shaft torque transmission system of the vehicle, and motor rotation with respect to the first torque command value T _m1 ^* . The vibration damping torque command value T _mfin ^* is calculated by performing vibration damping control called feedback control based on the number ω _rm .

FIG. 12 is a control block diagram of the vibration damping torque command value calculator 804. As shown in this figure, the vibration damping torque command value calculator 804 includes a feed forward compensator 1201, a control block 1202, a feedback compensator 1203, an adder 1204, and a subtractor 1205. The control block 1202, the feedback compensator 1203, and the subtractor 1205 may be collectively referred to as the feedback controller 1207.

When the feed forward compensator 1201 receives the input of the first torque command value T _m1 ^* , the feed forward compensator 1201 performs a filter process consisting of Gr (s) / Gp (s) on the first torque command value T _m1 ^* . Calculate the second torque command value T _m2 ^* . In addition, Gr (s) and Gp (s) will be described with reference to FIG. 13 described later.

The control block 1202 performs a filter process consisting of the transmission characteristic Gp (s) on the vibration damping torque command value T _mfin ^* output from the adder 1204, and calculates the motor angular velocity estimated value ω _m ^. In the adder 1204, the second torque command value T _m2 ^* output from the feed forward compensator 1201 and the third torque command value T _m3 ^* output from the feedback compensator 1203 are added. The vibration damping torque command value T _mfin ^* is calculated.

In the feedback controller 1207, first, in the subtractor 1205, the actual motor angular velocity ω _m is subtracted from the motor angular velocity estimated value ω _m ^. Then, the feedback compensator 1203 performs a filter process consisting of H (s) / Gp (s) on the difference between the motor angular velocity estimated value ω _m ^ output from the subtractor 1205 and the actual motor angular velocity ω _m. , Calculate the third torque command value T _m3 ^* .

Note that Gr (s), Gp (s), and H (s) will be described with reference to FIG. 13 described later. Further, H (s) is configured such that the difference between the denominator order and the numerator order is larger than the difference between the denominator order and the numerator order of Gp (s). The control system can be stabilized by H (s) / Gp (s).

First, the derivation of the filter (transfer function) Ginv (s) that removes the natural vibration frequency component of the drive shaft torque transmission system of the vehicle, which is a filter used for feed word control, will be described. First, the equation of motion of the vehicle will be described with reference to FIG.

FIG. 13 is a diagram in which the driving force transmission system of the vehicle is converted into a control block 601. Each parameter in the figure is as shown below.
J _m : Motor inertia J _w : Drive wheel inertia (for one axis)
M: Vehicle mass K _d : Torsional rigidity of drive shaft (drive shaft) K _t : Factor related to friction between tire and road surface N _al : Overall gear ratio r: Tire load radius ω _m : Motor angular velocity ω _w : Drive wheel angular velocity T _m : Motor torque T _d : Drive shaft torque F: Driving force (for 2 shafts)
V: Body speed

The following equations of motion (18) to (22) can be derived from FIG.

The Laplace transform of the above equations (20) to (22) to obtain the transmission characteristics from the motor torque T _m to the motor angular velocity ω _m can be expressed by the following equations (23) and (24).

However, a ₃ , a ₂ , a ₁ , a ₀ , b ₃ , b ₂ , b ₁ , b ₀ in the equations (23) and (24) are represented by the following equations (25), respectively.

By rearranging the equation (24), Gp (s) can be expressed as the following equation (26). However, ζ _p and ω _p in Eq. (26) are the damping coefficient and the natural vibration frequency of the drive shaft torsional vibration system, respectively.

Then, assuming that the ideal normative model Gr (s) indicating the response target of the motor rotation speed to the torque input to the vehicle is the following equation (27), the inverse characteristic of the linearly approximated vehicle transfer characteristic Gp (s) and The transfer function Ginv (s) having the characteristics of Gr (s) / Gp (s), which is composed of the normative model Gr (s), can be expressed by the following equation (28). However, ζ _m and ω _m in the equations (27) and (28) are the damping coefficient and the natural vibration frequency of the drive shaft torsional vibration system, respectively.

In addition, as a method of removing the natural vibration frequency component of the drive shaft torque transmission system of the vehicle in consideration of the influence of the backlash of the gear in the scene where the vehicle accelerates from the coast or the principle, as in the second embodiment. It is also possible to use a vehicle model composed of a dead zone model simulating gear backlash.

Next, the transfer function H (s) will be described. H (s) is a feedback element that reduces only vibration when a bandpass filter is used. At this time, if the characteristics of the filter are set as shown in FIG. 14, the greatest effect can be obtained. That is, the transfer function H (s) has substantially the same damping characteristics on both the low-pass side with a low frequency and the high-pass side with a high frequency, and the torsional resonance frequency of the drive system is on the logarithmic axis (log scale). It is set to be near the center of the pass band. Then, for example, when configuring H (s) is a first-order high-pass filter and a first-order low-pass filter, the frequency f _p and torsional resonance frequency of the drive system, the k as an arbitrary value by the following equation (29) Constitute. The parameters in the equation (29) are shown as in the equation (30).

The second q-axis current command value calculator 805 shown in FIG. 8 has a vibration damping torque command value T _mfin ^* output from the vibration damping torque command value calculator 804 and a magnetic flux output from the magnetic flux estimator 802. Using the estimated value φ ^ as an input, the q-axis current command value (second q-axis current command value) i _q2 ^* is calculated using the following equation (31). The calculated q-axis current command value i _q2 ^* is input to the q-axis current control unit 406 of the current control calculation processing unit 2a shown in FIG. As described above, the magnetic flux estimated value φ ^ is the d-axis current command value i _d1 ^* and the f-axis current command value i _f1 set according to the motor torque command value T _m ^* set based on the vehicle information. Calculated based on ^* and. That is, the q-axis current command value i _q2 ^* of the present embodiment is set according to the motor torque command value T _m ^* in consideration of the d-axis current command value i _d1 ^* and the f-axis current command value i _f1 ^*. It is calculated by correcting the q-axis current command value i _q1 ^* . As a result, the vibration damping control calculation unit 2b considers the influence of the reluctance torque generated by the d-axis current command value i _d1 ^* and the field magnetic flux generated by the f-axis current command value i _f1 ^* , and drives the drive shaft. It is possible to suppress the occurrence of torsional vibration of the torque transmission system.

In the following, with reference to FIG. 15, the action and effect of the electric vehicle control method (vibration control control process) of the first embodiment described above will be described.

FIG. 15 is a time chart showing the control results of the present embodiment and the comparative example. In the comparative example, it is assumed that the vibration damping control calculation in step S203 is not performed. The horizontal axis represents time, the vertical axis represents the motor torque command value [Nm], the vehicle front-rear acceleration [m / s2], and the f-axis voltage [V] in order from top to bottom on the left side, and from the top on the right side. In the order of the bottom, the q-axis current command value [A], the d-axis current command value [A], and the f-axis current command value [A] are shown. The solid line in the figure shows the present embodiment, and the dotted line shows a comparative example. In the control in this time chart, the f-axis current norm response time constant τ _f is set to a value at which f-axis voltage saturation does not occur.

As shown in the upper left part of the figure, the motor torque command value is changed (started up) in steps at the timing of time t1 from the state where the vehicle is stopped to accelerate. In such a case, when the f-current command value is increased stepwise as shown in the lower right part of the figure in order to change the rotor magnetic flux at time t1, the f-axis voltage is increased as shown in the lower left part of the figure. After rising, converges to a predetermined value. At this timing, as shown in the middle right of the figure, the d current command value for controlling the magnetic field component falls.

In the comparative example shown by the broken line, the f-axis current is not taken into consideration when calculating the q-axis current command value, and the influence of the field magnetic flux generated by the f-axis current and the reluctance torque generated by the d-axis current is further affected. Not considered. Therefore, as shown in the upper right part of the figure, the q-axis current command value for controlling the torque component increases in a stepwise manner, and as shown in the middle left part of the figure, the vehicle front-rear acceleration vibrates.

On the other hand, in the present embodiment shown by the solid line, the q-axis current i _q2 ^* is calculated in consideration of the f-axis current i _f1 ^* and the d-axis current i _d1 ^* as shown in FIG. Therefore, the q-axis current is controlled so as to decrease after the rise at time t1 and then converge to a predetermined value. As a result, the motor torque that suppresses the drive shaft torsional vibration is realized by the q-axis current command value i _q2 ^* calculated in consideration of the d-axis current _id and the f-axis current _if, which is shown in the middle left of the figure. As shown, the vibration of the vehicle front-rear acceleration is suppressed.

According to the electric vehicle control method of the first embodiment, the following effects can be obtained.

The method for controlling an electric vehicle according to the first embodiment is in an electric vehicle using a winding field type synchronous motor 4 as a drive source, which includes a rotor having a rotor winding and a stator having a stator winding. This is a control method for an electric vehicle that controls a stator current flowing through a stator winding and a rotor current flowing through a rotor winding.

In the control method, the motor torque command value T _m ^* is set based on the operation information, and the d-axis current command for the stator current is based on the motor torque command value T _m ^* and the motor angular speed ω _m which is the vehicle information. The values i _d1 ^*, the first q-axis current command value i _q1 ^*, and the f-axis current command value i _f1 ^* for the rotor current are calculated, and the d-axis current command value i _d1 ^* and the f-axis current command value i are calculated. _The magnetic flux estimated value φ ^, which is an estimated value of the magnetic flux generated in the rotor, is calculated based on _f1 ^*, and the first torque command is calculated based on the first q-axis current command value i _q1 ^* and the magnetic flux estimated value φ ^. calculates the value T _m1 ^*, with respect to the first torque command value T _m1 ^*, based on the vehicle information omega _m, by performing the suppressing operation of the vibration of the drive shaft torque transmitting system of the electric vehicle, damping The torque command value T _mfin ^* is calculated, and the second q-axis current command value i _q2 ^* is calculated based on the magnetic flux estimated value φ ^ and the vibration damping torque command value T _mfin ^* . Then, the stator current and the rotor current are controlled based on the second q-axis current command value i _q2 ^* , the d-axis current command value i _d1 ^*, and the f-axis current command value i _f1 ^* .

As a result, the second q-axis current command value i _q2 ^* can be calculated in consideration of the d-axis current command value i _d1 ^* and the f-axis current command value i _f1 ^*. Therefore, the d-axis current command value i _d1 and the reluctance torque generated by the ^*, taking into account the influence of the field magnetic flux generated by the f-axis current command value i _f1 ^*, the electric vehicle as a driving source for winding field synchronous motor 4 driving shaft torque transmission Vibration suppression control that suppresses the torsional vibration of the system can be applied.

Further, according to the control method of the electric vehicle of the first embodiment, in the calculation for suppressing the vibration of the drive shaft torque transmission system performed by the vibration damping torque command value calculator 804, the feed forward compensator 1201 is used for the first calculation. The second torque command value T _m2 ^* is calculated by performing feed forward control with respect to the torque command value T _m1 ^* of.

Further, in the feedback controller 1207, feedback control based on the actual vehicle information ω _m is performed with respect to the motor angular velocity estimated value ω _m ^ which is the vehicle information estimated according to the second torque command value T _m2 ^*. To calculate the third torque command value T _m3 ^* .

Then, in the adder 1204, the vibration damping torque command value T _mfin ^* is calculated by adding the second torque command value T _m2 ^* and the third torque command value T _m3 ^* .

In this way, in the calculation for suppressing the vibration of the drive shaft torque transmission system performed in the vibration damping torque command value calculator 804, the second torque command value T _m2 ^* calculated by the feed forward control and the calculation by the feedback control are performed. By adding the third torque command value T _m3 ^* to be performed, the vibration damping torque command value T _mfin ^* is calculated, so that the vibration that may occur in the vehicle can be suppressed accurately.

Further, according to the electric vehicle control method of the first embodiment, in the feed forward compensator 1201, the transmission characteristic Gp (s) of the linear approximation model in the electric vehicle and the transmission characteristic Gp (s) of the linear approximation model with respect to the first torque command value T _m1 ^* The second torque command value T _m2 ^* is calculated by performing Gr (s) / Gp (s) filtering based on the transmission characteristic Gr (s) of the normative model.

Then, in the feedback controller 1207, the control block 1202 _{filters the} vibration damping torque command value T _mfin ^* including the second torque command value T _m2 ^* based on the transmission characteristic Gp (s) of the linear approximation model. By performing the processing, the standard motor angular velocity ω _m ^ is obtained. In the feedback compensator 1203, the linear approximation model Gp (s) and the bandpass filter H (for the difference between the reference motor angular velocity ω _m ^ calculated by the subtractor 1205 and the measured motor angular velocity ω _m By performing the H (s) / Gp (s) filtering process based on s), the third torque command value T _m3 ^* is calculated.

In the system configured in this way, the second torque command value T _m2 ^* calculated by feedforward control and the third torque command value T _m3 ^* calculated by feedback control are added to each other. The vibration damping torque command value T _mfin ^* is calculated. Therefore, even when a disturbance or a model error occurs, the feedforward control and the feedback control composed of a plurality of transfer functions are performed, so that the vibration of the drive shaft torque transmission system of the vehicle can be suppressed.

Further, according to the control method of the electric vehicle of the first embodiment, the second q-axis current command value i _q2 ^* is calculated by dividing the vibration damping torque command value T _mfin ^* by the magnetic flux estimated value φ ^. To. As a result, it is possible to calculate the q-axis current command value i _q2 ^* that realizes the vibration control torque command value T _mfin ^* with vibration control control.

According to the control method of an electric vehicle of the first embodiment, to calculate the estimated value is the field flux estimate of the magnetic field flux of the rotor .phi.f ^ based on the f-axis current command value i _f1 ^*, d based on the axis current value i _d1 ^*, and calculates the equivalent flux estimation value of the reluctance torque generated in the rotor [phi] r ^, the magnetic flux estimation value phi ^ is the field flux estimate .phi.f ^ equivalent flux estimation value [phi] r ^ and Is calculated by adding. Thus, taking into account the influence of the d-axis current i _d and f-axis current i _f, damping control calculates the q-axis current command value i _q2 ^* for realizing the damping torque command value T _MFIN ^* subjected be able to.

Further, according to the control method of the electric vehicle of the first embodiment, the field magnetic flux estimated value φ ^ simulates the response delay of the f-axis current _if constituting the rotor current with _{respect to} the f-axis current command value i _f1 ^*. It is calculated using a transmission characteristic (phase advance compensator 1101) configured to compensate the q-axis current response with respect to the f-axis current transfer characteristic. Thus, it is possible to calculate the d-axis current i q-axis current command value q-axis current response delay is considered for _d i _q2 ^*.

Further, according to the control method of the electric vehicle of the first embodiment, the f-axis current transfer characteristic is a transfer function of the first-order lag. Thereby, the f-axis current response when the f-axis voltage saturation does not occur can be appropriately simulated.

Further, according to the control method of the electric vehicle of the first embodiment, the equivalent magnetic flux estimated value φr ^ simulates the response delay of the d-axis current id constituting the stator current to the d-axis current command value i _d1 ^* . It is calculated using a transmission characteristic (phase advance compensator 1001) configured to compensate the q-axis current response with respect to the shaft current transmission characteristic. As a result, the q-axis current command value i _q2 ^* can be calculated in consideration of the q-axis current response delay with respect to the f-axis current _if .

<Second Embodiment>
Hereinafter, the control method of the electric vehicle of the second embodiment will be described. In the first embodiment, when the non-interference control by the non-interference control unit 405 functions ideally, the characteristics from the voltage to the current of the d-axis, q-axis, and f-axis are described in the above equations (4) and (4), respectively. It was explained that the primary delay is as shown in 5) and (6). However, when the f-axis voltage is saturated, the f-axis current response does not match the normative response of the first-order lag. The control method of the electric vehicle of the present embodiment is a control method applied on the premise that the f-axis current _if is controlled in consideration of the f-axis voltage saturation, and is particularly provided by the vibration damping control calculation unit 2b. The configuration related to feedforward and the configuration related to feedback of the magnetic flux estimator 802 are different from those of the first embodiment.

Prior to the description of the flux estimator 802 of the present embodiment, in consideration of the f-axis voltage saturation describes a method of controlling the f-axis current i _f. Since the control on the d-axis and the q-axis is the same as that on the f-axis, the description thereof will be omitted, and only the control on the f-axis will be described below.

FIG. 16 is a diagram showing a configuration example of the motor control system 200 of the second embodiment. In the motor control system 200 of the present embodiment, the power supply voltage V _dc of the battery 1 and the non-interference voltage v _{f_dcpl} which is the output of the non-interference control unit 405 are input to the f-axis current control unit 408. 1 Different from the embodiment.

The details of the f-axis current control unit 408 will be described with reference to FIG. FIG. 17 is a control block diagram of the f-axis current control unit 408.

In the f-axis current control unit 408, the first _f is such that the f-axis current if input from the A / D conversion unit 411 follows the f-axis current command value if _f1 ^* with a desired response without steady deviation. Calculate the shaft voltage command value v _{f_dsh} . Further, the f-axis current control unit 408 calculates the f-axis current normative response _{if_ref} and the differential values s · _{if_ref} of the f-axis current normative response to be used in the subsequent processing. The f-axis current control unit 408 is composed of an f-axis F / F (feed forward) compensator 1701, an f-axis F / B compensator 1702, an f-axis robust compensator 1703, and an f-axis limit processing unit 1704. , The details of each will be described below.

The f-axis F / F compensator 1701 takes the f-axis current command value i _f1 ^* as an input, and in addition to the f-axis F / F compensation voltage v _{f_ff} , the f-axis current normative response _{if_ref} and its differential value. Calculate the differential value s · _{if_ref} of the f-axis current normative response. The f-axis F / F compensator 1701 _outputs the f-axis current normative response _{if_ref} and the differential value s · _{if_ref} of the f-axis current normative response, which is the differential value _thereof, to the non-interference control unit 405, and f. The shaft current norm response _{if_ref} is _output to the f-axis F / B compensator 1702. Details of the f-axis F / F compensator 1701 will be described later with reference to FIG. Although not shown, the power supply voltage V _dc output from the battery 1 and the non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 are input to the f-axis F / F compensator 1701. ..

The f-axis F / B compensator 1702 is a compensator that performs general feedback compensation. f-axis F / B compensator 1702, with respect to the f-axis current nominal response i _{F_REF} calculated in the f-axis F / F compensator 1701, is negatively fed back to the f-axis current i _f that is measured by the current sensor 7 F By performing the / B processing, the f-axis F / B compensation voltage v _{f_fb} is calculated so that the f-axis current _if follows the f-axis current normative response _{if_ref} . The f-axis F / B compensator 1702 outputs the f-axis F / B compensation voltage v _{f_fb} to the adder 1705. Details of the f-axis F / B compensator 1702 will be described later with reference to FIG. The f-axis F / B compensator 1702 is an example of a block that executes the F / B compensation step.

The f-axis robust compensator 1703 includes a first f-axis voltage command value v _{f_dsh} calculated by the f-axis limit processing unit 1704 described later and finally output from the f-axis current control unit 408, and an f-axis current _if . Based on, the f-axis robust compensation voltage v _{f_rbst} for ensuring the robustness of the system is calculated. The f-axis robust compensator 1703 outputs the f-axis robust compensation voltage v _{f_rbst} to the adder 1706. Details of the f-axis robust compensator 1703 will be described later with reference to FIG. 24.

Two

adders

1705 and 1706 are provided in front of the f-axis limit processing unit 1704. The f-axis F / B compensation voltage v _{f_fb} is added by the adder 1705 to the f-axis F / F compensation voltage v _{f_ff} calculated by the f-axis F / F compensator 1701, and further, the f-axis by the adder 1706. The robust compensation voltage v _{f_rbst} is added. Then, the final addition value is input to the f-axis limit processing unit 1704. Therefore, the f-axis limiting processor 1704, with respect to the f-axis F / F compensation voltage v _{F_ff} a F / F command value, F / B is a compensation value f-axis F / B compensation voltage v _{F_fb} and, The sum of the f-axis robust compensation voltage v _{f_rbst,} which is the f-axis robust compensation value, is input.

Then, the f-axis limit processing unit 1704 limits the input voltage command value and calculates the first f-axis voltage command value v _{f_dsh} . The f-axis limit processing unit 1704 outputs the f-axis voltage command value v _{f_dsh} to the voltage command value calculation unit 409 and the f-axis robust compensator 1703. The f-axis limit processing unit 1704 performs the same processing as the f-axis limit processing unit 303 described later with reference to FIGS. 22 and 23.

Next, the detailed configuration of the f-axis F / F compensator 1701 will be described with reference to FIG. FIG. 18 is a detailed block diagram of the f-axis F / F compensator 1701. The f-axis F / F compensator 1701 has an f-axis current model 1801, an f-axis current pseudo F / B model 1802, and an f-axis limit processing unit 1803.

The f-axis current model 1801 is a filter that models the normative response characteristics from the f-axis voltage to the f-axis current. The f-axis current model 1801 performs filtering processing on the f-axis F / F compensation voltage v _{f_ff} output from the f-axis limit processing unit 1803 described later using a normative response model from voltage to current on the f-axis. The f-axis current normative response _{if_ref} , which is the normative response, is calculated and output to the non-interference control unit 405 and the f-axis F / B compensator 202. Further, the f-axis current model 1801 _outputs the differential value s · _{if_ref} of the f-axis current normative response, which is the differential value of the f-axis current normative response _{if_ref,} to the non-interference control unit 405 for use in the subsequent processing. To do. Details of the f-axis current model 1801 will be described later with reference to FIG.

In the f-axis current pseudo F / B model 1802, the f-axis current normative response i output from the f-axis current model 1801 to the f-axis current command value i _f1 ^* calculated by the vibration damping control calculation unit 2b. _{f_ref} is negatively _fed back. The f-axis current pseudo F / B model 1802 has a pseudo FB voltage command value v _{f_pse_fb in} order to make the f-axis current normative response _{if_ref} follow the f-axis current command value if _f1 ^* with a desired response without steady deviation. Is calculated and output to the f-axis limit processing unit 1803. Details of the f-axis current pseudo F / B model 1802 will be described later with reference to FIG.

The f-axis limit processing unit 1803 limits the pseudo FB voltage command value v _{f_fb_psu} output from the f-axis current pseudo F / B model 1802, calculates the f-axis F / F compensation voltage v _{f_ff,} and _addser . Output to 205 and f-axis current model 1801. Details of the f-axis limit processing unit 1803 will be described later with reference to FIGS. 21 and 22.

Although not shown, the power supply voltage V _dc output from the battery 1 and the non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 are input to the f-axis limit processing unit 1803. As shown in FIG. 17, the f-axis F / F compensation voltage v _{f_ff} output from the f-axis F / F compensator 1701 _{passes through} the adder 1705, the adder 1706, and the f-axis limit processing unit 1704. The first f-axis voltage command value v _{f_dsh} is calculated.

In this way, in the f-axis F / F compensator 1701, the f-axis current model is not the F / B system in which the measured f-axis current _if is negatively fed back to the f-axis current pseudo F / B model 1802. A pseudo F / B system is constructed in which the f-axis current normative response _{if_ref} calculated in 1801 is negatively _fed back. By realizing a pseudo F / B system in this way, it is possible to avoid F / B control having poor responsiveness, so that responsiveness can be improved.

Further, as shown in FIG. 16, since the f-axis voltage v _f is generated by the battery 1, the upper limit of the f-axis voltage v _f is limited by the power supply voltage V _dc of the battery 1 and saturated. Therefore, in the f-axis F / F compensator 1701 shown in FIG. 18, an f-axis limit processing unit 1803 that models saturation at the power supply voltage V _dc is provided to limit the first f-axis voltage command value v _{f_dsh} . Then, the f-axis F / F compensation voltage v _{f_ff} is calculated. By _feeding back the f-axis F / F compensation voltage v _{f_ff in} consideration of voltage saturation to the f-axis current pseudo F / B model 1802, the accuracy of rotation control can be improved.

Next, the detailed configuration of the f-axis current model 1801 will be described with reference to FIG. FIG. 197 is a detailed block diagram of the f-axis current model 1801. The f-axis current model 1801 has a multiplier 1901, a subtractor 1902, a divider 1903, and an integrator 1904.

The multiplier 1901 is one of the final outputs of the f-axis current model 1801, and the rotor winding resistance R _f is multiplied by the f-axis current normative response _{if_ref} output from the integrator 1904 described later. , The multiplication result is output to the subtractor 1902. The result of this multiplication corresponds to the voltage value of the normative response.

The subtractor 1902 subtracts the voltage value of the normative response output from the multiplier 1901 from the f-axis F / F compensation voltage v _{f_ff} output from the f-axis limit processing unit 1803, and outputs the subtracted value to the divider 1903. To do.

The divider 1903 divides the difference calculated by the subtractor 1902 with the f-axis dynamic inductance L _f ^' , and outputs the division result to the non-interference control unit 405 and the integrator 1904. In this way, the differential value s · _{if_ref of} the f-axis current normative response is calculated.

Integrator 1904, the differential value s · i _{F_REF} of f-axis current norms response outputted from the divider 1903 integration processing to calculate the f-axis current nominal response i _{F_REF,} non-interference of the f-axis current nominal response i _{F_REF} It outputs to the control unit 405, the f-axis F / B compensator 1702, and the multiplier 1901.

Thus, in the f-axis current model 301, the f-axis current normative response _{if_ref,} which is one of the final outputs, is multiplied by the rotor winding resistance R _f by the multiplier 1901, and the f-axis is the input. Negative feedback is _given to the F / F compensation voltage v _{f_ff} . By this negative feedback divider 1903 the result value of is divided by the f-axis dynamic inductance L _f ^', f axis current nominal response i _{F_REF} based on the f-axis F / F compensation voltage v _{F_ff,} and its differential value s・_{If_ref} can be obtained.

Next, the detailed configuration of the f-axis current pseudo F / B model 1802 will be described with reference to FIG. FIG. 20 is a detailed block diagram of the f-axis current pseudo F / B model 1802. The f-axis current pseudo-F / B model 1802 has a filter 2001, a filter 2002, and a subtractor 2003.

Filter 2001, the gain G _af multiplied to the vibration control operation f-axis current command value output from unit 2b i _f1 ^*, and outputs a value after the filtering process to the subtractor 2003.

The filter 2002 multiplies the f-axis current norm response _{if_ref} output from the f-axis current model 301 by the gain G _bf , and outputs the filtered value to the subtractor 2003.

Then, the subtractor 2003 calculates the pseudo F / B voltage command value v _{f_fb_psu} by subtracting the output value of the filter 2002 from the output value of the filter 2001, and _sends the pseudo FB voltage command value v _{f_fb_psu} to the f-axis limit processing unit 1803. Is output. That is, a pseudo F / B control is configured by negatively feeding back the f-axis current norm response _{if_ref,} which is not a measured value.

However, the gain G _af and the gain G _bf can be expressed by the following equation (32). However, τ _f is an f-axis current control norm response time constant (f-axis current norm response time constant).

By such a configuration, in the f-axis current pseudo F / B model 1802, with respect to the f-axis current command value i _f1 ^*, rather than the actual measured f-axis current i _f f axis current nominal response Pseudo-F / B control can be realized by using _{if_ref} as an F / B component.

Next, the detailed configuration of the f-axis limit processing unit 1803 will be described with reference to FIG. FIG. 21 is a detailed block diagram of the f-axis limit processing unit 1803. The f-axis limit processing unit 1803 includes a comparator 2101, a reversing device 2102, a comparator 2103, and a

subtractor

2104, 2105.

In the subtractor 2104 provided in front of the comparator 2101, a subtraction value obtained by subtracting the f-axis non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 from the power supply voltage V _dc of the battery 1 is obtained. Then, the comparator 2101 compares the pseudo FB voltage command value v _{f_pse_fb} , which is the output value from the f-axis current pseudo F / B model 1802, with the subtracted value in the subtractor 2104, and transfers a smaller value to the comparator 2103. Is output.

The inverting device 2102 inverts the sign of the power supply voltage V _dc .

A subtractor 2105 is provided in front of the comparator _{2103. In} the subtractor ₂₁₀₅ , a subtraction value obtained by subtracting the f-axis non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 from the output of the reversing device 2102. Is required. Then, the comparator 2103 compares the output value of the comparator 2101 with the subtracted value of the subtractor 2105, and outputs a larger value to the f-axis current model 1801 and the adder 1705.

With such a configuration, the f-axis limit processing unit 1803 _adds the f-axis non-interference voltage v _{f_dcpl} to the pseudo FB voltage command value v _{f_pse_fb} , which is the output value of the f-axis current pseudo F / B model 1802. Limit processing based on the power supply voltage V _dc offset to the minus by the f-axis non-interference voltage v _{f_dcpl} , specifically, the upper limit is "V _dc -v _{f_dcpl} " and the lower limit is " -V _dc -v _{f_dcpl} "is performed.

Further, the f-axis limit processing unit 1803 may be configured as shown in FIG. FIG. 22 is another example of a detailed block diagram of the f-axis limit processing unit 1803. In this example, the f-axis limit processing unit 1803 has a comparator 2201, a reversing device 2202, a comparator 2203, a subtractor 2204, and an adder 2205.

An adder 2205 is provided in front of the _{comparer 2201. In} the adder 2205, the f-axis non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 and the f-axis current pseudo F / B model 1802 output. The pseudo FB voltage command value _{vf_pse_fb} to be added is added. Then, the comparator 2201 compares the power supply voltage V _dc of the battery 1 with the addition result in the adder 2205, and outputs a smaller value to the comparator 2203.

The inverting device 2202 inverts the sign of the power supply voltage V _dc .

The comparator 2203 compares the output from the comparator 2201 with the output from the inverting device 2202 and outputs a large value to the subtractor 2204.

The subtractor 2204 calculates the f-axis F / F compensation voltage v _{f_ff} by subtracting the f-axis non-interference voltage v _{f_dcpl} output from the non-interference control unit 405 from the output value of the comparator 2203. The subtractor 2204 outputs the f-axis F / F compensation voltage v _{f_ff} to the f-axis current model 1801 and the adder 1706 constituting the f-axis current control unit 408.

Even with such a configuration, in the f-axis limit processing unit 1803, the f-axis non-interference voltage v _{f_dcpl} is added to the pseudo FB voltage command value v _{f_pse_fb} which is the output value of the f-axis current pseudo F / B model 1802. Limit processing based on the power supply voltage V _dc offset negatively by the f-axis non-interference voltage v _{f_dcpl in} order to obtain a margin, specifically, the upper limit is "V _dc -v _{f_dcpl} " and the lower limit is "-". Restriction processing _{such as} "V _dc -v _{f_dcpl} " is performed.

Next, the details of the f-axis F / B compensator 1702 will be described. FIG. 23 is a detailed block diagram of the f-axis F / B compensator 1702. The f-axis F / B compensator 1702 has a block 2301, a multiplier 2302, and a subtractor 2303.

The block 2301 is a delay filter, and performs delay processing for the dead time L of the control system. The block 2301 _delays the f-axis current normative response _{if_ref} with respect to the input of the f-axis current normative response _{if_ref} output from the f-axis F / F compensator 1701, and the f-axis current normative response _{if_ref} and the f-axis current. calculating a dead time after treatment f axis current nominal response i _{F_REF} ^'in order to match the phase of the i _f, and outputs to the subtractor 2303 provided upstream of the multiplier 2302. Here, it is assumed that the waste time L of the control system corresponds to the control calculation delay. Block 2301 is an example of a block that executes a delay step.

Subtractor 2303, a dead time after treatment f axis current nominal response i _{F_REF} ^'output from block 2301 subtracts the f axis current i _f which is output from the A / D conversion unit 411 calculates the subtraction result.

The multiplier 2302 is input with the subtraction result in the subtracter 2303 calculates the f-axis F / B compensation voltage v _{F_fb} by multiplying the f-axis F / B gain K _f, the f-axis F / B compensation voltage v _{F_fb} Is output to the adder 1705. The value of the f-axis F / B gain K _f is determined by adjusting it experimentally so that the stability such as the gain margin and the phase margin satisfies a predetermined standard.

With this configuration, the f-axis F / B compensator 1702 calculates the f-axis F / B compensation voltage v _{f_fb} based on the f-axis current _if .

FIG. 24 is a detailed block diagram of the f-axis robust compensator 1703. The f-axis robust compensator 1703 is composed of a block 2401, a block 2402, a block 2403, and a subtractor 2404.

Block 2401 calculates the first f-axis voltage estimated value v _{F_est1} filtering process on the input of the f-axis current i _f which is output from the A / D conversion unit 411, the f-axis voltage estimated value v _{F_est1} Output to the subtractor 2404. The block 2401 is a delay filter having the characteristics of (L _f ^'・ s + R _f ) / (τ _{h_f}・ s + 1) including the low-pass filter 1 / (τ _{h_f}・ s + 1) of the block 2403 described later.

Block 2402 is the same delay filter as block 1801. The block 2402 _delays the first f-axis voltage command value v _{f_dsh} output from the f-axis limit processing unit 204 by the dead time L of the control system, and delays the second f-axis voltage estimated value v _{f_est2.} Is calculated. Then, the block 2402 outputs the second f-axis voltage estimated value v _{f_est2} to the block 2403.

Block 2403 is a low-pass filter having a characteristic of 1 / (τ _{h_f} · s + 1). The block 2403 performs a low-pass filter process on the second f-axis voltage estimated value v _{f_est2} output from the block 2402, and calculates the third f-axis voltage estimated value v _{f_est3} . Then, the block 2403 outputs the third f-axis voltage estimated value v _{f_est3} to the subtractor 2404.

The subtractor 2404 calculates the f-axis robust compensation voltage v _{f_rbst} into the adder 1706 by subtracting the first f-axis voltage estimate v _{f_est1} from the third f-axis voltage estimate v _{f_est3} .

In this way, the first f-axis voltage command value v _{f_dsh} is processed to process the delay filter block 2401 and the low-pass filter block 2403, and the first f-axis based on the measured value is processed. By subtracting the voltage estimate v _{f_est1} , the f-axis low- _pass compensation voltage v _{f_rbst} for further improving stability is calculated.

FIG. 25 is a flowchart showing the control process of the motor 4 described with reference to FIGS. 16 to 24 described above. These controls are performed by the controller 2 executing a predetermined program.

In step S1, the A / D conversion unit 411 acquires the current values (u-phase current i _us , v-phase current i _vs , and f-axis current _if ) and the electric angle θ _{re of the} motor 4.

In step S2, the motor rotation speed ω _rm , which is the mechanical angular velocity, and the electric angular velocity ω _re are calculated based on the electric angle θ _re acquired in step S1.

In step S3, the prefetch compensator 403 based on the electric angle theta _re calculated at step S2, and calculates the post-prefetch compensation electrical angle theta _re ^'.

In step S4, the coordinate conversion unit 404 calculates the d-axis current i _d and the q-axis current i _q based on the u-phase current i _u and v-phase current i _v calculated in step S1.

In step S5, the d-axis current command value _id ^* , the q-axis current command value i _q ^* , and the f-axis current command are based on the motor rotation speed ω _rm , the torque command value T ^* , and the power supply voltage V _dc. The value _if ^* is calculated.

In step S6, the q-axis current control unit 406, the d-axis current control unit 407, and the f-axis current control unit 408 _perform the first d-axis voltage command value v _{d_dsh} , the d-axis current normative response _{id_ref} , and the d-axis current. Differential value of normative response s · i _{d_ref} , first q-axis voltage command value v _{q_dsh} , q-axis current normative response i _{q_ref} , first f-axis voltage command value v _{f_dsh} , f-axis current normative response i _{f_ref} , and The differential value s · _{if_ref} of the f-axis current normative response is calculated.

In step S7, the non-interference control section 405, and the electrical angular velocity omega _re calculated in step S2, the d-axis current nominal response i _{d_ref,} the differential value s · i _{d_ref} the d-axis current nominal response calculated in step S6, The non-interference voltages v _{d_dcpl} , v _{q_dcpl} , and v _{f_dcpl} are calculated according to the q-axis current normative response i _{q_ref} , the f-axis current normative response _{if_ref} , and the differential values s · _{if_ref} of the f-axis current normative response.

In step S8, the voltage command value calculation unit 409 uses the first d-axis voltage command value v _{d_dsh} , the first q-axis voltage command value v _{q_dsh} , and the first f-axis voltage command calculated in step S6. for each value v _{F_dsh,} step S7 incoherent voltage v _{D_dcpl} calculated, v _{Q_dcpl,} and, v _{F_dcpl} by adding the second d-axis voltage command value v _d ^*, a second q-axis The voltage command value v _q ^* and the second f-axis voltage command value v _f ^* are calculated.

In step S9, the coordinate conversion unit 410 sets the second d-axis voltage command value v _d ^* , the second q-axis voltage command value v _q ^* , and the second f-axis voltage command calculated in step S8. By performing the coordinate change processing on the value v _f ^* , the voltage command values v _u ^* , v _v ^* , and v _w ^* of each phase of uvw are calculated.

In this way, the controller 2 executes the processes of steps S1 to S9 to generate a command value for controlling the motor 4. Of the generated command values, the voltage command values v _u ^* , v _v ^* , and v _w ^* calculated in step S9 are fixed to the motor 4 via the stator PWM converter 401 and the inverter 3. It is applied to the winding on the child side. The second f-axis voltage command value v _f ^* calculated in step S8 is applied to the winding on the rotor side of the motor 4 via the rotor PWM conversion unit 402. In this way, the rotation control of the motor 4 is performed.

Or a motor control method for controlling the f-axis current i _f in consideration of the f-axis voltage saturation. When vibration damping control processing is applied to such motor control, the reluctance torque equivalent magnetic flux estimator 901 and the field reluctance estimator 902 constituting the magnetic flux estimator 802 included in the vibration damping control calculation unit 2b are f. It is necessary to perform processing in consideration of shaft voltage saturation.

FIG. 26 shows a control block diagram of the reluctance torque equivalent magnetic flux estimator 901 in the magnetic flux estimator 802 of the present embodiment. The relaxation torque equivalent magnetic flux estimator 901 of the present embodiment is composed of a multiplier 2601.

The multiplier 2601 has a difference L between the d-axis inductance L _d and the q-axis inductance L _q with respect to the d-axis current command value _id1 ^* output from the first current command value calculator 801 of the second embodiment. Multiply _{d −} L _q to calculate the reluctance torque equivalent magnetic flux estimated value φr ^. As the d-axis inductance L _d and the q-axis inductance L _q , the values at arbitrary operating points (representative operating points) of the motor 4 may be used, or may be obtained by referring to the map data stored in advance. When the current is controlled so that the d-axis current response time constant and the q-axis current response time constant match, the q-axis current response is phased with respect to the transmission characteristic of the first-order delay simulating the d-axis current response delay. When the advance compensation is performed, the transmission characteristic becomes 1. Therefore, the configuration of the reluctance torque equivalent magnetic flux estimator 901 shown in the present embodiment can be simplified as compared with the configuration of FIG. 10 shown in the first embodiment.

FIG. 27 is a control block diagram of the field magnetic flux estimator 902 of the second embodiment. The field magnetic flux estimator 902 of the present embodiment includes

control blocks

2701 and 2704, a multiplier 2702, a control block 2703, a limiter 2705, and an adder 2706.

The control block 2701 is an f-axis model that models the transmission characteristics from the f-axis voltage v _f to the f-axis current _if . The f-axis model has a characteristic of (τ _q s + 1) / (L _f s + R _f ). Control block 2701, enter the f axis current nominal response v _{Fc_lim} considering f-axis voltage saturation characteristic outputted from the limiter 2705, in consideration of the transfer characteristic from the f-axis voltage v _f to f axis current i _f f The shaft current reference response _{if_ref} is calculated and output to the multiplier 2702 and the control block 2704.

The multiplier 2702 calculates the field magnetic flux estimated value φf ^ by multiplying the f-axis current norm response _{if_ref} by the mutual inductance M _f between the stator and the rotor. The mutual inductance Mf may be a value at an arbitrary operating point (representative operating point) of the motor 4, or may be obtained by referring to map data stored in advance.

The control block 2703 is composed of a gain G _af . The gain G _af is represented by the above equation (32). Control block 2703 outputs obtained by multiplying the gain G _af the f-axis current command value i _f1 ^* input values to the adder 2706.

The control block 2704 is a filter composed of a gain G _bf and 1 / (τ _qs +1). The gain G _bf is shown in the above equation (32). The control block 2704 outputs the value obtained by filtering the f-axis current norm response _{if_ref} to the adder 2706.

The adder 2706 calculates the f-axis voltage command value v _fc by adding the output values of the control blocks 2703 and 2704. The calculated f-axis voltage command value v _fc is output to the f-axis limiter 2705.

As described above, the field magnetic flux estimator 902 of the present embodiment includes the control block 2703 and the control block 2704, and the gain G _af is multiplied by the f-axis current command value i _f1 ^* and the f-axis current normative response i. _The current F / B system (f-axis current F / B model) is constructed by multiplying _{f_ref} by the gain G _bf . Thereby, in the absence of f-axis voltage saturation, the f-axis current response can be matched with the transmission characteristic of the first-order lag (see equation (6)).

The f-axis limiter 2705 simulates the f-axis voltage saturation characteristic by limiting the f-axis current command value v _{fc according} to the power supply voltage V _dc . As a result, the field magnetic flux estimator 902 can calculate the f-axis current norm response _{if_ref} in consideration of the f-axis voltage saturation characteristic in the control block 2701 arranged at the _{subsequent stage} .

Further, in the vibration suppression control calculation unit 2b of the present embodiment, the q-axis current is provided by the phase lead compensation (τ _qs +1) of the q-axis current response on the control blocks 2701 and 2704 included in the magnetic flux estimator 802. The q-axis current command value i _q2 ^* can be calculated in consideration of the response delay. That is, the field magnetic flux estimated value φf ^ of the present embodiment is an f-axis model that models the characteristics from the f-axis voltage vf to the f-axis current if that constitutes the rotor current, and the f-axis current command value if _f1 ^*. A pseudo F / B system composed of an f-axis current F / B model in which and the output of the f-axis model are input, and an f-axis limiter 2705 that limits the output of the f-axis current F / B model. In, the q-axis current response is phase-advanced and compensated for the f-axis model and the f-axis current F / B model. As a result, the vibration damping control calculation unit 2b can appropriately simulate the f-axis current response when there is f-axis voltage saturation.

FIG. 28 is a control block diagram of the vibration damping control calculation unit 2b of the second embodiment. The vibration damping control calculation process is composed of a feedforward compensator 2801 and a feedback compensator 2802.

Specifically, the feed forward compensator 2801 subtracts the value obtained by integrating the F / B gain from the torque command value and the pseudo torsion angular velocity with the vehicle model 2804 composed of the vehicle parameter and the dead zone model simulating the gear back crash. It is composed of a drive shaft torsional velocity F / B model 2805, and calculates a second torque command value T _m2 ^* and a first motor angular velocity estimated value ω _m1 ^.

In the vehicle model 2804, a vehicle model composed of a dead zone model simulating a gear back crash and modeling a torque transmission system is used, and the first motor angular velocity estimated value ω _m1 ^ is set according to the motor torque T _m . The pseudo drive shaft torsional velocity ω _d ^ is obtained. The pseudo drive shaft torsional velocity ω _d ^ is obtained by simulating the torsional angular velocity generated in the drive shaft 8 to which the torque of the motor 4 is transmitted.

In the drive shaft torsion angle speed F / B model 2805, feedback control is performed in which the pseudo drive shaft torsion angle velocity ω _d ^ multiplied by the gain is subtracted from the first torque command value T _m1 ^* before the vibration suppression control. , The second torque command value T _m2 ^* is obtained. In the drive shaft torsional velocity F / B model 2805, the torsional angular velocity FB model is used, and this torsional angular velocity FB model is obtained by using a part of the vehicle model.

In the feedback compensator 2802, the first motor angular velocity estimated value ω _m1 ^ and the motor angular velocity ω _m are input. Then, the feedback compensator 2802 outputs a third torque command value T _m3 ^* from these inputs.

The feedback compensator 2802 uses the transmission characteristic Gp (s) of the vehicle model shown in the block 2806 with respect to the third torque command value T _m3 ^* output from itself in order to perform feedback control. The motor angular velocity estimated value (second motor angular velocity estimated value) ω _m2 ^ corresponding to the second torque command value T _m2 ^* is calculated. Then, in the adder 2807, the first motor angular velocity estimated value ω _m1 ^ and the second motor angular velocity estimated value ω _m2 ^ are added to calculate the third motor angular velocity estimated value ω _m3 ^.

In the subtractor 2808, the deviation between the estimated third motor angular velocity estimated value ω _m3 ^ and the detected motor angular velocity ω _m is obtained. The third torque command value T _m3 ^* is calculated by processing the filter H (s) / Gp (s) shown in the block 2809 with respect to this deviation. The filter H (s) / Gp (s) is composed of an inverse characteristic of the transmission characteristic Gp (s) of the vehicle model and a bandpass filter H (s).

Finally, in the adder 2803, the second torque command value T _m2 ^* output from the feedforward compensator 2801 and the third torque command value T _m3 ^* output from the feedback compensator 2802 are added. Then, the vibration damping torque command value T _mfin ^* is calculated.

In the following, with reference to FIG. 29, the action and effect of the control method (vibration control control process) of the electric vehicle of the second embodiment will be described.

FIG. 29 is a time chart showing the control results of the present embodiment and the comparative example. In the comparative example, it is assumed that the vibration damping control calculation in step S203 is not performed. The horizontal axis represents time, the vertical axis represents the motor torque command value [Nm], the vehicle front-rear acceleration [m / s2], and the f-axis voltage [V] in order from top to bottom on the left side, and from the top on the right side. In the order of the bottom, the q-axis current command value [A], the d-axis current command value [A], and the f-axis current command value [A] are shown. The solid line in the figure shows the present embodiment, and the dotted line shows a comparative example. In the control in this time chart, the f-axis current norm response time constant τ _f is set to a value at which f-axis voltage saturation occurs.

FIG. 29 shows a scene in which the motor torque command value is changed (started up) in steps at the timing of time t1 while the vehicle is decelerating due to the regenerative torque of the motor 4. .. In the control represented by this time chart (the present embodiment and the comparative example), the motor torque command value (vibration damping torque command value) that suppresses the drive shaft torsional vibration is calculated in consideration of the influence of the backlash of the gear. are doing.

As shown in the upper left part of the figure, from the state where the vehicle is decelerating, the motor torque command value is changed (started up) in steps at the timing of time t1 to accelerate. In such a case, at time t1, when the f current command value is increased stepwise as shown in the lower right part of the figure, the f-axis voltage rises as shown in the lower left part of the figure.

Here, the f-axis current normative response time constant τ _f is set to a value at which f-axis voltage saturation occurs. Therefore, the f-axis voltage is saturated as shown in the lower left part of the figure. Then, as shown in the middle right of the figure, the d current command value changes stepwise from a positive value to a negative value.

In the comparative example shown by the broken line, the f-axis current is not taken into consideration when calculating the q-axis current command value, and the influence of the field magnetic flux generated by the f-axis current and the reluctance torque generated by the d-axis current is further affected. Not considered. Therefore, as shown in the upper right part of the figure, the q-axis current command value also increases stepwise. Then, as shown in the middle left of the figure, the vehicle front-rear acceleration vibrates after increasing due to the influence of gear back crash.

On the other hand, in the present embodiment shown by the solid line, the q-axis current i _q2 ^* is calculated in consideration of the f-axis current i _f1 ^* and the d-axis current i _d1 ^* as shown in FIG. .. Therefore, the q-axis current is controlled so as to decrease after the rise at time t1 and then converge to a predetermined value.

As a result, the motor torque that suppresses the drive shaft torsional vibration is realized by the q-axis current command value i _q2 ^* calculated in consideration of the d-axis current _id and the f-axis current _if, which is shown in the middle left of the figure. As you can see, the vehicle front-rear acceleration vibration is suppressed.

According to the electric vehicle control method of the second embodiment, the following effects can be obtained.

According to the control method of the electric vehicle of the second embodiment, the feed forward compensator 2801 is configured by a dead zone model simulating a gear back crash with respect to the output second torque command value T _m2 ^* , and torque is transmitted. By using the vehicle model 2804 that models the system, the first motor angular velocity estimated value ω _m1 ^ and the pseudo drive shaft torsional velocity ω _d ^ can be obtained according to the motor torque T _m . Then, in the drive shaft torsional velocity F / B model 2805, feedback control is performed in which the pseudo drive shaft torsional velocity ω _d ^ multiplied by the gain is subtracted from the first torque command value T _m1 ^* before the vibration damping control. Therefore, the second torque command value T _m2 ^* is obtained.

Then, in the feedback compensator 2802, in the block 2806, the output third torque command value T _m3 ^{* is} filtered by Gp (s) based on the linear approximation model in the electric vehicle. Estimate the motor angular velocity estimation value ω _m2 ^ of 2. Then, in the adder 2807, the first motor angular velocity estimated value ω _m1 ^ calculated in the feed forward control and the second motor angular velocity estimated value ω _m2 ^ are added to obtain a third motor angular velocity estimated value. Calculate ω _m3 ^. The difference between the third motor angular velocity estimated value ω _m3 ^ and the measured motor angular velocity ω _m is filtered based on a linear approximation model and a bandpass filter to perform a third filter process. Calculate the torque command value T _m3 ^* .

Finally, the vibration damping torque command value T _mfin ^* is calculated by adding the second torque command value T _m2 ^* and the third torque command value T _m3 ^* .

In the system configured in this way, the second torque command value T _m2 ^* calculated by feedforward control using the dead zone model and the third torque command value T _m3 ^* calculated by feedback control are added. By doing so, the vibration damping torque command value T _mfin ^* is calculated. Therefore, in the case of re-acceleration during deceleration due to regenerative torque, vibration in the drive shaft torque transmission system of the vehicle can be suppressed even if gear backlash or the like occurs.

Further, according to the control method of the electric vehicle of the second embodiment, the vibration control calculation unit 2b provides phase advance compensation (τ _qs +1) for the q-axis current response to the control blocks 2701 and 2704 provided in the magnetic flux estimator 902. The q-axis current command value i _q2 ^* can be calculated in consideration of the q-axis current response delay. That is, the magnetic field flux estimate φf of the present embodiment ^ includes a f-axis model 2701 models the characteristics of to f axis current i _f that constitute the rotor current from the f-axis voltage vf, f-axis current command value i _A pseudo F composed of an f-axis current F / B model 2704 in which _f1 ^* and the output of the f-axis model are input, and an f-axis limiter 2705 that limits the output of the f-axis current F / B model. In the / B system, it is calculated by compensating the q-axis current response with respect to the f-axis model and the f-axis current F / B model. As a result, the vibration damping control calculation unit 2b can appropriately simulate the f-axis current response when the f-axis voltage is saturated.

Although the embodiments of the present invention have been described above, the above embodiments are only a part of the application examples of the present invention, and the technical scope of the present invention is limited to the specific configuration of the above embodiments. Absent.

Claims

In an electric vehicle driven by a winding field type synchronous motor including a rotor having a rotor winding and a stator having a stator winding, the stator current flowing through the stator winding and the rotation thereof. It is a control method of an electric vehicle that controls the rotor current flowing through the child winding.
Set the basic torque command value based on the operating condition,
Based on the basic torque command value and the vehicle information, the d-axis current command value and the first q-axis current command value for the stator current and the f-axis current command value for the rotor current are calculated.
A magnetic flux estimated value, which is an estimated value of the magnetic flux generated in the rotor, is calculated based on the d-axis current command value and the f-axis current command value.
The first torque command value is calculated based on the first q-axis current command value and the magnetic flux estimated value, and the electric vehicle of the electric vehicle is based on the vehicle information with respect to the first torque command value. The vibration damping torque command value is calculated by performing the calculation to suppress the vibration of the drive shaft torque transmission system.
The second q-axis current command value is calculated based on the magnetic flux estimated value and the vibration damping torque command value.
A method for controlling an electric vehicle, which controls the stator current and the rotor current based on the second q-axis current command value, the d-axis current command value, and the f-axis current command value.
In the method for controlling an electric vehicle according to claim 1,
In the calculation for suppressing the vibration of the drive shaft torque transmission system,
By performing feed forward control with respect to the first torque command value, the second torque command value is calculated.
The third torque command value is calculated by performing feedback control based on the actual vehicle information with respect to the vehicle information estimated according to the second torque command value.
A control method for an electric vehicle, which calculates the vibration damping torque command value by adding the second torque command value and the third torque command value.
In the method for controlling an electric vehicle according to claim 2,
In the feed forward control,
The second torque command value is calculated by performing a filter process on the first torque command value based on the transmission characteristics of the linear approximation model and the normative model in the electric vehicle.
In the feedback control
The linear approximation model for the second torque command value,
Control of an electric vehicle that calculates the third torque command value by performing a filter process based on the linear approximation model and a bandpass filter on the difference between the standard motor angular velocity and the measured motor angular velocity. Method.
In the method for controlling an electric vehicle according to claim 2,
In the feed forward control,
With respect to the output second torque command value, the first is used by using a vehicle model that models the characteristics from the motor torque to the drive shaft torsional angular velocity in the electric vehicle and the dead zone characteristics in the torque transmission system. Estimate the motor angular velocity and the pseudo drive shaft torsional velocity,
The second torque command value is calculated by subtracting the pseudo drive shaft torsion angle velocity obtained by multiplying the first torque command value by the gain.
In the feedback control
The second motor angular velocity estimated value is estimated by performing a filtering process based on the linear approximation model in the electric vehicle for the output third torque command value.
The third motor angular velocity estimated value is calculated by adding the first motor angular velocity estimated value calculated in the feed forward control and the second motor angular velocity estimated value.
The third torque command value is obtained by performing a filter process including a filter based on the linear approximation model and a bandpass filter on the difference between the third motor angular velocity estimated value and the measured motor angular velocity. A method of controlling an electric vehicle to calculate.
In the method for controlling an electric vehicle according to any one of claims 1 to 4,
The second q-axis current command value is a control method for an electric vehicle, which is calculated by dividing the vibration damping torque command value by the estimated magnetic flux value.
In the method for controlling an electric vehicle according to any one of claims 1 to 5,
Based on the f-axis current command value, the field magnetic flux estimated value, which is the estimated value of the field magnetic flux of the rotor, is calculated.
Based on the d-axis current command value, the equivalent magnetic flux estimated value of the reluctance torque generated in the rotor is calculated.
The magnetic flux estimated value is a control method for an electric vehicle, which is calculated by adding the field magnetic flux estimated value and the equivalent magnetic flux estimated value.
In the method for controlling an electric vehicle according to claim 6,
The field magnetic flux estimate is
Using a transmission characteristic configured to phase-advance and compensate the q-axis current response to the f-axis current transmission characteristic that simulates the response delay of the f-axis current that constitutes the rotor current to the f-axis current command value. Calculated electric vehicle control method.
In the method for controlling an electric vehicle according to claim 7,
The f-axis current transmission characteristic is a method for controlling an electric vehicle, which is a transmission function of a first-order lag.
In the method for controlling an electric vehicle according to claim 6,
The field magnetic flux estimated value is
An f-axis model that models the characteristics from the f-axis voltage to the f-axis current that constitutes the rotor current, and an f-axis model.
An f-axis current feedback (F / B) model in which the f-axis current command value and the output of the f-axis model are input, and
In a pseudo F / B system composed of an f-axis limiter that limits the output of the f-axis current F / B model, q with respect to the f-axis model and the f-axis current F / B model. A control method for an electric vehicle, which is calculated by compensating for the phase lead of the shaft current response.
In the method for controlling an electric vehicle according to claim 6,
The equivalent magnetic flux estimate is
Using a transmission characteristic configured to phase-advance and compensate the q-axis current response to the d-axis current transmission characteristic that simulates the response delay of the d-axis current that constitutes the stator current to the d-axis current command value. Calculated electric vehicle control method.
A field field synchronous motor including a rotor having a rotor winding and a stator having a stator winding, a stator current flowing through the stator winding, and a rotor flowing through the rotor winding. A control device for an electric vehicle including a controller for controlling a current.
The controller
Set the basic torque command value based on the operating condition,
Based on the basic torque command value and the vehicle information, the d-axis current command value and the first q-axis current command value for the stator current and the f-axis current command value for the rotor current are calculated.
A magnetic flux estimated value, which is an estimated value of the magnetic flux generated in the rotor, is calculated based on the d-axis current command value and the f-axis current command value.
The first torque command value is calculated based on the first q-axis current command value and the magnetic flux estimated value.
The vibration damping torque command value is calculated by performing a calculation for suppressing the vibration of the drive shaft torque transmission system of the electric vehicle based on the vehicle information with respect to the first torque command value.
The second q-axis current command value is calculated based on the magnetic flux estimated value and the vibration damping torque command value.
An electric vehicle control device that controls the stator current and the rotor current based on the second q-axis current command value, the d-axis current command value, and the f-axis current command value.