ELECTRIC MOTOR CONTROL METHOD AND ELECTRIC MOTOR SYSTEM

Abstract

An electric motor control method for controlling an electric motor by voltage phase control performed based on a voltage norm command value representing a magnitude of a voltage to be supplied to the electric motor and a voltage phase command value representing a phase of the voltage. The method including: acquiring a required time required for calculating a final command value of a voltage applied to the electric motor by the voltage phase control according to a command value calculation model for calculation by using a rotation speed parameter related to a rotation speed of the electric motor; detecting the rotation speed parameter; changing the detected rotation speed parameter based on the required time; and calculating the final command value by using the changed rotation speed parameter according to the command value calculation model.

Description

TECHNICAL FIELD

The present invention relates to an electric motor control method and an electric motor system.

BACKGROUND ART

An operation of an electric motor in recent years is controlled by pulse width modulation. In addition, the electric motor may be controlled in an overmodulation state in which a modulation rate of the pulse width modulation is set to a value larger than 1. The overmodulation state is a rectangular wave state in which a waveform of a voltage applied from an inverter to the electric motor is substantially a rectangular wave by increasing the modulation rate.

In the overmodulation state, since a voltage norm of electric power supplied from the inverter to the electric motor is saturated, a phase of the voltage applied from the inverter to the electric motor is substantially a controllable parameter. On the other hand, in a non-overmodulation state in which the modulation rate is 1 or less, both the voltage norm and the voltage phase are controllable. Therefore, when the electric motor is controlled in the overmodulation state, controllability for the electric motor is reduced as compared with a case where the electric motor is controlled in the non-overmodulation state. As a result, for example, when a rotation speed of the electric motor changes suddenly, an overcurrent may flow in the electric motor.

Accordingly, a technique of preventing the overcurrent that occurs when the rotation speed of the electric motor changes suddenly is known. For example, JP2006-320039A discloses that when an electric motor is controlled in an overmodulation state, a rotation speed of the electric motor is detected, and a voltage applied from an inverter to the electric motor is corrected according to the detected rotation speed of the electric motor.

SUMMARY OF INVENTION

The electric motor control method disclosed in JP2006-320039A is intended to appropriately control an amount of a current flowing through the electric motor when the rotation speed of the electric motor changes suddenly in the overmodulation state by correcting the voltage applied to the electric motor according to the rotation speed of the electric motor, as described above.

However, in the control method, a response delay that occurs between the detection of the rotation speed and the application of the voltage to the electric motor according to the detected rotation speed of the electric motor is not considered. Therefore, in the control method, when the rotation speed of the electric motor changes suddenly in the overmodulation state, the overcurrent may still occur in the electric motor. In addition, if excessive compensation is performed on the voltage applied to the electric motor in order to reliably prevent the overcurrent, the voltage applied to the electric motor is insufficient in a normal situation other than a case where the rotation speed of the electric motor changes suddenly. Therefore, in the electric motor control method disclosed in JP2006-320039A, the electric motor may not be appropriately controlled.

An object of the present invention is to provide an electric motor control method and an electric motor system capable of appropriately controlling an electric motor even in an overmodulation state by taking into consideration a response delay that occurs until a voltage is applied to the electric motor.

One aspect of the present invention is an electric motor control method for controlling an electric motor by voltage phase control performed based on a voltage norm command value representing a magnitude of a volage to be supplied to the electric motor and a voltage phase command value representing a phase of the voltage. The method including: acquiring a required time required for calculating a final command value of a voltage applied to the electric motor by the voltage phase control according to a command value calculation model for calculation by using a rotation speed parameter related to a rotation speed of the electric motor; detecting the rotation speed parameter; changing the detected rotation speed parameter based on the required time; and calculating the final command value by using the changed rotation speed parameter according to the command value calculation model.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a block diagram illustrating a configuration of an electric motor system according to a first embodiment.

FIG. 2 is a block diagram illustrating a partial configuration of a current vector control unit for generating a d-axis voltage command value in current vector control.

FIG. 3 is a block diagram illustrating a configuration of a voltage phase control unit.

FIG. 4 is a graph illustrating setting of a voltage phase range in voltage phase control.

FIG. 5 is a block diagram illustrating a configuration of a phase change unit.

FIG. 6 is a block diagram illustrating a configuration of a switching determination unit.

FIG. 7 is a diagram illustrating a control mode determination method performed by a control mode determination unit.

FIG. 8 is a block diagram illustrating a configuration of an output controller.

FIG. 9 is a flowchart of electric motor control according to the present embodiment.

FIG. 10 is a flowchart of the voltage phase control.

FIG. 11 is a graph illustrating (A) an electric angular velocity, (B) an electric angular acceleration, (C) a current flowing through an electric motor, and (D) a voltage applied to the electric motor.

FIG. 12 is a block diagram illustrating a configuration of a phase change unit according to a modification.

DESCRIPTION OF EMBODIMENTS

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a block diagram illustrating a configuration of an electric motor system according to a first embodiment. As illustrated in FIG. 1, an electric motor system 100 includes an electric motor 9 to be controlled, a circuitry that controls the electric motor 9, a detector that detects a parameter necessary for controlling the electric motor 9, and a battery 15. Specifically, the electric motor system 100 includes a current vector control unit 1, a voltage phase control unit 2, an output controller 3, a converter 4, a pulse width modulation signal generator 5, and an inverter 6. Further, the electric motor system 100 includes a battery voltage detector 7, a current detector 8, a rotor detector 10, a rotation speed calculator 11, a converter 12, a control mode switching determination unit 13, and the like.

Units constituting the electric motor system 100 excluding the electric motor 9 are control devices for the electric motor 9 configured to control the electric motor 9. The battery voltage detector 7, the current detector 8, and the rotor detector 10 are each a detector that detects a parameter necessary for controlling the electric motor 9. The current vector control unit 1, the voltage phase control unit 2, and the control mode switching determination unit 13 constitute a controller (control unit). The controller is, for example, a computer including a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), or an input/output interface (I/O interface). The controller is configured to determine a control mode of the electric motor 9, change the control mode of the electric motor 9 as necessary, and control an operation of the electric motor 9 in the determined control mode. In addition, the controller is programmed to execute processing of implementing the above in each predetermined calculation cycle. In addition, the controller, the output controller 3, the converter 4, the pulse width modulation signal generator 5, the inverter 6, the rotation speed calculator 11, and the converter 12 are circuitries that control the electric motor system 100.

The electric motor 9 according to the present embodiment is an interior permanent magnet (IPM) type synchronous motor including multi-phase windings. In the present embodiment, the windings of the electric motor 9 have three phases, i.e., U-phase, V-phase, and W-phase. That is, the electric motor 9 is a three-phase synchronous motor. In addition, the electric motor 9 is driven by pulse width modulation control. The electric motor system 100 is to be incorporated in, for example, an automobile. In an automobile in which the electric motor system 100 is incorporated, the electric motor 9 can function as a driving force source and/or a generator of the automobile.

The current vector control unit 1 executes current vector control. The current vector control is one control mode of the electric motor 9. In the current vector control, a vector related to a current supplied to the electric motor 9 (hereinafter referred to as a current vector) is controlled. In the current vector control, the current vector is adjusted such that a torque generated in the electric motor 9 converges to a target value (hereinafter referred to as a torque target value T*).

More specifically, the current vector control unit 1 receives inputs of the torque target value T*, a rotation speed N [rpm] of the electric motor 9, and a DC voltage V_dc of the battery 15. In addition, a d-axis current i_d and a q-axis current i_q of the electric motor 9 are fed back to the current vector control unit 1. Then, the current vector control unit 1 calculates a d-axis voltage command value V_{di_fin}* and a q-axis voltage command value V_{qi_fin}* based on these inputs. The d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}* in the current vector control are input to the output controller 3. In the current vector control, the electric motor 9 is controlled based on the d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}*, so that the torque generated in the electric motor 9 converges to the torque target value T*.

The d-axis voltage command value V_{di_fin}* is a command value for a d-axis voltage in the current vector control. Similarly, the q-axis voltage command value V_{qi_fin}* is a command value for a q-axis voltage in the current vector control.

The current vector control unit 1 calculates a d-axis current target value i_d* and a q-axis current target value i_q* in the process of generating the d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}*. The d-axis current target value i_d is a target value of a d-axis component in the current vector. Similarly, the q-axis current target value i_q* is a target value of a q-axis component in the current vector. The d-axis current target value i_d* and the q-axis current target value i_q* are output to the voltage phase control unit 2.

In addition, a specific configuration of the current vector control unit 1 and a method of calculating the d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}* in the current vector control will be described in detail later.

The voltage phase control unit 2 executes voltage phase control. The voltage phase control is one control mode of the electric motor 9. In the voltage phase control, a phase of a voltage supplied between phases of the electric motor 9 (hereinafter referred to as a phase-to-phase voltage) is controlled. In the voltage phase control, the phase of the phase-to-phase voltage is adjusted such that the torque generated in the electric motor 9 converges to the torque target value T*.

More specifically, the voltage phase control unit 2 receives inputs of the torque target value T*, the rotation speed N of the electric motor 9, the DC voltage V_dc of the battery 15, the d-axis current target value i_d*, and the q-axis current target value i_q*. In addition, the d-axis current i_d and the q-axis current i_q are fed back to the voltage phase control unit 2. Then, the voltage phase control unit 2 outputs a d-axis voltage command value V_{dv_fin}* and a q-axis voltage command value V_{qv_fin}* to the output controller 3 based on these inputs.

The d-axis voltage command value V_{dv_fin}* is a command value for a d-axis voltage in the voltage phase control. Similarly, the q-axis voltage command value V_{qv_fin}* is a command value for a q-axis voltage in the voltage phase control. A specific configuration of the voltage phase control unit 2 and a method of calculating the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* in the voltage phase control will be described in detail later.

The d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}* in the current vector control and the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* in the voltage phase control are input to the output controller 3. In addition, a control mode signal Sm is input to the output controller 3 as a determination result from the control mode switching determination unit 13. Then, the output controller 3 selects, based on the control mode signal Sm, any one of the current vector control and the voltage phase control as the control mode of the electric motor 9 to be executed. The output controller 3 outputs a d-axis voltage command value and a q-axis voltage command value in the selected control mode to the converter 4 as a d-axis final voltage command value V_{d_fin}* and a q-axis final voltage command value V_{q_fin}*, respectively. The d-axis final voltage command value V_{d_fin}* is a final command value of a voltage applied in a d-axis direction of the electric motor. Similarly, the q-axis final voltage command value V_{q_fin}* is a final command value of a voltage applied in a q-axis direction of the electric motor 9. A configuration of the output controller 3 and generation of the control mode signal Sm will be described in detail later.

The converter 4 converts a voltage command value in a dq-axis coordinate system into a voltage command value in a UVW three-phase coordinate system. That is, the converter 4 converts the d-axis final voltage command value V_{d_fin}* and the q-axis final voltage command value V_{q_fin}* into three-phase voltage command values (V_u*, V_v*, and V_w*). The three-phase voltage command values are command values for determining voltages of the UVW phases. The converter 4 performs coordinate conversion based on an electric angle θ of the electric motor 9 as shown in the following Equation (1). The converter 4 outputs the three-phase voltage command values to the pulse width modulation signal generator 5.

$\begin{array}{cc}\left[\mathrm{Equation}1\right]& \\ \left[\begin{array}{c}{V}_u^{*}\\ V_v^{*}\\ V_w^{*}\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{cc}1& 0\\ -\frac{1}{2}& \frac{\sqrt{3}}{2}\\ -\frac{1}{2}& -\frac{\sqrt{3}}{2}\end{array}\right]\left[\begin{array}{cc}\cos \theta & -\sin \theta \\ \sin \theta & \cos \theta \end{array}\right]\left[\begin{array}{c}{V}_{\mathrm{d\_fin}}^{*}\\ V_{\mathrm{q\_fin}}^{*}\end{array}\right]& \left(1\right)\end{array}$

The pulse width modulation signal generator 5 generates a pulse width modulation signal (D_uu*, D_ul*, D_vu*, D_vl*, D_wu*, and D_wl*) based on the three-phase voltage command values and the DC voltage V_dc of the battery 15. The pulse width modulation signal (hereinafter referred to as a PWM signal) is a signal of driving a power element provided in the inverter 6. Therefore, the pulse width modulation signal is input to the inverter 6.

The inverter 6 converts, based on the PWM signal, the DC voltage V_dc of the battery 15 into three-phase AC voltages (V_u, V_v, and V_w) for driving the electric motor 9. The three-phase AC voltages are applied to the electric motor 9. Accordingly, currents flow through the phases of the electric motor 9. The currents flowing through the U phase, the V phase, and the W phase are AC currents i_u, i_v, and i_w (hereinafter, referred to as three-phase AC currents), respectively. Accordingly, the electric motor 9 is driven based on the PWM signal and generates the torque.

The battery voltage detector 7 detects a voltage of the battery 15 connected to the inverter 6. In the present embodiment, the battery voltage detector 7 detects the DC voltage V_dc of the battery 15. The detected DC voltage V_dc is output to the current vector control unit 1, the voltage phase control unit 2, and the control mode switching determination unit 13.

The current detector 8 detects a current supplied from the inverter 6 to the electric motor 9. In the present embodiment, the current detector 8 detects at least two-phase AC currents among the three-phase AC currents (i_u, i_v, and i_w). In the present embodiment, the current detector 8 detects the U-phase AC current i_u and the V-phase AC current i_v. Detected values of the three-phase AC currents are input to the converter 12.

The rotor detector 10 detects the electric angle θ of the electric motor 9. The detected electric angle θ is input to the converter 4 and the converter 12. In addition, the electric angle θ is input to the rotation speed calculator 11.

The rotation speed calculator 11 calculates the rotation speed N (mechanical angle rotation speed) and an electric angular velocity ω_e of the electric motor 9 based on a change amount of the electric angle θ in a unit time Δt_Nave. The rotation speed N and the electric angular velocity ω_e are rotation speed parameters related to the rotation speed N of the electric motor 9. The rotation speed parameter is the rotation speed N or a parameter calculated by using the rotation speed N. More specifically, the rotation speed calculator 11 detects the rotation speed N of the electric motor 9. Then, the rotation speed calculator 11 calculates the electric angular velocity ω_e by performing unit conversion of the rotation speed N using the number of pole pairs of the electric motor 9. The unit time Δt_Nave is a calculation cycle (operation cycle) of the rotation speed calculator 11, and is a predetermined value determined in advance. The rotation speed N is input to the current vector control unit 1, the voltage phase control unit 2, and the control mode switching determination unit 13. The electric angular velocity ω_e is input to the voltage phase control unit 2.

The converter 12 converts the three-phase AC currents detected by the current detector 8 into a current in the dq-axis coordinate system by using the electric angle θ detected by the rotor detector 10. In the present embodiment, since the current detector 8 detects the U-phase AC current i_u and the V-phase AC current i_v, the converter 12 performs the above coordinate conversion according to the following Equation (2).

$\begin{array}{cc}\left[\mathrm{Equation}2\right]& \\ \left[\begin{array}{c}{i}_d\\ i_q\end{array}\right]=\left[\begin{array}{cc}\cos \theta & -\sin \theta \\ \sin \theta & \cos \theta \end{array}\right]\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -\frac{1}{2}& -\frac{1}{2}\\ 0& \begin{array}{c}\underset{¯}{\sqrt{3}}\\ 2\end{array}& \begin{array}{c}-\underset{¯}{\sqrt{3}}\\ 2\end{array}\end{array}\right]\left[\begin{array}{c}{i}_u\\ i_v\\ -i_u-i_v\end{array}\right]& \left(2\right)\end{array}$

The control mode switching determination unit 13 determines a control mode of the electric motor 9 suitable for an operating state (so-called operation point or operating point) of the electric motor 9 according to the operating state. Parameters indicating the operating state of the electric motor 9 include, for example, the d-axis current i_d, the d-axis current target value i_d*, the d-axis final voltage command value V_{d_fin}*, the q-axis current i_q, the q-axis current target value i_q*, and the q-axis final voltage command value V_{q_fin}*. In addition, the DC voltage V_dc of the battery 15, the rotation speed N, the electric angular velocity ω_e, and the like are also parameters indicating the operating state of the electric motor 9. In the present embodiment, the control mode switching determination unit 13 determines the control mode of the electric motor 9 based on the d-axis final voltage command value V_{d_fin}*, the q-axis final voltage command value V_{q_fin}*, the d-axis current i_d, the rotation speed N, and the DC voltage V_dc. The control mode switching determination unit 13 outputs the control mode signal Sm, which is the determination result of the control mode, to the current vector control unit 1, the voltage phase control unit 2, and the output controller 3.

As described above, the electric motor system 100 is a system that drives the electric motor 9 by the pulse width modulation control. In addition, the electric motor system 100 appropriately switches the control mode of the electric motor 9 between the current vector control and the voltage phase control according to the operating state of the electric motor 9.

Hereinafter, specific configurations and the like of the units of the electric motor system 100 will be described in detail.

[Current Vector Control]

FIG. 2 is a block diagram illustrating a partial configuration (d-axis voltage command value calculation unit) of the current vector control unit 1 for generating the d-axis voltage command value V_{di_fin}* in the current vector control. As illustrated in FIG. 2, the current vector control unit 1 includes a decoupling voltage calculation unit 21, a low-pass filter (LPF) 22, a current target value calculation unit 23, a subtractor 24, a PI control unit 25, and an addition unit 26.

The decoupling voltage calculation unit 21 calculates a decoupling voltage V_{d_dcpl}* based on the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. The decoupling voltage V_{d_dcpl}* is output to the LPF 22. The decoupling voltage V_{d_dcpl}* is a voltage value for canceling a voltage that interferes with each other between the d-axis and the q-axis (hereinafter referred to as an interference voltage).

The decoupling voltage calculation unit 21 stores, for example, a decoupling table (not illustrated). The decoupling table is a lookup table in which the decoupling voltage V_{d_dcpl}* is associated with each operating state of the electric motor 9 in advance. In the decoupling table, the operating state of the electric motor 9 is specified by, for example, the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. Accordingly, when the decoupling voltage calculation unit 21 has the decoupling table, the decoupling voltage calculation unit 21 acquires the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. Then, the decoupling voltage calculation unit 21 calculates (acquires) the decoupling voltage V_{d_dcpl}* according to the operating state of the electric motor 9 by referring to the decoupling table. The decoupling voltage V_{d_dcpl}* stored in the decoupling table is determined in advance according to the d-axis current target value i_d* when the electric motor 9 generates a torque designated by the torque target value T* at maximum efficiency. In addition, the decoupling voltage V_{d_dcpl}* is determined in advance based on an experiment or the like.

The LPF 22 is a low-pass filter adopted in consideration that the interference voltage depends on an amount of current. A time constant of the LPF 22 is set so as to ensure responsiveness of a target d-axis current. A decoupling voltage V_{d_dcpl_flt} processed by the LPF 22 is input to the addition unit 26.

The current target value calculation unit 23 calculates the d-axis current target value i_d* based on the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. The d-axis current target value i_d* is output to the subtractor 24. In addition, the d-axis current target value i_d* is output to the voltage phase control unit 2.

The current target value calculation unit 23 stores, for example, a current table (not illustrated). The current table is a table in which the d-axis current target value i_d* is associated with each operating state (operation point) of the electric motor 9 in advance. In the current table, the operating state of the electric motor 9 is specified by, for example, the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. Accordingly, when the current target value calculation unit 23 has the current table, the current target value calculation unit 23 acquires the torque target value T*, the rotation speed N, and the DC voltage V_dc of the battery 15. Then, the current target value calculation unit 23 calculates (acquires) the d-axis current target value i_d* according to the operating state of the electric motor 9 by referring to the current table. The d-axis current target value i_d* stored in the current table is a value for the electric motor 9 to generate a torque designated by the torque target value T* or the like at the maximum efficiency, and is determined in advance by an experiment or the like.

The subtractor 24 calculates a d-axis current deviation. The d-axis current deviation is a deviation between a d-axis current target value and an actual detected value. In the present embodiment, the subtractor 24 calculates a d-axis current deviation (i_d*−i_d) by subtracting the d-axis current i_d, which is a detected value, from the d-axis current target value i_d*. The d-axis current deviation is input to the PI control unit 25.

The PI control unit 25 calculates a d-axis current feedback voltage command value V_di′ by proportional-integral (PI) control for feeding back the d-axis current deviation. In the present embodiment, the PI control unit 25 calculates the current feedback voltage command value V_di′ according to the following Equation (3). In the Equation (3), “K_dp” is a proportional gain for the d-axis. In addition, in the Equation (3), “K_di” is an integral gain for the d-axis. These PI control gains are determined in advance based on an experiment and the like. The d-axis current feedback voltage command value V_di′ is input to the addition unit 26.

$\begin{array}{cc}\left[\mathrm{Equation}3\right]& \\ V_{\mathrm{di}}^{\prime }=\frac{K_{\mathrm{dp}}S-K_{di}}{S}\left(i_d^{*}-i_d\right)& \left(3\right)\end{array}$

The addition unit 26 calculates the d-axis voltage command value V_{di_fin}* in the current vector control by adding the decoupling voltage V_{d_dcpl_flt} processed by the LPF 22 and the current feedback voltage command value V_dc as represented by the following Equation (4).

[Equation 4]

V _{di_fin} *=V _{d_dcpl_flt} +V _di′  (4)

As described above, the current vector control unit 1 calculates the d-axis voltage command value V_{di_fin}* in the current vector control by feeding back the d-axis current i_d. The current vector control is suitable for controlling the electric motor 9 in a situation where a voltage norm of electric power supplied to the electric motor 9 is not saturated and the voltage norm can be changed according to a request.

Here, among the specific configuration of the current vector control unit 1, the d-axis voltage command value calculation unit that generates the d-axis voltage command value V_{di_fin}* in the current vector control has been described. Alternatively, a portion unit generating the q-axis voltage command value V_{qi_fin}* in the current vector control is configured in the same manner as the above d-axis voltage command value calculation unit. That is, the current vector control unit 1 includes a q-axis voltage command value calculation unit as a configuration for generating the q-axis voltage command value V_{qi_fin}* in the current vector control. The q-axis voltage command value calculation unit includes a decoupling voltage calculation unit, an LPF, a current target value calculation unit, a subtractor, a PI control unit, and an addition unit. Then, units of the q-axis voltage command value calculation unit have the same configuration as that of the d-axis voltage command value calculation unit except that the units are for the q-axis.

[Voltage Phase Control]

FIG. 3 is a block diagram illustrating a configuration of the voltage phase control unit 2. The voltage phase control unit 2 includes (a) a voltage norm determination unit, (b) a voltage phase determination unit, and (c) a voltage command value calculation unit.

• • (a) The voltage norm determination unit is configured to determine the voltage norm of the electric power supplied to the electric motor 9 in the voltage phase control. Specifically, the voltage norm determination unit includes a voltage norm command value generation unit, a magnetic flux feedback control unit, a voltage norm synthesis unit 40, and a voltage norm limitation unit 41 illustrated in FIG. 3. Specifically, the voltage norm command value generation unit includes an upper voltage norm limit value calculation unit 31, a phase change unit 32, an induced voltage calculation unit 33, and a feedforward voltage norm command value calculation unit 34 illustrated in FIG. 3. Specifically, the magnetic flux feedback control unit includes a reference current calculation unit 35, a magnetic flux calculation unit 36, a magnetic flux estimation unit 37, a magnetic flux deviation calculation unit 38, and a PI control unit 39 illustrated in FIG. 3.
  • (b) The voltage phase determination unit is configured to determine a phase of a voltage applied to the electric motor 9 in the voltage phase control. Specifically, the voltage phase determination unit includes a voltage phase command value generation unit 42, a torque feedback control unit, a phase synthesis unit 47, and a phase limitation unit 48 illustrated in FIG. 3. Specifically, the torque feedback control unit includes a reference torque generation unit 43, a torque estimation unit 44, a torque deviation calculation unit 45, and a PI control unit 46 illustrated in FIG. 3.
  • (c) The voltage command value calculation unit calculates the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* in the voltage phase control. These are calculated based on the voltage norm determined by the voltage norm determination unit and the phase of the voltage determined by the voltage phase determination unit. Specifically, the voltage command value calculation unit includes a vector conversion unit 49 and a stabilization filter 50 illustrated in FIG. 3.

(a-1) Voltage Norm Command Value Generation Unit

The voltage norm command value generation unit generates a feedforward voltage norm command value (hereinafter referred to as an FF voltage norm command value V_{a_ff}) by feedforward control. The FF voltage norm command value V_{a_ff} is a command value for determining a magnitude of the voltage applied to the electric motor 9 in the voltage phase control. In addition, the FF voltage norm command value V_{a_ff} is a basic target value of a voltage norm determined based on the operating state of the electric motor 9. In the present embodiment, the FF voltage norm command value V_{a_ff} generated by the voltage norm command value generation unit is a basic target value (so-called basic value) of a voltage norm in the voltage phase control. Therefore, a voltage norm actually applied to the electric motor 9 is limited by the voltage norm synthesis unit 40 and the voltage norm limitation unit 41.

The FF voltage norm command value V_{a_ff} corresponds to a command value that designates a modulation rate M of PWM control (hereinafter referred to as a modulation rate command value M*). Therefore, the FF voltage norm command value V_{a_ff} generated by the voltage norm command value generation unit corresponds to a basic target value of the modulation rate M (reference modulation rate). The modulation rate command value M* can be set freely. The modulation rate command value M* is calculated by using, for example, the DC voltage V_dc of the battery 15 and a voltage norm command value V_a* described later. For example, the modulation rate command value M* is calculated according to M*=(√2/V_dc)V_a*. In the present embodiment, the modulation rate M of the PWM control is a ratio of an amplitude “Amp” of a fundamental wave component to the DC voltage V_dc of the battery 15 (|Amp/V_dc|). The fundamental wave component is a fundamental frequency component among frequency components included in a waveform of the phase-to-phase voltage to be applied to the electric motor 9.

In a non-overmodulation state in which the modulation rate M is 1 or less (normal pulse width modulation state), a fundamental wave component of a correlation voltage is a pseudo-sine wave. A lower limit of the modulation rate M is 0.

On the other hand, in an overmodulation state in which the modulation rate M exceeds 1, the power element of the inverter 6 does not switch at a vicinity of a maximum value and a minimum value of the fundamental wave component. Therefore, in the overmodulation state, a waveform of the correlation voltage is a waveform in which the vicinity of the maximum value and the minimum value of the fundamental wave component is cut. In addition, in the overmodulation state, the waveform of the correlation voltage includes a harmonic component. Then, in the overmodulation state, when the modulation rate M is, for example, approximately 1.1 or more, the waveform of the correlation voltage is substantially a rectangular wave. That is, the overmodulation state extremely includes a rectangular wave state in which the waveform of the correlation voltage is substantially a rectangular wave.

As described above, the voltage norm command value generation unit includes the upper voltage norm limit value calculation unit 31, the phase change unit 32, the induced voltage calculation unit 33, and the feedforward voltage norm command value calculation unit 34.

The upper voltage norm limit value calculation unit 31 calculates an upper voltage norm limit value V_{a_upper} based on the DC voltage V_dc of the battery 15 and the modulation rate command value M*. The upper voltage norm limit value V_{a_upper} is an upper limit value of the FF voltage norm command value V_{a_ff} and the voltage norm command value V_a*. The upper voltage norm limit value V_{a_upper} is calculated, for example, according to V_{a_upper}=(V_dc/√2) M*. The upper voltage norm limit value V_{a_upper} is input to the feedforward voltage norm command value calculation unit 34.

The phase change unit 32 changes a phase of the electric angular velocity ω_e. More specifically, the phase change unit 32 acquires, for a command value calculation model (configuration of the voltage phase control unit 2) determined in advance in order to calculate the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}*, a predetermined required time T_s including a time required from detection of the electric angular velocity ω_e to calculation of these command values. The required time T_s is generally minimized at a time point when the command value calculation model is designed. Then, the phase of the electric angular velocity ω_e is changed based on the required time T_s. More specifically, the phase change unit 32 calculates an absolute value |ω_e′| of the electric angular velocity after the phase change (hereinafter, simply referred to as an “electric angular velocity after phase change” for simplicity) by advancing a phase of an absolute value |ω_e| of the electric angular velocity ω_e. A configuration of the phase change unit 32 and a phase change method will be described in detail later. The electric angular velocity |ω_e′| after the phase change is input to the induced voltage calculation unit 33.

The induced voltage calculation unit 33 calculates a reference induced voltage V_a0 by using a reference magnetic flux norm φ_{0_ref} described later and the electric angular velocity ω_e′ after the phase change. In the present embodiment, the induced voltage calculation unit 33 calculates the reference induced voltage V_a0 by multiplying the reference magnetic flux norm φ_{0_ref} by the electric angular velocity ω_e′ after the phase change. The calculated reference induced voltage V_a0 is input to the feedforward voltage norm command value calculation unit 34.

The feedforward voltage norm command value calculation unit 34 calculates the FF voltage norm command value V_{a_ff} by using the upper voltage norm limit value V_{a_upper} and the reference induced voltage V_a0. Specifically, the feedforward voltage norm command value calculation unit 34 is configured by a minimum function (min) using the upper voltage norm limit value V_{a_upper} and the reference induced voltage V_a0 as arguments, and outputs a smaller value in the upper voltage norm limit value V_{a_upper} and the reference induced voltage V_a0 as the FF voltage norm command value V_{a_ff}. Therefore, in principle, the reference induced voltage V_a0 is the FF voltage norm command value V_{a_ff}, but when the reference induced voltage V_a0 exceeds the upper voltage norm limit value V_{a_upper}, the upper voltage norm limit value V_{a_upper} is the FF voltage norm command value V_{a_ff}. The FF voltage norm command value V_{a_ff} calculated in this way is input to the voltage norm synthesis unit 40 and the voltage norm limitation unit 41.

(a-2) Magnetic Flux Feedback Control Unit

As described above, the magnetic flux feedback control unit includes the reference current calculation unit 35, the magnetic flux calculation unit 36, the magnetic flux estimation unit 37, the magnetic flux deviation calculation unit 38, and the PI control unit 39. In addition, the magnetic flux feedback control unit calculates a magnetic flux state quantity representing a magnetic flux generated in the electric motor 9 based on the current supplied to the electric motor 9 as a whole. The magnetic flux generated in the electric motor 9 is a synthetic magnetic flux obtained by synthesizing a magnetic flux (hereinafter referred to as a magnetic flux of magnet) Φ_a of a permanent magnet provided in the electric motor 9 and a magnetic flux (hereinafter referred to as a winding magnetic flux) generated by a current supplied to a winding (coil) of the electric motor 9. The magnetic flux state quantity is, for example, a reference magnetic flux norm φ_{0_ref} representing a target response of a norm of a vector representing the synthetic magnetic flux (hereinafter referred to as a magnetic flux norm), an estimated magnetic flux norm value φ_{0_est} representing an estimated value of the magnetic flux norm, and a magnetic flux deviation φ_{0_err} representing these deviations. In addition, the magnetic flux feedback control unit calculates a feedback norm command value (FB voltage norm command value V_{a_fb}) to feed back to the FF voltage norm command value V_{a_ff} according to the magnetic flux state quantity.

The reference current calculation unit 35 calculates a d-axis reference current i_{d_ref}* representing a target response of the d-axis current by using the d-axis current target value i_d*. In addition, the reference current calculation unit 35 calculates a q-axis reference current i_{q_ref}* representing a target response of a q-axis current by using the q-axis current target value i_q*. The d-axis reference current i_{d_ref}* and the q-axis reference current i_{q_ref}* are input to the magnetic flux calculation unit 36. The reference current calculation unit 35 is, for example, a low-pass filter (LPF) having the same time constant as that of the LPF 22 (see FIG. 2).

The magnetic flux calculation unit 36 calculates the reference magnetic flux norm φ_{0_ref} based on the d-axis reference current i_{d_ref}* and the q-axis reference current i_{q_ref}*. The reference magnetic flux norm φ_{0_ref} represents a target value of the magnetic flux norm in the voltage phase control. The magnetic flux calculation unit 36 calculates the reference magnetic flux norm φ_{0_ref} by using a d-axis inductance L_d, a q-axis inductance L_q, and the magnetic flux Φ_a of magnet according to the following Equation (5). The reference magnetic flux norm φ_{0_ref} is input to the magnetic flux deviation calculation unit 38.

[Equation 5]

ϕ_{0_ref}=√{square root over ((Φ_a+L_di_{d_ref}*)²+(L_qi_{q_ref}*)²)}  (5)

The magnetic flux Φ_a of magnet, the d-axis inductance L_d, and the q-axis inductance L_q are constants. The d-axis reference current i_{d_ref}* and the q-axis reference current i_{q_ref}* are variables. However, the respective constants, i.e., the magnetic flux Φ_a of magnet, the d-axis inductance L_d, and the q-axis inductance L_q may change depending on a magnet temperature, the d-axis current i_d, the q-axis current i_q, the d-axis current target value i_d*, and/or the q-axis current target value i_q* of the electric motor 9. Therefore, the magnetic flux calculation unit 36 can appropriately change some or all of the respective constants depending on the magnet temperature or the like. In this case, the magnetic flux calculation unit 36 uses an appropriate constant corresponding to the magnet temperature or the like with reference to a map in which the above respective constants are determined according to the magnet temperature or the like. This map is determined in advance by an experiment, a simulation, or the like.

The magnetic flux estimation unit 37 calculates the estimated magnetic flux norm value φ_{0_est} based on the d-axis current i_d and the q-axis current i_q, which are detected values. The estimated magnetic flux norm value φ_{0_est} is an estimated value of an actual magnetic flux norm in the electric motor 9. A method of calculating the estimated magnetic flux norm value φ_{0_est} is the same as a method of calculating the reference magnetic flux norm φ_{0_ref} except that the d-axis current i_d and the q-axis current i_q are used as variables. That is, the magnetic flux estimation unit 37 calculates the estimated magnetic flux norm value φ_{0_est} according to the following Equation (6). The estimated magnetic flux norm value φ_{0_est} is input to the magnetic flux deviation calculation unit 38.

[Equation 6]

ϕ_{0_est}=√{square root over ((Φ_a+L_di_d)²+(L_qi_q)²)}  (6)

The magnetic flux deviation calculation unit 38 calculates the magnetic flux deviation φ_{0_err} In the present embodiment, the magnetic flux deviation calculation unit 38 calculates the magnetic flux deviation φ_{0_err} by subtracting the estimated magnetic flux norm value φ_{0_est} from the reference magnetic flux norm φ_{0_ref}. The magnetic flux deviation φ_{0_err} is input to the PI control unit 39.

The PI control unit 39 calculates the feedback voltage norm command value (hereinafter referred to as a FB voltage norm command value) V_{a_fb} by PI control. The FB voltage norm command value V_{a_fb} is a voltage norm command value for feeding back to the FF voltage norm command value V_{a_ff}. Specifically, the PI control unit 39 calculates the FB voltage norm command value V_{a_fb} by using the electric angular velocity ω_e and the magnetic flux deviation φ_{0_err} according to the following Equation (7). In the Equation (7), “Kφ_p” is a proportional gain. “Kφ_i” is an integral gain. These gains are determined in advance by an experiment, a simulation, or the like. In addition, the electric angular velocity ω_e fluctuates as a variable gain that varies according to the rotation speed N. The FB voltage norm command value V_{a_fb} is input to the voltage norm synthesis unit 40.

$\begin{array}{cc}\left[\mathrm{Equation}7\right]& \\ V_{\mathrm{a\_fb}}={\omega }_e{K}_{\varphi p}\left(1+\frac{K_{\varphi i}}{s}\right){\varphi }_{0\mathrm{\_err}}& \left(7\right)\end{array}$

(a-3) Voltage Norm Synthesis Unit

The voltage norm synthesis unit 40 generates the voltage norm command value V_a* by feeding back the FB voltage norm command value V_{a_fb} to the FF voltage norm command value V_{a_ff}. The voltage norm command value V_a* is a continuous value. In the present embodiment, the voltage norm synthesis unit 40 calculates the voltage norm command value V_a* by adding the FB voltage norm command value V_{a_fb} to the FF voltage norm command value V_{a_ff}. The voltage norm command value V_a* is a target value of a voltage norm corrected based on the magnetic flux state quantity (or a current amount that is a calculation source of the magnetic flux state quantity). Accordingly, the voltage norm command value (voltage norm command value V_a*) in the voltage phase control can be adjusted by the feedback of the FB voltage norm command value V_{a_fb}. As a result, the voltage norm command value (voltage norm command value V_a*) in the voltage phase control is controlled such that the magnetic flux deviation φ_{0_err} approaches zero.

As described above, the magnetic flux deviation φ_{0_err} is a value obtained by subtracting the estimated magnetic flux norm value φ_{0_est} from the reference magnetic flux norm φ_{0_ref}. Therefore, for example, when the rotation speed N suddenly decreases due to a high load such as a sudden braking from a state in which the rotation speed N is high (a state in which the electric motor 9 is rotating at a medium speed to a high speed), the estimated magnetic flux norm value φ_{0_est} steeply increases with respect to the reference magnetic flux norm φ_{0_ref}. Then, the magnetic flux deviation φ_{0_err} becomes a negative value, and the FB voltage norm command value V_{a_fb}, also becomes a negative value. In this case, the voltage norm command value V_a* is a value smaller than the FF voltage norm command value V_{a_ff}, which is an initial target value. As a result, the voltage norm command value V_a* becomes a value close to the voltage norm of the electric power applied to the electric motor 9 in the current vector control. When the current vector control is performed, the same applies to a case where the voltage norm of the electric power applied to the electric motor 9 suddenly increases due to a sudden acceleration or the like, and the voltage norm command value V_a* becomes the value close to the voltage norm of the electric power applied to the electric motor 9 in the current vector control. That is, the above feedback control based on the magnetic flux state quantity has an action of reducing a gap between the voltage norms in the voltage phase control and the current vector control. Therefore, switching between the voltage phase control and the current vector control is performed substantially seamlessly by the above feedback control based on the magnetic flux state quantity.

(a-4) Voltage Norm Limitation Unit

The voltage norm limitation unit 41 limits the voltage norm command value V_a* to a value between a predetermined lower limit value and the upper voltage norm limit value V_{a_upper} (voltage norm limitation processing). The lower limit value is, for example, zero. When the voltage norm command value V_a* is limited to the lower limit value or an upper limit value V_{a_max}, the voltage norm limitation unit 41 outputs a notification signal indicating the above to the PI control unit 39. When receiving the notification signal, the PI control unit 39 initializes an integrating unit included therein for anti-windup.

(b-1) Voltage Phase Command Value Generation Unit

The voltage phase command value generation unit 42 generates a feed forward voltage phase command value (hereinafter referred to as a FF voltage phase command value) α_ff according to the operating state of the electric motor 9 by feed forward control. The FF voltage phase command value α_ff represents a phase of a voltage to be supplied to the electric motor 9. In addition, the FF voltage phase command value α_ff is a basic target value of a voltage phase determined according to the operating state of the electric motor 9. The voltage phase command value generation unit 42 includes a voltage phase table 42a that is a lookup table for generating the FF voltage phase command value α_ff. The voltage phase table 42a is a lookup table in which the FF voltage phase command value α_ff is stored in association with each operating state (operation point) of the electric motor 9. The FF voltage phase command value α_ff stored in the voltage phase table 42a is, for example, a value of a voltage phase measured in a nominal state for each operating state of the electric motor 9 in an experiment or the like. In the present embodiment, the FF voltage norm command value V_{a_ff}, the rotation speed N, the torque target value T*, and the FF voltage phase command value α_ff are stored in association with each other in the voltage phase table 42a. Therefore, when acquiring these parameters, the voltage phase command value generation unit 42 generates the FF voltage phase command value α_ff corresponding to a combination of these parameters by referring to the voltage phase table 42a. The generated FF voltage phase command value α_ff is input to the phase synthesis unit 47.

(b-2) Torque Feedback Control Unit

As described above, the torque feedback control unit includes the reference torque generation unit 43, the torque estimation unit 44, the torque deviation calculation unit 45, and the PI control unit 46. The torque feedback control unit calculates a feedback voltage phase command value (hereinafter referred to as a FB voltage phase command value) α_fb as a whole. The FB voltage phase command value α_fb is a voltage phase command value for feeding back to the FF voltage phase command value α_ff.

The reference torque generation unit 43 generates a reference torque T_ref representing a target response of the torque in the electric motor 9 by using the torque target value T*. The reference torque generation unit 43 is, for example, a low-pass filter having the same time constant as that of the LPF 22 (see FIG. 2). The generated reference torque T_ref is input to the torque deviation calculation unit 45.

The torque estimation unit 44 calculates an estimated torque value T_est. The estimated torque value T_est is an estimated value of a torque of the electric motor 9 in a specific operating state. The torque estimation unit 44 includes a torque table (not illustrated) for estimating the estimated torque value T_est. The torque table is a lookup table in which the estimated torque value T_est is stored in association with each operating state (operation point) of the electric motor 9. In the present embodiment, the d-axis current i_d, the q-axis current i_q, and the estimated torque value T_est are stored in association with each other in the torque table. Therefore, when acquiring these parameters, the torque estimation unit 44 refers to the torque table and calculates the estimated torque value T_est corresponding to a combination of these parameters. The estimated torque value T_est is input to the torque deviation calculation unit 45.

The torque deviation calculation unit 45 calculates a torque deviation T_err. The torque deviation T_err is a deviation between the reference torque T_ref and the estimated torque value T_est. In the present embodiment, the torque deviation calculation unit 45 calculates the torque deviation T_err by subtracting the estimated torque value T_est from the reference torque T_ref. The torque deviation T_err is input to the PI control unit 46.

The PI control unit 46 calculates the FB voltage phase command value α_fb by using the torque deviation T_err (=T_ref−T_est). Specifically, the PI control unit 46 calculates the FB voltage phase command value α_fb according to the following Equation (8). In the Equation (8), “Kα_p” is a proportional gain. In addition, “Kα_i” is an integral gain. These gains are determined in advance by an experiment, a simulation, or the like. The FB voltage phase command value α_fb is input to the phase synthesis unit 47.

$\begin{array}{cc}\left[\mathrm{Equation}8\right]& \\ {\alpha }_{fb}=\frac{K_{\alpha p}s+K_{\alpha i}}{s}\left(T_{\mathrm{ref}-}{T}_{est}\right)& \left(8\right)\end{array}$

(b-3) Phase Synthesis Unit

The phase synthesis unit 47 generates a voltage phase command value α* by feeding back the FB voltage phase command value α_fb to the FF voltage phase command value α_ff. In the present embodiment, the phase synthesis unit 47 generates the voltage phase command value α* by adding the FB voltage phase command value α_fb to the FF voltage phase command value α_ff. The voltage phase command value α* is a target value of a voltage phase corrected based on the torque deviation T_err (or a current amount used to calculate the torque deviation T_err). The correction converges the torque deviation T_err to zero. Accordingly, the voltage phase command value α* generated by the phase synthesis unit 47 is input to the phase limitation unit 48.

(b-4) Phase Limitation Unit

The phase limitation unit 48 limits the voltage phase command value α* to a predetermined voltage phase range (voltage phase limitation processing). The predetermined voltage phase range is a range from a predetermined lower limit value of the voltage phase (hereinafter referred to as a lower voltage phase limit value) α_min to a predetermined upper limit value of the voltage phase (hereinafter referred to as an upper voltage phase limit value) α_max. Setting of the lower voltage phase limit value α_min and the upper voltage phase limit value α_max will be described later in detail. When the voltage phase command value α* is limited by the voltage phase limitation processing, the phase limitation unit 48 outputs a notification signal indicating the above to the PI control unit 46. When receiving the notification signal, the PI control unit 46 initializes an integrating unit included therein for anti-windup.

(c-1) Vector Conversion Unit

The vector conversion unit 49 converts the voltage norm command value V_a* and the voltage phase command value α* into a d-axis voltage command value V_dv* and a q-axis voltage command value V_qv* according to the following Equation (9).

$\begin{array}{cc}\left[\mathrm{Equation}9\right]& \\ \{\begin{array}{c}{V}_{dv}^{*}=-V_{\alpha }^{*}\sin {\alpha }^{*}\\ V_{\mathrm{qv}}^{*}=V_{\alpha }^{*}\cos {\alpha }^{*}\end{array}& \left(9\right)\end{array}$

The stabilization filter 50 calculates the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}*, which are voltage command values in the voltage phase control, by using the above d-axis voltage command value V_dv* and q-axis voltage command value V_qv*. Specifically, the stabilization filter 50 calculates the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* according to an Equation (10). The stabilization filter 50 cancels resonance characteristics of the d-axis current i_d and the q-axis current i_q with respect to the d-axis voltage command value V_dv* and the q-axis voltage command value V_qv* after the vector conversion. Accordingly, the stabilization filter 50 stabilizes a feedback loop.

$\begin{array}{cc}\left[\mathrm{Equation}10\right]& \\ \left(\begin{array}{c}{V}_{\mathrm{dv\_fin}}^{*}\\ V_{\mathrm{qv\_fin}}^{*}\end{array}\right)=\left(\begin{array}{cc}\frac{k_{11}s+1}{{\tau }_{m}s+1}& \frac{k_{12}s}{{\tau }_{m}s+1}\\ -\frac{k_{21}s}{{\tau }_{m}s+1}& \frac{k_{22}s+1}{{\tau }_{m}s+1}\end{array}\right)\left(\begin{array}{c}{V}_{\mathrm{dv}}^{*}\\ V_{\mathrm{qv}}^{*}\end{array}\right)& \left(10\right)\end{array}$

In the Equation (10), “k₁₁”, “k₁₂”, “k₂₁”, and “k₂₂” are variable gains that change depending on the electric angular velocity ω_e. These variable gains (k₁₁, k₁₂, k₂₁, k₂₂) are determined according to the following Equation (11). In the Equation (11), “L_d′” is a dynamic inductance of the d-axis. “L_q” is a dynamic inductance of the q-axis.

$\begin{array}{cc}\left[\mathrm{Equation}11\right]& \\ \left(\begin{array}{cc}{k}_{11}& k_{12}\\ k_{21}& k_{22}\end{array}\right)=\frac{1}{R^2+{\omega }_e^2{L}_d{L}_q}\left(\begin{array}{cc}R{L}_d^{\prime }& {\omega }_e{L}_d^{\prime }{L}_q\\ {\omega }_e{L}_d{L}_q^{\prime }& R{L}_q^{\prime }\end{array}\right)& \left(11\right)\end{array}$

As can be seen from the Equation (10), the stabilization filter 50 is designed such that the d-axis current i_d and the q-axis current i_q transiently exhibit a first-order delay response of a predetermined time constant τ_m. The phase of the voltage applied to the electric motor 9 is advanced by the inductances of the d-axis and the q-axis, but the phase advance is small enough to be ignored when viewed as a whole system. Therefore, when the phase advance due to the inductances of the d-axis and the q-axis is ignored, transient dynamics of the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* in the voltage phase control can be regarded as the first-order delay response of the time constant τ_m.

As described above, the voltage phase control unit 2 changes the voltage phase command value α* such that the torque deviation T_err converges to zero. Therefore, in the voltage phase control, even in a situation where the voltage norm of the electric power supplied to the electric motor 9 is saturated, the torque of the electric motor 9 can be changed according to the request. In particular, even in the overmodulation state or the extreme rectangular wave state, the torque of the electric motor 9 can be appropriately controlled by the voltage phase control. In addition, in the above voltage phase control, since the voltage norm command value V_a* is changed by feeding back the magnetic flux deviation φ_{0_err}, switching between the voltage phase control and the current vector control is seamless.

As described above, the voltage phase control unit 2 calculates a final command value of the voltage applied to the electric motor 9 (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) by the voltage phase control according to a “command value calculation model” (so-called transfer function) determined in advance for calculation by using a rotation parameter (electric angular velocity ω_e) related to the rotation speed N of the electric motor 9. The voltage phase control unit 2 acquires the required time T_s required for calculating the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) in the voltage phase control according to the command value calculation model. Then, the detected rotation speed parameter (electric angular velocity ω_e) is changed based on the required time T_s, and the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) is calculated by using the changed rotation speed parameter (electric angular velocity |ω_e′|) according to the command value calculation model.

The required time T_s corresponding to the “command value calculation model” includes, for example, a time required from the detection of the rotation speed parameter (electric angular velocity ω_e) to the calculation of the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*). In addition, the required time T_s is determined at a time point when the command value calculation model is set (a time point when the voltage phase control unit 2 is designed). The command value calculation model is usually set such that the required time T_s is as small as possible. However, the required time T_s does not become zero no matter what system the command value calculation model is implemented specifically, since there are delay times due to various types of detection, calculation, and the like included therein. In addition, the required time T_s can include not only the delay time generated by the calculation or the like of the voltage phase control unit 2, but also a delay time required for acquiring a parameter necessary for the voltage phase control (for example, the rotation speed parameter). For example, a time lag (detection period) in the detection of the rotation speed N can be included in the required time T_s. In addition, when there is a delay time until the voltage is actually applied from the inverter 6 to the electric motor 9 based on the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*), the delay time can also be included in the required time T_s. That is, the required time T_s represents a part or the whole of a delay time of a response included in a period from the detection of the rotation speed parameter related to the rotation speed N to the application of the voltage to the electric motor 9. In the present embodiment, the required time T_s is known by setting the command value calculation model as described above. Therefore, information related to the required time T_s is held in advance in a memory (not illustrated), for example.

[Setting of Upper Voltage Phase Limit Value and Lower Voltage Phase Limit Value]

FIG. 4 is a graph illustrating setting of the voltage phase range in the voltage phase control. As illustrated in FIG. 4, in a certain electric motor, a correlation between a voltage phase α and a torque T is maintained generally within a range of the voltage phase α of −105 degrees to +105 degrees. In this case, the upper voltage phase limit value α_max is set to +105 degrees, and the lower voltage phase limit value α_min is set to −105 degrees. Therefore, when an electric motor having characteristics in FIG. 4 is used as the electric motor 9 of the electric motor system 100, the phase limitation unit 48 limits the voltage phase command value α* in the range of −105 degrees to +105 degrees. This is to ensure controllability for the electric motor 9 by maintaining the correlation with the torque of the electric motor 9.

[Configuration and Phase Change Method of Phase Change Unit]

FIG. 5 is a block diagram illustrating a configuration of the phase change unit 32. As illustrated in FIG. 5, the phase change unit 32 includes an acceleration calculation unit 51, a change limitation unit 52, a phase change amount calculation unit 53, an addition unit 54, and a final limitation unit 55.

The acceleration calculation unit 51 differentiates the absolute value |ω_e| of the electric angular velocity ω_e with respect to time, and calculates an electric angular acceleration A_e. In the present embodiment, the acceleration calculation unit 51 is a so-called incomplete differentiator, and is a combination of a differentiator “s” and a low-pass filter “1/(τ_ws+1)”. Therefore, the acceleration calculation unit 51 approximately time-differentiates the absolute value |ω_e| of the electric angular velocity ω_e. The time constant “τ_w” is a constant determined in advance in consideration of resistance to high-frequency noise included in a detected value of the rotation speed N and responsiveness to a change of the rotation speed N. The electric angular acceleration A_e is input to the change limitation unit 52.

The change limitation unit 52 is a switch that switches whether to change the phase of the electric angular velocity ω_e based on the electric angular acceleration A_e. In the present embodiment, the change limitation unit 52 determines whether to change the phase of the electric angular velocity ω_e by comparing the electric angular acceleration A_e with a predetermined threshold A_e0 determined in advance. Specifically, when the electric angular acceleration A_e is smaller than the threshold A_e0, the change limitation unit 52 determines to execute the change of the phase of the electric angular velocity ω_e (hereinafter, simply referred to as a phase change). On the other hand, when the electric angular acceleration A_e is equal to or larger than the threshold A_e0, the change limitation unit 52 determines not to execute the phase change of the electric angular velocity ω_e. The threshold A_e0 is determined in advance based on an experiment, a simulation, or the like. In addition, the threshold A_e0 is freely determined in order to correspond to a specific scene.

When it is determined that the electric angular acceleration A_e is smaller than the threshold A_e0, the change limitation unit 52 outputs the input electric angular acceleration A_e. On the other hand, when it is determined that the electric angular acceleration A_e is equal to or larger than the threshold A_e0, the change limitation unit 52 outputs an initial value determined in advance (zero in the present embodiment). The output by the change limitation unit 52 is input to the phase change amount calculation unit 53.

The phase change amount calculation unit 53 calculates a change amount of the phase of the electric angular velocity ω_e (hereinafter referred to as a phase change amount) ω_ecmp by using the output of the change limitation unit 52. In addition, when calculating the phase change amount ω_ecmp, the phase change amount calculation unit 53 acquires the required time T_s required until the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) is calculated according to the command value calculation model in the voltage phase control. Then, the phase change amount calculation unit 53 calculates the phase change amount ω_ecmp in consideration of the required time T_s. The calculated phase change amount ω_ecmp is input to the addition unit 54.

In the present embodiment, the phase change amount calculation unit 53 calculates the phase change amount ω_ecmp by multiplying the output of the change limitation unit 52 by the required time T_s. Therefore, when the electric angular acceleration A_e is smaller than the threshold A_e0, the phase change amount ω_ecmp is a product (A_eT_s) of the electric angular acceleration A_e and the required time T_s. Accordingly, the phase change amount calculation unit 53 calculates the phase change amount ω_ecmp corresponding to the electric angular velocity ω_e that changes during the required time T_s. However, when the electric angular acceleration A_e is equal to or larger than the threshold A_e0, since the output of the change limitation unit 52 is zero, the phase change amount ω_ecmp is also zero. This is a scene in which the phase of the electric angular velocity ω_e is not changed.

In the present embodiment, as described above, the rotation speed calculator 11 calculates the electric angular velocity ω_e by calculating the change amount of the electric angle θ in the unit time Δt_Nave. Therefore, in a certain electric angular acceleration A_e, a wasted time required for detecting the rotation speed N can be regarded as approximately Δt_Nave/2. In addition, as described above, in the stabilization filter 50, responses of the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* with respect to the d-axis voltage command value V_dv* and the q-axis voltage command value V_qv* can be regarded as a first-order response delay of the time constant τ_m. Therefore, in a certain electric angular acceleration A_e, a wasted time in the output of the voltage to the electric motor 9 can be regarded as the time constant τ_m. Therefore, in the present embodiment, the required time T_s is expressed by the sum of the above. That is, the required time T_s is calculated according to the following Equation (12).

$\begin{array}{cc}\left[\mathrm{Equation}12\right]& \\ T_S=\frac{\Delta t_{Na\nu e}}{2}+{\tau }_m& \left(12\right)\end{array}$

By setting the required time T_s in this way, the phase of the electric angular velocity ω_e is changed within a range that does not exceed a change amount that can be generated in the electric angular velocity ω_e per required time T_s within a time range in which the electric angular acceleration A_e can be regarded as substantially constant. As a result, even when the voltage norm is operated by using the electric angular velocity |ω_e′| after the phase change, an excessive change exceeding the required time T_s is prevented.

The addition unit 54 adds the phase change amount ω_ecmp to the electric angular velocity ω_e. Therefore, the phase of the electric angular velocity ω_e is advanced. An output of the addition unit 54 is the electric angular velocity |ω_e′| after the phase change. The electric angular velocity |ω_e′| after the phase change is output via the final limitation unit 55.

The final limitation unit 55 is configured by a maximum function (max) using the electric angular velocity |ω_e′| after the phase change, which is the output of the addition unit 54, and a fixed value “zero (0)” as arguments. Therefore, the final limitation unit 55 limits the lower limit value such that the electric angular velocity |ω_e′| after the phase change is not a negative value.

As described above, the phase change unit 32 acquires the required time T_s required until the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) is calculated according to the command value calculation in the voltage phase control. Then, the phase change unit 32 changes the phase of the electric angular velocity ω_e based on the required time T_s. Thereafter, the calculation of the final command value according to the command value calculation model in the voltage phase control is executed based on the electric angular velocity |ω_e′| whose phase is changed. In the phase change processing, the phase of the electric angular velocity ω_e is changed in advance such that the electric angular velocity |ω_e′| after the phase change approaches the electric angular velocity ω_e after the required time T_s is elapsed.

[Specific Configuration of Output Controller and Generation of Control Mode Signal]

FIG. 6 is a block diagram illustrating a configuration of the control mode switching determination unit 13. As illustrated in FIG. 6, the control mode switching determination unit 13 includes a first norm threshold calculation unit 94, a second norm threshold calculation unit 95, an averaging filter 96, an averaging filter 97, a norm calculation unit 98, and a control mode determination unit 99.

The first norm threshold calculation unit 94 calculates a first norm threshold V_{a_th1} based on a first modulation rate threshold Mail. The first modulation rate threshold M_th1 is a threshold determined for the modulation rate M in the PWM control, and represents a modulation rate M serving as a reference for switching the control mode from the voltage phase control to the current vector control. The first norm threshold V_{a_th1} is a threshold to be compared with an averaged voltage norm command value V_{a_fin_flt}* described later, and is used as a condition for switching from the voltage phase control to the current vector control. The first norm threshold V_{a_th1} is calculated by using the DC voltage V_dc of the battery 15 according to a first formula of the following Equation (13).

The second norm threshold calculation unit 95 calculates a second norm threshold V_{a_th2} based on a second modulation rate threshold M_th2. The second modulation rate threshold M_th2 is a threshold determined for the modulation rate M in the PWM control, and represents a modulation rate M serving as a reference for switching the control mode from the current vector control to the voltage phase control. The second norm threshold V_{a_th2} is a threshold to be compared with the averaged voltage norm command value V_{a_fin_flt}* described later, and is used as a condition for switching from the current vector control to the voltage phase control. The second norm threshold V_{a_th2} is calculated by using the DC voltage V_dc of the battery 15 according to a second formula of the following Equation (13).

$\begin{array}{cc}\left[\mathrm{Equation}13\right]& \\ \{\begin{array}{c}{V}_{\mathrm{a\_th}1}=\frac{V_{dc}}{\sqrt{2}}{M}_{th1}\\ V_{\mathrm{a\_th}2}=\frac{V_{dc}}{\sqrt{2}}{M}_{th2}\end{array}& \left(13\right)\end{array}$

In the present embodiment, a magnitude relationship between an upper limit value M_max* of the modulation rate M, the first modulation rate threshold M_th1, and the second modulation rate threshold M_th2 is as shown in the following Equation (14). That is, the first modulation rate threshold M_th1 related to the switching from the voltage phase control to the current vector control is the smallest among them. The second modulation rate threshold M_th2 related to the switching from the current vector control to the voltage phase control is a value between the upper limit value M_max* and the first modulation rate threshold M_th1. That is, in the present embodiment, the first modulation rate threshold M_th1 and the second modulation rate threshold M_th2 are at least different values. Therefore, the first norm threshold V_{a_th1} calculated by using the first modulation rate threshold M_th1 and the second norm threshold V_{a_th2} calculated by using the second modulation rate threshold M_th2 are also different values. As a result, hysteresis is provided for switching between the current vector control and the voltage phase control. In the present embodiment, since the switching between the current vector control and the voltage phase control is performed seamlessly, a phenomenon in which these switchings are repeated (so-called chattering) can be suppressed by providing the hysteresis in these switchings.

[Equation 14]

M _max *>M _th2 >M _th1  (14)

The upper limit value M_max* of the modulation rate M, the first modulation rate threshold Mail, and the second modulation rate threshold M_th2 are set as follows, for example. That is, the upper limit value M_max* is a value of the modulation rate M in the rectangular wave state, i.e., “1.1”. The second modulation rate threshold M_th2 is a value of the modulation rate M of a boundary between the non-overmodulation state and the overmodulation state, i.e., “1.0”. The first modulation rate threshold M_th1 is a value of the modulation rate M that can be regarded as in the non-overmodulation state, i.e., “0.9”.

The averaging filter 96 reduces noise of the d-axis final voltage command value V_{d_fin}* by averaging the d-axis final voltage command value V_{d_fin}*. The averaging filter 96 is, for example, a low-pass filter (LPF). An output of the averaging filter 96 is an averaged d-axis final voltage command value V_{d_fin_flt}*. The averaged d-axis final voltage command value V_{d_fin_flt}* is input to the norm calculation unit 98.

The averaging filter 97 reduces noise of the q-axis final voltage command value V_{q_fin}* by averaging the q-axis final voltage command value V_{q_fin}*. The averaging filter 97 is, for example, a low-pass filter (LPF). An output of the averaging filter 97 is an averaged q-axis final voltage command value V_{q_fin_flt}*. The averaged q-axis final voltage command value V_{q_fin_flt}* is input to the norm calculation unit 98.

The norm calculation unit 98 calculates the averaged voltage norm command value V_{a_fin_flt}* by using the averaged d-axis final voltage command value V_{d_fin_flt}* and the averaged q-axis final voltage command value V_{q_fin_flt}* according to the following Equation (15). Specifically, the norm calculation unit 98 calculates the averaged voltage norm command value V_{a_fin_flt}* according to the following Equation (15). The averaged voltage norm command value V_{a_fin_flt}* is input to the control mode determination unit 99.

[Equation 15]

V _{a_fin_flt}*=√{square root over (V_{d_fin_flt}*²+V_{q_fin_flt}*²)}  (15)

The output controller 3 may output the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}* in the voltage phase control as the d-axis final voltage command value V_{d_fin}* and the q-axis final voltage command value V_{q_fin}*. In this case, instead of the above averaged voltage norm command value V_{a_fin_flt}*, a final voltage norm command value V_{a_fin}* input to the vector conversion unit 49 may be input to the control mode determination unit 99.

The control mode determination unit 99 determines a control mode to be applied according to the operating state of the electric motor 9 by using the first norm threshold V_{a_th1}, the second norm threshold V_{a_th2}, and the averaged voltage norm command value V_{a_fin_flt}*. A determination result of the control mode determination unit 99 is output as the control mode signal Sm to the current vector control unit 1, the voltage phase control unit 2, and the output controller 3.

FIG. 7 is a diagram illustrating a control mode determination method performed by the control mode determination unit 99. As illustrated in FIG. 7, the control mode determination unit 99 compares the averaged voltage norm command value V_{a_fin_flt}* with the first norm threshold V_{a_th1} and the second norm threshold V_{a_th2}. As a result of the comparison, when it is detected that the averaged voltage norm command value V_{a_fin_flt}* is equal to or smaller than the first norm threshold V_{a_th1} when the electric motor 9 is controlled by the voltage phase control, the switching to the current vector control is determined based on the detection result. Then, the control mode signal Sm (for example, a control mode switching command) indicating the above is output. In addition, when it is detected that the averaged voltage norm command value V_{a_fin_flt}* is equal to or larger than the second norm threshold V_{a_th2} when the electric motor 9 is controlled by the voltage phase control, the switching to the voltage phase control is determined based on the detection result. Then, the control mode signal Sm indicating the above is output.

[Specific Configuration of Output Controller]

FIG. 8 is a block diagram illustrating a configuration of the output controller 3. FIG. 8 illustrates a switch that switches the control mode between the current vector control and the voltage phase control according to the control mode signal Sm. When the control mode is set to the current vector control based on the control mode signal Sm, the output controller 3 outputs the d-axis voltage command value V_{di_fin}* in the current vector control as the d-axis final voltage command value V_{d_fin}*. At the same time, the output controller 3 outputs the q-axis voltage command value V_{qi_fin}* in the current vector control as the q-axis final voltage command value V_{q_fin}*. On the other hand, when the control mode is set to the voltage phase control based on the control mode signal Sm, the output controller 3 outputs the d-axis voltage command value V_{dv_fin}* in the voltage phase control as the d-axis final voltage command value V_{d_fin}*. At the same time, the output controller 3 outputs the q-axis voltage command value V_{qv_fin}* in the voltage phase control as the q-axis final voltage command value V_{q_fin}*.

Effect of Present Embodiment

Effects of the electric motor system 100 and the electric motor 9 according to the present embodiment configured as described above will be described below.

FIG. 9 is a flowchart of electric motor control according to the present embodiment. As illustrated in FIG. 9, first, in step S101, the three-phase AC currents of the electric motor 9 are detected by the current detector 8, and the detected three-phase AC currents are converted into the d-axis current i_d and the q-axis current i_q by coordinate conversion processing of the converter 12. In step S102, the electric angle θ of the electric motor 9 is detected by the rotor detector 10, and the rotation speed N is calculated based on the detected electric angle θ by rotation speed calculation processing of the rotation speed calculator 11. In step S103, the torque target value T* is acquired, and the DC voltage V_dc of the battery 15 is also acquired. That is, various parameters necessary for the control of the electric motor 9 are acquired or calculated.

When the various parameters are acquired in this way, in step S104, control mode switching determination is performed by the control mode switching determination unit 13. In step S105, when it is determined that a result of the control mode switching determination is to control the electric motor 9 by the current vector control, step S106 and subsequent steps are executed. In step S106, the d-axis current target value i_d*, the q-axis current target value i_q*, and the decoupling voltage V_{d_dcpl}* are calculated by the current vector control unit 1. Thereafter, in step S107, the d-axis current feedback voltage command value V_di′ and a q-axis current feedback voltage command value V_qi′ are calculated by current feedback processing of feeding back the d-axis current target value i_d* and the q-axis current target value i_q*, respectively. Further, in step S108, the d-axis voltage command value V_{di_fin}* in the current vector control is calculated by performing decoupling control in which the decoupling voltage V_{d_dcpl_flt} processed by the LPF 22 and the current feedback voltage command value V_di′ are added. In step S108, the q-axis voltage command value V_{qi_fin}* in the current vector control is similarly calculated.

When the d-axis voltage command value V_{di_fin}* and the q-axis voltage command value V_{qi_fin}* in the current vector control are calculated in this way, the three-phase voltage command values are calculated by the converter 4 in step S110 through output control processing performed by the output controller 3 in step S109. Thereafter, the PWM signal based on the three-phase voltage command values is generated by the pulse width modulation signal generator 5, and the control of the electric motor 9 by the pulse width modulation is executed.

On the other hand, as a result of the control mode switching determination in step S104, when it is determined in step S105 that the operating state of the electric motor 9 is not suitable for the current vector control, the current phase control is executed in step S111.

FIG. 10 is a flowchart of the voltage phase control. As illustrated in FIG. 10, in the voltage phase control, the phase change processing of the electric angular velocity ω_e is executed in step S201. In step S202, the d-axis current target value i_d* and the q-axis current target value i_q* are acquired from the current vector control unit 1. In step S203, the reference magnetic flux norm φ_{0_ref} and the reference torque T_ref are calculated.

Thereafter, in step S204, the induced voltage V_a0 is calculated based on the electric angular velocity |ω_e′| after the phase change processing and the reference magnetic flux norm φ_{0_ref}. Then, in step S205, the upper voltage norm limit value V_{a_upper}, the FF voltage norm command value V_{a_ff}, and the FF voltage phase command value α_ff are calculated. In step S206, the estimated magnetic flux norm value φ_{0_est} and the estimated torque value T_est are calculated by using the d-axis current target value i_d* and the q-axis current target value i_q*. Then, in step S207, the FB voltage norm command value V_{a_fb} is calculated.

Next, the voltage norm command value V_a* is calculated by adding the FF voltage norm command value V_{a_ff} and the FB voltage norm command value V_{a_fb}. In step S208, the voltage norm command value V_a* calculated in this way is limited to the value between the predetermined lower limit value and the upper voltage norm limit value V_{a_upper}.

In step S209, when the voltage norm command value V_a* is limited by the voltage norm limitation processing, step S210 is executed, and the PI control unit 39 is initialized. On the other hand, in step S209, when the voltage norm command value V_a* is not limited by the above voltage norm limitation processing, step S210 is omitted. As a result, the voltage norm command value V_a* after the voltage norm limitation processing is input to the vector conversion unit 49.

On the other hand, in step S211, the FB voltage phase command value α_fb, is calculated. Next, the voltage phase command value α* is calculated by adding the FF voltage phase command value α_ff and the FB voltage phase command value α_fb. In step S212, the voltage phase command value α* calculated in this way is subjected to the voltage phase limitation processing. In step S213, when the voltage phase command value α* is limited by the voltage phase limitation processing, the PI control unit 46 is initialized in step S214. On the other hand, when the voltage phase command value α* is not limited by the voltage phase limitation processing in step S213, step S214 is omitted. As a result, the voltage phase command value α* after the phase limitation processing is input to the vector conversion unit 49.

In step S215, the vector conversion unit 49 converts the voltage norm command value V_a* and the voltage phase command value α* into the d-axis voltage command value V_dv* and the q-axis voltage command value V_qv*. Then, the d-axis voltage command value V_dv* and the q-axis voltage command value V_qv* are output through stabilization filter processing in step S216. Thereafter, as in the current vector control, as illustrated in FIG. 9, the three-phase voltage command values are calculated by the converter 4 in step S110 through the output control processing by the output controller 3 in step S109. Then, the PWM signal based on the three-phase voltage command values is generated by the pulse width modulation signal generator 5, and the control of the electric motor 9 by the pulse width modulation is executed.

As described above, in the voltage phase control of the present embodiment, the electric motor 9 may be controlled in the overmodulation state depending on the operating state. In addition, when the rotation speed N is detected, the rotation speed N is converted into the electric angular velocity ω_e. Then, in the command value calculation in the voltage phase control, the electric angular velocity ω_e is used. However, instead of using the electric angular velocity ω_e corresponding to the detected rotation speed N as it is, the phase change processing is executed in consideration of the required time T_s with respect to the electric angular velocity ω_e. Therefore, in the voltage phase control of the present embodiment, the calculation of the final command value according to the command value calculation model in the voltage phase control is executed by using the electric angular velocity |ω_e′| after the phase change processing. As a result, in the voltage phase control of the present embodiment, the electric motor 9 can be appropriately controlled regardless of the operating situation. That is, according to the voltage phase control of the present embodiment, as will be described below, it is possible to appropriately prevent an overcurrent that is a concern in a scene in which the voltage applied to the electric motor 9 is neither excessive nor deficient and the rotation speed N of the electric motor 9 suddenly changes when the electric motor 9 is controlled in the overmodulation state.

FIG. 11 is a graph illustrating (A) the electric angular velocity ω_e, (B) the electric angular acceleration A_e, (C) a current Im flowing through the electric motor 9, and (D) a voltage Vm applied to the electric motor 9. FIG. 11 illustrates parameters in situations where the electric motor 9 is controlled in the overmodulation state by the voltage phase control over the entire period from a time point t0 to a time point t5. In addition, during the period from the time point t0 to the time point t1, the electric motor 9 is operated at a constant rotation speed N, and the rotation speed N suddenly decreases from the time point t1. The current Im flowing through the electric motor 9 is, for example, the d-axis current i_d, the q-axis current i_q, or a square average of the d-axis current i_d and the q-axis current i_q. In addition, the voltage Vm applied to the electric motor 9 is a voltage norm corresponding to the d-axis voltage command value V_{dv_fin}* and the q-axis voltage command value V_{qv_fin}*.

As illustrated in (A) in FIG. 11, the electric angular velocity ω_e obtained by converting the detected rotation speed N (detected electric angular velocity) deviates from an actual angular velocity (hereinafter, referred to as an actual angular velocity) 201 of the electric motor 9 after the time point t1 at which the rotation speed N suddenly decreases. This is because, as the rotation speed N of the electric motor 9 suddenly decreases, a slight response delay of about the required time T_s has an influence that cannot be ignored for the control of the electric motor 9.

As indicated by an alternate long and short dash line in (B) in FIG. 11, as the rotation speed N suddenly decreases from the time point t1, an actual electric angular acceleration (hereinafter referred to as an actual angular acceleration) 202 of the electric motor 9 decreases stepwise at the time point t1. At this time, as indicated by a solid line in (B) in FIG. 11, the electric angular acceleration A_e obtained by differentiating the electric angular velocity ω_e (detected electric angular velocity) is delayed in tracking due to the required time T_s. Therefore, the electric angular acceleration A_e substantially matches the actual angular acceleration 202 at the time point t3.

Therefore, in the voltage phase control of the present embodiment, the phase of the electric angular velocity ω_e is changed in consideration of the required time T_s. The phase change of the electric angular velocity ω_e is performed after the time point t2 at which the detected electric angular acceleration A_e is compared with the threshold A_e0 and the electric angular acceleration A_e is smaller than the threshold A_e0. As a result, as illustrated in (A) in FIG. 11, the electric angular velocity |ω_e′| after the phase change decreases from the time point t2 with respect to the detected electric angular velocity ω_e. In addition, as illustrated in (A) in FIG. 11, the electric angular velocity |ω_e′| after the phase change further decreases from a detected electric angular velocity ω_e side beyond the actual electric angular velocity 201. This is because the phase change is performed in consideration of the required time T_s. That is, the phase change executed in the voltage phase control of the present embodiment is not a correction or compensation for matching the detected electric angular velocity ω_e with the actual electric angular velocity 201, but is a change of advancing the phase more than the actual electric angular velocity 201. As illustrated in (A) in FIG. 11, in the present embodiment, the phase of the electric angular velocity |ω_e′| after the phase change is substantially advanced from the time point t3 to the time point t4 with respect to the actual electric angular velocity 201.

When the phase of the electric angular velocity ω_e is changed in this way and the electric angular velocity |ω_e′| after the phase change is used for the command value calculation in the voltage phase control, the current Im flowing through the electric motor 9 is suppressed to be equal to or smaller than an upper limit value of the current that can flow to the electric motor 9 (hereinafter referred to as an upper current limit I_lim). Specifically, as illustrated in a graph GI1 in (C) in FIG. 11, when the command value calculation in the voltage phase control is performed by using the detected electric angular velocity ω_e without changing the phase, the current flowing through the electric motor 9 continues to increase even after the time point t3 at which the electric angular acceleration A_e is settled to the actual angular acceleration 202. As a result, the current flowing through the electric motor 9 becomes an overcurrent exceeding the upper current limit I_lim. On the other hand, as illustrated in a graph GI2 in (C) in FIG. 11, when the final command value is calculated according to the command value calculation model in the voltage phase control by using the electric angular velocity ω_e after the phase change, the current flowing through the electric motor 9 after the time point t3 is suppressed from increasing and exhibits a generally flat transition. Therefore, the current flowing through the electric motor 9 does not exceed the upper current limit I_lim. The upper current limit I_lim is determined in advance according to durability and the like of the electric motor 9.

As illustrated in a graph GV1 in (D) in FIG. 11, when the final command value is calculated according to the command value calculation in the voltage phase control by using the detected electric angular velocity ω_e without changing the phase, a decrease of the voltage Vm applied to the electric motor 9 is gradual even though the rotation speed N suddenly decreases. This is one of reasons why the current flowing through the electric motor 9 continues to increase as described above (see graph GI1). On the other hand, as illustrated in a graph GV2 in (C) in FIG. 11, when the command value calculation in the voltage phase control is performed by using the electric angular velocity ω_e after the phase change, the voltage Vm applied to the electric motor 9 is decreased more quickly than in the graph GV1. Accordingly, the increase in the current flowing through the electric motor 9 is suppressed (see graph GI2).

The phase change processing is included in the voltage phase control. That is, instead of adding a change to the result of the command value calculation in the voltage phase control, the voltage applied to the electric motor 9 is appropriately adjusted in a state including the phase change processing in the voltage phase control. Therefore, in FIG. 11, transition of each parameter when the rotation speed N suddenly decreases has been described. When the rotation speed N changes gradually, naturally, the electric motor 9 is appropriately controlled without causing the excess or deficiency of the voltage or the like due to the voltage phase control.

In the phase change of the electric angular velocity ω_e performed in the present embodiment, the phase is advanced more than the actual electric angular velocity 201. However, in order to prevent an adverse effect caused by excessively advancing the phase, the phase of the electric angular velocity ω_e is advanced in a range that does not exceed the change amount that can be generated in the electric angular velocity ω_e per required time T_s. That is, the change amount of the phase of the electric angular velocity ω_e is set to be within a range of a delay time that is included from the detection of the rotation speed N to the application of the voltage to the electric motor 9. For example, in an ideal system in which the required time T_s is completely zero, if the voltage to be applied to the electric motor 9 can be output without a delay, the overcurrent can be suppressed, which cannot be achieved. On the other hand, if an excessive phase change is performed to exceed the delay time included in the system, the excess or deficiency of the voltage occurs, and a risk of an overcurrent may increase. Therefore, as described above, the phase change processing is preferably performed only within the range of the delay time included in the system.

In the above embodiment, the electric angular velocity ω_e is used for the command value calculation in the voltage phase control, and the command value calculation using the rotation speed N may be set without converting to the electric angular velocity ω_e. In this case, similar to the above embodiment, the phase change processing is performed on the rotation speed N. That is, the phase change of the electric angular velocity ω_e is an example of the change of the rotation speed N. Similarly, the above phase change processing can be performed on a rotation speed parameter other than the rotation speed N or the electric angular velocity ω_e. That is, instead of performing the phase change processing on the rotation speed N or the electric angular velocity ω_e, the phase change processing may be performed on a rotation speed parameter having a correlation with the rotation speed N or the electric angular velocity ω_e. For example, when the induced voltage V_a0 is calculated by using the electric angular velocity ω_e before the phase change, the phase change processing may be performed on the induced voltage V_a0.

In the above embodiment, the required time T_s is estimated by using the wasted time related to the detection of the rotation speed N and the time constant τ_m of the stabilization filter 50 (see Equation (12)), which is an example of the required time T_s. For example, a plurality of delay elements that can be considered in design can be selected and taken into consideration for the required time T_s. For example, in addition to the above, the required time T_s can be set by using some or all of a sampling period of the electric angle θ, a calculation period of the rotation speed N, a wasted time generated in the pulse width modulation, a wasted time related to the output of the PWM pulse, a response delay included in the phase change unit 32 itself, and the like.

In addition, the various constants in the above embodiment can be changed to variables. For example, various constants in the above embodiment can be varied according to various parameters related to the operating states of the electric motor 9 and the inverter 6, such as the current, the rotation speed N, the torque, the voltage, or the temperature. When a risk such as a failure of hardware due to an overcurrent is high, various constants may be varied such that a correction amount of the phase change in the above embodiment increases.

In the above embodiment, the change limitation unit 52 compares the electric angular acceleration A_e with the threshold A_e0, and determines whether to change the phase of the electric angular velocity ω_e (see FIG. 5). In the determination, when the threshold A_e0 is set to zero, the phase change processing is substantially not performed when the rotation speed N increases, and the phase change processing is performed only when the rotation speed N decreases. The risk of an overcurrent is greater when the rotation speed N decreases than when the rotation speed N increases. Therefore, when the threshold A_e0 is set to zero, the phase change processing is accurately performed only in a scene in which the risk of an overcurrent is high, and thus the risk of an overcurrent is efficiently reduced. In addition, when the threshold A_e0 is set to zero, in a scene in which the rotation speed N increases, original voltage phase control in which the phase change processing is not performed is maintained. Therefore, in the scene in which the rotation speed N increases, original responsiveness of the voltage phase control is maintained. As a result, the phase change processing is performed by an excessive temporary voltage, and control efficiency of the electric motor 9 is prevented from deteriorating. This is an example of a case where the change amount of the phase change processing (phase change amount ω_ecmp) when an absolute value of the rotation speed N decreases is larger than the change amount of the phase change processing (phase change amount ω_ecmp) when the absolute value of the rotation speed N increases.

When the threshold A_e0 is set to a predetermined negative value determined in advance according to the characteristics of the electric motor 9 or the like, the phase change processing is substantially not performed at the time of a slow acceleration or deceleration. Then, the phase change processing is appropriately performed only at the time of a sudden acceleration or deceleration in which the risk of an overcurrent is high. Therefore, when the threshold A_e0 is set to a predetermined negative value, the overcurrent is efficiently prevented only at the time of a sudden acceleration or deceleration, such as a case where wheels of a vehicle equipped with the electric motor system 100 slip, a case where the wheels grip again on a road surface after slipping, or a case where the wheels are locked. In addition, at the time of a slow acceleration or deceleration assumed in a normal driver operation, the original responsiveness of the voltage phase control in which the phase change processing is not performed is maintained. Therefore, the phase change processing is performed by an excessive temporary voltage, and the control efficiency of the electric motor 9 is prevented from deteriorating.

In the above embodiment, the change limitation unit 52 compares the electric angular acceleration A_e with the threshold A_e0, and determines whether to change the phase of the electric angular velocity ω_e, but the change limitation unit 52 may be omitted. FIG. 12 is a block diagram illustrating a configuration of the phase change unit 32 according to a modification. As illustrated in FIG. 12, in the phase change unit 32, the change limitation unit 52 is omitted and the phase change amount calculation unit 53 is variable according to the electric angular acceleration A_e. Specifically, the phase change amount calculation unit 53 according to the present modification calculates the phase change amount ω_ecmp by, for example, multiplying the product of the required time T_s and the electric angular acceleration A_e by a coefficient γ (not illustrated) according to the electric angular acceleration A_e. If the coefficient γ is set to “0” when the electric angular acceleration A_e is positive and is set to “1” when the electric angular acceleration A_e is negative, the coefficient functions similarly to the case of setting the threshold A_e0 to zero in the above embodiment. In addition, if the coefficient γ is set to “0” when the electric angular acceleration A_e is larger than a predetermined negative value and is set to “1” when the electric angular acceleration A_e is equal to or smaller than the predetermined negative value, the coefficient functions similarly to the case of setting the threshold A_e0 to a predetermined negative value in the above embodiment.

Further, as described above, when the change limitation unit 52 is omitted and the phase change amount calculation unit 53 is variable according to the electric angular acceleration A_e, the electric motor 9 is controlled more flexibly and appropriately.

For example, “a coefficient γ whose value increases as the electric angular acceleration A_e decreases” can be set such that the coefficient γ when the rotation speed N increases is smaller than the coefficient γ when the rotation speed N decreases. In this case, the phase change amount ω_ecmp at the time of a deceleration is larger than the phase change amount ω_ecmp at the time of acceleration. Therefore, it is easy to appropriately achieve both the prevention of the overcurrent and the maintenance of the original responsiveness of the voltage phase control according to the operating state of the electric motor 9. This is an example of the case where the change amount of the phase change processing (phase change amount ω_ecmp) when an absolute value of the rotation speed N decreases is larger than the change amount of the phase change processing (phase change amount ω_ecmp) when the absolute value of the rotation speed N increases.

For example, “a coefficient γ whose value decreases as the ‘absolute value’ of the electric angular acceleration A_e decreases” can be set. In this case, the original responsiveness of the voltage phase control is maintained at the time of a slow acceleration or deceleration in which the absolute value of the electric angular acceleration A_e is small, and the overcurrent is prevented at the time of a sudden acceleration or deceleration in which the absolute value of the electric angular acceleration A_e is large.

As described above, the control method for the electric motor 9 in the present embodiment and/or the modification is a method of controlling the electric motor 9 that controls the electric motor 9 by the voltage phase control performed based on the voltage norm command value V_a* representing the magnitude of the voltage to be supplied to the electric motor 9 and the voltage phase command value α* representing the phase of the voltage. In the control method for the electric motor 9, the required time T_s required for calculating the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) of the voltage (voltage norm command value V_a*) applied to the electric motor 9 by the voltage phase control according to the command value calculation model for calculation by using the rotation speed parameter (electric angular velocity ω_e) related to the rotation speed N of the electric motor 9 is acquired. Then, the detected rotation speed parameter (electric angular velocity ω_e) is changed based on the required time T_s. Thereafter, the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) is calculated by using the changed rotation speed parameter (electric angular velocity ω_e) according to the command value calculation model.

Accordingly, the control method for the electric motor 9 according to the present embodiment and/or the modification can appropriately control the electric motor 9 when the electric motor 9 is controlled in the overmodulation state by the voltage phase control. That is, the overcurrent is prevented when the rotation speed N of the electric motor 9 suddenly decreases. In addition, in the change of the rotation speed parameter, the command value calculation model or the final command value calculated according to the command value calculation model is not changed. That is, the command value calculation model includes the change of the rotation speed parameter, and is set to calculate an appropriate final command value including the change of the rotation speed parameter. Therefore, when the rotation speed N of the electric motor 9 gradually increases or decreases, the voltage applied to the electric motor 9 does not become excess or deficiency.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, specifically, the rotational speed N is detected, and the rotational speed parameter is detected by calculating the electric angular velocity ω_e based on the rotational speed N. Then, the phase of the electric angular velocity ω_e is changed based on the required time T_s. Thereafter, the final command value (d-axis voltage command value V_{dv_fin}* and q-axis voltage command value V_{qv_fin}*) is calculated by using the electric angular velocity |ω_e′| after the phase change and the command value calculation model.

In this way, when the electric angular velocity ω_e is detected as the rotation speed parameter and the electric angular velocity |ω_e′| after the phase change of the electric angular velocity ω_e is used for the calculation of the final command value according to the command value calculation model, the electric motor 9 is particularly controlled appropriately. This is because, when the phase of the electric angular velocity ω_e is changed among the plurality of rotation speed parameters that can be taken into consideration, other composite elements need not to be considered, and an overcurrent prevention effect is most directly facilitated.

Specifically, in the control method for the electric motor 9 according to the present embodiment and/or the modification, the phase of the electric angular velocity ω_e is changed such that the electric angular velocity |ω_e′| after the phase change approaches the electric angular velocity ω_e after the required time T_s is elapsed. That is, instead of changing the phase such that the electric angular velocity |ω_e′| after the phase change approaches the actual electric angular velocity 201, the phase of the electric angular velocity ω_e is changed such that the electric angular velocity |ω_e′| after the phase change approaches the electric angular velocity ω_e predicted to be acquired after the required time T_s is elapsed. By changing the phase of the electric angular velocity ω_e in this way, the electric motor 9 can be particularly controlled appropriately when the electric motor 9 is controlled in the overmodulation state by the voltage phase control.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, the phase of the electric angular velocity ω_e is advanced based on the required time T_s. Therefore, a case where the phase of the electric angular velocity ω_e is advanced from the detected electric angular velocity ω_e side beyond the actual electric angular velocity 201 is included. Accordingly, when the electric motor 9 is controlled in the overmodulation state by the voltage phase control, the electric motor 9 can be particularly controlled appropriately.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, the change amount (phase change amount ω_ecmp) of the rotation speed parameter (electric angular velocity ω_e) when the absolute value of the rotation speed N decreases is set to be larger than the change amount (phase change amount ω_ecmp) of the rotation speed parameter (electric angular velocity ω_e) when the absolute value of the rotation speed N increases. Accordingly, when the electric motor 9 is controlled in the overmodulation state by the voltage phase control, the electric motor 9 can be particularly controlled appropriately. For example, since the overcurrent is likely to occur at the time of a deceleration due to a sudden change in the rotation speed N, the risk of an overcurrent can be particularly effectively reduced by performing the phase change processing specialized in the deceleration. In addition, at the time of an acceleration, the original response without phase change is maintained, and the phase change processing is performed by an instantaneous voltage fluctuation or the like, so that the control efficiency is prevented from decreasing.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, specifically, a temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is calculated, and the temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is compared with the predetermined threshold A_e0. Then, a change amount (phase change amount ω_ecmp) of the change (phase change processing) when the temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is smaller than the threshold A_e0 is set to be smaller than a change amount (phase change amount ω_ecmp) of the change (phase change processing) when the temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is equal to or larger than the threshold A_e0. Accordingly, when the electric motor 9 is controlled in the overmodulation state by the voltage phase control, the electric motor 9 can be particularly controlled appropriately. For example, since the phase change processing is performed only at the time of a sudden acceleration or deceleration such as slipping of wheels connected to the electric motor 9, the risk of an overcurrent can be effectively reduced. In addition, since the original response when the phase change processing is not performed is maintained at the time of a slow acceleration or deceleration, a risk of efficiency deterioration due to an excessive temporary voltage and a risk of controllability deterioration caused by amplification of the noise included in the detected rotation speed N and the voltage applied to the electric motor 9 becoming vibration can be reduced.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, a temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is calculated, the temporal change (electric angular acceleration A_e) of the rotation speed parameter (electric angular velocity ω_e) is compared with the predetermined threshold A_e0, and the change (phase change processing) of the rotation speed parameter (electric angular velocity ω_e) is substantially stopped when the temporal change (electric angular acceleration A_e) of the rotation speed N is smaller than the threshold A_e0. Accordingly, in the same way as described above, when the electric motor 9 is controlled in the overmodulation state by the voltage phase control, the electric motor 9 can be particularly controlled appropriately.

In the control method for the electric motor 9 according to the present embodiment and/or the modification, the change (phase change of the electric angular velocity ω_e) of the rotation speed parameter (electric angular velocity ω_e) is performed within a range that does not exceed the change amount of the rotation speed parameter (electric angular velocity ω_e) per required time T_s. Accordingly, an excessive change can be prevented, and a risk that the risk of an overcurrent is high can be reduced.

Although the embodiment of the present invention has been described above, configurations described in the above embodiment and the modification are only a part of application examples of the present invention, and are not intended to limit the technical scope of the present invention.

Claims

1. An electric motor control method for controlling an electric motor by voltage phase control performed based on a voltage norm command value representing a magnitude of a voltage to be supplied to the electric motor and a voltage phase command value representing a phase of the voltage, the method comprising: acquiring a required time required for calculating a final command value of a voltage applied to the electric motor by the voltage phase control according to a command value calculation model for calculation by using a rotation speed parameter related to a rotation speed of the electric motor;detecting the rotation speed parameter;changing the rotation speed parameter by advancing a phase of the detected rotation speed parameter based on the required time such that the rotation speed parameter approaches the rotation speed parameter after the required time is elapsed; andcalculating the final command value by using the changed rotation speed parameter according to the command value calculation model.

2. The electric motor control method according to claim 1, wherein the rotation speed parameter is an electric angular velocity.

3. (canceled)

4. (canceled)

5. The electric motor control method according to claim 1, wherein setting a change amount of the rotation speed parameter when an absolute value of the rotation speed decreases to be larger than a change amount of the rotation speed parameter when the absolute value of the rotation speed increases.

6. The electric motor control method according to claim 1, wherein calculating a temporal change in the rotation speed parameter;comparing the temporal change in the rotation speed parameter with a predetermined threshold; andsetting a change amount of the rotation speed parameter when the temporal change of the rotation speed parameter is smaller than the threshold to be smaller than a change amount of the rotation speed parameter when the temporal change of the rotation speed parameter is equal to or larger than the threshold.

7. The electric motor control method according to claim 1, wherein calculating a temporal change in the rotation speed parameter,comparing the temporal change in the rotation speed parameter with a predetermined threshold, andstopping to change the rotation speed parameter when the temporal change in the rotation speed parameter is smaller than the threshold.

8. The electric motor control method according to claim 1, wherein changing the rotation speed parameter within a range that does not exceed a change amount of the rotation speed parameter per the required time.

9. An electric motor system comprising: an electric motor; anda controller configured to control the electric motor by voltage phase control performed based on a voltage norm command value representing a magnitude of a voltage to be supplied to the electric motor and a voltage phase command value representing a phase of the voltage, whereinthe controller is configured to acquire a required time required for calculating a final command value of a voltage applied to the electric motor by the voltage phase control according to a command value calculation model for calculation by using a rotation speed parameter related to a rotation speed of the electric motor,detect the rotation speed parameter,change the rotation speed parameter by advancing a phase of the detected rotation speed parameter based on the required time such that the rotation speed parameter approaches the rotation speed parameter after the required time is elapsed, andcalculate the final command value by using the changed rotation speed parameter according to the command value calculation model.