JP7779176B2 - Control method for electric vehicle and control device for electric vehicle - Google Patents

Description

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

Patent Document 1 discloses technology that suppresses vibrations generated in a vehicle by taking into account the transmission characteristics of the driving force transmission mechanism connected between the motor output shaft and the drive wheels.

International Publication No. 2013/157315

However, in Patent Document 1, motor control is performed based on a vehicle model that models the transmission characteristics of the driving force transmission mechanism and the motor characteristics. Therefore, in a configuration where multiple motors are connected to the same drive shaft, if control to suppress vibrations individually is applied to each motor, the output torque of each motor will affect each other, making it difficult to suppress vibrations.

The present invention aims to provide a control method and control device for an electric vehicle that can suppress vibrations generated in the vehicle, even when the vehicle is equipped with a driving force transmission mechanism that combines the power of multiple electric motors on a single drive shaft and transmits the power to the drive wheels.

The control method for an electric vehicle according to the present invention is equipped with a driving force transmission mechanism that combines the power of multiple electric motors on one drive shaft and transmits the power to the drive wheels. The method sets a torque command value for the electric motor based on vehicle information and controls the torque of the electric motor based on the torque command value. This control method inputs a total torque amount, which is the sum of the torque command values, into a vehicle model that models the characteristics of the multiple electric motors in the driving force transmission mechanism when they are assumed to be an inertial body rotating as a unit, and calculates parameters correlated with the rotational motion of the vehicle model. Then, when correcting the total torque amount based on a correction amount obtained by multiplying the parameter by a predetermined gain so that the rotational motion becomes a reference response, a distribution gain, which is the sum of the gains, is set individually for each electric motor, and the torque command value is individually corrected based on the distribution correction amount obtained by multiplying the parameter by the distribution gain.

According to the present invention, even in vehicles equipped with a driving force transmission mechanism that combines the power of multiple electric motors on a drive shaft, simulating an electric motor of a single inertial body that is rigidly connected and rotates roughly as one unit makes it possible to perform vibration suppression control using a controller that individually controls each electric motor, just as in the case of a single electric motor on a drive shaft. In this case, parameters correlated with the rotational motion of the driving force transmission system (e.g., motor rotation speed, motor angular acceleration, drive shaft torsional angular velocity, drive shaft torque, etc.) are calculated based on the total torque generated by the multiple electric motors, rather than the torque of each individual electric motor. This enables accurate parameter calculation, preventing the torque generated by each electric motor from acting as a disturbance, and effectively suppressing vibrations occurring in the vehicle (particularly torsional vibration of the drive shaft).

FIG. 1 is a diagram showing an example of a basic configuration of a control device for an electric vehicle according to a first embodiment. FIG. 2 is a diagram showing an example of an accelerator opening-torque table. FIG. 3 is a flowchart of the motor current control executed by the first motor controller and the second motor controller. FIG. 4 is a block diagram showing an example of a functional configuration for executing vibration damping control processing in the control device for the electric vehicle of the first embodiment. FIG. 5 is a diagram showing a model of a driving force transmission system of an electric vehicle. FIG. 6 is a block diagram of the rotational motion parameter calculator. FIG. 7 is a block diagram of the F/F compensator. FIG. 8 is a block diagram of the F/B compensator. FIG. 9 is a table showing the first drive shaft torsional angular velocity gain and first F/B gain allocated to the first motor controller, and the second drive shaft torsional angular velocity gain and second F/B gain allocated to the second motor controller in the control device for an electric vehicle of the first embodiment. FIG. 10 is a diagram showing an example of a basic configuration of a control device for an electric vehicle according to the second embodiment. FIG. 11 is a block diagram showing an example of a functional configuration for executing vibration damping control processing in a control device for an electric vehicle according to the second embodiment. FIG. 12 is a table showing the first drive shaft torsional angular velocity gain and first F/B gain allocated to the first motor controller, and the second drive shaft torsional angular velocity gain and second F/B gain allocated to the second motor controller in the control device for an electric vehicle of the third embodiment. FIG. 13 is a table showing the first drive shaft torsional angular velocity gain and first F/B gain allocated to the first motor controller, and the second drive shaft torsional angular velocity gain and second F/B gain allocated to the second motor controller in the control device for an electric vehicle of the fourth embodiment. FIG. 14 is a graph comparing the time charts of the prior art and the present embodiment when switching from negative torque to positive torque in the control devices for electric vehicles of the first, third, and fourth embodiments, in which [A] is the total torque amount and torque command value, [B] is the drive shaft torsion angular velocity estimated value, [C] is the motor rotational angular velocity estimated value, [D] is the F/F torque, [E] is the final torque command value, [F] is the drive shaft torque, [G] is the motor rotational angular velocity detected value, and [H] is the F/B torque. FIG. 15 is a graph comparing the time charts for switching from a small positive torque to a large positive torque in the control device for an electric vehicle of the second embodiment with those of the prior art, in which [A] is the total torque amount and torque command value, [B] is the motor rotational angular velocity estimated value, [C] is the F/B torque, [D] is the drive shaft torque, and [G] is the motor rotational angular velocity detected value.

Embodiments of the present invention will be described below with reference to the accompanying drawings.

[Basic configuration of the first embodiment]
FIG. 1 is a diagram showing an example of the basic configuration of a control device for an electric vehicle according to the first embodiment. The electric vehicle to which the control device for an electric vehicle according to the first embodiment is applied includes a driving force transmission mechanism in which two motors (a first motor 8A and a second motor 8B) are coupled to a single drive shaft 10 via a reduction gear 9, and drives itself by transmitting torque from the first motor 8A and the second motor 8B to drive wheels 11A and 11B attached to the drive shaft 10. Note that electric vehicles include not only electric vehicles using the first motor 8A and the second motor 8B as driving force, but also hybrid vehicles and fuel cell vehicles. The control device for an electric vehicle according to the first embodiment controls the first motor 8A and the second motor 8B.

The control device for an electric vehicle in the first embodiment includes a vehicle controller 1, a first motor controller 2A, a second motor controller 2B, a first inverter 3A, a second inverter 3B, a battery 4, a first current sensor 5A, a second current sensor 5B, a first rotation sensor 6A, a second rotation sensor 6B, and a wheel rotation sensor 7.

The vehicle controller 1 calculates the total torque amount T mall *, which corresponds to the sum of the torques to be commanded to the first motor controller 2A and the second motor controller 2B, from the accelerator opening-torque table shown in Figure 2 based on the vehicle speed V [km/h] obtained from the wheel rotation sensor 7 and the accelerator opening θ [%] obtained from the accelerator opening sensor _{(not shown), and then distributes the total torque amount T mall * to a first torque command value T m1 *} ^and _a ^{second torque} command value T _m2 ^* based on a predetermined torque distribution ratio, and commands the first motor controller 2A, which corresponds to the lower controller, to the first torque command value T _m1 ^* and the total torque amount T _mall ^* , and commands the second motor controller 2B to the second torque command value T _m2 ^* and the total torque amount T _mall ^* .

The first motor controller 2A receives signals of various vehicle variables such as the current of the first motor 8A ( _iu , _iv , iw in the case of three-phase AC) and the rotor phase α1 as digital signals, generates a PWM signal that controls the first motor 8A in accordance with the various vehicle variables, and generates a drive signal for the first inverter 3A via a drive circuit in accordance with this PWM signal.

The second motor controller 2B operates independently of the first motor controller 2A, but its configuration is similar to that of the first motor controller 2A.

The first inverter 3A is composed of, for example, two pairs of switching elements (e.g., power semiconductor elements such as IGBTs and MOS-FETs) for each phase, and by turning the switching elements ON/OFF in response to a drive signal, it converts the direct current supplied from the battery 4 into alternating current and supplies the desired current to the first motor 8A. Conversely, when regenerative power is generated in the first motor 8A, the first inverter 3A converts the regenerative power into direct current and charges the battery 4.

The second inverter 3B operates independently of the first inverter 3A, but its configuration is similar to that of the first inverter 3A.

The battery 4 supplies drive power to the first motor 8A via the first inverter 3A and to the second motor 8B via the second inverter 3B. The battery 4 is also charged with regenerative power generated by the first motor 8A via the first inverter 3A, and with regenerative power generated by the second motor 8B via the second inverter 4B.

The reducer 9 combines the driving force generated by the first motor 8A and the driving force generated by the second motor 8B, and drives the drive shaft 10 at a predetermined reduction ratio, thereby transmitting the driving force to the drive wheels 11A and 11B.

When the first motor 8A and second motor 8B rotate in conjunction with the drive wheels 11A, 11B while the vehicle is running, they generate regenerative driving force, allowing the vehicle's kinetic energy to be recovered as electrical energy (regenerative power).

In addition, the first motor 8A is equipped with a first current sensor 5A that detects each phase current and a first rotation sensor 6A (resolver or encoder) that detects the rotor phase α1 of the first motor 8A, while the second motor 8B is equipped with a second current sensor 5B that detects each phase current and a second rotation sensor 6B that detects the rotor phase α2 of the second motor 8B.

[Motor current control]
3 is a flowchart of the motor current control executed by the first motor controller 2A and the second motor controller 2B, respectively. Here, the motor current control by the first motor controller 2A will be explained as an example.

In the input processing of step S301, the signals required for the control calculations described below are obtained via sensor input or communication from another controller.

The three-phase current values _iu1 , _iv1 , and _iw1 flowing through the first motor 8A are acquired by the first current sensor 5A. Note that since the sum of the three-phase current values is 0, for example, _iw1 may be calculated from the values of _iu1 and _iv1 without being input from a sensor.

The rotor phase α1 (electrical angle) [rad] of the first motor 8A is obtained by the first rotation sensor 6A, such as a resolver or encoder.

The rotor angular speed ω (electrical angle) [rad/s] is calculated by differentiating the rotor phase α1 (electrical angle).

The motor rotation speed Nm [rpm] is calculated by dividing the rotor angular velocity ω (electrical angle) by the number of pole pairs of the electric motor to obtain the motor rotation angular velocity detection value ω _m [rad/s], which is the mechanical angular velocity of the electric motor, and then multiplying this by the unit conversion coefficient (60/2π) from [rad/s] to [rpm].

The DC voltage value Vdc [V] is determined from a voltage sensor installed on the DC power line or the power supply voltage value transmitted from the battery controller.

In the vibration suppression control calculation process of step S302, the vibration suppression control calculation of the present invention is performed. As will be described in detail later, the vibration suppression control calculation of the present invention calculates a first final torque command value _Tmf1 ^* that suppresses vibrations in the driving force transmission system (such as torsional vibrations of the drive shaft 10) based on the first torque command value _Tm1 ^{* (} and the total torque amount _Tmall ^* ) and the like.

In the current command value calculation process in step S303, first dq-axis current target values i d1 *, i _q1 ^* are found from a table based on the first final torque command value _{T mf1} ^* calculated in step S302, the motor rotational angular velocity detection value ωm, and the DC voltage value ^Vdc .

In the current control of step S304, first, first dq-axis current values i _d1 and i _q1 are calculated from the three-phase current values i _u1 , i _v1 and i _w1 of the first motor 8A and the rotor phase α1 of the first motor 8A.

Next, the first dq-axis voltage command values _vd1 , _vq1 are calculated from the deviation between the first dq-axis current target values _id1 ^* , _iq1 ^* calculated in step S304 and the first dq-axis current values _id1 , _iq1 . Note that decoupling control may be applied to this part.

Next, three-phase voltage command values _vu1 , _vv1 , and _vw1 are calculated from the first dq-axis voltage command values _vd1 , _vq1 and the rotor phase α1 of the first motor 8A. First PWM signals (on duty) _tu1 [%], _tv1 [%], and _tw1 [% _] are calculated from the three-phase voltage command values _vu1 , _vv1 , and vw1 and the DC voltage value Vdc.

By controlling the opening and closing of the switching elements of the first inverter 3A using the first PWM signal obtained in this manner, the first motor 8A can be driven at the desired torque indicated by the first final torque command value T _mf1 ^* .

The processing of the first motor controller 2A has been described above, but the second motor controller 2B also performs similar processing independently for the second motor 8B. That is, the second motor controller 2B controls the opening and closing of the switching elements of the second inverter 3AB using a second PWM signal calculated using the same processing as above, thereby driving the second motor 8B at the desired torque indicated by the second final torque command value _Tmf2 ^* .

[Vibration suppression control calculation overview]
Figure 4 is a block diagram showing an example of the functional configuration for executing vibration damping control processing in the control device for an electric vehicle of the first embodiment. As shown in Figure 4, the control device for an electric vehicle of the first embodiment includes, as components (programs) that execute the vibration damping control processing, a rotational motion parameter calculator 21, an F/F compensator 22 (feedforward compensator), an F/B compensator 23 (feedback compensator), and an adder 24. Note that although Figure 4 shows the components implemented in the first motor controller 2A, similar components are also implemented in the second motor controller 2B.

The total torque amount T _mall ^* is input to the rotational motion parameter calculator 21. The rotational motion parameter calculator 21, details of which will be described later, also has a vehicle model that represents the response of the driving force transmission mechanism including the first motor 8A, the second motor 8B, the speed reducer 9, and the drive shaft 10, and assumes that the first motor 8A and the second motor 8B are an inertial body that rotates integrally. When the total torque amount T _mall ^* is input, the rotational motion parameter calculator 21 calculates, based on the vehicle model, parameters related to the rotational motion of the driving force transmission mechanism, such as an estimated drive shaft torsion angular velocity value ω _d ^ and an estimated motor rotational angular velocity value ω _m ^.

The F/F compensator 22, details of which will be described later, calculates a first F/F torque command value T mff1 ^* when the first torque command value T _m1 *, the total torque amount T _mall ^* , and the drive shaft torsion angular velocity estimated value ω _{d ^} are input, taking into account ^the relationship between the first torque command value T _m1 ^* and the total torque amount T _mall ^* .

The F/B compensator 23, details of which will be described later, calculates a first F/B torque command value T _mfb1 ^* when the first torque command value T _m1 ^* , the total torque amount T _mall ^* , the motor rotational angular velocity estimated value ω _m ^, and the motor rotational angular velocity detected value ω _m are input.

The adder 24 adds the first F/F torque command value T _mff1 ^* and the first F/B torque command value T _mfb1 ^* together to output a first final torque command value T _mf1 ^* .

[Drive force transmission system model]
5 is a diagram showing a model of the driving force transmission system of an electric vehicle. The equations of motion of the vehicle are expressed by equations (1) to (6).

Here, the parameters are as follows:

J _m : Motor inertia J _w : Drive shaft inertia (for one shaft)
_Kd : Torsional rigidity of drive shaft _Kf : Coefficient of friction between drive wheels and road surface N: Overall gear ratio r: Tire load radius _ωm : Motor rotational angular velocity _θm : Motor rotational angle _ωw : Drive wheel rotational angular velocity _θw : Drive wheel rotational angle _Tm : Motor torque _Td : Drive shaft torque F: Driving force (for two shafts)
V: Vehicle speed θ _d : Drive shaft twist angle

_Jm is the motor inertia when the first motor 8A and the second motor 8B are regarded as an inertial body that rotates substantially as one unit in the driving force transmission system, and has the relationship of equation (7).

J _m1 : Inertia of the first motor 8A J _m2 : Inertia of the second motor 8B When equations (1) to (6) are Laplace transformed to find the transfer characteristics from the motor torque T _m to the motor rotational angular velocity ω _m, the following equation is obtained.

However, the parameters are as follows:

The transfer characteristic from the motor torque _Tm to the drive shaft torque _Td is expressed by the following equation:

The transfer characteristic from the motor rotation angular velocity ω _m to the drive shaft angular velocity ω _w can be calculated from equations (2), (4), (5), and (6) to obtain the following equation.

From equations (8), (9), and (14), the transfer characteristic from the motor torque T _m to the drive shaft angular velocity ω _w is expressed by the following equation:

From equations (11) and (15), the transfer characteristic from the drive shaft torque _Td to the drive shaft angular velocity _ωw is expressed by the following equation:

Here, equation (1) is transformed to
Therefore, from equations (16) and (17), the drive shaft torsional angular velocity _ωd can be expressed as follows:

however,

is.

Furthermore, equation (9) is
Here, ζ _p is the damping coefficient of the drive shaft torque transmission system, and ω _p is the natural vibration frequency of the drive shaft torque transmission system.

Furthermore, when the poles and zeros of equation (24) are examined, α≈c ₀ /c ₁ is obtained, and therefore, when the poles and zeros are cancelled out, the following equation is obtained.

Here, the vibration damping torque standard value T _mr ^* is
Then, equations (4) and (6) can be rewritten as follows:

Here, when T _m =T _mr ^* and equation (26) is substituted into equation (24), the following equation can be obtained.

When the reference response from the motor torque to the drive shaft torque is expressed as follows:
The condition under which equations (28) and (29) match is as follows:

where ζ _r is the viscosity coefficient of the reference response.

Next, apply equations (1) to (30) to model the backlash characteristics from the motor to the drive shaft using a dead zone.

The drive shaft torque _Td is expressed by the following equation instead of equation (4).
Here, θ _dead is the overall backlash amount from the motor to the drive shaft.

The motor rotational angular velocity estimate ω _m ^ is obtained by integrating equation (1) as follows:
By calculating it in this way, it is possible to take into account the backlash characteristics from the motor to the drive shaft.

The drive shaft torsional angular velocity estimate ω _d ^ is expressed as follows, similarly to equation (18):
By calculating it in this way, it is possible to take into account the backlash characteristics from the motor to the drive shaft.

In the vibration damping model 2113 (FIG. 6), the vibration damping torque reference value T _mr ^* for realizing the reference response of the equation (29) is calculated by the following equation, similarly to the equation (26):
and input it to the driving force transmission system model 2101 (FIG. 6).

[Rotational motion parameter calculator 21]
6 is a block diagram of the rotational motion parameter calculator 21. The rotational motion parameter calculator 21 includes a driving force transmission system model 2101 and a vibration damping model 2113 as vehicle models.

The driving force transmission system model 2101 includes a proportional element 2102, a proportional element 2103, a subtractor 2104, an integral element 2105, an integral element 2106, a dead band element 2107, a proportional element 2108, a proportional element 2109, a filter 2110, a subtractor 2111, and an integral element 2112.

The vibration control model 2113 includes a subtractor 2114 and a proportional element 2115.

When proportional element 2102 receives the output (vibration-damping torque reference value T _mr ^* ) from subtractor 2114 , it multiplies the output by a gain (1/Jm) and outputs the result to subtractor 2111 .

When proportional element 2103 receives the output (vibration-damping torque reference value T _mr ^* ) from subtractor 2114 , it multiplies the output by a gain (1/(J _m ·N)) and outputs the result to subtractor 2104 .

Subtractor 2104 outputs the difference obtained by subtracting the output of filter 2110 from the output of proportional element 2103 to integral element 2105.

The integral element 2105 integrates the output of the subtractor 2104 and outputs the output (drive shaft torsional angular velocity estimated value ω _d ^: equation (33)) to the integral element 2106, the F/F compensator 22 (FIGS. 4 and 7), and the proportional element 2115.

The integral element 2106 integrates the output of the integral element 2105 (the drive shaft torsional angular velocity estimated value ω _d ^) and outputs the output (θ _d : equation (31)) to the dead band element 2107 .

The dead band element 2107 outputs to the proportional element 2108 an output obtained by adding the backlash amount to the output (θ _d ) from the integral element 2016 .

The proportional element 2108 multiplies the output of the dead band element 2107 by a gain (K _d : torsional rigidity of the drive shaft 10 ), and supplies the output (drive shaft torque T _d : equation (31)) to the proportional element 2109 and the filter 2110 .

Proportional element 2109 multiplies the output from proportional element 2108 by a gain (1/(Jm·N)) and outputs the result to subtractor 2111.

The filter 2110 multiplies the output from the proportional element 2108 by a filter (H _w (s): equation (19)) and outputs the result to the subtractor 2104 .

Subtractor 2111 calculates the difference by subtracting the output of proportional element 2109 from the output of proportional element 2102 and outputs it to integral element 2112.

The integral element 2112 integrates the output from the subtractor 2111 and outputs the output (motor rotational angular velocity estimated value ω _m ^: equation (32)) to the F/B compensator 23 (FIGS. 4 and 8).

The proportional element 2115 multiplies the output of the integral element 2105 (drive shaft torsion angular velocity estimated value ω _d ^) by the drive shaft torsion angular velocity gain k _all and outputs the result to the subtractor 2114 .

Subtractor 2114 calculates the difference between the total torque amount T _mall ^* and the output of proportional element 2115 , and outputs the output (oscillation-damping torque reference value T _mr ^* : equation (34)) to proportional elements 2102 and 2103 .

[F/F compensator 22]
7 is a block diagram of the F/F compensator 22. The F/F compensator 22 includes a drive shaft torsion angular velocity gain setting unit 221 and a subtractor 222.

The first torque command value T _m1 ^* , the total torque amount T _mall ^* , and the drive shaft torsional angular velocity estimated value ω _d ^ are input to the drive shaft torsional angular velocity gain setting unit 221. Then, the drive shaft torsional angular velocity gain setting unit 221 calculates a first drive shaft torsional angular velocity gain k _ff1 based on the following equation (36), and outputs a value obtained by multiplying the drive shaft torsional angular velocity estimated value ω _d ^ by the first drive shaft torsional angular velocity gain K _ff1 to the subtractor 222.

The subtractor 222 calculates a first F/F torque command value T _mff1 ^* from the difference between the first torque command value T _m1 ^* and the output of the drive shaft torsion angular velocity gain setting unit 221 based on the following equation (35).

In equation (34), the reference response of equation (29) is realized by feeding back a value obtained by multiplying the drive shaft torsion angular velocity estimate value ω _d ^ generated by the combined torque of the first motor 8A and the second motor 8B by the drive shaft torsion angular velocity gain k _all . This pseudo feedback torque component can be fed back to only one of the first motor 8A and the second motor 8B, or can be distributed to the first motor 8A and the second motor 8B. In the first embodiment, the feedback torque component is distributed according to the torque distribution ratio between the first motor 8A and the second motor 8B (the ratio between the magnitudes of the first torque command value T _m1 ^* and the second torque command value T _m2 ^* ).

Therefore, the first F/F torque command value T _mff1 ^* is calculated using the following equation:
Here, the first drive shaft torsional angular velocity gain k _ff1 is calculated by
If T _mall ^* is 0, a process to prevent division by zero is performed, such as holding a past value that is not 0.

[F/B compensator]
8 is a block diagram of the F/B compensator 23. The F/B compensator 23 includes an adder 231
, a subtractor 232 , a filter 233 , a filter 234 , and an F/B gain setting unit 235 .

The adder 231 adds the motor rotational angular velocity estimated value ω _m ^ calculated by the rotational motion parameter calculator 21 and the output of the filter 234 (motor rotational angular velocity estimated value for the F/B torque command value), and outputs the output (final motor rotational angular velocity estimated value ω _mf ^) to the subtractor 232.

The subtractor 232 calculates a difference by subtracting the motor rotation angular velocity detection value ω _m from the output of the adder 231 (the final motor rotation angular velocity estimate value ω _mf ^ or the motor rotation angular velocity estimate value ω _m ^), and outputs the difference to the filter 233 .

The filter 233 superimposes a transfer characteristic (H(s)/G _p (s)) described below on the output of the subtractor 232 , and outputs the resultant output to a filter 234 and an F/B gain setting section 235 .

The filter 234 superimposes a transfer characteristic (G _p (s)) described below on the output of the filter 234 , and outputs the output (a motor rotation angular velocity estimate value for the F/B torque command value) to the adder 231 .

The first torque command value T _m1 ^* , the total torque amount T _mall ^* , and the output of the filter 233 are input to the F/B gain setting section 235 .

The F/B gain setting unit 235 calculates the first F/B gain k _fb1 according to the following equation.
Here, the gain _kfball is a value that satisfies the predetermined stability margin (gain margin, phase margin) of the F/B control system (step S304 in FIG. 3) of the control device for an electric vehicle of this embodiment. If _Tmall ^* is 0, a process to prevent division by zero is performed, such as holding a past value that is not 0.

Furthermore, the F/B gain setting unit 235 calculates the first F/B torque command value T _mfb1 ^* according to the following equation.
Here, ω _mf ^=ω _m ^+H(s)(ω _mf ^(previous value)-ω _m (previous value)).

As described above, the F/B compensator 23 calculates the first F/B torque command value T mfb1 * by taking the deviation between the motor rotational angular velocity estimated value ω _m ^ (or the final motor rotational angular velocity estimated value ω _mf ^) calculated by the rotational motion parameter calculator 21 and the motor rotational angular velocity detected value ω _m , and multiplying it by the first F/B gain k _fb1 through a filter H(s)/Gp(s) (filter 233) consisting of the inverse characteristic of the transfer characteristic Gp(s) of the controlled object and a band-pass filter H _(s ⁾ .

Here, the transfer characteristic G _p (s) of the controlled object is expressed by equation (9). The band-pass filter H(s) is set so that the attenuation characteristics on the low-pass side and the high-pass side are approximately the same and the torsional resonance frequency of the drivetrain is near the center of the passband on a logarithmic axis (log scale). For example, when H(s) is configured with a first-order high-pass filter and a first-order low-pass filter, it is configured as shown in equation (39) where frequency f _p is the torsional resonance frequency of the drivetrain and k is an arbitrary value.
However, τ _L =1/(2πf _HC ), f _HC =k·f _p , τ _H =1/(2πf _LC ), and f _LC = f _p /k.

Incidentally, the above-mentioned F/B compensation based on the actual motor rotational angular velocity detection value _ωm can be fed back to only one of the first motor 8A and the second motor 8B, or can be distributed simultaneously to the first motor 8A and the second motor 8B. In this embodiment, as described above, the F/B torque portion is distributed according to the torque distribution ratio between the first motor 8A and the second motor 8B.

The vibration suppression control calculation process performed by the first motor controller 2A has been described above, but the second motor controller 2B also performs similar processing.

Figure 9 is a table showing the first drive shaft torsional angular velocity gain _kff1 and first F/B gain _kfb1 allocated to the first motor controller 2A, and the second drive shaft torsional angular velocity gain kff2 and second F/B _{gain kfb2 allocated} to the second motor controller 2B in the control device for an electric vehicle of the first embodiment.

As shown in Fig. 9, the sum of the first torque command value _Tm1 ^* and the second torque command value _Tm2 ^* sent to the first motor controller 2A is defined as total torque _Tmall ^* . Furthermore, kall represents the response of the driving force transmission mechanism including the first motor 8A, the second motor 8B, the speed reducer 9, and the drive shaft 10, and is a drive shaft torsional angular velocity gain for realizing a reference response (reducing drive shaft torsional vibration) in a vehicle model in which the first motor 8A and the second motor 8B are assumed to be inertial bodies rotating integrally. Furthermore, _kfball is a F/B gain that satisfies predetermined stability margins (gain margin, phase margin) of the F/B control system (step S304 in Fig. 3) of the control device for an electric vehicle according to this _embodiment .

At this time, as described above, the first drive shaft torsional angular velocity gain k _ff1 applied to the first motor controller 2A is set to k _ff1 = k _all (T _m1 ^* / T _mall ^* ), and the first F/B gain k _fb1 is set to k _fb1 = k _fball (T _m1 ^* / T _mall ^* ).

Furthermore, the second drive shaft torsional angular velocity gain _kff2 applied to the second motor controller 2B is set to _kff2 = _kall ( _Tm2 ^* / _Tmall ^* ), and the second F/B gain _kfb2 is set to _kfb2 = _kfball ( _Tm2 ^* / _Tmall ^* ), where _kall = _kff1 + _kff2 and _kfball = _kfb1 + _kfb2 .

As is clear from Figure 9, when there is a difference between the first torque command value T _m1 ^* and the second torque command value T _m2 ^* , the F/B gain becomes variable depending on the distribution ratio, but the total amount of F/B torque when F/B control is performed by the first motor 8A and the second motor 8B becomes constant regardless of the distribution ratio, and therefore the stability margin also becomes constant.

Second Embodiment
10 is a diagram showing an example of the basic configuration of a control device for an electric vehicle according to the second embodiment. The second embodiment is based on the premise that the outputs of the first motor 8A and the second motor 8B are the same, and that the magnitudes of the first torque command value T _m1 ^* and the second torque command value T _m2 ^* are the same.

Based on the vehicle speed V [km/h] obtained from the wheel rotation sensor 7 and the accelerator opening θ [%] obtained from the accelerator opening sensor (not shown), the vehicle controller 1 calculates the first torque command value T _m1 ^* to be commanded to the first motor controller 2A and the second torque command value T _m2 * to be commanded to the second motor controller 2B from the accelerator opening-torque table shown in Figure 2, and commands the first torque command value T _m1 ^* to the first motor controller 2A and the second torque command value T _m2 ^* to the second motor controller 2B.

The current control (Figure 3) performed by the first motor controller 2A and the second motor controller 2B is the same as in the first embodiment.

Figure 11 is a block diagram showing an example of the functional configuration for executing vibration damping control processing in a control device for an electric vehicle according to the second embodiment. The first motor controller 2A (and the second motor controller 2B) includes an F/F compensator 22, a rotational motion parameter calculator 21, an F/B compensator 23, and an adder 24 as components that also execute vibration damping control processing, similar to the first embodiment.

The F/F compensator 22 calculates a first F/F torque command value T _mff1 ^* based on the following equation to which the first torque command value T _m1 ^* is input, and outputs the calculated value to the adder 24 .

When the first F/F torque command value T _mff1 calculated by the F/F compensator 22 is input, the rotational motion parameter calculator 21 calculates the motor rotational angular velocity estimate value ω _m ^ by the following equation based on equation (8).
Here, 2T _ff1 ^* is used in order to calculate the rotational angular velocity when the total torque T _mall ^* (first torque command value T _m1 ^* × 2) is input to the vehicle model (G _p (s)). G _p (s) follows equation (9).

The F/B compensator 23 receives the motor rotation angular velocity estimated value ω _m ^ and the motor rotation angular velocity detected value ω _m .

As in the first embodiment, the F/B compensator 23 calculates the first F/B torque command value T mfb1 * by taking the deviation between the motor rotational angular velocity estimated value ωm^ and the motor rotational angular velocity detected value _ωm , and multiplying the deviation by a gain k _fball /2 through a filter H(s)/Gp(s) consisting of the inverse characteristic of the transfer characteristic Gp( _s ) of the controlled object and a band-pass filter H(s ⁾ . As described above, the gain k _fball is a value that satisfies the stability margin (gain margin, phase margin) of the F/B control system (step S304 in FIG. 3). Therefore, when setting the F/B gain in the first motor controller 2A and the second motor controller 2B, it is necessary to set a value that is half the F/B gain k _fball to avoid excessive gain.

As in the first embodiment, the adder 24 adds the first F/F torque command value T _mff1 ^* and the first F/B torque command value T _mfb1 ^* together to output a first final torque command value T _mf1 ^* .

[Third embodiment]
FIG. 12 is a table showing the first drive shaft torsional angular velocity gain k _ff1 and first F/B gain k _fb1 allocated to the first motor controller 2A, and the second drive shaft torsional angular velocity gain k _ff2 and second F/B gain k _fb2 allocated to the second motor controller 2B in the control device for an electric vehicle of the third embodiment.

The control device for an electric vehicle of the third embodiment and the electric vehicle to which it is applied are similar to those of the first embodiment, but the drive shaft torsional angular velocity gain and F/B gain are allocated according to the ratio between the maximum output Pmax1 (maximum torque) of the first motor 8A and the maximum output Pmax2 (maximum torque) of the second motor 8B.

That is, regardless of the first torque command value T _m1 ^* commanded to the first motor controller 2A, the first drive shaft torsional angular velocity gain k _ff1 applied to the first motor controller 2A is set to k _ff1 = k _all (P _max1 / (P _max1 + P _max2 )), and the first F/B gain k _fb1 is set to k _fb1 = k _fball (P _max1 / (P _max1 + P _max2 )).

In addition, regardless of the second torque command value T _m2 ^* commanded to the second motor controller 2B, the second drive shaft torsional angular velocity gain k _ff2 applied to the second motor controller 2B is set to k _ff2 = k _all (P _max2 /(P _max1 + P _max2 )), and the second F/B gain k _fb2 is set to k _fb2 = k _fball (P _max2 /(P _max1 + P _max2 )).

By allocating the drive shaft torsional angular velocity gain and F/B gain as described above, the drive shaft torsional angular velocity gain and F/B gain can be set to fixed values, reducing the computational burden on the first motor controller 2A and second motor controller 2B.

[Fourth embodiment]
FIG. 13 is a table showing the first drive shaft torsional angular velocity gain k _ff1 and first F/B gain k _fb1 allocated to the first motor controller 2A, and the second drive shaft torsional angular velocity gain k _ff2 and second F/B gain k _fb2 allocated to the second motor controller 2B in the control device for an electric vehicle of the fourth embodiment.

The electric vehicle control device of the fourth embodiment and the electric vehicle to which it is applied are similar to those of the first embodiment, but are applicable, for example, to cases in which motors with different rotation detection accuracies are combined and used.

When combining two types of electric motors, if there are differences in the type of rotation sensor used or in the detection method (actually measured values versus estimated values), using an electric motor with higher rotation detection accuracy will improve torque accuracy and make it easier to demonstrate performance, at least when using the motor rotation angular velocity detection value for feedback compensation. Therefore, for example, if the rotation detection accuracy of the first motor 8A is higher than that of the second motor 8B, the drive shaft torsional angular velocity gain and F/B gain to be applied to the first motor controller 2A and the second motor controller 2B are allocated as shown in Figure 13.

That is, regardless of the first torque command value T _m1 ^* commanded to the first motor controller 2A, the first drive shaft torsional angular velocity gain k _ff1 applied to the first motor controller 2A is set to k _ff1 = k _all , and the first F/B gain k _fb1 is set to k _fb1 = k _fball .

Furthermore, regardless of the second torque command value T _m2 ^* commanded to the second motor controller 2B, the second drive shaft torsional angular velocity gain k _ff2 applied to the second motor controller 2B is set to k _ff2 =0, and the second F/B gain k _fb2 is set to k _fb2 =0.

[Time Chart 1]
FIG. 14 is a graph comparing the time charts of the prior art and the present embodiment when switching from negative torque to positive torque in the control devices for electric vehicles of the first, third, and fourth embodiments, in which [A] is the total torque amount and torque command value, [B] is the drive shaft torsion angular velocity estimated value, [C] is the motor rotational angular velocity estimated value, [D] is the F/F torque, [E] is the final torque command value, [F] is the drive shaft torque, [G] is the motor rotational angular velocity detected value, and [H] is the F/B torque.

Figure 14 shows the response when the torque command value is changed in a stepwise manner from a state in which regenerative torque is being generated and the vehicle is decelerating, resulting in re-acceleration.

The control device for an electric vehicle of the prior art (see Patent Document 1) that is compared to the present invention is applied to an electric vehicle in which a pair of motors are attached to a reducer and are capable of transmitting the torque of the pair of motors to a drive shaft, and the drive shaft torsional angular velocity gain and F/B gain applied to one of the pair of motors (and the same for the other motor) are set so that the drive force transmission system, which includes one motor, reducer, and drive shaft, exhibits a nominal response to the torque command value given to one of the motors (controller).

In other words, in the prior art, the drive shaft torsional angular velocity gain is set to a value that estimates the drive shaft torsional angular velocity when torque is input to only one motor (and the other motor as well), and is used to calculate F/F torque that offsets drive shaft torsional vibration related to the estimated drive shaft torsional angular velocity. Similarly, the F/B gain is set to a value that satisfies the stability margin when F/B control is performed based on the difference between the estimated motor rotational angular velocity and the detected motor rotational angular velocity.

Therefore, in conventional technology, the case where two motors are driven simultaneously is not taken into consideration when calculating the drive shaft torsion angular velocity estimate and the motor rotational angular velocity estimate.

14A, the torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) change stepwise at time t0. For simplicity, the first torque command value T _m1 ^* = second torque command value T _m2 ^* = (total torque amount T _mall ^* )/2.

As shown in Figure 14 [B], in the prior art (dashed line), the drive shaft torsional angular velocity estimate is calculated based on the torque of one motor, which deviates from the value calculated based on the torque of two motors. In contrast, in the present invention (solid line), the drive shaft torsional angular velocity estimate is calculated based on the torque of two motors.

As shown in Figure 14 [C], in the present invention (solid line), the estimated motor rotational angular velocity is calculated based on the torque of two motors. Therefore, in the present invention, the estimated motor rotational angular velocity and the detected motor rotational angular velocity (solid line) shown in Figure 14 [G] nearly match, and as shown in Figure 14 [H], almost no F/F torque is generated.

On the other hand, as shown in Figure 14 [C], in the prior art (dashed line), the estimated motor rotational angular velocity is calculated based on the torque of a single motor. Therefore, in the prior art, the estimated motor rotational angular velocity deviates significantly from the detected motor rotational angular velocity value (solid line) shown in Figure 14 [G], and as shown in Figure 14 [H], after time t0, unnecessary F/B torque oscillates in response to the drive shaft torsional angular velocity. For this reason, as shown in Figure 14 [G], oscillations also occur in the detected motor rotational angular velocity value.

As described above, in the prior art, the F/B gain setting value is set for each of the two motors to satisfy the stability margin when driven by one motor, resulting in excessive gain and oscillation. On the other hand, in the present invention, the gain is set so that the sum of the F/B gains of the two motors (k _fb1 +k _fb2 ) matches the gain (k _fball ) that satisfies the stability margin when driven by one motor, so oscillation does not occur.

As shown in Figure 14 [D], in the prior art (dashed line), the F/F torque is calculated based on the torque of one motor. On the other hand, in this embodiment (solid line), the F/F torque is calculated based on the torque of two motors. There is no difference between the prior art and the present invention regarding the F/F torque from time t0 to time t1.

In both cases, torque is applied to two motors, and the gear backlash dead zone is passed at time t1. However, in conventional technology, F/F torque is calculated based on the torque of one motor, so the dead zone is determined to have been passed at time t2, which is delayed from time t1, and the F/F torque is increased. In contrast, in this embodiment, F/F torque is calculated based on the torque of two motors, and the F/F torque is increased when the dead zone is actually passed.

The final torque command value (solid line) of the present invention shown in Figure 14[E] corresponds to the sum of the solid curve shown in Figure 14[D] and the solid curve shown in Figure 14[H]. Similarly, the final torque command value (dashed line) of the prior art shown in Figure 14[E] corresponds to the sum of the dashed curve shown in Figure 14[D] and the dashed curve shown in Figure 14[H]. Furthermore, the drive shaft torque (solid line, dashed line) in Figure 14[F] is calculated using the curves (solid line, dashed line) in Figure 14[E], equations (28), and (31).

As shown by the final torque command value (dashed line) in Figure 14 [E] and the drive shaft torque (dashed line) in Figure 14 [F], with conventional technology, torque begins to rise after time t2, which is after the time t1 when the dead zone is passed, causing a feeling of stagnation in acceleration. Furthermore, with the drive shaft torque (dashed line) shown in Figure 14 [F], the vibration component appearing in the dashed curve in Figure 14 [G] is offset by the vibration component of the dashed curve in Figure 14 [E]. However, a vibrational response remains in the drive shaft torque (dashed line), which causes discomfort to the driver. Furthermore, it takes approximately time t4 for the torque to reach a steady value.

On the other hand, as shown by the final torque command value (solid line) in Figure 14[E] and the drive shaft torque (solid line) in Figure 14[F], in the present invention, torque begins to rise after time t1 when the dead zone is passed, significantly improving the feeling of sluggish acceleration. Also, by hastening the time at which the torque reaches a steady value (from time t4 to time t3), the feeling of sluggish acceleration is significantly reduced. Furthermore, the vibrational response caused by excessive F/B gain is also improved, improving ride comfort. As a result, even when two motors are used, it is possible to accelerate again without causing any discomfort to the driver.

[Time Chart 2]
FIG. 15 is a graph comparing the time charts for switching from a small positive torque to a large positive torque in the control device for an electric vehicle of the second embodiment with those of the prior art, in which [A] is the total torque amount and torque command value, [B] is the motor rotational angular velocity estimated value, [C] is the F/B torque, [D] is the drive shaft torque, and [G] is the motor rotational angular velocity detected value.

Here, we show the response when the accelerator is pressed during gentle acceleration with power torque being generated.

As shown in Figure 15 [A], at time t0, the torque command value and total torque amount change in a step-like manner.

As shown in Figure 15[B], in the prior art (dashed line), the estimated motor rotational angular velocity is calculated based on the torque of a single motor, which deviates from the detected motor rotational angular velocity value (solid line) shown in Figure 15[E]. Therefore, as shown by the dashed curve in Figure 15[C], unnecessary F/B torque is generated after time t0.

On the other hand, in the present invention (solid line), the estimated motor rotational angular velocity is calculated based on a torque twice that of a single motor, which roughly matches the detected motor rotational angular velocity value (solid line) shown in Figure 15[E]. Therefore, as shown by the solid curve in Figure 15[C], almost no unnecessary F/B torque is generated even after time t0.

As shown in Figure 15 [D], the time at which the drive shaft torque reaches a steady value is delayed until t2 in the prior art, whereas in the present invention it is advanced to t1, significantly reducing the feeling of sluggishness during acceleration. As a result, even when two motors are used, acceleration is possible without causing any discomfort to the driver.

[Effects of this embodiment]
According to the control method for an electric vehicle of this embodiment, the electric vehicle is provided with a driving force transmission mechanism (reduction gear 9, drive shaft 10) that combines the power of multiple electric motors (first motor 8A, second motor 8B) on one drive shaft 10 and transmits the power to drive wheels 11A, 11B, and the method sets torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) for the electric motors (first motor 8A, second motor 8B) based on vehicle information (vehicle speed _V , accelerator opening _θ ), and controls the torque of the electric motors (first motor 8A, second motor 8B) using the torque command values ⁽ first torque command value ^T m1 *, second torque command value T m2 *). In the control method for an electric vehicle of this embodiment, a vehicle model (G _p In the case where a total torque amount T _mall ^*, which is the sum of torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ), is input to a vehicle model (G _p (s)), parameters (drive shaft torsion angular velocity estimated value ω _d ^, motor rotational angular velocity estimated value ω _m ^) correlated with the rotational motion of the vehicle model (G p (s)) are calculated, and the total torque amount T _mall ^* is corrected based on a correction amount (k _fall ω _d ^, k _fball ( _H (s)/G _p (s)(ω _m ^-ω _m ))) obtained by multiplying the _parameters (drive shaft _torsion angular velocity estimated value ω _d ^, motor rotational angular velocity estimated value ω m ^) by predetermined gains (k _all , k _fball ) so that the rotational motion becomes a reference response, Parameters (( _kff1 , _kff2 ), ( _kfb1 , _kfb2 )) are set individually for the electric motors (first motor 8A, second motor 8B), and the torque command values (first torque command value _Tm1 *, second torque command value _Tm2 *) are individually corrected based on distribution correction amounts (first _F / _F torque command value _Tmff1 ^* , first F/ _B torque command value _Tmfb1 ^* , second F/F torque command value _Tmff2 *, second F/B torque command value _Tmfb2 ^*) obtained by multiplying the parameters (drive shaft torsional angular velocity estimated value ωd^, motor rotational angular velocity estimated value ωm ^{^} ) by the distribution gains (( _kff1 , ^kff2 ), ( _kfb1 , _kfb2 ⁾ ).

The above method allows a vehicle equipped with a driving force transmission mechanism (reduction gear 9, drive shaft 10) that combines the power of multiple electric motors (first motor 8A, second motor 8B) on the drive shaft 10 to simulate a single inertial body of electric motors that are rigidly connected and rotate generally as one unit. This allows vibration control to be performed using controllers (first motor controller 2A, second motor controller 2B) that individually control the electric motors (first motor 8A, second motor 8B), just as in the case of a single electric motor for the drive shaft 10. In this case, parameters correlated with the rotational motion of the driving force transmission system (e.g., motor rotation speed, motor angular acceleration, drive shaft torsional angular velocity, drive shaft torque, etc.) are calculated based on the total torque generated by the multiple electric motors, rather than the torque of each individual electric motor. This allows for accurate parameter calculation, preventing the torque generated by each electric motor from acting as a disturbance, and effectively suppressing vibrations occurring in the vehicle (particularly torsional vibration of the drive shaft 10).

In the above control method, the parameters (drive shaft torsion angular velocity estimated value ω _d ^, motor rotational angular velocity estimated value ω _m ^) include rotational angular velocity estimated values (motor rotational angular velocity estimated value ω _m ^) of the electric motors (first motor 8A, second motor 8B), the gains (k _all , k _fball ) include a feedback gain (k fbll ) multiplied by a difference between the rotational angular velocity estimated value (motor rotational angular velocity estimated value ω _m ^) and the rotational angular velocity detected value (motor rotational angular velocity detected value ω _m ) of the electric motors (first motor 8A, second motor 8B), and when the total torque amount T _mall ^* is corrected based on the product of the feedback gain (k _fbll ) and the difference so as to reduce the difference, the _distribution gains include distribution feedback gains (k _fb1 , k _fb2 ) which are the sum of the feedback gains (k _fbll ), and the difference is multiplied by the distribution feedback gain (k _{fball )} . The torque command values (first torque command value T _m1 ^* , second torque command value T m2 *) are individually corrected based on the distribution correction amounts (first F/B torque command value T _mfb1 *, second F/B torque command value T _mfb2 ^* ) obtained by multiplying the torque command values (first F/B torque command value T _mfb1 ^* , second F/B torque command value T mfb2 *) by the torque command values (first F/B torque command value T m1 *, second F/B torque command value T _mfb2 ^* ).

If the above method were to be used to perform the same feedback compensation using multiple motors as when the drive shaft 10 is driven by a single motor, overcompensation could occur, potentially making it impossible to ensure the necessary stability margin. Therefore, by individually setting the feedback gains of the multiple motors so that the necessary stability margin is ensured, feedback compensation can be performed without compromising stability.

In the above control method, the distribution feedback gains (k _fb1 , f _fb2 ) are set higher as the maximum outputs (P _max1 , P _max2 ) of the electric motors (first motor 8A, second motor 8B) are higher.

Using the above method, when combining electric motors with different maximum outputs, the motor with the higher maximum output can perform more torque correction. Therefore, by setting a larger F/B gain for the electric motor with the higher maximum output, it is possible to avoid torque shortages in the electric motor with a smaller range of torque correction possible, while still performing the necessary F/B compensation.

In the above control method, the distribution feedback gains (k _fb1 , f _fb2 ) are set based on the ratio of the torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) to the total _torque ^amount T mall *. This makes it possible to set the distribution F/B gains (k _fb1 , f _fb2 ) in a simple manner.

In the above control method, the distribution feedback gains (k _fb1 , f _fb2 ) are set higher as the detection accuracy of the rotational speed detection value (motor rotational angular velocity detection value ω _m ) of the electric motors (first motor 8A, second motor 8B) increases.

When using the above method in combination with electric motors having angle detectors such as resolvers with different performance (or electric motors with and without angle detectors), the accuracy of F/B compensation based on the rotation speed (motor rotation angular velocity detection value ω _m ) can be improved by setting a larger F/B gain for the electric motor with higher detection accuracy of the rotation speed (motor rotation angular velocity detection value ω _m ).

In the above control method, the parameters (drive shaft torsional angular velocity estimate ω _d ^, motor rotational angular velocity estimate ω _m ^) further include a torsional angular velocity estimate of the drive shaft 10 (drive shaft torsional angular velocity estimate ω _d ^), the gains (k _all , k _fball ) further include a drive shaft torsional angular velocity gain k _all by which the torsional angular velocity estimate (drive shaft torsional angular velocity estimate ω _d ^) is multiplied, and when the total torque amount T _mall ^* is corrected so as to reduce torsional vibration of the drive shaft 10 based on the product of the drive shaft torsional angular velocity gain k _all and the torsional angular velocity estimate (drive shaft torsional angular velocity estimate ω _d ^), the distribution gain is a distribution drive shaft torsional angular velocity gain (first drive shaft torsional angular velocity gain k _ff1 , second drive shaft torsional angular velocity gain k _ff2 ) which is the sum of the drive shaft torsional angular velocity gain k _all ), and the torque command values (first torque command value T m1 *, second torque command value _{T m2} *) are individually corrected based on distribution correction amounts (first F/F torque command value T _mff1 ^* , second F/F torque command value T _mff2 ^* ) obtained by multiplying the torsional angular velocity estimated value (drive shaft torsional angular velocity estimated value _{ω d} ^) by distribution drive shaft torsional angular velocity gains (first drive _shaft torsional angular velocity gain ^k _ff1 , second drive shaft torsional angular velocity gain ^k _ff2 ).

The above method allows for accurate simulation of nonlinear responses, including dead zones such as gear backlash, and allows for appropriate F/F compensation.

In the above control method, the distributed feedforward gains (k _ff1 , k _ff2 ) are set based on the ratio of the torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) to the total _torque ^amount T mall *. This makes it possible to set the distributed drive shaft torsional angular velocity gains (k _ff1 , k _ff2 ) in a simple manner.

Furthermore, according to the control device for an electric vehicle of the present invention, the control device for an electric vehicle is provided with a plurality of electric motors (first motor 8A, second motor 8B) and a driving force transmission mechanism (reduction gear 9, drive shaft 10) that combines the power of the plurality of electric motors (first motor 8A, second motor 8B) on one drive shaft 10 and transmits the power to drive wheels 11A, 11B, and includes a vehicle controller 1 that sets torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) for the electric motors (first motor 8A, second motor 8B) based on vehicle information (vehicle speed V, accelerator opening θ), and a control unit that controls the torque command values (first torque command value T _m1 ^* , second torque command value T _m2 * ⁾ . The vehicle model (G p (s)) is a vehicle model that models the characteristics of the plurality of electric motors (first motor 8A, second motor 8B) assumed to be an inertial body that rotates integrally in a driving force transmission mechanism (reduction gear 9, drive shaft 10). The motor controllers (first motor controller 2A, second motor controller 2B) include a parameter calculation means (rotational motion parameter calculator 21) that inputs a total torque amount T mall *, which is the sum of torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^{* )} , to the vehicle model (G _p (s)), which models the characteristics of the plurality of electric motors (first motor 8A, second _motor 8B) assumed to be an inertial body that rotates integrally in a driving force transmission mechanism (reduction _gear 9, drive shaft 10), and calculates parameters (drive shaft torsion angular velocity estimated value ω _d ^, motor rotational angular velocity estimated value ω _m ^) that are correlated with the rotational motion of the vehicle model (G p (s) ₎ . ^* is corrected based on the correction amount (k _fall ω _d ^, k _fball (H(s)/G p (s)(ω _m ^-ω _m ))) obtained by multiplying the parameters (drive shaft torsion angular velocity estimated value ω _d ^, motor _rotation angular velocity estimated value ω _m ^) by a predetermined gain, distribution gains ((k _ff1 , k _ff2 ), (k _fb1 , k _fb2 )) which are the sum of the gains (k _all , k _fball ) are set individually to the electric motors (first motor 8A, second motor 8B), and the distribution gains ((k _ff1 , k _ff2 ), (k _fb1 , k _fb2 )) are set to the parameters (drive shaft torsion angular velocity estimated value ω _d ^, motor rotation angular velocity estimated value ω _m ^). and torque command value correction means (F/F compensator 22, F/B compensator 23) that individually corrects the torque command values (first torque command value T _m1 ^* , second torque command value T _m2 ^* ) based on distribution correction amounts (first F/F torque command value T _mff1 ^* , first F/B torque command value T _mfb1 ^* , second F/F torque command value T _mff2 ^* , second F/B torque command value T _mfb2 ^* ) obtained by multiplying the torque command values by the torque command values (first F/F torque command value T mff1 *, first F/B torque command value T mfb1 *, second F/F torque command value T mff2 *, second F/B torque command value T mfb2 *).

With the above configuration, even in a vehicle equipped with a driving force transmission mechanism (reduction gear 9, drive shaft 10) that combines the power of multiple electric motors (first motor 8A, second motor 8B) on the drive shaft 10, simulating an electric motor of a single inertial body that is rigidly connected and rotates generally as one unit enables vibration control using controllers (first motor controller 2A, second motor controller 2B) that individually control the electric motors (first motor 8A, second motor 8B), just as in the case of a single electric motor for the drive shaft 10. In this case, parameters correlated with the rotational motion of the driving force transmission system (e.g., motor rotation speed, motor angular acceleration, drive shaft torsional angular velocity, drive shaft torque, etc.) are calculated based on the total torque generated by the multiple electric motors, rather than the torque of each individual electric motor. This enables accurate parameter calculation, preventing the torque generated by each electric motor from acting as a disturbance, and effectively suppressing vibrations occurring in the vehicle (particularly torsional vibration of the drive shaft 10).

The above describes embodiments of the present invention, but these embodiments merely illustrate some of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments. Furthermore, the above embodiments can be combined as appropriate.

1 Vehicle controller, 2A First motor controller, 2B Second motor controller, 8A First motor, 8B Second motor, 9 Reducer, 10 Drive shaft, 11A Drive wheel, 11B Drive wheel, Rotational motion parameter calculator 21, 22 F/F compensator, 23 F/B compensator

Claims (8)

A control method for an electric vehicle including a driving force transmission mechanism that combines power from a plurality of electric motors on one drive shaft and transmits the combined power to drive wheels, setting a torque command value for the electric motor based on vehicle information, and controlling torque of the electric motor based on the torque command value,
a total torque amount, which is the sum of the torque command values, is input to a vehicle model that models the characteristics of the plurality of electric motors in the driving force transmission mechanism when they are assumed to be inertial bodies that rotate together, and a parameter correlated with the rotational motion of the vehicle model is calculated;
When the total torque amount is corrected based on a correction amount obtained by multiplying the parameter by a predetermined gain so that the rotational motion becomes a reference response,
A distribution gain, which is the sum of the gains, is set individually for each of the electric motors;
a control method for an electric vehicle, the torque command value being individually corrected based on a distribution correction amount obtained by multiplying the parameter by the distribution gain; the parameters include an estimated rotational angular velocity of the electric motor;
the gain includes a feedback gain that is multiplied by a difference between the rotational angular velocity estimation value and the rotational angular velocity detection value of the electric motor,
When the total torque amount is corrected based on the product of the feedback gain and the difference so as to reduce the difference,
the distributed gain includes a distributed feedback gain that is a sum of the feedback gains,
2. The method for controlling an electric vehicle according to claim 1, wherein the torque command values are individually corrected based on the distribution correction amount obtained by multiplying the difference by the distribution feedback gain. The control method for an electric vehicle described in claim 2, wherein the distribution feedback gain is set higher as the maximum output of the electric motor increases. The control method for an electric vehicle described in claim 2, wherein the distribution feedback gain is set based on the ratio of the torque command value to the total torque amount. The control method for an electric vehicle described in claim 2, wherein the distribution feedback gain is set higher as the detection accuracy of the rotational speed detection value of the electric motor increases. the parameters further include an estimated torsional angular velocity of the drive shaft;
the gain further includes a drive shaft torsional angular velocity gain by which the torsional angular velocity estimate value is multiplied;
In a case where the total torque amount is corrected based on the product of the drive shaft torsional angular velocity gain and the torsional angular velocity estimation value so as to reduce torsional vibration of the drive shaft,
the distribution gain includes a distribution drive shaft torsional angular velocity gain that is a sum of the drive shaft torsional angular velocity gains,
6. The method for controlling an electric vehicle according to claim 2, wherein the torque command values are individually corrected based on the distribution correction amount obtained by multiplying the torsional angular velocity estimate value by the distribution drive shaft torsional angular velocity gain. The control method for an electric vehicle described in claim 6, wherein the distributed drive shaft torsional angular velocity gain is set based on the ratio of the torque command value to the total torque amount. A control device for an electric vehicle including a plurality of electric motors and a driving force transmission mechanism that combines power of the plurality of electric motors on one drive shaft and transmits the power to a drive wheel,
a vehicle controller that sets a torque command value for the electric motor based on vehicle information;
a plurality of motor controllers that individually control the torque of the electric motors according to the torque command values;
The motor controller
a parameter calculation means for calculating a parameter correlated with rotational motion of a vehicle model obtained by inputting a total torque amount, which is the sum of the torque command values, into the vehicle model, which models the characteristics of the plurality of electric motors in the driving force transmission mechanism when they are assumed to be an inertial body rotating as a unit;
a torque command value correcting means for individually setting a distribution gain, which is a sum of the gains, for the electric motors, and individually correcting the torque command value based on a distribution correction amount obtained by multiplying the parameter by the distribution gain, when correcting the total torque amount based on a correction amount obtained by multiplying the parameter by a predetermined gain so that the rotational motion becomes a reference response.