Motor control device and electric power steering device equipped with same

文档序号:1449603 发布日期:2020-02-18 浏览:29次 中文

阅读说明:本技术 电动机控制装置以及搭载了该电动机控制装置的电动助力转向装置 (Motor control device and electric power steering device equipped with same ) 是由 高濑博明 皆木亮 泽田英树 菅原孝义 于 2018-06-12 设计创作,主要内容包括:本发明提供一种电动机控制装置,其不需要调节操作,对逆变器的死区时间进行补偿,改善电流波形的失真,提高电流控制的响应性,抑制声音、振动和扭矩脉动。本发明的基于矢量控制方式的电动机控制装置运算出dq轴的控制辅助指令值,基于控制辅助指令值运算出dq轴电流指令值,将dq轴电流指令值变换成3个相的占空比指令值,通过PWM控制的逆变器对三相无刷电动机进行驱动控制,其基于经“使用dq轴电流指令值来进行的”相位补正后的电动机旋转角运算出3相死区时间基准补偿值,通过使“通过增益以及符号等来对3相死区时间基准补偿值进行处理后得到”的死区时间补偿值与dq轴电压指令值或3相电压指令值相加来进行逆变器的死区时间补偿。(The invention provides a motor control device which does not need adjustment operation, compensates the dead time of an inverter, improves the distortion of current waveform, improves the responsiveness of current control, and restrains sound, vibration and torque ripple. A motor control device based on a vector control method calculates a control assist command value of a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value of 3 phases, and drives and controls a three-phase brushless motor by a PWM-controlled inverter, wherein dead time compensation of the inverter is performed by adding a dead time compensation value obtained by "processing a 3-phase dead time reference compensation value by gain and sign or the like" to a dq axis voltage command value or a 3-phase voltage command value based on a motor rotation angle after "phase correction performed using the dq axis current command value".)

1. A motor control device based on a vector control method, which calculates a control assist command value for a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value for 3 phases, and controls driving of a three-phase brushless motor by a PWM-controlled inverter, characterized in that:

the dead time compensation of the inverter is performed by performing phase correction on a motor rotation angle using the dq-axis current command value, calculating a 3-phase dead time reference compensation value based on the phase-corrected motor rotation angle, processing the 3-phase dead time reference compensation value by a gain and a sign, and then performing 3-phase/dq-axis conversion on the processed value to obtain a dq-axis dead time compensation value, and adding the dq-axis dead time compensation value to a dq-axis voltage command value "obtained by processing the dq-axis current command value".

2. A motor control device based on a vector control method, which calculates a control assist command value for a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value for 3 phases, and controls driving of a three-phase brushless motor by a PWM-controlled inverter, characterized in that:

the dead time compensation of the inverter is performed by performing phase correction on a motor rotation angle using the dq-axis current command value, calculating a 3-phase dead time reference compensation value based on the phase-corrected motor rotation angle, processing the 3-phase dead time reference compensation value by a gain and a sign to obtain a 3-phase dead time compensation value, and adding the 3-phase dead time compensation value to a 3-phase voltage command value "obtained by dq-axis space vector modulation".

3. The motor control device according to claim 1, characterized in that:

has a function of adjusting the dq-axis dead time compensation value based on the control assist command value.

4. The motor control device according to claim 2, characterized in that:

has a function of adjusting the 3-phase dead time compensation value based on the control assist command value.

5. The motor control device according to any one of claims 1 to 4, characterized in that:

the phase of the motor rotation angle is changed according to the motor rotation speed.

6. A motor control device based on a vector control method, which calculates a control assist command value for a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value for 3 phases, and controls driving of a three-phase brushless motor by a PWM-controlled inverter, characterized in that:

comprises an axial current factor phase correction arithmetic unit, an angle-dead time compensation value function unit, an inverter applied voltage induction gain unit and a dead time compensation value output unit,

the shaft current factor phase correction computing means computes a phase correction rotation angle by performing phase correction on a motor rotation angle using the dq-axis current command value,

the angle-dead time compensation value function means calculates a 3-phase dead time reference compensation value based on the phase correction rotation angle,

the inverter applied voltage induction gain unit calculates a voltage induction gain based on an inverter applied voltage,

the dead time compensation value output unit obtains a dq axis dead time compensation value by multiplying the voltage induced gain by the 3-phase dead time reference compensation value and then performing 3-phase/dq-axis conversion on the multiplication result, and outputs the dq axis dead time compensation value,

the motor control device performs dead time compensation of the inverter by adding the dq-axis dead time compensation value to a dq-axis voltage command value obtained by processing the dq-axis current command value.

7. The motor control device according to claim 6, characterized in that:

the dead time compensation value output unit is composed of a multiplication unit and a 3-phase alternating current/dq-axis conversion unit,

the multiplication unit multiplies the voltage induced gain by the 3-phase dead time reference compensation value,

the 3-phase AC/dq-axis conversion unit converts the 3-phase output of the multiplication unit into the dq-axis dead time compensation value.

8. The motor control device according to claim 6 or 7, characterized in that:

a current instruction value induction gain operation unit is provided,

the current command value induction gain operation unit calculates a current command value induction gain of "changing a compensation amount of the dq-axis dead time compensation value according to the control assist command value".

9. A motor control device based on a vector control method, which calculates a control assist command value for a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value for 3 phases, and controls driving of a three-phase brushless motor by a PWM-controlled inverter, characterized in that:

comprises a space vector modulation unit, an axial current factor phase correction operation unit, an angle-dead time compensation value function unit, an inverter applied voltage induction gain unit, a dead time compensation value output unit, and a current command value induction gain operation unit,

the space vector modulation unit obtains a 3-phase voltage command value by space vector modulating the dq-axis current command value,

the shaft current factor phase correction computing means computes a phase correction rotation angle by performing phase correction on a motor rotation angle using the dq-axis current command value,

the angle-dead time compensation value function means calculates a 3-phase dead time reference compensation value based on the phase correction rotation angle,

the inverter applied voltage induction gain unit calculates a voltage induction gain based on an inverter applied voltage,

the dead time compensation value output unit finds a first 3-phase dead time compensation value by multiplying the voltage induced gain by the 3-phase dead time reference compensation value, and outputs the first 3-phase dead time compensation value,

the current command value induction gain operation unit calculates a current command value induction gain of changing the compensation amount of the first 3-phase dead time compensation value according to the control assist command value,

the motor control device obtains a second 3-phase dead time compensation value by multiplying the current command value induction gain by the first 3-phase dead time compensation value, and performs dead time compensation of the inverter by adding the second 3-phase dead time compensation value to the 3-phase voltage command value.

10. The motor control device according to claim 9, characterized in that:

the current instruction value induction gain operation unit is composed of a current control delay model, a compensation sign estimation unit, a current instruction value induction gain unit and a multiplication unit,

the current control delay model inputs the control auxiliary command value to compensate the delay of the current,

the compensation sign estimation unit estimates a sign of an output of the current control delay model,

the current command value induction gain unit outputs the current command value induction gain based on an output of the current control delay model,

the multiplication unit multiplies the sign by the current instruction value induction gain.

11. The motor control device according to any one of claims 6 to 10, characterized in that:

the axial current factor phase correction computing means includes dq-axis current ratio computing means and fluctuation angle computing means,

the dq-axis current ratio arithmetic unit calculates a current ratio of the dq-axis current command value,

the variation angle calculation means obtains a variation angle based on the current ratio,

the shaft current factor phase correction arithmetic means performs phase correction on the motor rotation angle by using the variation angle.

12. An electric power steering apparatus characterized in that:

the motor control device according to any one of claims 1 to 11 is mounted, and the assist torque is applied to a steering mechanism of the vehicle.

Technical Field

The present invention relates to a motor control device that performs vector control of driving a three-phase brushless motor using a dq-axis rotation coordinate system and compensates for a dead time of an inverter based on a function of a motor rotation angle (electrical angle) (dq-axis angle or 3-phase angle — dead time compensation value reference table) to realize "smooth and abnormal noise-suppressed" control, and an electric power steering device equipped with the motor control device.

Background

Examples of the "device having the motor control device" include an Electric Power Steering (EPS) device, an electric vehicle, and a machine tool, which "applies a steering assist force (assist force) to a steering mechanism of a vehicle by using a rotational force of a motor". The electric power steering apparatus applies a steering assist force to a steering shaft or a rack shaft by a transmission mechanism such as a gear or a belt via a reduction gear of a driving force of an electric motor as an actuator. Such a conventional electric power steering apparatus performs feedback control of a motor current in order to accurately generate a torque of a steering assist force. The feedback control adjusts the motor applied voltage so as to reduce the difference between the steering assist command value (current command value) and the motor current detection value, and generally adjusts the motor applied voltage by adjusting the duty ratio of PWM (pulse width modulation) control.

A general structure of the electric power steering apparatus will be described with reference to fig. 1. As shown in fig. 1, a column shaft (steering shaft or steering wheel shaft) 2 of a steering wheel (steering wheel) 1 is connected to steered wheels 8L and 8R via a reduction gear 3, universal joints 4a and 4b, a rack and pinion mechanism 5, tie rods 6a and 6b, and hub units 7a and 7 b. Further, a steering angle sensor 14 for detecting a steering angle θ of the steering wheel 1 and a torque sensor 10 for detecting a steering torque Th of the steering wheel 1 are provided on the column shaft 2, and an electric motor 20 for assisting a steering force of the steering wheel 1 is connected to the column shaft 2 through a reduction gear 3. The battery 13 supplies power to a control unit (ECU)30 for controlling the electric power steering apparatus, and an ignition signal is input to the control unit 30 via an ignition switch 11. The control unit 30 calculates a current command value of an assist (steering assist) command based on the steering torque Th detected by the torque sensor 10 and the vehicle speed Vs detected by the vehicle speed sensor 12, and controls the current supplied to the electric motor 20 by a voltage control command value Vref obtained by compensating the calculated current command value. Further, the steering angle sensor 14 is not essential, the steering angle sensor 14 may not be provided, and the steering angle (motor rotation angle) θ may be obtained from a rotation sensor such as a resolver connected to the motor 20.

In addition, a CAN (Controller Area Network) 40 for transmitting and receiving various information of the vehicle is connected to the control unit 30, and the vehicle speed Vs CAN also be obtained from the CAN 40. Further, a non-CAN 41 for transmitting and receiving communication other than CAN40, analog/digital signals, electric waves, and the like may also be connected to the control unit 30.

In such an electric power steering apparatus, the control Unit 30 is mainly configured by a CPU (Central processing Unit) (including an MPU (Micro Processor Unit), an MCU (Micro controller Unit), and the like), and general functions executed by a program in the CPU are, for example, as shown in the configuration of fig. 2.

The function and operation of the control unit 30 will be described with reference to fig. 2. As shown in fig. 2, the steering torque Th from the torque sensor 10 and the vehicle speed Vs from the vehicle speed sensor 12 are input to the steering assist command value calculation unit 31. The steering assist command value calculation unit 31 calculates a steering assist command value Iref1 using an assist map (assist map) or the like based on the steering torque Th and the vehicle speed Vs. The calculated steering assist command value Iref1 is added to the compensation signal CM from the compensation unit 34 for improving the characteristics in the addition unit 32A, the maximum value of the added steering assist command value Iref2 is limited in the current limiting unit 33, and the current command value Irefm limited in the maximum value is input to the subtraction unit 32B so as to be subtracted from the motor current detection value Im in the subtraction unit 32B.

A PI (Proportional-Integral) control unit 35 performs current control such as PI on a deviation Δ I (Irefm-Im) that is a subtraction result obtained in the subtraction unit 32B, a voltage control command value Vref obtained after the current control is input to the PWM control unit 36 together with a modulation signal (triangular wave carrier) CF to calculate a duty command value, and PWM driving of the motor 20 is performed by a PWM signal for which the duty command value has been calculated and through an inverter 37. The motor current detector 38 detects a motor current value Im of the motor 20, and the motor current value Im detected by the motor current detector 38 is fed back to the subtracting section 32B.

In addition, the compensation unit 34 first adds the detected or estimated self-alignment torque (SAT)343 to the inertia compensation value 342 at the addition unit 344, then adds the addition result obtained at the addition unit 344 to the convergence control value 341 at the addition unit 345, and finally inputs the addition result obtained at the addition unit 345 as the compensation signal CM to the addition unit 32A so as to perform characteristic improvement.

In recent years, as an actuator of an electric power steering apparatus, a three-phase brushless motor has become the mainstream, and since the electric power steering apparatus is an in-vehicle product, its operating temperature range is wide, and from the viewpoint of fail-safe, an inverter for driving a motor in the electric power steering apparatus requires a longer dead time (that is, the dead time of a general inverter for industrial equipment < the dead time of an inverter for EPS) as compared with a general inverter for industrial use typified by a household electric appliance product. In general, since there is a delay time when a switching element (e.g., a Field-effect transistor (FET)) is turned OFF (OFF), if switching elements of upper and lower arms are turned OFF/ON simultaneously, a short circuit occurs in a dc link, and in order to prevent this, a time (dead time) is set in which both switching elements of the upper and lower arms are in an OFF (OFF) state.

As a result, the current waveform is distorted, and the responsiveness and steering feeling of the current control deteriorate. For example, when the steering wheel is in a state near the center (on-center) and the steering is performed slowly, a discontinuous steering feeling or the like due to torque pulsation or the like occurs. Further, since the interference voltage between the back electromotive force of the motor and the winding, which occurs during the middle or high speed steering, acts as external interference on the current control, the steering following performance and the steering feeling during the reverse steering are deteriorated.

A vector control method is known in which "a q-axis for controlling torque and a d-axis for controlling magnetic field strength, which are coordinate axes of a rotor of a three-phase brushless motor, are independently set, and since each axis has a relationship of 90 °, a current (a d-axis current command value and a q-axis current command value) corresponding to each axis is controlled by the vector.

Fig. 3 shows a configuration example of "a case where the three-phase brushless motor 100 is drive-controlled by a vector control method". As shown in fig. 3, steering assist command values (Iref2(idref and iqref)) (control assist command values) for two axes (dq axis coordinate system) are calculated based on the steering torque Th, the vehicle speed Vs, and the like, and the d-axis current command value id and the q-axis current command value iq for the two axes whose maximum values are restricted are input to the subtracting unit 131d and the subtracting unit 131q, respectively, and the current deviation Δ id obtained by the subtracting unit 131d and the current deviation Δ iq obtained by the subtracting unit 131q are input to the PI control unit 120d and the PI control unit 120q, respectively. The voltage command value vd obtained by PI control in the PI control unit 120d and the voltage command value vq obtained by PI control in the PI control unit 120q are input to the subtracting unit 141d and the adding unit 141q, respectively, and the command voltage Δ vd obtained by the subtracting unit 141d and the command voltage Δ vq obtained by the adding unit 141q are input to the dq-axis/3-phase alternating current converting unit 150. The voltage command values Vu, Vv, and Vw converted into 3 phases in the dq-axis/3-phase ac conversion means 150 are input to the PWM control means 160, and the motor 100 is driven by PWM signals UPWM, VPWM, and WPWM based on the calculated duty command values (Dutyu, Dutyv, and Dutyw) of the 3 phases via an inverter (inverter applied voltage VR)161 "configured by a bridge configuration having upper and lower arms as shown in fig. 4". The upper arm is composed of fet q1, fet q3, and fet q5 as switching elements, and the lower arm is composed of fet q2, fet q4, and fet q 6.

The current detector 162 detects 3-phase motor currents iu, iv, and iw of the motor 100, and the detected 3-phase motor currents iu, iv, and iw are input to the 3-phase ac/dq-axis conversion unit 130, and the feedback currents id and iq converted into 2 phases by the 3-phase ac/dq-axis conversion unit 130 are subtracted by the subtraction unit 131d and the subtraction unit 131q, respectively, and are also input to the d-q decoupling control (decoupling control) unit 140. The rotation sensor or the like is attached to the motor 100, and the "angle detection unit 110 for processing the sensor signal" outputs a motor rotation angle θ and a motor rotation speed (motor rotation speed) ω. The motor rotation angle θ is input to the dq-axis/3-phase ac conversion unit 150 and the 3-phase ac/dq-axis conversion unit 130. Also, the motor speed ω is input to the d-q decoupling control unit 140. The voltages vd1 and vq1 of the 2 phases from the d-q decoupling control unit 140 are input to the subtraction unit 141d and the addition unit 141q, respectively, and the subtraction unit 141d calculates the command voltage Δ vd and the addition unit 141q calculates the command voltage Δ vq. The command voltage Δ vd and the command voltage Δ vq are input to the dq-axis/3-phase ac conversion unit 150, and the motor 100 is driven via the PWM control unit 160 and the inverter 161.

Such an electric power steering apparatus based on the vector control method is an apparatus for "assisting the steering of the driver", and the sound, vibration, torque pulsation, and the like of the motor are transmitted as a force feeling to the driver via the steering wheel. As a power device for "driving an inverter", an FET is generally used, and in the case of a three-phase motor, FETs connected in series in upper and lower arms need to be used for each phase as shown in fig. 4 in order to energize the motor. Although ON/OFF (ON/OFF) of the FETs of the upper and lower arms is alternately repeated, an ON time and an OFF time are required since the FETs are not ideal switches and it is impossible to instantaneously perform ON/OFF of the FETs in accordance with a command of a gate signal. Therefore, if an ON command (ON command) to the FET of the upper arm and an OFF command (OFF command) to the FET of the lower arm are simultaneously input, there is a problem that "the FET of the upper arm and the FET of the lower arm are simultaneously in an ON state (ON state) and the upper and lower arms are short-circuited". Since the ON time and the OFF time of the FETs are different, when the FETs are simultaneously instructed, if "ON instruction is issued to the FET of the upper arm and the ON time is short (for example, the ON time is 100 ns)", the FETs are immediately turned ON, but if "OFF instruction is issued to the FET of the lower arm and the OFF time is long (for example, the OFF time is 400 ns)", the FETs are not immediately turned OFF, and therefore, there is a possibility that "the FET of the upper arm is instantaneously turned ON and the FET of the lower arm is turned ON (for example, ON-ON between 400ns and 100 ns)".

Therefore, by applying the ON signal to the gate drive circuit after a predetermined time "as a dead time" has elapsed, the phenomenon "the FET of the upper arm and the FET of the lower arm are simultaneously in the ON state" does not occur. Since the dead time is nonlinear, the current waveform is distorted, the response performance of the control is deteriorated, and sound, vibration, and torque ripple occur. In the case of a column shaft assist type electric power steering apparatus, since the electric motor directly connected to the gear box connected via the steering wheel and the steel column shaft is disposed very close to the driver in structure, it is necessary to take special consideration of sound, vibration, torque pulsation, and the like caused by the electric motor as compared with a downstream assist type electric power steering apparatus.

As a method of "compensating for the dead time of the inverter", in the related art, a timing of "occurrence of the dead time" is detected, a compensation value is added, and the dead time is compensated for by an external disturbance observer on the dq axis of the current control.

For example, japanese patent No. 4681453 (patent document 1) and japanese patent application laid-open No. 2015-171251 (patent document 2) disclose electric power steering apparatuses for "compensating for the dead time of the inverter". Patent document 1 discloses a dead time compensation circuit that inputs a current command value to a reference model circuit of a current control loop including a motor and an inverter, generates a model current based on the current command value, and compensates for the influence of the dead time of the inverter based on the model current. Patent document 2 includes dead time compensation means for "correcting a duty command value based on a dead time compensation value", and further includes basic compensation value calculation means for calculating a basic compensation value "which is a base value of the dead time compensation value" based on a current command value, and Filter means for performing Filter processing "corresponding to a Low Pass Filter (LPF)" on the basic compensation value.

Disclosure of Invention

Technical problem to be solved by the invention

The device of patent document 1 calculates a dead time compensation amount based on the magnitude of the q-axis current command value, and estimates a compensation sign using a 3-phase current reference model. In the case of being equal to or smaller than a prescribed fixed value, the output value of the compensation circuit is a variation value "proportional to the model current"; when the output value is equal to or larger than a predetermined fixed value, the output value of the compensation circuit is an added value obtained by adding the fixed value and the "change value proportional to the model current". Although the voltage command is output based on the current command, it is necessary to perform an adjustment operation of "determining hysteresis characteristics for outputting a prescribed fixed value".

In addition, the device of patent document 2 has a problem that when determining the dead time compensation value, dead time compensation is performed by using the q-axis current command value and the "compensation value obtained by low-pass filtering the q-axis current command value", which causes a delay, and the "final voltage command to be input to the motor is not a voltage command for operating the dead time compensation value".

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a motor control device based on a vector control method, which compensates for a dead time of an inverter so as not to require an adjustment operation, improves distortion of a current waveform, improves responsiveness of current control, and suppresses sound, vibration, and torque ripple, and an electric power steering device equipped with the motor control device.

Technical scheme for solving technical problem

The present invention relates to a motor control device based on a vector control method, which calculates a control assist command value for a dq axis, calculates a dq axis current command value based on the control assist command value, converts the dq axis current command value into a duty command value for 3 phases, and controls driving of a three-phase brushless motor by a PWM-controlled inverter, and the above object of the present invention can be achieved by: performing phase correction on a motor rotation angle using the dq-axis current command value, calculating a 3-phase dead time reference compensation value based on the phase-corrected motor rotation angle, processing the 3-phase dead time reference compensation value by a gain and a sign, and performing 3-phase/dq-axis conversion on the processed value to obtain a dq-axis dead time compensation value, and performing dead time compensation of the inverter by adding the dq-axis dead time compensation value to a dq-axis voltage command value obtained by processing the dq-axis current command value; or, the motor rotation angle is phase-corrected using the dq-axis current command value, a 3-phase dead time reference compensation value is calculated based on the phase-corrected motor rotation angle, the 3-phase dead time reference compensation value is processed by a gain and a sign to obtain a 3-phase dead time compensation value, and the 3-phase dead time compensation value is added to a 3-phase voltage command value "obtained by dq-axis space vector modulation" to perform dead time compensation of the inverter; or, the motor drive device is provided with an axial current factor phase correction arithmetic unit that calculates a phase correction rotation angle by phase correcting a motor rotation angle using the dq-axis current command value, an angle-dead time compensation value function unit that calculates a 3-phase dead time reference compensation value based on the phase correction rotation angle, an inverter applied voltage induction gain unit that calculates a voltage induction gain based on an inverter applied voltage, and a dead time compensation value output unit that obtains a dq-axis dead time compensation value by multiplying the voltage induction gain by the 3-phase dead time reference compensation value and then performing 3-phase/dq-axis conversion on the multiplication result, and outputting the dq-axis dead time compensation value, the motor control device performing dead time compensation of the inverter by adding the dq-axis dead time compensation value to a dq-axis voltage command value obtained by processing the dq-axis current command value; or, the motor control device is provided with a space vector modulation means for obtaining a 3-phase voltage command value by space vector modulation of the dq-axis current command value, an axis current factor phase correction arithmetic means for calculating a 3-phase dead time reference compensation value based on the phase correction rotation angle by phase correction of a motor rotation angle using the dq-axis current command value, an angle-dead time compensation value function means for calculating a voltage induction gain based on an inverter applied voltage, an inverter applied voltage induction gain means for calculating a voltage induction gain based on an inverter applied voltage, the dead time compensation value output unit calculates a first 3-phase dead time compensation value by multiplying the voltage induced gain by the 3-phase dead time reference compensation value, and outputs the first 3-phase dead time compensation value, the current command value induced gain operation unit calculates a current command value induced gain of "changing a compensation amount of the first 3-phase dead time compensation value according to the control assist command value", the motor control device calculates a second 3-phase dead time compensation value by multiplying the current command value induced gain by the first 3-phase dead time compensation value, and the motor control device performs dead time compensation of the inverter by adding the second 3-phase dead time compensation value to the 3-phase voltage command value.

Also, the above object of the present invention can be achieved by: any one of the motor control devices is mounted on an electric power steering device of a "steering mechanism for applying assist torque to a vehicle".

ADVANTAGEOUS EFFECTS OF INVENTION

According to the motor control device and the electric power steering device equipped with the motor control device of the present invention, the dead time compensation value of the inverter is calculated based on the function of the motor rotation angle (electric angle), and then the dead time compensation value is added to the voltage command value on the dq axis to perform compensation (that is, compensation is performed by feedforward control for the dead time compensation value and the voltage command value on the dq axis). By so doing, it is possible to compensate for the dead time of the inverter in such a manner that no adjustment operation is required, so that the distortion of the current waveform can be improved, and the responsiveness of the current control can also be improved. Also, the magnitude and direction of the dead time compensation value are adjusted and changed by controlling the auxiliary command value (iqref) so that it is not overcompensated.

Since the dead time compensation based on the function of the motor rotation angle (electrical angle) is performed, the control becomes smooth, so that the sound, vibration, and torque ripple of the motor can be suppressed. The present invention has high compensation accuracy in the low-speed and medium-speed steering regions where the phase of the motor angle matches the phase of the 3-phase current, and can compensate even when the compensation waveform of the 3 phases is not a rectangular wave.

Drawings

Fig. 1 is a schematic configuration diagram showing a general electric power steering apparatus.

Fig. 2 is a block diagram showing a configuration example of a control system of the electric power steering apparatus.

Fig. 3 is a block diagram showing an example of the configuration of the vector control method.

Fig. 4 is a wiring diagram showing a configuration example of a general inverter.

Fig. 5 is a block diagram showing a configuration example (embodiment 1) of the present invention.

Fig. 6 is a block diagram showing a configuration example (embodiment 1) of the dead time compensation unit of the present invention in detail.

Fig. 7 is a block diagram showing an example of the configuration of the current command value sensing gain unit.

Fig. 8 is a characteristic diagram showing an example of characteristics of a gain cell in the current command value sensing gain cell.

Fig. 9 is a characteristic diagram showing an example of characteristics of the current command value sense gain section.

Fig. 10 is a waveform diagram showing an example of the operation of the compensated symbol estimation unit.

Fig. 11 is a block diagram showing a configuration example of the inverter applied voltage induction gain unit.

Fig. 12 is a characteristic diagram showing an example of characteristics of the inverter applied voltage induction gain unit.

Fig. 13 is a characteristic diagram showing an example of characteristics of the phase adjusting means.

Fig. 14 is a diagram showing an example of the operation of the angle-dead time compensation value function unit (embodiment 1).

Fig. 15 is a block diagram showing an example of the structure of the space vector modulation unit.

Fig. 16 is a diagram showing an example of the operation of the space vector modulation section.

Fig. 17 is a diagram showing an example of the operation of the space vector modulation section.

Fig. 18 is a timing chart showing an example of the operation of the space vector modulation section.

Fig. 19 is a waveform diagram showing the effect of space vector modulation.

Fig. 20 is a waveform diagram showing an effect of the present invention (embodiment 1).

Fig. 21 is a waveform diagram showing an effect of the present invention (embodiment 1).

Fig. 22 is a block diagram showing a configuration example (embodiment 2) of the present invention.

Fig. 23 is a block diagram showing a configuration example (embodiment 2) of the dead time compensation unit of the present invention in detail.

Fig. 24 is a diagram showing an example of the operation of the angle-dead time compensation value function unit (embodiment 2).

Fig. 25 is a characteristic diagram showing an example of the output voltage characteristic of the angle-dead time compensation value reference table.

Fig. 26 is a waveform diagram showing an effect of the present invention (embodiment 2).

Fig. 27 is a waveform diagram showing an effect of the present invention (embodiment 2).

Fig. 28 is a block diagram showing a configuration example (embodiment 3) of the present invention.

Fig. 29 is a block diagram showing a configuration example (embodiment 3) of the dead time compensation unit of the present invention in detail.

Fig. 30 is a waveform diagram showing an effect of the present invention (embodiment 3).

Fig. 31 is a waveform diagram showing an effect of the present invention (embodiment 3).

Fig. 32 is a block diagram showing a configuration example (embodiment 4) of the present invention.

Fig. 33 is a block diagram showing a configuration example (embodiment 4) of the dead time compensation unit of the present invention in detail.

Fig. 34 is a block diagram showing an example of the configuration of the axial current factor phase correction arithmetic unit.

Fig. 35 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 36 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 37 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 38 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 39 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 40 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 41 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 42 is a waveform diagram showing an effect of the present invention (embodiment 4).

Fig. 43 is a block diagram showing a configuration example (embodiment 5) of the present invention.

Fig. 44 is a block diagram showing a configuration example (embodiment 5) of the dead time compensation unit of the present invention in detail.

Detailed Description

In order to solve the problem of "current distortion, torque ripple, steering noise deterioration, etc., due to the influence of the dead time of an inverter of a control unit (ECU, etc.)", the dead time compensation value is set to a function "corresponding to the motor rotation angle (electrical angle)", and then, the compensation is performed by adding the dq-axis dead time compensation value and the dq-axis voltage command value in a Feed Forward (FF) manner, or the compensation is performed by adding the 3-phase dead time compensation value and the 3-phase voltage command value "obtained by space vector modulation" in a feed forward manner. Obtaining a dq-axis dead time compensation value or a 3-phase dead time compensation value in advance in an off-line manner by a function corresponding to a motor rotation angle (electrical angle); creating a dq-axis angle (3-phase angle) -dead time compensation value reference table based on an output waveform of the dq-axis dead time compensation value or an output waveform of the 3-phase dead time compensation value; dead time compensation is performed by adding a dq-axis dead time compensation value to a dq-axis voltage command value in a feed-forward manner based on a dq-axis angle-dead time compensation value reference table, or dead time compensation is performed by adding a 3-phase dead time compensation value to a 3-phase voltage command value in a feed-forward manner based on a 3-phase angle-dead time compensation value reference table.

The adjustment of the appropriate dead time compensation amount and the estimation of the compensation direction can be performed by the control assist command value of the dq-axis or 3-phase command unit, and the adjustment of the appropriate dead time compensation amount can be performed based on the inverter applied voltage. In addition, the dead time compensation value "based on the motor rotation angle" can be calculated in real time, and the dead time compensation value "corresponding to the motor rotation angle" can be compensated for the dq-axis voltage command value or the 3-phase voltage command value.

In the low-speed and medium-speed control regions, the conventional 3-phase dead time compensation has a problem of "offset compensation due to a specific phase current amplitude and offset compensation due to a specific rotation speed" (in the case of EPS, there is a problem of "steering noise deterioration, steering feeling deterioration, etc.). In the conventional 3-phase dead time compensation, it is necessary to consider the magnitude of the rotational speed and the magnitude of the phase current for time adjustment, but it is difficult to obtain an optimal adjustment "capable of satisfying both" of them. In the conventional 3-phase dead time compensation, there is a problem that "accurate compensation cannot be performed" when the 3-phase compensation waveform is not a rectangular wave. In order to solve these problems, the present invention has been proposed to have an obvious effect "in a low-speed and medium-speed control state (in the case of EPS, in a low-speed and medium-speed steering state).

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

Fig. 5 corresponding to fig. 3 shows the overall structure of the present invention (embodiment 1). As shown in fig. 5, "dead time compensation unit 200 for calculating dead time compensation values vd and vq" on the dq axis "is provided. A steering assist command value (control assist command value) iqref of "q-axis equivalent to the steering assist command value Iref 2" of fig. 2 is input to the dead time compensation unit 200, and the motor rotation angle θ and the motor rotation speed ω are also input to the dead time compensation unit 200. In addition, the inverter applied voltage VR "applied to the inverter 161" is input to the dead time compensation unit 200. Also, PWM signals (UPWM, VPWM, and WPWM) from a PWM control circuit (not shown in the figure) within the PWM control unit 160 are input to the inverter 161.

The d-axis current reference value id limited by the maximum value of the steering assist reference value idref and the q-axis current reference value iq limited by the maximum value of the steering assist reference value iqref are input to the subtracting unit 131d and the subtracting unit 131q, respectively. The subtracting unit 131d calculates a current deviation Δ id between the d-axis current command value id and the feedback current id, and the subtracting unit 131q calculates a current deviation Δ iq between the q-axis current command value iq and the feedback current iq. The calculated current deviation Δ id is input to the PI control unit 120d, and the calculated current deviation Δ iq is input to the PI control unit 120 q. The d-axis voltage command value vd "obtained through PI control" and the q-axis voltage command value vq "obtained through PI control" are input to the adding means 121d and the adding means 121q, respectively. The adding unit 121d performs compensation by adding the d-axis voltage command value vd to a dead time compensation value vd from the dead time compensation unit 200 described later, and the adding unit 121q performs compensation by adding the q-axis voltage command value vq to a dead time compensation value vq from the dead time compensation unit 200 described later. The voltage value "compensated in the adding unit 121 d" is input to the subtracting unit 141d, and the voltage value "compensated in the adding unit 121 q" is input to the adding unit 141 q. The voltage vd1 from the d-q decoupling control unit 140 is input to the subtraction unit 141d, the voltage command value vd which is the difference between the d-axis voltage command value vd and the voltage vd1 is obtained by the subtraction unit 141d, the voltage vq1 from the d-q decoupling control unit 140 is input to the addition unit 141q, and the voltage command value vq which is the addition result of the addition of the q-axis voltage command value vq and the voltage vq1 is obtained by the addition unit 141 q. The "voltage command values vd and vq compensated for the dead time" are input to the "space vector modulation unit 300 for converting 2 phases of the dq axis into 3 phases of the U phase, the V phase, and the W phase, and superimposing the third harmonic". The 3-phase voltage command values Vu, Vv, and Vw "obtained by vector modulation in the space vector modulation unit 300 are input to the PWM control unit 160, and the motor 100 is drive-controlled via the PWM control unit 160 and the inverter 161 in the same manner as described above.

Next, the dead time compensation unit 200 will be explained.

Dead time compensation section 200 is composed of current control delay model 201, compensation sign estimation section 202, multiplication section 203, multiplication section 204d, multiplication section 204q, addition section 221, phase adjustment section 210, inverter applied voltage induction gain section 220, angle-dead time compensation value function section 230U, angle-dead time compensation value function section 230V, angle-dead time compensation value function section 230W, multiplication section 231U, multiplication section 231V, multiplication section 231W, 3-phase ac/dq axis conversion section 240, and current command value induction gain section 250.

Further, dead time compensation value output means is configured by multiplication means 231U, multiplication means 231V, multiplication means 231W, and 3-phase ac/dq-axis conversion means 240. Further, current command value inductive gain operation means is configured by current control delay model 201, compensation sign estimation means 202, current command value inductive gain means 250, and multiplication means 203.

Fig. 6 shows a detailed structure of the dead time compensation unit 200. Next, the dead time compensation unit 200 will be described with reference to fig. 6.

As shown in fig. 6, the q-axis steering assist command value iqref is input to the current control delay model 201. A delay occurs due to a noise filter of the ECU or the like, and the current command values id and iq up to the dq axis are reflected on the actual current. Therefore, if the sign is determined based directly on the current command value iq, timing may be deviated. To solve this problem, in the present embodiment, the phase difference is improved by approximating the delay of the entire current control to a first order filter model. The current control delay model 201 of the present embodiment is a first-order filter "represented by the following formula 1 with T as a filter time constant". Also, the current control delay model 201 may be a model of a second or higher order filter.

Formula 1

Figure BDA0002312799260000121

The current command value Icm "output from the current control delay model 201" is input to the current command value sensing gain unit 250 and the compensation sign estimation unit 202. In a low current region, the dead time compensation amount may be overcompensated, and the current command value sensing gain unit 250 has a function of calculating a gain for reducing the compensation amount in accordance with the magnitude of the current command value Icm (steering assist command value iqref). The noise reduction process is performed by using a weighted average filter so that the gain for "reducing the compensation amount" does not fluctuate due to noise or the like from the current command value Icm (steering assist command value iqref).

The current command value sensing gain unit 250 has a structure as shown in fig. 7. As shown in fig. 7, the absolute value unit 251 calculates the absolute value of the current command value Icm, and the absolute value from the absolute value unit 251 is limited to the maximum value by the input limiting unit 252, and the "maximum value limited" absolute value current command value is input to the weighted average filter 254 via the scaling unit 253. The current instruction value Iam with noise reduced in the weighted average filter 254 is added and input to the subtraction unit 255. The subtracting unit 255 subtracts the specified offset OS from the current command value Iam. The reason why the offset OS is subtracted from the current command value Iam is to fix the input value equal to or smaller than the offset OS at the minimum gain in order to prevent contact chattering (chattering) caused by the minute current command value. The offset OS is a fixed value. The "current command value Ias obtained by subtracting the offset OS from the current command value Iam in the subtraction unit 255" is input to the gain unit 256. The gain section 256 outputs a current command value induction gain Gc according to the gain characteristic as shown in fig. 8.

The current command value induction gain Gc output from the current command value induction gain unit 250 has characteristics for the input current command value Icm as shown in fig. 9, for example. That is, the current command value induction gain Gc has a characteristic that, when the current command value Icm is equal to or smaller than the prescribed current Icm1, the current command value induction gain Gc is a fixed gain Gcc 1; when the current command value Icm is larger than the specified current Icm1 and smaller than the specified current Icm2 (> the specified current Icm1), the current command value induction gain Gc increases linearly (or nonlinearly) with an increase in the current command value Icm; when the current command value Icm is equal to or greater than the prescribed current Icm2, the current command value induction gain Gc is a fixed gain Gcc2 and remains unchanged. The predetermined current Icm1 may be 0.

The compensation sign estimation unit 202 outputs a positive (+1) or negative (-1) compensation sign SN in accordance with the hysteresis characteristics shown in fig. 10 (a) and 10 (B) for the input current command value Icm. Although the compensation symbol SN is estimated with reference to the "zero cross point of the current command value Icm", the compensation symbol SN has a hysteresis characteristic in order to suppress contact chattering. The estimated compensation symbol SN is input to the multiplication unit 202. Further, the positive/negative threshold of the hysteresis characteristic may be appropriately changed.

In the case where the sign of the dead time compensation value is decided simply based on the sign of the current of the phase current command value model, contact chattering occurs at low load. When the driver turns the steering wheel slightly to the left or right at the center, torque pulsation occurs. To solve this problem, hysteresis is set in the symbol determination. In cases other than the case where "the sign changes due to exceeding the set current value", the current sign is held, thereby suppressing contact chattering.

The current command value induction gain Gc from the current command value induction gain unit 250 is input to the multiplication unit 203. Multiplication section 203 outputs "current command value induction gain Gcs (Gc × SN)" obtained by multiplying "current command value induction gain Gc by compensation symbol SN. The current command value induction gain Gcs is input to the multiplication unit 204d and the multiplication unit 204 q.

Since the most appropriate dead time compensation amount changes depending on the inverter applied voltage VR, the dead time compensation amount is changed by calculating the voltage induced gain Gv "corresponding to the inverter applied voltage VR" in the present embodiment. The inverter applied voltage induction gain unit 220 for "inputting the inverter applied voltage VR and outputting the voltage induction gain Gv" has the configuration shown in fig. 11. As shown in fig. 11, the inverter applied voltage VR is limited to the positive/negative maximum value by the input limiting unit 221, and the "maximum value limited" inverter applied voltage VR1 is input to the inverter applied voltage/dead time compensation gain conversion table 222. The characteristics of the inverter applied voltage/dead time compensation gain conversion table 222 are, for example, the characteristics shown in fig. 12. The inverter applied voltages of 9.0V and 15.0V and the voltage induction gains "0.7" and "1.2" at the inflection points are merely an example, and they may be appropriately modified. The calculated voltage-induced gain Gv is input to the multiplication section 231U, the multiplication section 231V, and the multiplication section 231W.

In the present embodiment, when it is desired to advance or retard the timing of the dead time compensation in accordance with the motor rotation speed ω, the phase adjustment means 210 has a function of "calculating the adjustment angle in accordance with the motor rotation speed ω". In the case of lead angle control, the phase adjustment unit 210 has the characteristics as shown in fig. 13. The phase adjustment angle Δ θ calculated by the phase adjustment unit 210 is input to the addition unit 221, and the addition unit 221 adds the phase adjustment angle Δ θ to the detected motor rotation angle θ. The motor rotation angle θ m (═ θ + Δ θ) which is the addition result of the addition unit 221 is input to the angle-dead-time compensation value function unit 230U, the angle-dead-time compensation value function unit 230V, and the angle-dead-time compensation value function unit 230W, and is also input to the 3-phase alternating current/dq-axis conversion unit 240.

There is a time lag of several tens [ mu ] s to hundreds [ mu ] s from "detecting the motor rotation angle and then calculating the duty command value" to "the calculated duty command value is actually reflected in the PWM signal". Since the motor is rotating during this period, a phase shift is generated between "the motor rotation angle at the time of calculation" and "the motor rotation angle at the time of reflection". In order to compensate for this phase shift, the phase is adjusted by performing advance angle control in accordance with the motor rotation speed ω.

As shown in detail in fig. 14, the angle-dead-time compensation value function unit 230U, the angle-dead-time compensation value function unit 230V, and the angle-dead-time compensation value function unit 230W output the respective phase dead-time reference compensation values Udt, Vdt, and Wdt of rectangular waves having phases different by 120[ deg ] from each other in the range of the electrical angle 0[ deg ] to 359[ deg ] with respect to the phase-adjusted motor rotation angle θ m. The angle-dead-time compensation value function unit 230U, the angle-dead-time compensation value function unit 230V, and the angle-dead-time compensation value function unit 230W set the dead-time reference compensation values required for all of the 3 phases as angle-based functions, perform real-time calculation in the ECU, and output the dead-time reference compensation values Udt, Vdt, and Wdt for the 3 phases. Also, the angular function of the dead time reference compensation value differs depending on the characteristic of the dead time of the ECU.

The dead time reference compensation values Udt, Vdt, and Wdt are input to the multiplication unit 231U, the multiplication unit 231V, and the multiplication unit 231W, respectively, so as to be multiplied by the voltage induced gain Gv, respectively. The 3-phase dead time compensation values Udtc (═ Gv · Udt), Vdtc (═ Gv · Vdt), and Wdtc (═ Gv · Wdt) obtained by multiplying the voltage-induced gain Gv are input to the 3-phase ac/dq-axis conversion unit 240. The 3-phase ac/dq-axis conversion unit 240 converts the dead time compensation values Udtc, Vdtc, and Wdtc of the 3 phases into dead time compensation values vda × and vqa × of the dq-axes of the 2 phases in synchronization with the motor rotation angle θ m. Dead time compensation values vda and vqa are input to the multiplication unit 204d and the multiplication unit 204q, respectively, so that they are multiplied by the current command value induction gain Gcs, respectively. The multiplication result obtained in the multiplication unit 204d is a dead time compensation value vd, and the multiplication result obtained in the multiplication unit 204q is a dead time compensation value vq. The dead time compensation value vd is added to the voltage command value vd in the adding unit 121d, and then is inputted to the space vector modulation unit 300 as the voltage command value vd via the subtracting unit 141d, and the dead time compensation value vq is added to the voltage command value vq in the adding unit 121q, and then is inputted to the space vector modulation unit 300 as the voltage command value vq via the adding unit 141 q.

As described above, embodiment 1 employs a configuration in which the dead time compensation value is set as a function of 3 phases "corresponding to the motor rotation angle (electrical angle)," and compensation is performed by performing 3-phase/dq-axis conversion on the dead time compensation value, and then adding the dead time compensation value of the dq axis "obtained by the 3-phase/dq-axis conversion" to the voltage command value on the dq axis in a feed-forward manner. When estimating the "compensation sign of dead time", the steering assist command value of the dq axis is used, and also, the amount of compensation is changed so as to have the most appropriate magnitude in accordance with the magnitude of the steering assist command value iqref and the magnitude of the inverter applied voltage VR.

Next, the space vector modulation will be explained.

As shown in fig. 15, the space vector modulation unit 300 may have a function of converting 2-phase voltages (vd ×, vq ×) in the dq-axis space into 3-phase voltages (Vua, Vva, Vwa) and superimposing a third harmonic on the 3-phase voltages (Vua, Vva, Vwa). For example, the space vector modulation unit 300 may use the methods of space vector modulation described in japanese patent laid-open publication nos. 2017-70066 and WO/2017/098840 previously proposed by the present applicant.

That is, the space vector modulation has a function of performing coordinate conversion as described below based ON voltage command values vd and vq, a motor rotation angle θ, and sector numbers n (#1 to #6) in dq axis space, and controlling ON/OFF of "FETs for inverter of bridge configuration (upper arm FET q1, FET q3, and FET q5 and lower arm FET q2, FET q4, and FET q6), and supplying switching patterns S1 to S6 corresponding to sectors #1 to #6 to the motor so as to control rotation of the motor, and regarding the coordinate conversion, in the space vector modulation, the voltage command values vd and vq are converted into voltage vectors V α and V β in a coordinate system α to β based ON the following equation 2 and by coordinate conversion, and the coordinate conversion is used for coordinate conversion between the coordinates and coordinates of the motor 3616.

Formula 2

Figure BDA0002312799260000161

The relationship between the target voltage vector in the d-q coordinate system and the target voltage vector in the α - β coordinate system is expressed by the following equation 3, and the absolute value of the target voltage vector V is stored.

Formula 3

Figure BDA0002312799260000162

In the switching pattern of the space vector control, the output voltages of the inverter are defined by the switching patterns S1 to S6 of the FETs (Q1 to Q6) and by the 8 kinds of discrete reference output voltage vectors V0 to V7 (non-zero voltage vectors V1 to V6 and zero voltage vectors V0 and V7 which differ in phase by pi/3 [ rad ]) as shown in the space vector diagram of fig. 17, and the selection of these reference output voltage vectors V0 to V7 and the generation times thereof are controlled, and by using 6 regions sandwiched by adjacent reference output voltage vectors, the space vector can be divided into 6 sectors #1 to #6, the target voltage vector V belongs to one of the sectors #1 to #6, and a sector number can be assigned, and based on the rotation angle γ in the coordinate system of α to-38 of the target voltage vector V, it is possible to determine the rotation angle γ as "the target voltage vector V α and V β is a target voltage vector V + Q" which is present in the space vector, and the rotation angle δ -V + Q is determined by the hexagonal voltage vector δ -V + Q "and the rotation angle" which is represented by the hexagonal voltage d-V + Q "and" indicated by the hexagonal voltage vector.

Fig. 18 shows a basic timing chart of "the switching pulse width and the time in the ON/OFF signals S1 to S6 (switching pattern) for the FETs are determined so that the target voltage vector V is output from the inverter by digital control of the switching patterns S1, S3, and S5 of the inverter by space vector control". The space vector modulation performs an operation or the like in each predetermined sampling period Ts, and converts the operation result into a switching pulse width and the time in the switching patterns S1 to S6 in the next sampling period Ts, and outputs them.

The space vector modulation generates the switching patterns S1 to S6 corresponding to the sector number "found based on the target voltage vector V". Fig. 18 shows an example of switching patterns S1 to S6 of FETs of the inverter in the case of sector number #1(n ═ 1). Signals S1, S3, and S5 indicate gate signals of fet q1, fet q3, and fet q5 "corresponding to the upper arm". The horizontal axis represents time, Ts is a period "corresponding to the switching period, divided into 8 periods, consisting of T0/4, T1/2, T2/2, T0/4, T0/4, T2/2, T1/2, and T0/4". The time periods T1 and T2 are times dependent on the sector number n and the rotation angle γ, respectively.

In the case where there is no space vector modulation, the dead time compensation value waveform (U-phase waveform) in which "the dead time compensation of the present embodiment is applied to the dq axis and only the dead time compensation value is subjected to dq axis/3 phase conversion" becomes a waveform from which the third harmonic component is removed as shown by a broken line in fig. 19. The same applies to the V phase and the W phase. By applying space vector modulation instead of dq-axis/3-phase transformation, it is possible to superimpose the third harmonic on the 3-phase signal, and to compensate for the third harmonic component lost by 3-phase transformation, and it is possible to generate an ideal dead time compensation waveform like the solid line of fig. 19.

Fig. 20 and 21 are simulation results showing the effect of the present embodiment, and fig. 20 shows the U-axis current, the d-axis current, and the q-axis current in the case of "no compensation of dead time". By applying the dead time compensation of the present embodiment, it can be confirmed that "in the steering operation state of the low-speed and medium-speed steering, as shown in fig. 21, the waveform distortion of the phase current and the dq-axis current is improved (the ripple in the waveform of the dq-axis current is small, the waveform of the phase current is close to a sine wave)", and "the torque ripple and the steering noise at the time of steering are both improved".

Note that, in fig. 20 and 21, only the U-phase current is representatively shown.

Next, embodiment 2 of the present invention will be explained.

Fig. 22 corresponding to fig. 5 shows the overall structure of the present invention (embodiment 2). As shown in fig. 22, "dead time compensation unit 200A for calculating dead time compensation values vd and vq" on the dq axis "is provided. Fig. 23 shows a detailed structure of the dead time compensation unit 200A, and the dead time compensation unit 200A will be described below with reference to fig. 23.

Dead time compensation section 200A is provided with "current control delay model 201, compensation sign estimation section 202, phase adjustment section 210, inverter applied voltage induced gain section 220, addition section 221, multiplication section 203, multiplication section 204d, multiplication section 204q, and current command value induced gain section 250, which have the same configuration as that of embodiment 1 and perform the same operation as that of embodiment 1". In embodiment 2, a d-axis angle-dead-time compensation value reference table 260d in which "the motor rotation angle θ m from the addition unit 221 is input and the d-axis dead-time reference compensation value vda is output" and a q-axis angle-dead-time compensation value reference table 260q in which "the motor rotation angle θ m from the addition unit 221 is input and the q-axis dead-time reference compensation value vqa" is output are provided. The dead time reference compensation values vda and vqa are input to the multiplication unit 205d and the multiplication unit 205q, respectively, so as to be multiplied by the voltage induced gain Gv from the inverter applied voltage induced gain unit 220, respectively. The dead time compensation values vdb and vqb "obtained by multiplying the voltage sense gain Gv" are input to the multiplication units 204d and 204q, respectively. The current command value induction gain Gcs is input to the multiplication unit 204d and the multiplication unit 204 q. The multiplying unit 204d outputs a dead time compensation value vd which is a multiplication result of the current command value sensing gain Gcs multiplied by the dead time compensation value vdb, and the multiplying unit 204q outputs a dead time compensation value vq which is a multiplication result of the current command value sensing gain Gcs multiplied by the dead time compensation value vqb.

As shown in detail in fig. 24, the angle-dead time compensation value reference tables 260d and 260q calculate the dead time compensation value "as a function of the angle required for 3 phases" in an off-line manner, and convert it into the dead time compensation value on the dq axis. That is, as described in embodiment 1, the angle-dead-time compensation value function units 230U, 230V, and 230W output the respective phase dead-time reference compensation values Udt, Vdt, and Wdt of rectangular waves having phases different by 120[ deg ] within the range of the electrical angle 0[ deg ] to 359[ deg ] with respect to the phase-adjusted motor rotation angle θ m. The angle-dead-time compensation value function units 230U, 230V, and 230W set the dead-time reference compensation values required for all of the 3 phases as angle-based functions, perform calculations in an off-line manner, and output dead-time reference compensation values Udt, Vdt, and Wdt. The angular functions of the dead time reference compensation values Udt, Vdt, and Wdt differ according to the characteristics of the dead time of the ECU.

The dead time reference compensation values Udt, Vdt, and Wdt are input to the 3-phase alternating current/dq-axis conversion unit 261, and then converted into a d-axis dead time compensation value DTd and a q-axis dead time compensation value DTq "having an output waveform as shown in fig. 24". The "angle (θ m) -based input" angle-dead time compensation value reference tables 260d and 260q are generated based on the dq-axis output waveform of fig. 24. As shown in fig. 25 a, the d-axis angle-dead time compensation value reference table 260d has an output voltage characteristic (d-axis dead time reference compensation value) of "a waveform in a zigzag shape with respect to the motor rotation angle θ m", and as shown in fig. 25B, the q-axis angle-dead time compensation value reference table 260q has an output voltage characteristic (q-axis dead time reference compensation value) of "a waveform in a wavy shape with an offset voltage added to the motor rotation angle θ m".

Dead time reference compensation values vda and vqa from angle-dead time compensation value reference tables 260d and 260q are input to the multiplication unit 205d and the multiplication unit 205q, respectively, so as to be multiplied by the voltage sense gain Gv, respectively. The dead time compensation values vdb and vqb of the dq axis "obtained by multiplying the voltage-induced gain Gv" are input to the multiplication unit 204d and the multiplication unit 204q, respectively, so as to be multiplied by the current command value-induced gain Gcs, respectively. The dead time compensation value vd from the multiplying unit 204d is added to the voltage command value vd in the adding unit 121d, and then is inputted to the space vector modulation unit 300 as the voltage command value vd via the subtracting unit 141d, and the dead time compensation value vq from the multiplying unit 204q is added to the voltage command value vq in the adding unit 121q, and then is inputted to the space vector modulation unit 300 as the voltage command value vq via the adding unit 141 q.

As described above, the present embodiment employs a configuration in which the dead time compensation value is calculated from the angle-dead time compensation value reference table of the function "corresponding to the motor rotation angle (electrical angle)", and then compensation is performed by directly adding the "calculated" dead time compensation value to the voltage command value on the dq axis in a feed-forward manner. When estimating the "compensation sign of dead time", the steering assist command value (iqref) is used, and the amount of compensation is changed so as to have the most appropriate magnitude in accordance with the magnitude of the steering assist command value and the magnitude of the inverter applied voltage.

Fig. 26 and 27 are simulation results based on a bench test apparatus that simulates an actual vehicle, showing the effect of embodiment 2 of "U-phase as an example". Fig. 26 shows the U-phase current, the d-axis current, and the q-axis current in the case of "no compensation of dead time". By applying the dead time compensation according to the present embodiment, it can be confirmed that "in the steering operation state of the low-speed and medium-speed steering, as shown in fig. 27, the waveform distortion of the phase current and the dq-axis current is improved (the ripple in the waveform of the dq-axis current is small and the waveform of the phase current is close to a sine wave)", and "the torque ripple and the steering noise at the time of steering are improved".

Next, fig. 28 corresponding to fig. 5 shows embodiment 3 of the present invention. Fig. 29 shows a detailed structure of the dead time compensation unit 200B. In embodiment 3, dead time compensation values Vum, Vvm, and Vwm of the 3 phases are calculated by the dead time compensation unit 200B, and then dead time compensation is performed by adding the dead time compensation values Vum, Vvm, and Vwm "calculated by the dead time compensation unit 200B" to the voltage command values Vu, Vv, and Vw of the 3 phases from the space vector modulation unit 300, respectively.

In embodiment 3, compensation value adjusting section 270 "configured by multiplying section 271U, multiplying section 271V, and multiplying section 271W" is provided, and dead time compensation value Udtc from multiplying section 231U, dead time compensation value Vdtc from multiplying section 231V, and dead time compensation value Wdtc from multiplying section 231W are inputted to multiplying section 271U, multiplying section 271V, and multiplying section 271W, respectively, so as to be multiplied by current command value sensing gain Gcs, respectively. The multiplication result of "multiplied by the current command value sensing gain Gcs" is output as dead time compensation values Vum, Vvm, and Vwm, and the addition unit 310U, the addition unit 310V, and the addition unit 310W add the dead time compensation values Vum, Vvm, and Vwm to the 3-phase voltage command values Vu, Vv, and Vw "obtained by" space vector modulation ", respectively. The voltage command values Vuc, Vvc, and Vwc "which are the addition results obtained by the addition units 310U, 310V, and 310W are input to the PWM control unit 160.

As described above, the present embodiment adopts a configuration in which the dead time compensation value is set as a function of 3 phases "corresponding to the motor rotation angle (electrical angle"), and then compensation is performed by directly adding the dead time compensation value to the voltage command values of the 3 phases in a feed-forward manner. When estimating the "compensation sign of dead time", the steering assist command value of the dq axis is used, and further, the amount of compensation is changed so as to have the most appropriate magnitude in accordance with the magnitude of the steering assist command value and the magnitude of the inverter applied voltage.

Fig. 30 and 31 show simulation results of the effect of the present embodiment "taking U phase as an example". Fig. 30 shows the U-phase current, the d-axis current, and the q-axis current in the case of "compensation without dead time". By applying the dead time compensation according to the present embodiment, it can be confirmed that "in the steering operation state of the low-speed and medium-speed steering, as shown in fig. 31, the waveform distortion of the phase current and the dq-axis current is improved (the ripple in the waveform of the dq-axis current is small and the waveform of the phase current is close to a sine wave)", and "the torque ripple and the steering noise at the time of steering are both improved".

Next, an embodiment in which a function of "performing phase correction on the motor rotation angle by using the d-axis current command value and the q-axis current command value" is added will be described.

In the middle-speed and high-speed steering regions, it is necessary to pass a current through the d-axis in order to improve the steering following performance. At this time, the phase of the phase current fluctuates with respect to the motor rotation angle (electrical angle) according to the amount of current in the d-axis. If dead time compensation is applied in a feed-forward manner in accordance with the motor rotation angle in a state where a current flows through the d-axis, the timing of the dead time compensation for the phase current may deviate, thereby causing torque ripple. In order to solve this problem, a function of "calculating a phase difference (a variation angle) between a motor rotation angle and a phase current based on a d-axis current command value and a q-axis current command value to perform phase correction on the motor rotation angle" (hereinafter, this function is referred to as a "phase correction function") is added, and dead time compensation is performed using the motor rotation angle (hereinafter, this function is referred to as a "phase correction rotation angle") after the phase correction by the phase correction function, thereby realizing dead time compensation with "almost no time shift".

Fig. 32 corresponding to fig. 5 shows a configuration example (embodiment 4) of "phase correction function is added to embodiment 1". As shown in fig. 32, dead time compensation section 200C is provided with an axis current factor phase correction arithmetic section 280 for "performing a phase correction function". Also, fig. 33 shows a detailed structure of the dead time compensation unit 200C. As shown in fig. 33, the motor rotation angle θ m, the d-axis current command value id, and the q-axis current command value iq are input to the axis current factor phase correction computing means 280, and the phase correction rotation angle θ d "output from the axis current factor phase correction computing means 280" is input to the angle-dead time compensation value function means 230U, 230V, and 230W.

The axis current factor phase correction arithmetic unit 280 calculates a current ratio based on the d-axis current command value id and the q-axis current command value iq, and calculates a fluctuation angle of the phase current "caused by the d-axis current" by performing an operation "based on an arc tangent function" (hereinafter, this operation is referred to as "arctan operation"). Next, the axis current factor phase correction arithmetic unit 280 adds the calculated variation angle to the motor rotation angle θ m, inverts the addition result obtained by the addition, and outputs the phase correction rotation angle θ d. Fig. 34 shows an example of the configuration of the axis current factor phase correction arithmetic unit 280. As shown in fig. 34, the axis current factor phase correction arithmetic unit 280 includes an absolute value unit 281, an absolute value unit 282, a divide-by-zero prevention limiting unit 283, a dq axis current ratio arithmetic unit 284, an input limiting unit 285, a variation angle arithmetic unit 286, a sign determination unit 287, an inversion processing unit 288, a multiplication unit 289A, and an addition unit 289B.

The dq-axis current ratio operation unit 284 calculates a current ratio Rdq based on the d-axis current command value id and the q-axis current command value iq. Therefore, in order to calculate the current ratio Rdq based on the d-axis current command value id and the q-axis current command value iq, the absolute value unit 281 calculates the absolute value of the d-axis current command value id and the absolute value unit 282 calculates the absolute value of the q-axis current command value iq. In addition, in order to prevent "the absolute value of the q-axis current command value iq which becomes the denominator from becoming zero when the current ratio Rdq is calculated", the absolute value of the q-axis current command value iq is limited by dividing by the zero prevention limiting unit 283. For example, the division-by-zero prevention limiting unit 283 limits the absolute value of the q-axis current command value iq by "setting the upper limit value to the maximum detection value (e.g., 200A) and the lower limit value to the non-zero minimum detection value (e.g., 0.05A)". Then, the dq-axis current ratio operation unit 284 calculates the current ratio Rdq by dividing the absolute value of the d-axis current command value id by the absolute value of the "limited" q-axis current command value iq.

The input limiting unit 285 limits the current ratio Rdq of "the object of arctan operation" in such a manner that "the operation range is set to the limit value" so that the current ratio Rdq does not exceed the operation range.

Variation angle calculation section 286 calculates a variation angle d θ by performing arctan calculation with respect to current ratio Rdq "limited by input limiting section 285". In consideration of the processing load of the CPU mounted on the ECU, the arctan operation is performed in an off-line manner, and a table "for converting the unit of the arithmetic result of the arctan operation from rad (radians) to deg (degrees)" is created in advance and stored in the variation angle arithmetic unit 286. Variation angle calculation section 286 finds variation angle d θ from current ratio Rdq on the basis of a table created in advance by an on-line method. When the processing load of the CPU does not need to be taken into consideration, the variation angle calculation unit 286 may calculate the variation angle d θ by performing the arctan calculation and unit conversion of the calculation result of the arctan calculation in an on-line manner. The variation angle d θ is input to the multiplication unit 289A.

The sign determination unit 287 inputs the d-axis current command value id and the q-axis current command value iq, and outputs "-" when the sign of the d-axis current command value id is different from the sign of the q-axis current command value iq; when the sign of the d-axis current command value id is the same as the sign of the q-axis current command value iq, "+" is output. The output value from the sign determination unit 287 is input to the multiplication unit 289A so as to be multiplied by the variation angle d θ in the multiplication unit 289A. The multiplication result obtained by the multiplication section 289A is input to the addition section 289B as the variation angle d θ'.

In the addition unit 289B, the phase correction is performed on the motor rotation angle θ m by adding the variation angle d θ' to the motor rotation angle θ m, thereby calculating a phase correction rotation angle θ d 0. When the phase correction rotation angle θ d0 is equal to or greater than 360deg, the phase correction rotation angle θ d0 is input to the inversion processing unit 288 in order to perform the inversion processing of "converting to an angle of 0deg to 360 deg". The inversion processing unit 288 performs inversion processing "in units of 360 deg" on the phase correction rotation angle θ d0, that is, processing for obtaining "the remainder of the division of the value of the phase correction rotation angle θ d0 by 360" on the phase correction rotation angle θ d0, and outputs the processing result as the phase correction rotation angle θ d.

The axial current factor phase correction arithmetic unit 280 performs phase correction on the motor rotation angle θ m using the d-axis current command value id and the q-axis current command value iq by the configuration and the processing of each unit, and calculates a phase correction rotation angle θ d.

Fig. 35 to 42 show the results of "simulation performed with phase correction" and "simulation performed without phase correction" under the medium-speed steering condition (inverter applied voltage 12[ V ], q-axis current command value 45[ a ], d-axis current command value 15[ a ], and motor rotation speed 1200[ rpm ]) "in order to confirm the effect of phase correction in the present embodiment. Fig. 35 to 38 show the results of the "case without phase correction", and fig. 39 to 42 show the results of the "case with phase correction". In fig. 35 and 39, the motor rotation angle (fig. 35) or the phase correction rotation angle (fig. 39), the U-phase current, the d-axis current, and the q-axis current are shown; in fig. 36 and 40, the U-phase current and the U-phase dead time compensation value are shown; fig. 37 and 41 show the U-phase current and the d-axis dead time compensation value; fig. 38 and 42 show the U-phase current and the q-axis dead time compensation value.

As shown in fig. 35, when a current flows through the d-axis, distortion occurs in all of the U-phase current, the d-axis current, and the q-axis current. Although the zero-crossing point (the point crossing the horizontal axis and having a value of zero) of the U-phase current coincides with the point at which the motor rotation angle becomes zero when the d-axis current is zero, it can be seen that the phase of the U-phase current is advanced with respect to the motor rotation angle when the current flows through the d-axis. Although the U-phase dead time compensation value shown in fig. 36 is a signal during calculation, it can be seen that the U-phase dead time compensation value is deviated in timing with respect to the U-phase current "phase lead due to current flowing through the d-axis". As shown in fig. 37 and 38, the d-axis dead time compensation value and the q-axis dead time compensation value are similarly deviated.

In contrast, in the dead time compensation of the feedforward method "corresponding to the motor rotation angle", by performing the phase correction "corresponding to the d-axis current", even when the current flows through the d-axis, as shown in fig. 40, it can be confirmed that the timing of the dead time compensation is appropriate. As shown in fig. 41 and 42, the dead time compensation time is also preferably set for the d-axis dead time compensation value and the q-axis dead time compensation value. As described above, by performing the phase correction according to the present embodiment, the timing of the dead time compensation of the feedforward method "corresponding to the motor rotation angle" can be improved, and as shown in fig. 39, it is confirmed that the distortion of the current waveform is improved, the ripple in the waveforms of the d-axis current and the q-axis current is reduced, and the waveform of the U-phase current becomes a waveform close to a sine wave.

Fig. 43 corresponding to fig. 28 shows a configuration example (embodiment 5) of "phase correction function is added to embodiment 3". As shown in fig. 43, as in embodiment 4, the dead time compensation means 200D is also provided with an axis current factor phase correction arithmetic means 280 for "performing a phase correction function". Also, fig. 44 shows a detailed structure of the dead time compensation unit 200D. As shown in fig. 44, the axial current factor phase correction arithmetic unit 280 performs the phase correction function by the "same configuration and operation as in the case of embodiment 4". Further, the dead time compensation value output unit is configured by the multiplication unit 231U, the multiplication unit 231V, and the multiplication unit 231W.

In embodiment 5 as well, compensation is performed by directly adding the dead time compensation value shown in fig. 40 to the 3-phase voltage command value in the feed-forward manner, and therefore the same effect as embodiment 4 can be obtained.

In the above-described embodiments, the motor control device "mounted in the electric power steering device" has been described, but it is needless to say that the motor control device of the present invention may be mounted in an electric vehicle, a machine tool, or the like.

Description of the reference numerals

1 steering wheel (steering wheel)

2 column shaft (steering shaft or steering wheel shaft)

10 torque sensor

20. 100 motor

30 control unit (ECU)

31 steering assist command value arithmetic unit

35. 120d, 120q PI control unit

36. 160 PWM control unit

37. 161 inverter

130. 240, 2613 AC/dq axis conversion unit

200. 200A, 200B, 200C, 200D dead time compensation unit

201 current control delay model

202 compensating symbol estimation unit

210 phase adjustment unit

220 inverter applied voltage induction gain unit

230U, 230V, 230W angle-dead time compensation value function unit

250 current command value induction gain unit

270 compensation value adjusting unit

280-axis current factor phase correction arithmetic unit

284 dq-axis current ratio operation unit

286 variable angle arithmetic unit

300 space vector modulation unit

3012 phase/3 phase conversion unit

302 third harmonic wave superposition unit

50页详细技术资料下载
上一篇:一种医用注射器针头装配设备
下一篇:用于功率放大器的限流电路

网友询问留言

已有0条留言

还没有人留言评论。精彩留言会获得点赞!

精彩留言,会给你点赞!