JP7415772B2 - Electric vehicle control method and electric vehicle control device - Google Patents

Description

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

Patent Document 1 discloses, in an electric vehicle using an electric motor as a power source, torque input to the motor and motor rotation in order to prevent vibrations generated in a vehicle drive system connected between the electric motor and drive wheels. A control method for an electric vehicle is disclosed that controls a target torque of an electric motor using a transmission model with respect to speed.

More specifically, the electric vehicle control method disclosed in Patent Document 1 first calculates the drive torque target value Tm from the accelerator opening, vehicle speed, and the like. Then, the calculated drive torque target value Tm is passed through a damping filter having a characteristic of reducing a frequency component equal to the torsional vibration of the vehicle drive system, thereby obtaining the first torque target value Tm1. On the other hand, a second torque target value Tm2 is calculated based on the deviation between the estimated value of the motor rotation speed and the measured value of the motor rotation speed. Then, the final torque command value Tmf is calculated by adding the first torque target value Tm1 and the second torque target value Tm2. Thereafter, the current of the electric motor is controlled so that the torque of the electric motor matches the final torque command value Tmf.

That is, the electric vehicle control method disclosed in Patent Document 1 suppresses vibrations caused by resonance between the electric motor and the vehicle drive system extending from the electric motor to the drive wheels by controlling the torque of the electric motor. As a result, a vibration damping effect is achieved that suppresses rotational vibration of the electric vehicle that occurs when the accelerator is depressed from a stopped state or a decelerated state.

JP2003-009566A

When driving on a sandy road surface or a deep snowy road surface, an electric vehicle may experience greater vibration than when driving on a normal flat road surface. This is caused by the drive wheels sinking into the road surface, such as on sandy roads.

Specifically, on sandy roads and the like, running resistance increases due to the sinking characteristics of the drive wheels. In addition, due to the spring mass characteristics of the suspension and the sinking characteristics of the drive wheels, electric vehicles are subject to not only rotational vibrations in the longitudinal direction of the vehicle due to resonance points unique to the vehicle drive system, but also sustained vibrations in the vertical direction of the vehicle. Vibration occurs. For this reason, when the accelerator is depressed from a stopped or decelerated state on a road surface with high running resistance, such as a sandy road surface, vertical vibrations occur in the vehicle, giving the driver a sense of discomfort.

As described above, when the vehicle is traveling on a sandy road surface, vibrations that are larger than those caused by the characteristics of the vehicle drive system that occur when the vehicle is traveling on a flat road surface may occur. However, in the control method described in Patent Document 1, vibration damping control is performed using uniquely determined damping parameters in the sense that the running resistance of the road surface is not considered. Therefore, with the control method described in Patent Document 1, it is difficult to perform accurate and sufficient vibration damping control when traveling on a sandy road surface with high running resistance.

In view of the above circumstances, an object of the present invention is to provide a control method and a control device for an electric vehicle that can perform accurate and sufficient vibration damping control even when driving on a sandy road surface or the like.

According to one aspect of the present invention, a method for controlling an electric vehicle using an electric motor as a power source is provided. This electric vehicle control method includes a driving torque target value setting step of setting a driving torque target value based on vehicle information of the electric vehicle, and an estimated value of the rotational speed of the electric motor based on a model of the electric vehicle. The estimated motor rotational speed calculation step calculates the estimated motor rotational speed, and calculates the amount of disturbance applied to the electric vehicle based on the deviation between the motor rotational speed, which is the actual rotational speed of the electric motor, and the estimated motor rotational speed. An estimated disturbance calculation step that calculates an estimated disturbance that is an estimated value, and a final torque command value calculation that calculates a final torque command value that suppresses resonance of the vehicle drive system by correcting the drive torque target value based on the estimated disturbance . a running resistance detection step of detecting running resistance of a road surface on which an electric vehicle runs; and a correction amount changing step of changing the correction amount of the drive torque target value Tm based on the running resistance.

According to the present invention, accurate and sufficient vibration damping control can be performed even when driving on a sandy road surface or the like.

FIG. 1 is a block diagram showing the main configuration of an electric vehicle. FIG. 2 is a flowchart illustrating the main processes of the electric vehicle control method. FIG. 3 is a diagram showing an example of an accelerator opening-torque table. FIG. 4 is an explanatory diagram showing a dynamic model of an electric vehicle. FIG. 5 is a block diagram of disturbance torque estimation processing. FIG. 6 is an explanatory diagram showing a mechanical model of the gradient resistance estimated value calculation step. FIG. 7 is a block diagram of vibration damping control processing. FIG. 8 is an explanatory diagram of the frequency band changing step. FIG. 9 is an explanatory diagram of the gain changing step. FIG. 10 is a time chart showing changes over time in various parameters of the comparative example. FIG. 11 is a time chart showing changes over time in various parameters of the present invention. FIG. 12 is a block diagram showing the configuration of a modified example of an electric vehicle.

As shown in FIG. 1, the electric vehicle 100 includes a battery 1, a motor controller 2, an inverter 3, an electric motor 4, a reduction gear 5, a rotation sensor 6, a current sensor 7, a drive shaft 8, and a drive wheel 9a and a drive wheel. 9b etc. Note that an electric vehicle refers to a vehicle that uses an electric motor as a power source. A vehicle that can use an electric motor as part or all of its driving source or braking source is an electric vehicle. That is, electric vehicles include not only electric vehicles but also hybrid vehicles, fuel cell vehicles, and the like. The electric vehicle 100 of this embodiment is an electric vehicle that uses an electric motor 4 as a power source.

The battery 1 is connected to an electric motor 4 via an inverter 3, and supplies driving power to the electric motor 4 by discharging. Further, the battery 1 can be charged by receiving regenerative power from the electric motor 4.

The motor controller 2 is a computer that includes, for example, a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), an input/output interface (I/O interface), and the like. Motor controller 2 generates a control signal for directly or indirectly controlling electric motor 4 based on vehicle information of electric vehicle 100. Motor controller 2 comprehensively controls the operation of electric vehicle 100 by controlling electric motor 4 .

Vehicle information is information indicating the operating state or control state of the entire electric vehicle 100 or each part that constitutes the electric vehicle 100, and is a so-called vehicle variable. Vehicle information can be obtained by detection, measurement, generation, calculation, estimation, or the like. For example, the longitudinal acceleration of the electric vehicle 100 (hereinafter referred to as vehicle longitudinal acceleration) Ac [m/s ² ], the vehicle speed V [km/h], the accelerator opening θ [%], the rotor phase α of the electric motor 4 [rad], currents iu, iv, iw [A] of electric motor 4, DC voltage value Vdc [V] of battery 1, etc. are vehicle information of electric vehicle 100. The motor controller 2 controls the electric motor 4 using, for example, this vehicle information input as a digital signal.

The control signal for controlling the electric motor 4 is, for example, a PWM signal (Pulse Width Modulation signal) that controls the current of the electric motor 4. Further, the motor controller 2 generates a drive signal for the inverter 3 according to the PWM signal. The drive signal for the inverter 3 is also a control signal for controlling the electric motor 4.

The inverter 3 includes, for example, two switching elements (for example, power semiconductor elements such as IGBTs (Insulated Gate Bipolar Transistors) and MOS-FETs (metal-oxide-semiconductor field-effect transistors)) corresponding to each phase. . The inverter 3 converts the direct current supplied from the battery 1 into alternating current by turning on/off switching elements according to the drive signal generated by the motor controller 2, and adjusts the current supplied to the electric motor 4. . Further, the inverter 3 reversely converts the alternating current generated by the electric motor 4 due to the regenerative braking force into direct current, and adjusts the current supplied to the battery 1.

The electric motor 4 is, for example, a three-phase AC motor, and generates driving force using an AC current supplied from the inverter 3. The driving force generated by the electric motor 4 is transmitted via the reducer 5 and the drive shaft 8 to a pair of left and right drive wheels 9a and 9b. Furthermore, when the electric motor 4 rotates while being rotated by the drive wheels 9a and 9b, it generates regenerative braking force and recovers the kinetic energy of the electric vehicle 100 as electrical energy.

The speed reducer 5 is composed of, for example, a plurality of gears. The speed reducer 5 reduces the rotational speed of the electric motor 4 and transmits it to the drive shaft 8, thereby generating driving torque or braking torque (hereinafter simply referred to as torque) proportional to the reduction ratio.

Rotation sensor 6 detects rotor phase α of electric motor 4 and outputs it to motor controller 2 . The rotation sensor 6 is, for example, a resolver or an encoder.

Current sensor 7 detects the current flowing through electric motor 4 and outputs it to motor controller 2 . In this embodiment, the current sensor 7 detects three-phase alternating current currents iu, iv, and iw of the electric motor 4, respectively. Note that the current of two arbitrary phases may be detected using the current sensor 7, and the current of the remaining one phase may be calculated.

Note that the vehicle longitudinal acceleration Ac, which is one of the vehicle information, can be detected at any timing by using an acceleration sensor or other controller (not shown). The accelerator opening θ can be detected using an accelerator opening sensor or other controller (not shown). The same applies to other vehicle information such as the vehicle speed V and the DC voltage value Vdc of the battery 1, and these various vehicle information can be detected at any timing using a sensor or other controller (not shown).

As shown in FIG. 2, the motor controller 2 performs input processing (step S201), torque target value calculation processing (step S202), disturbance torque estimation processing (step S203), vibration damping control processing (step S204), current target value The calculation process (step S205) and the current control process (step S206) are executed in this order at every predetermined calculation cycle.

The input process in step S201 is a process in which the motor controller 2 receives input of vehicle information, and calculates various parameters to be used in the processes after step S202, as necessary. In this embodiment, the motor controller 2 receives vehicle longitudinal acceleration Ac, accelerator opening θ, rotor phase α, electric motor 4 currents iu, iv, iw, and battery 1 DC voltage value Vdc from various sensors. get.

In addition, the motor controller 2 uses some or all of the directly acquired vehicle information to determine the motor electrical angular velocity ωe [rad/s], which is the electrical angular velocity of the electric motor 4, and the mechanical angular velocity of the electric motor 4. The motor rotational speed ωm [rad/s], the unit-converted motor rotational speed Nm [rpm] of the electric motor 4, the angular velocity ωw [rad/s] of the drive wheels 9a and 9b, and the vehicle speed V, etc. calculate.

Specifically, the motor controller 2 calculates the motor electrical angular velocity ωe by time-differentiating the rotor phase α. Thereafter, the motor controller 2 calculates the motor rotational speed ωm by dividing the motor electrical angular velocity ωe by the number of pole pairs of the electric motor 4. Further, the motor controller 2 calculates the motor rotation speed Nm by multiplying the motor rotation speed ωm by a unit conversion coefficient (60/2π).

Further, the motor controller 2 calculates the angular velocity ωw of the drive wheels 9a, 9b by dividing the motor rotational speed ωm or the motor rotational speed Nm by the gear ratio of the final gear of the reduction gear 5. Then, the motor controller 2 calculates the vehicle speed V by multiplying the angular velocity ωw by the load radius r [m] of the driving wheels 9a, 9b, and then by the unit conversion coefficient (3600/1000).

Note that instead of being calculated as described above, the vehicle speed V may be directly obtained by communicating with another controller such as a meter or a brake controller. Moreover, when a plurality of drive wheels 9a and 9b are each provided with a wheel speed sensor, the average value of each wheel speed sensor value can be used as the vehicle speed V. In addition, the vehicle speed V is an estimated vehicle speed value calculated using a value obtained from a sensor such as a GPS (Global Positioning System), a value selected from a wheel speed sensor, etc., or a longitudinal acceleration sensor, etc. (Refer to Japanese Unexamined Patent Publication No. 2002-127881, etc.) may be used.

The torque target value setting process in step S202 is a process for setting the drive torque target value Tm, which is the target value of the drive torque. The motor controller 2 sets the drive torque target value Tm based on the accelerator opening θ and the motor rotational speed ωm by referring to the accelerator opening-torque table shown in FIG. 3, for example. That is, step S202 is a drive torque target value setting step that sets the drive torque target value Tm based on the vehicle information of the electric vehicle 100.

The disturbance torque estimation process in step S203 is a process for calculating the estimated disturbance torque value Td^. The disturbance torque is the amount of increase or decrease in torque that is increased or decreased due to disturbance to electric vehicle 100. The estimated disturbance torque value Td^ is an estimated value representing the presence or absence of disturbance torque and its amount. A disturbance to electric vehicle 100 is an external factor that increases or decreases the running resistance of electric vehicle 100. Specifically, the disturbances to the electric vehicle 100 include air resistance, modeling errors due to variations in vehicle weight depending on the number of passengers and loading capacity, rolling resistance of the drive wheels 9a and 9b, gradient resistance of the road surface, and the drive wheel 9a. , 9b sinking into the road surface. The motor controller 2 calculates the estimated disturbance torque value Td^ using the motor rotational speed ωm and the previous value of the final torque command value Tmf calculated in step S204. Details of the disturbance torque estimation process will be described later.

The vibration damping control process in step S204 is a process for calculating a final torque command value Tmf that suppresses resonance in the vehicle drive system. The motor controller 2 uses the drive torque target value Tm and the motor rotation speed ωm to calculate the final torque command value Tmf without sacrificing the torque response of the drive shaft 8. Resonance in the vehicle drive system typically refers to rotational vibration in the longitudinal direction of electric vehicle 100 (hereinafter referred to as longitudinal vibration) caused by torsional vibration of drive shaft 8 and the like. In addition, resonance of the vehicle drive system includes vertical vibration of electric vehicle 100 (hereinafter referred to as vertical vibration) caused by the driving wheels 9a, 9b sinking into the road surface. Drive control of electric vehicle 100 according to final torque command value Tmf of this embodiment suppresses both longitudinal vibration and vertical vibration of electric vehicle 100. Details of the damping control process will be described later.

The current target value calculation process in step S205 calculates a d-axis current target value id*, which is the target value of the d-axis current id of the electric motor 4, and a q-axis current target value iq ^*, which is the target value of the q-axis current ^iq . It is processing. The motor controller 2 stores a ^dq -axis current target value table ⁽ (not shown) is held in advance. Therefore, by referring to this dq-axis current target value table, the motor controller 2 determines the final torque command value Tmf, motor rotational speed ωm, and d-axis current target value id ^* and q-axis corresponding to the DC voltage value Vdc. Calculate the current target value iq ^* .

The current control process in step S206 is a process of generating torque for driving or braking electric vehicle 100 by controlling the current of electric motor 4. In the current control process, the motor controller 2 first calculates the d-axis current id and the q-axis current iq based on the currents iu, iv, and iw of the electric motor 4 and the rotor phase α. Next, the motor controller 2 sets d-axis voltage command values vd and q based on the deviation between the d-axis current target value id* and the q-axis current target value iq*, and the d-axis current id and q-axis current iq. Calculate the shaft voltage command value vq. Further, the motor controller 2 calculates three-phase voltage command values vu, vv, and vw based on the d-axis voltage command value vd, the q-axis voltage command value vq, and the rotor phase α. Then, the motor controller 2 calculates the duty ratios tu, tv, tw [%] of the PWM signals input to each phase based on the three-phase voltage command values vu, vv, vw and the DC voltage value Vdc.

The motor controller 2 controls the electric motor 4 by opening and closing the switching elements of the inverter 3 according to the PWM signal obtained in this way. As a result, motor controller 2 drives or brakes electric vehicle 100 with the desired torque specified by final torque command value Tmf.

<Disturbance torque estimation process>
The motor controller 2 uses the transfer characteristic Gp(s) from the drive torque target value Tm to the motor rotational speed ωm in the disturbance torque estimation process. Transfer characteristic Gp(s) is calculated using an equation of motion derived from the mechanical model of electric vehicle 100 shown in FIG. 4. Each symbol in the dynamic model and/or equation of motion of electric vehicle 100 in FIG. 4 is as follows.

Jm: Inertia of electric motor Jw: Inertia of driving wheels M: Mass of vehicle KD: Torsional rigidity of vehicle drive system Kt: Coefficient of friction between driving wheels and road surface N: Overall gear ratio r: Load radius of driving wheels ωm: Motor Rotational speed Tm: Drive torque target value TD: Drive wheel torque F: Force applied to the vehicle V: Vehicle speed (vehicle speed)
ωw: Angular velocity of driving wheels

From the mechanical model of electric vehicle 100 shown in FIG. 4, the following equation of motion can be derived. Note that the symbol "*" in equations (1) to (3) represents time differentiation.

When the transfer characteristic Gp(s) is determined based on the above equations of motion (1) to (5), it can be expressed by equation (6). Furthermore, the coefficients a ₁ to a ₄ and the coefficients b ₀ to b ₃ in equation (6) are expressed by equations (7) to (14).

When examining the poles and zeros of the transfer characteristic Gp(s) shown in the above equation (6), one pole and one zero point show extremely close values. This means that α and β in Equation (15) below have extremely close values.

Therefore, by performing pole-zero cancellation that approximates α=β in equation (15), the transfer characteristic Gp(s) in the (quadratic)/(cubic) form is obtained as shown in equation (16) below. be able to.

Furthermore, the natural vibration angular velocity ωp can be expressed by the following equation (17) using the coefficient a ₁ ' and the coefficient a ₃ ' used in the denominator of equation (16). Further, the natural vibration angular velocity ωp can be converted into a resonance frequency (natural vibration frequency) fp using the following equation (18).

In this embodiment, the motor controller 2 uses the transfer characteristic Gp(s) of equation (16) in the disturbance torque estimation process, but instead of the transfer characteristic Gp(s) of equation (16), the motor controller 2 uses the following equation ( 19) may be used. The equivalent mass Mv used in the transmission characteristic Gp(s) of equation (19) is determined from the mass M of the vehicle, the inertia Jm of the electric motor 4, and the inertia Jw of the drive wheels 9a and 9b, as shown in equation (20). . Further, the coefficient K _M used in the transfer characteristic Gp(s) of equation (19) is expressed by equation (21).

As shown in FIG. 5, the disturbance torque estimation process includes a first motor torque estimation value calculation step, a second motor torque estimation value calculation step, a torque estimation deviation calculation step, a slope resistance estimation value calculation step, and a gradient resistance subtraction step.

In the first motor torque estimated value calculation step, the first motor torque estimated value T1^ is calculated by filtering the motor rotation speed ωm using the first motor torque estimation filter 501. The first motor torque estimation filter 501 is configured using a low-pass filter H1(s) expressed by the following equation (22) using a time constant τv and a transfer characteristic Gp(s), and H1( s)/Gp(s).

In the second motor torque estimated value calculation step, the second motor torque estimated value T2^ is calculated by filtering the final torque command value Tmf using the second motor torque estimation filter 502. The second motor torque estimation filter 502 is, for example, the same low-pass filter H1(s) that constitutes the first motor torque estimation filter 501.

In the torque estimation deviation calculation step, the torque estimation deviation Tδ^ is calculated by subtracting the first motor torque estimation value T1^ from the second motor torque estimation value T2^ using the subtractor 503.

In the slope resistance estimated value calculation step, the slope resistance estimated value Tg^ is calculated using the multiplier 504 using the vehicle longitudinal acceleration Ac and the coefficient Kg. Gradient resistance is running resistance caused by the slope of the road surface. As shown in FIG. 6, by multiplying the slope ψ by a coefficient Kg expressed by the following equation (23), the estimated slope resistance value Tg^ required for the electric vehicle 100 to balance on the slope can be calculated. . The symbol "r" is the load radius of the drive wheels 9a, 9b, the symbol "M" is the mass of the vehicle, the symbol "g" is the gravitational acceleration, and the symbol "N" is the overall gear ratio.

In the gradient resistance subtraction step, the estimated disturbance torque value Td^ is calculated by subtracting the estimated gradient resistance value Tg^ from the estimated torque deviation Tδ^ using the subtracter 505.

As can be seen from the above calculation method, the disturbance torque estimated value Td^ is calculated by subtracting the gradient resistance estimated value Tg^ from the torque estimation deviation Tδ^, which comprehensively estimates the disturbance torque caused by all disturbances. reducing or eliminating the impact of Therefore, although the disturbance factors differ depending on the driving conditions, the disturbance torque estimation process can collectively estimate other disturbance torques other than gradient resistance. Other disturbance torques other than this gradient resistance include an increase in running resistance (hereinafter referred to as road running resistance) due to the driving wheels 9a, 9b sinking into the road surface. Furthermore, among a plurality of disturbance factors, the gradient of the road surface and the sinking of the drive wheels 9a, 9b into the road surface are factors that cause particularly large fluctuations in running resistance. However, the disturbance torque estimate Td^ reduces or excludes the contribution of slope resistance. Therefore, when driving on a road surface where the drive wheels 9a, 9b sink into the road surface, such as a sandy road surface, a deep snowy road surface, or a muddy road surface (hereinafter referred to as a soft road surface), the estimated disturbance torque value Td^ is substantially This represents the fact that the road surface is soft, the driving wheels 9a and 9b sink into the soft road surface, and the increase in running resistance of the road surface due to the soft road surface. On the other hand, when driving on a normal road surface (hereinafter referred to as a hard road surface) in which the drive wheels 9a and 9b do not substantially sink into the road surface, no large fluctuation occurs in the estimated disturbance torque value Td^. This fact essentially indicates that the road surface is a hard road surface, that the drive wheels 9a and 9b do not sink, and that there is no increase in the running resistance of the road surface. That is, global fluctuations in the estimated disturbance torque value Td^ represent fluctuations in the running resistance of the road surface. Therefore, the above-mentioned disturbance torque estimation process (step S203) functions as a running resistance detection step that detects the running resistance of the road surface on which the electric vehicle 100 runs by calculating the estimated disturbance torque value Td^.

<Vibration control processing>
As shown in FIG. 7, the damping control process includes a final torque command that calculates a final torque command value Tmf that suppresses resonance of the vehicle drive system by correcting the drive torque target value Tm based on the motor rotational speed ωm. Includes a value calculation step. In this embodiment, the final torque command value calculation step includes a first torque target value calculation step, a second torque target value calculation step, and an addition step.

In the first torque target value calculation step, the first torque target value Tm1 is calculated using the damping filter 701. The damping filter 701 is a feedforward compensator, and outputs a first torque target value Tm1 by filtering the drive torque target value Tm. The damping filter 701 is configured using an ideal transfer characteristic Gm(s) and a transfer characteristic Gp(s), and has a characteristic of Gm(s)/Gp(s). The ideal transmission characteristic Gm(s) is an ideal model of the transmission characteristic from the drive torque target value Tm to the motor rotational speed ωm. On the other hand, the transmission characteristic Gp(s) is a practical transmission characteristic model that takes into account torsional vibration of the vehicle drive system, as described above. Therefore, the first torque target value Tm1 is a torque target value that reduces the factors that cause torsional vibrations in the vehicle drive system.

In the second torque target value calculation step, the second torque target value Tm2 is calculated using the disturbance suppression filter 702. The disturbance suppression filter 702 is a feedback compensator, receives the previous value of the final torque command value Tmf, the motor rotation speed ωm, and the estimated disturbance torque value Td^, and outputs the second torque target value Tm2.

More specifically, the second torque target value calculation step includes a motor rotational speed estimated value calculation step, a deviation calculation step, an estimated disturbance calculation step, and a gain multiplication step.

In the motor rotational speed estimated value calculation step, the motor rotational speed estimated value ωm^ is calculated from the previous value of the final torque command value Tmf using the vehicle model 711 of the electric vehicle 100. Vehicle model 711 is a transfer characteristic Gp(s) that is a realistic model of electric vehicle 100.

In the deviation calculation step, the subtractor 712 is used to calculate the deviation Δ between the motor rotational speed ωm and the motor rotational speed estimated value ωm^. In this embodiment, the subtracter 712 calculates the deviation Δ by subtracting the motor rotation speed ωm from the motor rotation speed estimated value ωm^.

In the estimated disturbance calculation step, the estimated disturbance d^ is calculated by extracting a disturbance component that is a component in a predetermined frequency band from the deviation Δ using the disturbance estimation filter 713. The disturbance estimation filter 713 is configured using a transfer characteristic H2(s) and a transfer characteristic Gp(s), and has a characteristic of H2(s)/Gp(s).

The transfer characteristic H2(s) is set such that the difference between the denominator order and the numerator order is greater than or equal to the difference between the denominator order and the numerator order of the transfer characteristic Gp(s). Furthermore, the transfer characteristic H2(s) can be configured by, for example, a bandpass filter. When the transfer characteristic H2(s) is a bandpass filter, the disturbance suppression filter 702 becomes a feedback element that selectively reduces the resonance of the vehicle drive system that occurs in the electric vehicle 100.

As shown in FIG. 8(A), the transmission characteristic H2(s) is such that, in principle, the damping characteristics on the low-pass side and the high-pass side are almost the same, and the resonance frequency fp of the torsional vibration of the vehicle drive system is on the logarithmic axis ( (log scale) so that it is near the median of the passband. In this case, the disturbance suppression filter 702 has the greatest vibration suppression effect on resonance of torsional vibration of the vehicle drive system.

Note that the attenuation characteristic on the low-pass side refers to the manner in which the pass characteristic changes in the vicinity of the cutoff frequency f _LC on the low frequency side of the transfer characteristic H2(s). Furthermore, the attenuation characteristic on the high-pass side refers to the manner in which the pass characteristic changes in the vicinity of the cutoff frequency f _HC on the high-frequency side of the transfer characteristic H2(s). Regarding the attenuation characteristics on the low-pass side and the high-pass side, "matching" means that the attenuation characteristics on the low-pass side and the attenuation characteristics on the high-pass side are symmetrical with respect to the median value on the logarithmic axis of the cutoff frequency fLC and the cutoff frequency fHC. It means that. Furthermore, “near” means that a specific value variation can be tolerated to the extent that there is no substantial difference in actual driving of electric vehicle 100.

When the transfer characteristic H2(s) is composed of a first-order high-pass filter and a first-order low-pass filter, the transfer characteristic H2(s) is calculated using the time constant τ _H of the high-pass filter and the time constant τ _L of the low-pass filter. The configuration can be expressed by the following equation (24). In addition, the time constants τ _H , τ _L and the cutoff frequencies f _HC , f _LC are determined using the resonance frequency fp of longitudinal vibration caused by torsional vibration of the vehicle drive system, a predetermined frequency fq, and a predetermined coefficient k. , is expressed by the following formulas (25) to (27). The resonance frequency fp of the longitudinal vibration is a constant determined in advance by the vehicle model of the electric vehicle 100, as described above. Further, the coefficient k is set by experiment, simulation, or the like depending on the specific electric vehicle 100 so that the disturbance suppression filter 702 functions.

Furthermore, in the transfer characteristic H2(s), the frequency fq (see equation (28)) that determines the cutoff frequency _fLC of the low-pass filter is a variable that can be switched according to the estimated disturbance torque value Td^. If the estimated disturbance torque value Td^ is less than the predetermined value C0, set fq=fp. In this case, the center frequency of each cutoff frequency f _HC , f _LC of the high-pass filter and the low-pass filter becomes the resonance frequency fp of longitudinal vibration caused by torsional vibration of the vehicle drive system. On the other hand, when the estimated disturbance torque value Td^ is greater than or equal to the predetermined value C0, the frequency fq, which is a variable, is calculated using the resonant frequency fp1 of the vertical vibration caused by the sinking of the drive wheels 9a and 9b into the road surface, and then fq= Set to fp1. As a result, as shown in FIG. 8(B), the cutoff frequency _fLC on the low frequency side of the transfer characteristic H2(s) is switched to the low frequency side, and the passband is made to include the resonant frequency fp1 of the vertical vibration. Become.

The resonant frequency fp1 of the vertical vibration depends on the type of soft road surface on which the electric vehicle 100 runs, such as a sandy road surface, a deep snow road surface, a muddy road surface, etc., and/or the actual driving wheel 9a, It may vary depending on the degree of subduction of 9b. However, the resonant frequency fp1 of the vertical vibration is a low frequency at least compared to the resonant frequency fp of the longitudinal vibration, regardless of the specific type of soft road surface or the actual degree of sinking of the drive wheels 9a, 9b. . Further, compared to the gap between the resonance frequency fp of the longitudinal vibration and the resonance frequency fp1 of the vertical vibration, the variation in the resonance frequency fp1 of the vertical vibration due to the type of soft road surface and the degree of subsidence of the drive wheels 9a, 9b is small. Therefore, the resonant frequency fp1 of the vertical vibration can be set in advance based on the vehicle model of the electric vehicle 100.

The predetermined value C0 used for switching the cutoff frequency _fLC on the low frequency side is a standard for determining whether the estimated disturbance torque value Td^ includes a component due to vertical vibration in addition to a component due to longitudinal vibration. Furthermore, when vertical vibration occurs due to running resistance on the road surface, the estimated disturbance torque value Td^ increases significantly. Therefore, the predetermined value C0 can be arbitrarily set in advance for the estimated disturbance torque value Td^ within a range in which it is possible to determine whether or not vertical vibration occurs. By setting the predetermined value C0 in this way, when the estimated disturbance torque value Td^ is greater than or equal to the predetermined value C0, the motor controller 2 determines that the road surface on which the electric vehicle 100 is running is a soft road surface with high running resistance. I can judge. Then, the motor controller 2 changes the frequency fq, which is a variable, to a low frequency that takes vertical vibration into consideration.

In the gain multiplication step, the second torque target value Tm2 is calculated by using the multiplier 714 to multiply the estimated disturbance d^ calculated as described above by the feedback gain Kfb. As shown in FIG. 9, the feedback gain Kfb is variable according to the estimated disturbance torque value Td^. Further, the feedback gain Kfb is set to be large according to the estimated disturbance torque value Td^. For example, the feedback gain Kfb is set to increase monotonically according to the estimated disturbance torque value Td^.

In the addition step, the final torque command value Tmf is calculated by using the adder 703 to add the first torque target value Tm1 and the second torque target value Tm2. That is, in the final torque command value calculation step, feedforward correction is performed to correct the drive torque target value Tm based on the deviation from the ideal transfer characteristic Gm(s), and the value of the drive torque target value Tm is corrected based on the estimated disturbance d^. The final torque command value Tmf is calculated by performing two correction processes: feedback correction and correction. Note that since the estimated disturbance d^ is calculated based on the motor rotational speed ωm, it can be said that the final torque command value Tmf is calculated based on the motor rotational speed ωm by the feedback correction described above.

Motor controller 2 drives electric vehicle 100 according to this final torque command value Tmf. However, when the electric vehicle 100 actually travels, a disturbance is applied to the electric vehicle 100, so there is a difference between the motor rotation speed ωm of the electric vehicle 100 and the estimated motor rotation speed ωm^ due to the disturbance. Therefore, since the deviation Δ is not always zero, the feedback correction using the disturbance suppression filter 702 exerts a damping effect until the deviation Δ becomes substantially zero.

In addition to the vibration damping control process described above, the method for controlling electric vehicle 100 of the present invention further includes a correction amount changing step. In the correction amount changing step, the correction amount of the drive torque target value Tm is changed based on the running resistance of the road surface. As mentioned above, there are two types of correction for the drive torque target value Tm: feedforward correction using the damping filter 701 and feedback correction using the disturbance suppression filter 702. The correction amount of feedback correction using the disturbance suppression filter 702 is changed. Further, since the estimated disturbance torque value Td^ represents the running resistance of the road surface, in the correction amount changing step, the estimated disturbance torque value Td^ is used to change the correction amount based on the running resistance of the road surface.

The correction amount changing step can be comprised of a gain changing step and/or a frequency band changing step. In this embodiment, the correction amount changing step includes both a gain changing step and a frequency band changing step.

In the gain changing step, the value of the feedback gain Kfb by which the estimated disturbance d^ is multiplied is changed based on the estimated disturbance torque value Td^ representing the running resistance of the road surface. Furthermore, in the gain changing step, the larger the estimated disturbance torque value Td^ is, the larger the value of the feedback gain Kfb is set.

In the frequency band changing step, a predetermined frequency band, which is a passband of the disturbance estimation filter 713, is changed based on the estimated disturbance torque value Td^ representing running resistance on the road surface. The predetermined frequency band is changed by changing the lower limit value of the predetermined frequency band, that is, the cutoff frequency f _LC of the disturbance estimation filter 713. Further, this change in the predetermined frequency band is to expand the predetermined frequency band by changing the cutoff frequency f _LC to a lower frequency side.

As described above, one aspect of the control method for the electric vehicle 100 is a control method for the electric vehicle 100 using the electric motor 4 as a power source, in which the drive torque target value Tm is determined based on the vehicle information of the electric vehicle 100. By correcting the drive torque target value Tm based on the drive torque target value setting step and the motor rotation speed ωm, which is the rotation speed of the electric motor 4, the final torque command value Tmf that suppresses the resonance of the vehicle drive system is determined. a final torque command value calculation step, a running resistance detection step of detecting the running resistance of the road surface on which the electric vehicle 100 runs, and a correction amount changing step of changing the correction amount of the driving torque target value Tm based on the running resistance. ,including. Therefore, even when the running resistance changes due to the sinking of the drive wheels 9a and 9b into the soft road surface, both the frequency component of the resonance point of longitudinal vibration and the frequency component of the resonance point of vertical vibration are adjusted to the drive torque target value Tm. It is possible to set a final torque command value Tmf that is reduced or removed from. Therefore, even when the accelerator is depressed from a stopped state or a decelerated state on a soft road surface such as a sandy road surface, an accurate and sufficient vibration damping effect can be obtained for both longitudinal vibration and vertical vibration. As a result, electric vehicle 100 can be smoothly accelerated even on a sandy road surface or the like.

For example, the electric vehicle of the comparative example calculates the final torque command value Tmf that suppresses torsional vibration of the vehicle drive system by feedback correcting the drive torque target value Tm based on the motor rotation speed ωm. However, the electric vehicle of the comparative example does not perform the running resistance detection step and the correction amount changing step. In this case, as shown in FIG. 10(A), when the electric vehicle of the comparative example is stopped from time t0 to time t1 and starts by depressing the accelerator a certain amount at time t1, Accordingly, the drive torque target value Tm becomes a constant value. Further, as the drive wheels 9a and 9b begin to rotate, they sink into the sandy road surface, so the running resistance in the forward direction increases, and the estimated disturbance torque value Td^ becomes a constant positive value. Furthermore, although the estimated disturbance torque value Td^ is a larger value than in the case of a hard road surface due to an increase in running resistance on the road surface, the electric vehicle of the comparative example does not detect this and performs constant control. On the other hand, in the electric vehicle of the comparative example, the drive torque target value Tm is corrected by feedback control based on the motor rotation speed ωm. A certain amount of feedback correction is performed to suppress longitudinal vibration caused by torsional vibration of the drive system, and thereafter, the final torque command value Tmf changes oscillatingly.

As a result, as shown in FIG. 10(B), the motor rotational speed Nm also changes oscillatingly, so that the sinking of the drive wheel also changes oscillatingly, as shown in FIG. 10(C). As a result, as shown in FIG. 10(D), vertical vibration occurs in the electric vehicle of the comparative example due to sinking of the drive wheels. Further, this vertical vibration is hardly attenuated by the vibration damping control process performed in the electric vehicle of the comparative example, and continues for a long time.

On the other hand, in electric vehicle 100, in addition to feedback correction control of drive torque target value Tm based on motor rotational speed ωm, a running resistance detection step and a correction amount changing step are performed. When this electric vehicle 100 is stopped on a sandy road surface from time t0 to time t1 and starts by depressing the accelerator a certain amount at time t1, as shown in FIG. Accordingly, the drive torque target value Tm becomes a constant value, as in the electric vehicle of the comparative example. Also, as the drive wheels 9a and 9b begin to rotate, they sink into the sandy road surface, so the running resistance in the forward direction increases, and the estimated disturbance torque value Td^ becomes a constant positive value. This is similar to the example electric vehicle.

However, in the electric vehicle 100, running resistance on the road surface is detected by the estimated disturbance torque value Td^. Then, by comparing the estimated disturbance torque value Td^ with a predetermined value C0, if it is determined that the vehicle is traveling on a soft road surface with high running resistance, the feedback gain Kfb is changed according to the estimated disturbance torque value Td^. . In addition, the predetermined frequency band of the disturbance estimation filter 713 is expanded according to the estimated disturbance torque value Td^, and in addition to the resonant frequency fp of the longitudinal vibration, the resonant frequency fp1 of the vertical vibration is extracted and used as the final torque command value Tmf. Feedback will be given.

As a result, the final torque command value Tmf changes in an oscillatory manner from the time t1 when the accelerator pedal depression starts until around the very early time t2. However, the vibration of the final torque command value Tmf attenuates early and converges to a constant value from time t3 to time t5 and thereafter. As a result, as shown in FIG. 11(B), the vibration of the motor rotation speed Nm is also suppressed, so that the amount of depression of the drive wheels 9a, 9b also converges as shown in FIG. 11(C). Therefore, in the electric vehicle 100, longitudinal vibration due to torsional vibration of the vehicle drive system is suppressed as in the electric vehicle of the comparative example, and as shown in FIG. 11(D), compared to the electric vehicle of the comparative example, Almost no vertical vibration occurs in electric vehicle 100 due to sinking of drive wheels 9a and 9b. In addition, even slight vertical vibrations that occur in electric vehicle 100 are quickly attenuated. Further, in the electric vehicle 100, the amount of depression of the drive wheels 9a and 9b is smaller than that of the electric vehicle of the comparative example. Furthermore, in the electric vehicle 100, the amount of depression of the drive wheels 9a, 9b stabilizes to a substantially constant value at an early stage. Therefore, the electric vehicle 100 can minimize the increase in running resistance even on a sandy road surface, and can obtain higher acceleration faster than the electric vehicle of the comparative example.

In addition to the above, one aspect of the control method for the electric vehicle 100 is to calculate a motor rotational speed estimated value ωm^ which is an estimated value of the rotational speed of the electric motor 4 based on a transfer characteristic Gp(s) which is a model of the electric vehicle 100. The disturbance applied to the electric vehicle 100 is calculated based on the deviation Δ between the motor rotation speed estimate ωm, which is the actual rotation speed of the electric motor 4, and the motor rotation speed estimate ωm^. In the final torque command value calculation step, the driving torque target value Tm is corrected based on the estimated disturbance d^. This estimated disturbance d^ includes not only the resonant frequency fp of the longitudinal vibration of the vehicle drive system but also the resonant frequency fp1 of the vertical vibration when vertical vibration occurs. Therefore, according to the above control method, a more specific and particularly effective control logic that can accurately and sufficiently suppress not only the longitudinal vibration of electric vehicle 100 but also the vertical vibration is realized.

Further, one aspect of the control method for electric vehicle 100 includes a gain multiplication step of multiplying the estimated disturbance d^ by feedback gain Kfb, and the correction amount changing step includes a gain multiplication step of changing the value of feedback gain Kfb based on running resistance. Contains modification steps. As a result, it is easy to obtain a particularly accurate and sufficient vibration damping effect on longitudinal vibrations and vertical vibrations of electric vehicle 100.

In particular, one aspect of the control method for electric vehicle 100 is to set the value of feedback gain Kfb to be larger as the running resistance is larger in the gain changing step. Thereby, electric vehicle 100 can exhibit sound vibration suppression performance according to the running resistance of the road surface. For example, it is also possible to uniformly multiply the feedback gain Kfb, which is larger than that of the conventional electric vehicle, regardless of the estimated disturbance torque value Td^. However, if the feedback gain Kfb is set large over the entire range regardless of the estimated disturbance torque value Td^, etc., sound vibrations containing high frequency noise in the audible range may occur depending on the vehicle speed V, etc., which may give a sense of discomfort to the driver. Therefore, as described above, a driving scene in which vertical vibration occurs and driving resistance is large is determined. By switching not to use the high feedback gain Kfb for soft roads at least on hard roads, it is possible to obtain accurate and sufficient damping effects against vertical vibrations while maintaining appropriate sound and vibration suppression performance.

One aspect of the control method for the electric vehicle 100 is that in the estimated disturbance calculation step, the estimated disturbance d^ is calculated by extracting components in a predetermined frequency band from the motor rotational speed ωm, the estimated motor rotational speed value ωm^, and the deviation Δ. The correction amount changing step includes a frequency band changing step of changing a predetermined frequency band based on running resistance. As a result, when vertical vibration occurs, the estimated disturbance d^ surely includes a component of the resonant frequency fp1 of the vertical vibration, so that it is particularly easy to obtain an accurate and sufficient damping effect against the vertical vibration.

Further, in one aspect of the control method for electric vehicle 100, in the frequency band changing step, the lower limit value of the predetermined frequency band is changed based on running resistance. Normally, the resonant frequency fp1 of the vertical vibration is a lower frequency than the resonant frequency fp of the longitudinal vibration, so it is not the cutoff frequency _fHC on the high frequency side, which is the upper limit value of the predetermined frequency band, but the lower limit value. It is sufficient to adjust the cutoff frequency _fLC on the low frequency side. Therefore, by changing the lower limit value of the predetermined frequency band, an accurate and sufficient upper limit vibration suppression effect can be obtained with a slight adjustment.

Further, in one aspect of the control method for electric vehicle 100, in the frequency band changing step, the predetermined frequency band is expanded by changing the lower limit value of the predetermined frequency band to a lower frequency side based on running resistance. In this way, by changing the lower limit value of the predetermined frequency band to the lower frequency side and increasing the sensitivity (gain) on the low frequency side of the disturbance estimation filter 713, it is possible to generate vertical vibration resonance with the estimated disturbance d^ with the minimum necessary adjustment. Frequency fp1 can be included. As a result, a particularly accurate and sufficient damping effect on vertical vibration can be obtained.

One aspect of the control method for electric vehicle 100 is to use a first motor torque estimated value T1^ estimated using motor rotational speed ωm and a final torque command value Tmf input to electric vehicle 100 in the running resistance detection step. The disturbance torque estimated value Td^ representing running resistance is calculated based on the torque estimated deviation Tδ^ which is the deviation between the second motor torque estimated value T2^ estimated by the above-described method. Thereby, it is possible to accurately estimate running resistance on sandy ground, deep snow, mud, etc. Since the correction amount of the feedback correction of the drive torque target value Tm is changed according to the estimated disturbance torque value Td^, a particularly accurate and sufficient effect of suppressing vertical vibration can be obtained.

Further, in one aspect of the control method for the electric vehicle 100, the running resistance detection step includes calculating an estimated slope resistance value Tg^ based on a vehicle longitudinal acceleration Ac that is an acceleration of the electric vehicle 100 in the longitudinal direction. step, and a gradient resistance that calculates running resistance by subtracting the estimated gradient resistance value Tg^ from the estimated torque deviation Tδ^, which is the deviation between the first estimated motor torque value T1^ and the second estimated motor torque value T2^. a subtraction step. That is, in calculating the estimated disturbance torque value Td^ representing running resistance, the estimated gradient resistance value Tg^ is subtracted. Thereby, the running resistance of the road surface is evaluated by reducing or eliminating the influence of the slope of the road surface. Therefore, since the correction amount of the feedback correction of the drive torque target value Tm is changed according to the estimated disturbance torque value Td^ which reduces the influence of gradient resistance, etc., it is possible to obtain a particularly accurate and sufficient vertical vibration suppression effect. can.

The above embodiment is a control device for an electric vehicle 100 using an electric motor 4 as a power source, and includes a drive torque target value setting section that sets a drive torque target value Tm based on vehicle information of the electric vehicle 100; A final torque command value calculation unit that calculates a final torque command value Tmf that suppresses resonance of the vehicle drive system by correcting the drive torque target value Tm based on the motor rotational speed ωm that is the rotational speed of the motor 4; The control device for an electric vehicle includes a running resistance detection section that detects running resistance of a road surface on which vehicle 100 runs, and a correction amount changing section that changes the correction amount of drive torque target value Tm based on the running resistance. The motor controller 2 functions as a drive torque target value setting section, a running resistance detection section, a final torque command value calculation section, and a correction amount changing section.

Note that the method for controlling the electric vehicle 100 of the above embodiment may be performed by changing or omitting some configurations or adding new configurations within the range in which an accurate and sufficient vibration damping effect on sandy road surfaces etc. can be obtained. It's okay. For example, in the embodiment described above, the correction amount changing step includes both the gain changing step and the frequency band changing step, but the correction amount changing step is configured with only either the gain changing step or the frequency band changing step. You can. This is because even if only one of the gain changing step and the frequency band changing step is performed, the effect of suppressing not only the longitudinal vibration of electric vehicle 100 but also the vertical vibration on a soft road surface such as a sandy road surface can be obtained. However, when performing both the gain change step and the frequency band change step as in the above embodiment, electric vehicle 100 is particularly likely to obtain an accurate and sufficient vertical vibration suppression effect.

Furthermore, in the above embodiment, in the frequency band changing step, the frequency fq that determines the low-frequency side cutoff frequency _fLC of the disturbance estimation filter 713 is switched between the resonance frequency fp of longitudinal vibration and the resonance frequency fp1 of vertical vibration. However, the frequency fq can be changed continuously or stepwise. For example, in the above embodiment, the resonant frequency fp1 of the vertical vibration is almost constant, but strictly speaking, the resonant frequency fp1 generated differs depending on the type and condition of the road surface even on a soft road surface. . Further, depending on the type of road surface and the condition of the road surface, a plurality of vertical vibrations having different resonance frequencies may occur. Therefore, based on the type of road surface and the condition of the road surface, the frequency fq is not limited to the two values of the resonance frequency fp and the resonance frequency fp1, but may be changed continuously or stepwise to suppress vertical vibration more accurately. There is. In acquiring such road surface type and road surface condition, the motor controller 2 can use map information, position information, and/or weather-related information in addition to the estimated disturbance torque value Td^ in the above embodiment.

Furthermore, in the above embodiment, in the frequency band changing step, the _motor controller 2 changes the cutoff frequency f _LC on the low frequency side of the disturbance estimation filter 713; Alternatively, or in addition to changing the cutoff frequency f _LC on the low frequency side, the cutoff frequency f _HC on the high frequency side may be changed. In many cases, the resonant frequency fp1 of vertical vibration is lower than the resonant frequency fp of longitudinal vibration, but depending on the specific characteristics of the soft road surface, vertical vibration with a higher frequency than the resonant frequency fp of longitudinal vibration may occur. This may occur. Furthermore, depending on the specific characteristics of the soft road surface, vertical vibrations with a frequency lower than the resonance frequency fp of longitudinal vibrations and vertical vibrations with a higher frequency than the resonance frequency fp of longitudinal vibrations may occur. In such a case, if the motor controller 2 changes the cutoff frequency f _HC on the high frequency side in the frequency band changing step, the high frequency vertical vibration can be accurately and sufficiently suppressed. Note that in order for the motor controller 2 to be able to change the cutoff frequency f _HC on the high frequency side, the resonant frequency fp is changed in equation (26) in the same manner as the switching of the cutoff frequency f _LC on the low frequency side in the above embodiment. This may be changed according to the value of the estimated disturbance torque value Td^, taking into consideration the resonant frequency of the generated high-frequency vertical vibration.

Further, in the above embodiment, in the frequency band changing step, when vertical vibration occurs, the cutoff frequency f _LC is further switched to the lower frequency side. This is because when fq=fp, the cutoff frequency _fLC is higher than the resonant frequency fp1 of vertical vibration. However, in the frequency band changing step, whether the motor controller 2 switches the cutoff frequency fLC to the low frequency side or to the high frequency side is arbitrary. That is, in the frequency band changing step, the cutoff frequency f _LC can be switched to the high frequency side. For example, when fq=fp, if a disturbance estimation filter is used whose cutoff frequency _fLC is smaller than the resonant frequency fp1 of vertical vibration, then vertical vibration will not occur, contrary to the above embodiment. By determining this and switching the cutoff frequency _fLC to the high frequency side, the resonant frequency fp1 of the vertical vibration can be removed. Such control ensures accurate and sufficient suppression of vertical vibration and sound vibration suppression performance in accordance with the running resistance of the road surface. The same applies to the case where the cutoff frequency f _HC on the high frequency side is made variable.

Although the control method for the electric vehicle 100 of the above embodiment detects the running resistance on the road surface by calculating the estimated disturbance torque value Td^, if the running resistance on the road surface can be appropriately grasped, the disturbance torque estimation in the above embodiment The running resistance on the road surface may be detected by a method other than calculating the value Td^. Further, the same applies to the estimation of gradient resistance, etc., and the estimation of gradient resistance, etc. may be performed by a method other than the method of the above embodiment.

Although the above embodiment has been described using a two-wheel drive (2WD) electric vehicle 100 that uses one electric motor 4 to drive a pair of left and right drive wheels 9a, 9b, the present invention is applicable to a four-wheel drive vehicle. It is also suitable for a (4WD) electric vehicle 1200. As shown in FIG. 12, the 4WD electric vehicle 1200 includes, for example, a front drive system 1201f, a rear drive system 1201r, and a battery 1210 and a motor controller 1220 that are common to the front drive system 1201f and the rear drive system 1201r. .

The front drive system 1201f includes a front inverter 3f, a front drive motor 4f, a front reduction gear 5f, a front rotation sensor 6f, a front current sensor 7f, a front drive shaft 8f, and a pair of front drive wheels 9af, 9bf. Further, the rear drive system 1201r includes a rear inverter 3r, a rear drive motor 4r, a rear reduction gear 5r, a rear rotation sensor 6r, a rear current sensor 7r, a rear drive shaft 8r, and a pair of rear drive wheels 9ar and 9br.

Front inverter 3f and rear inverter 3r correspond to inverter 3 of electric vehicle 100. Front drive motor 4f and rear drive motor 4r correspond to electric motor 4 of electric vehicle 100. The front speed reducer 5f and the rear speed reducer 5r correspond to the speed reducer 5 of the electric vehicle 100. Front rotation sensor 6f and rear rotation sensor 6r correspond to rotation sensor 6 of electric vehicle 100, and input front rotor phase αf and rear rotor phase αr to motor controller 1220, respectively. Front current sensor 7f and rear current sensor 7r correspond to current sensor 7 of electric vehicle 100, and input front currents iuf, ivf, iwf and rear currents iur, ivr, iwr to motor controller 1220, respectively. Front drive shaft 8f and rear drive shaft 8r correspond to drive shaft 8 of electric vehicle 100. Front drive wheels 9af, 9bf and rear drive wheels 9ar, 9br correspond to drive wheels 9a, 9b of electric vehicle 100.

Battery 1210 is one or more batteries corresponding to battery 1 of electric vehicle 100. The battery 1210 supplies power to the front drive motor 4f and rear drive motor 4r via the front inverter 3f and rear inverter 3r, respectively, or receives input of regenerated power.

Motor controller 1220 corresponds to motor controller 2 of electric vehicle 100, and similarly to motor controller 2 of electric vehicle 100, controls front drive system 1201f and rear drive system 1201r. However, the motor controller 1220 may detect the running resistance on the road surface in either the front drive system 1201f or the rear drive system 1201r, similarly to the electric vehicle 100. In addition, the motor controller 1220 adjusts the amount of correction for the drive torque target value Tm of either the front drive system 1201f or the rear drive system 1201r based on the detected running resistance of the road surface, thereby reducing vertical vibrations on a sandy road surface or the like. can be suppressed. In addition, when changing the correction amount of the drive torque target value Tm based on the running resistance of the road surface in both the front drive system 1201f and the rear drive system 1201r, the motor controller 1220 is independent in the front drive system 1201f and the rear drive system 1201r. The correction amount of the drive torque target value Tm can be changed. Of course, the motor controller 1220 may change the correction amount of the drive torque target value Tm in common for the front drive system 1201f and the rear drive system 1201r.

1: Battery 2: Motor controller 3: Inverter 3f: Front inverter 3r: Rear inverter 4: Electric motor 4f: Front drive motor 4r: Rear drive motor 5: Reducer 5f: Front reducer 5r: Rear reducer 6: Rotation sensor 6f: Front rotation sensor 6r: Rear rotation sensor 7: Current sensor 7f: Front current sensor 7r: Rear current sensor 8: Drive shaft 8f: Front drive shaft 8r: Rear drive shaft 9a: Drive wheel 9af: Front drive wheel 9ar: Rear Drive wheel 9b: Drive wheel 9bf: Front drive wheel 9br: Rear drive wheel 100: Electric vehicle 501: First motor torque estimation filter 502: Second motor torque estimation filter 503: Subtractor 504: Multiplier 505: Subtractor 701: Damping filter 702: Disturbance suppression filter 703: Adder 711: Vehicle model 712: Subtractor 713: Disturbance estimation filter 714: Multiplier 1200: Electric vehicle 1201f: Front drive system 1201r: Rear drive system 1210: Battery 1220: motor controller

Claims (9)

A method for controlling an electric vehicle using an electric motor as a power source, the method comprising:
a driving torque target value setting step of setting a driving torque target value based on vehicle information of the electric vehicle;
a step of calculating an estimated motor rotational speed, which is an estimated value of the rotational speed of the electric motor, based on a model of the electric vehicle;
an estimated disturbance calculation step of calculating an estimated disturbance that is an estimated value of disturbance applied to the electric vehicle based on a deviation between a motor rotation speed that is the actual rotation speed of the electric motor and the estimated motor rotation speed; ,
a final torque command value calculation step of calculating a final torque command value that suppresses resonance of the vehicle drive system by correcting the drive torque target value based on the estimated disturbance ;
a running resistance detection step of detecting running resistance of a road surface on which the electric vehicle runs;
a correction amount changing step of changing the correction amount of the drive torque target value based on the running resistance;
A control method for an electric vehicle including A method for controlling an electric vehicle according to claim 1 ,
a gain multiplication step of multiplying the estimated disturbance by a gain;
The correction amount changing step includes a gain changing step of changing the gain value based on the running resistance.
How to control electric vehicles. A method for controlling an electric vehicle according to claim 2 , comprising:
In the gain changing step, the larger the running resistance, the larger the gain value is set;
How to control electric vehicles. A method for controlling an electric vehicle according to any one of claims 1 to 3 , comprising:
In the estimated disturbance calculation step, the estimated disturbance is calculated by extracting a disturbance component in a predetermined frequency band from the motor rotation speed and the motor rotation speed estimated value and the deviation;
The correction amount changing step includes a frequency band changing step of changing the predetermined frequency band based on the running resistance.
How to control electric vehicles. The method for controlling an electric vehicle according to claim 4 ,
In the frequency band changing step, the lower limit value of the predetermined frequency band is changed based on the running resistance.
How to control electric vehicles. The method for controlling an electric vehicle according to claim 5 ,
In the frequency band changing step, the lower limit of the predetermined frequency band is changed to a lower frequency side based on the running resistance, thereby expanding the predetermined frequency band.
How to control electric vehicles. A method for controlling an electric vehicle according to any one of claims 1 to 6 , comprising:
In the running resistance detection step, a first estimated motor torque value is estimated using the motor rotation speed, and a second estimated motor torque value is estimated using a final torque command value input to the electric vehicle. calculating the running resistance based on the deviation;
How to control electric vehicles. The method for controlling an electric vehicle according to claim 7 ,
The running resistance detection step includes:
a slope resistance estimated value calculation step of calculating a slope resistance estimated value based on the longitudinal acceleration of the electric vehicle;
a slope resistance subtraction step of calculating the running resistance by subtracting the slope resistance estimate from the deviation between the first motor torque estimate and the second motor torque estimate;
A control method for an electric vehicle including A control device for an electric vehicle using an electric motor as a power source,
a drive torque target value setting unit that sets a drive torque target value based on vehicle information of the electric vehicle;
a motor rotation speed estimated value calculation unit that calculates a motor rotation speed estimated value that is an estimated value of the rotation speed of the electric motor based on a model of the electric vehicle;
an estimated disturbance calculation unit that calculates an estimated disturbance that is an estimated value of disturbance applied to the electric vehicle based on a deviation between a motor rotation speed that is the actual rotation speed of the electric motor and the estimated motor rotation speed; ,
a final torque command value calculation unit that calculates a final torque command value that suppresses resonance of the vehicle drive system by correcting the drive torque target value based on the estimated disturbance ;
a running resistance detection unit that detects running resistance of a road surface on which the electric vehicle runs;
a correction amount changing unit that changes the correction amount of the drive torque target value based on the running resistance;
A control device for an electric vehicle.

Images